Merge branch 'fix/prod-readiness-healthcheck-domain'

Two issues would each stop a clean production `docker compose up` for the event:

1. Healthchecks probed http://localhost:{3000,3001}, but the app and frontend
   bind IPv4 (0.0.0.0) while `localhost` resolves to ::1 (IPv6) first inside the
   container — so the probe got "connection refused" and neither container ever
   turned healthy. Caddy is gated on `condition: service_healthy` for both, so on
   a fresh boot it would block forever and nothing gets served. Switch both probes
   to 127.0.0.1. (Verified: both containers now report healthy.)

2. The prod caddy service never received DOMAIN, so the Caddyfile's `{$DOMAIN}`
   site address expanded to empty — malformed site block, no TLS, no serving. Add
   `environment: { DOMAIN: ${DOMAIN} }` to the caddy service.

Also make .env.example honest and event-ready:
- Add DATABASE_MAX_CONNECTIONS (real env lever; recommend 30 for ~100 guests).
- The DEFAULT_* upload/rate/capacity vars are NOT read from env — they are seeded
  into the DB config table and managed at runtime via the admin dashboard. Replace
  the misleading entries (e.g. upload rate showed 10; live value is 100 via
  migration 015) with a note pointing to the admin UI and the real seeded defaults.
- Document that raising COMPRESSION_WORKER_CONCURRENCY also needs the app memory
  limit raised (ffmpeg), so a video burst can't OOM the box.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

.claude/settings.json

@@ -0,0 +1,9 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(cargo check *)",
+      "Bash(cargo clippy *)",
+      "Bash(git --no-pager diff *)"
+    ]
+  }
+}

.env.example

@@ -4,12 +4,21 @@ DOMAIN=my-event.example.com
 
 # ── App server ────────────────────────────────────────────────────────────────
 APP_PORT=3000
+# Set to `production` in real deployments. This activates the secret guard that
+# refuses to boot with placeholder JWT_SECRET / ADMIN_PASSWORD_HASH values.
+# (docker-compose.yml already sets APP_ENV=production for the app service.)
+APP_ENV=production
 
 # ── Database ──────────────────────────────────────────────────────────────────
-DATABASE_URL=postgres://eventsnap:secret@db:5432/eventsnap
+# Set a strong password and keep it in sync between DATABASE_URL and
+# POSTGRES_PASSWORD. Generate one with: openssl rand -hex 24
+DATABASE_URL=postgres://eventsnap:CHANGE_ME_use_a_strong_password@db:5432/eventsnap
 POSTGRES_USER=eventsnap
-POSTGRES_PASSWORD=secret
+POSTGRES_PASSWORD=CHANGE_ME_use_a_strong_password
 POSTGRES_DB=eventsnap
+# Connection pool size. Default 10. For a busy event (~100 guests polling the feed
+# + SSE + uploads at once) raise to ~30 so requests don't queue on a pool permit.
+DATABASE_MAX_CONNECTIONS=30
 
 # ── Authentication ────────────────────────────────────────────────────────────
 # Generate with: openssl rand -hex 64
@@ -18,7 +27,11 @@ SESSION_EXPIRY_DAYS=30
 
 # Admin dashboard password (bcrypt hash).
 # Generate with: htpasswd -bnBC 12 "" yourpassword | tr -d ':\n'
-ADMIN_PASSWORD_HASH=$2y$12$placeholder_replace_me
+# IMPORTANT: keep the SINGLE QUOTES. A bcrypt hash is full of `$` (e.g. $2b$12$…$…),
+# and both Docker Compose's env_file interpolation and dotenvy's variable substitution
+# would otherwise eat the `$…` segments (reading them as unset vars) and corrupt the
+# hash — every admin login then 401s. Single quotes make both read it literally.
+ADMIN_PASSWORD_HASH='$2y$12$placeholder_replace_me'
 
 # ── Event ─────────────────────────────────────────────────────────────────────
 EVENT_NAME=Max & Maria's Wedding
@@ -26,20 +39,29 @@ EVENT_SLUG=max-maria-2026
 
 # ── Storage ───────────────────────────────────────────────────────────────────
 MEDIA_PATH=/media
+# Export archives (Gallery.zip / Memories.zip). MUST be outside MEDIA_PATH —
+# /media is publicly served, so exports here would be downloadable without auth.
+EXPORT_PATH=/exports
 
-# ── Upload limits ─────────────────────────────────────────────────────────────
-DEFAULT_MAX_IMAGE_SIZE_MB=20
-DEFAULT_MAX_VIDEO_SIZE_MB=500
-
-# ── Rate limiting ─────────────────────────────────────────────────────────────
-DEFAULT_UPLOAD_RATE_PER_HOUR=10
-DEFAULT_FEED_RATE_PER_MIN=60
-DEFAULT_EXPORT_RATE_PER_DAY=3
-
-# ── Capacity ──────────────────────────────────────────────────────────────────
-DEFAULT_ESTIMATED_GUEST_COUNT=100
-# Fraction of total storage that triggers the "low storage" warning (0.0–1.0)
-DEFAULT_QUOTA_TOLERANCE=0.75
+# ── Runtime settings (upload limits, rate limits, capacity) ───────────────────
+# NOTE: These are NOT environment variables. Upload size caps, rate limits, guest
+# count and quota tolerance are stored in the database `config` table (seeded once
+# at first boot) and changed at runtime from the ADMIN DASHBOARD — the backend does
+# not read them from .env. Setting them here has no effect. Current seeded defaults:
+#   upload rate      100 / hour / guest   (raised from 10 by migration 015)
+#   feed rate         60 / minute
+#   export rate        3 / day
+#   max image size    20 MB
+#   max video size   500 MB
+#   estimated guests 100
+#   quota tolerance  0.75  (fraction of disk that triggers the low-storage warning)
+# Adjust these in the admin UI before the event if needed.
 
 # ── Workers ───────────────────────────────────────────────────────────────────
+# Number of parallel image/video compression workers. Default 2. This is the main
+# throughput bottleneck: with 2 workers a burst of uploads can take ~5s to appear.
+# For a large event (100+ guests) 4 is a good target — but each worker can run an
+# ffmpeg transcode, so if you raise this ALSO raise the app container's memory limit
+# in docker-compose.yml (`app.deploy.resources.limits.memory`) from 1G to ~2G, or a
+# burst of large videos can OOM the box and take Postgres down with it.
 COMPRESSION_WORKER_CONCURRENCY=2

.env.test

@@ -1,45 +0,0 @@
-# ── Domain ────────────────────────────────────────────────────────────────────
-# Public domain Caddy will serve and obtain a TLS certificate for.
-DOMAIN=my-event.example.com
-
-# ── App server ────────────────────────────────────────────────────────────────
-APP_PORT=3000
-
-# ── Database ──────────────────────────────────────────────────────────────────
-DATABASE_URL=postgres://eventsnap:secret@db:5432/eventsnap
-POSTGRES_USER=eventsnap
-POSTGRES_PASSWORD=secret
-POSTGRES_DB=eventsnap
-
-# ── Authentication ────────────────────────────────────────────────────────────
-# Generate with: openssl rand -hex 64
-JWT_SECRET=change_me_to_a_random_64_byte_hex_string
-SESSION_EXPIRY_DAYS=30
-
-# Admin dashboard password (bcrypt hash).
-# Generate with: htpasswd -bnBC 12 "" yourpassword | tr -d ':\n'
-ADMIN_PASSWORD_HASH=$2y$12$placeholder_replace_me
-
-# ── Event ─────────────────────────────────────────────────────────────────────
-EVENT_NAME=Max & Maria's Wedding
-EVENT_SLUG=max-maria-2026
-
-# ── Storage ───────────────────────────────────────────────────────────────────
-MEDIA_PATH=/media
-
-# ── Upload limits ─────────────────────────────────────────────────────────────
-DEFAULT_MAX_IMAGE_SIZE_MB=20
-DEFAULT_MAX_VIDEO_SIZE_MB=500
-
-# ── Rate limiting ─────────────────────────────────────────────────────────────
-DEFAULT_UPLOAD_RATE_PER_HOUR=10
-DEFAULT_FEED_RATE_PER_MIN=60
-DEFAULT_EXPORT_RATE_PER_DAY=3
-
-# ── Capacity ──────────────────────────────────────────────────────────────────
-DEFAULT_ESTIMATED_GUEST_COUNT=100
-# Fraction of total storage that triggers the "low storage" warning (0.0–1.0)
-DEFAULT_QUOTA_TOLERANCE=0.75
-
-# ── Workers ───────────────────────────────────────────────────────────────────
-COMPRESSION_WORKER_CONCURRENCY=2

.github/workflows/audit.yml

@@ -0,0 +1,73 @@
+# Dependency advisory scan.
+#
+# Lives in .github/workflows/ because Gitea Actions scans BOTH .gitea/workflows and
+# .github/workflows, and e2e.yml already proves this instance resolves `actions/*` from GitHub
+# (DEFAULT_ACTIONS_URL). Keeping one directory avoids a split where half the CI is invisible
+# depending on which convention you look under.
+#
+# Unlike the GitHub-hosted runners this workflow does NOT assume a preinstalled Rust toolchain —
+# the common Gitea runner images (gitea/runner-images, catthehacker/ubuntu) ship Node and Docker
+# but not cargo. So we install it explicitly instead of relying on the ambient environment.
+name: Audit
+
+on:
+  pull_request:
+  push:
+    branches: [main]
+  schedule:
+    # New advisories land against unchanged code, so a push-only trigger would never see them.
+    - cron: '0 6 * * 1'
+
+jobs:
+  cargo-audit:
+    name: cargo audit (backend)
+    runs-on: ubuntu-latest
+    timeout-minutes: 15
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install Rust toolchain
+        run: |
+          if ! command -v cargo > /dev/null; then
+            curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal
+            echo "$HOME/.cargo/bin" >> "$GITHUB_PATH"
+          fi
+
+      - name: Cache cargo
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cargo/registry
+            ~/.cargo/bin
+          key: cargo-audit-${{ hashFiles('backend/Cargo.lock') }}
+          restore-keys: cargo-audit-
+
+      - name: Install cargo-audit
+        run: cargo install cargo-audit --locked || true
+
+      # `rsa` 0.9 (RUSTSEC-2023-0071, "Marvin") is a transitive dep of the SQLx MySQL driver that
+      # this app never exercises: auth is HS256 + bcrypt, and the only database is Postgres. There
+      # is no patched release, so failing the build on it would mean a permanently red pipeline
+      # that everyone learns to ignore — which is worse than not scanning at all. Ignore it
+      # SPECIFICALLY, so that any OTHER advisory still fails the job.
+      - name: Audit
+        working-directory: ./backend
+        run: cargo audit --deny warnings --ignore RUSTSEC-2023-0071
+
+  npm-audit:
+    name: npm audit (frontend)
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: '22'
+      - name: Install deps
+        working-directory: ./frontend
+        run: npm install
+      # Production dependencies only: a dev-only advisory (build tooling, test runners) can't be
+      # reached by a guest at the party, and gating merges on it just trains people to skip the gate.
+      - name: Audit
+        working-directory: ./frontend
+        run: npm audit --omit=dev --audit-level=high

.github/workflows/checks.yml

@@ -0,0 +1,129 @@
+# The checks that were only ever running on a laptop.
+#
+# Before this file, CI ran Playwright (chromium-desktop) and the dependency audit — and nothing
+# else. `cargo test` (40 tests), the frontend vitest suite (5 files, including the offline
+# upload-queue and auth-token logic), svelte-check, and the e2e typecheck were all green on a
+# developer's machine and gated NOTHING. A check that only ever runs locally is not running.
+#
+# In .github/workflows/ because Gitea Actions scans it too (see audit.yml). Rust is installed
+# explicitly: the common Gitea runner images ship Node and Docker, not cargo.
+name: Checks
+
+on:
+  pull_request:
+  push:
+    branches: [main]
+
+jobs:
+  backend:
+    name: Backend — cargo test + clippy + fmt
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install Rust toolchain
+        run: |
+          if ! command -v cargo > /dev/null; then
+            curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal
+            echo "$HOME/.cargo/bin" >> "$GITHUB_PATH"
+          fi
+          rustup component add clippy rustfmt
+
+      - uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cargo/registry
+            ~/.cargo/git
+            backend/target
+          key: cargo-${{ hashFiles('backend/Cargo.lock') }}
+          restore-keys: cargo-
+
+      # SQLx runs its queries against a live database at TEST time (see backend/tests/), so the
+      # DB-backed tests need one. The pure unit tests don't care, but starting it unconditionally
+      # keeps the job simple and honest about what it covers.
+      - name: Start Postgres
+        run: |
+          docker run -d --name ci-pg -p 5432:5432 \
+            -e POSTGRES_PASSWORD=postgres -e POSTGRES_DB=eventsnap_ci \
+            postgres:16-alpine
+          for _ in $(seq 1 30); do
+            docker exec ci-pg pg_isready -U postgres > /dev/null 2>&1 && break
+            sleep 1
+          done
+
+      - name: Test
+        working-directory: ./backend
+        env:
+          DATABASE_URL: postgres://postgres:postgres@localhost:5432/eventsnap_ci
+        # `#[sqlx::test]` creates a fresh database per test and opens its own pool; run at unbounded
+        # parallelism against a default `max_connections=100` Postgres, a full suite can exhaust the
+        # server's connection slots and fail with PoolTimedOut — a pure infra flake, not a real
+        # failure. Cap the concurrency so the gate stays trustworthy on a loaded runner.
+        run: cargo test --all-features -- --test-threads=8
+
+      - name: Clippy
+        working-directory: ./backend
+        run: cargo clippy --all-targets -- -D warnings
+
+      - name: Format
+        working-directory: ./backend
+        run: cargo fmt --check
+
+  frontend:
+    name: Frontend — vitest + svelte-check
+    runs-on: ubuntu-latest
+    timeout-minutes: 15
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: '22'
+          cache: 'npm'
+          cache-dependency-path: 'frontend/package-lock.json'
+
+      - name: Install deps
+        working-directory: ./frontend
+        run: npm ci || npm install
+
+      - name: Unit tests
+        working-directory: ./frontend
+        run: npm run test:unit
+
+      - name: svelte-check
+        working-directory: ./frontend
+        run: npx svelte-check --threshold error
+
+      - name: ESLint
+        working-directory: ./frontend
+        run: npm run lint
+
+      - name: Prettier
+        working-directory: ./frontend
+        run: npm run format:check
+
+  e2e-typecheck:
+    name: E2E — typecheck + lint
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: '22'
+      - name: Install deps
+        working-directory: ./e2e
+        run: npm install
+      # Playwright TRANSPILES specs without typechecking them, so a type error in a spec is
+      # invisible until the assertion it guards silently does the wrong thing at runtime.
+      - name: tsc --noEmit
+        working-directory: ./e2e
+        run: npx tsc --noEmit
+
+      - name: ESLint
+        working-directory: ./e2e
+        run: npm run lint
+
+      - name: Prettier
+        working-directory: ./e2e
+        run: npm run format:check

.github/workflows/e2e.yml

@@ -0,0 +1,118 @@
+name: E2E
+
+on:
+  pull_request:
+  push:
+    branches: [main]
+
+jobs:
+  e2e:
+    name: Playwright E2E (chromium-desktop)
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-node@v4
+        with:
+          node-version: '22'
+          cache: 'npm'
+          cache-dependency-path: 'e2e/package.json'
+
+      - name: Install e2e deps
+        working-directory: ./e2e
+        run: npm install
+
+      - name: Install Playwright browsers
+        working-directory: ./e2e
+        run: npx playwright install --with-deps chromium
+
+      - name: Bring up the test stack
+        working-directory: ./e2e
+        run: docker compose -f docker-compose.test.yml up -d --build
+
+      - name: Wait for stack health
+        working-directory: ./e2e
+        run: |
+          for i in $(seq 1 60); do
+            if curl -fsS http://localhost:3101/health > /dev/null; then exit 0; fi
+            sleep 2
+          done
+          echo "Stack never became healthy. Dumping logs:"
+          docker compose -f docker-compose.test.yml logs
+          exit 1
+
+      - name: Run E2E tests
+        working-directory: ./e2e
+        run: npm run test:e2e -- --project=chromium-desktop
+
+      # 09-mobile is `testIgnore`d on chromium-desktop (it needs hasTouch + a phone viewport), so
+      # running only that project left 22 tests — focus traps, touch targets, safe-area insets,
+      # viewport reflow, upload-cancel — never executing in CI. On a phone-first event app, where
+      # essentially every real guest is on a phone, that was the wrong half of the suite to skip.
+      - name: Run E2E tests (mobile)
+        working-directory: ./e2e
+        run: npm run test:e2e -- --project=chromium-mobile
+
+      - name: Upload Playwright report
+        if: failure()
+        uses: actions/upload-artifact@v4
+        with:
+          name: playwright-report
+          path: e2e/playwright-report/
+          retention-days: 14
+
+      - name: Upload traces
+        if: failure()
+        uses: actions/upload-artifact@v4
+        with:
+          name: traces
+          path: e2e/test-results/
+          retention-days: 14
+
+      - name: Tear down stack
+        if: always()
+        working-directory: ./e2e
+        run: docker compose -f docker-compose.test.yml down -v
+
+  smoke-ua-matrix:
+    name: Cross-UA smoke matrix
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: '22'
+      - name: Install e2e deps
+        working-directory: ./e2e
+        run: npm install
+      - name: Install Playwright browsers (full)
+        working-directory: ./e2e
+        run: npx playwright install --with-deps chromium firefox webkit
+      - name: Bring up the test stack
+        working-directory: ./e2e
+        run: docker compose -f docker-compose.test.yml up -d --build
+      - name: Wait for stack health
+        working-directory: ./e2e
+        run: |
+          for i in $(seq 1 60); do
+            if curl -fsS http://localhost:3101/health > /dev/null; then exit 0; fi
+            sleep 2
+          done
+          docker compose -f docker-compose.test.yml logs
+          exit 1
+      - name: Smoke matrix
+        working-directory: ./e2e
+        run: npm run test:e2e:smoke
+      - name: Upload smoke report
+        if: failure()
+        uses: actions/upload-artifact@v4
+        with:
+          name: playwright-smoke-report
+          path: e2e/playwright-report/
+          retention-days: 14
+      - name: Tear down
+        if: always()
+        working-directory: ./e2e
+        run: docker compose -f docker-compose.test.yml down -v

.gitignore

@@ -1,5 +1,7 @@
 # Environment secrets — never commit the real .env
 .env
+# Stale local scratch copy of .env.example; nothing in the test stack reads it.
+.env.test
 
 # Rust
 backend/target/
@@ -8,10 +10,25 @@ backend/target/
 frontend/node_modules/
 frontend/.svelte-kit/
 frontend/build/
+frontend/export-viewer/node_modules/
+frontend/export-viewer/.svelte-kit/
 
 # Media uploads (mounted volume in production)
 media/
 
+# Playwright E2E suite — runtime artifacts (the suite itself is committed)
+e2e/node_modules/
+e2e/playwright-report/
+e2e/test-results/
+e2e/.cache/
+e2e/.env.test
+# Playwright artifacts when run from the repo root instead of e2e/
+/test-results/
+/playwright-report/
+
 # OS
 .DS_Store
 Thumbs.db
+
+# Claude Code personal (per-user) settings — shared settings.json IS committed
+.claude/settings.local.json

Caddyfile

@@ -1,26 +1,42 @@
 {$DOMAIN} {
-    encode zstd gzip
+	# Compress everything EXCEPT the SSE stream — gzip buffering delays
+	# "real-time" likes/comments until the ~30s keep-alive tick.
+	@compressible not path /api/v1/stream
+	encode @compressible zstd gzip
 
-    # SvelteKit frontend — static assets with long-lived cache (content-hashed filenames)
-    @hashed_assets path_regexp hashed /_app/immutable/.*\.[a-f0-9]{8,}\.(js|css|woff2)$
-    header @hashed_assets Cache-Control "public, max-age=31536000, immutable"
+	# Site-wide security headers (defense-in-depth). HSTS is free since Caddy
+	# already terminates TLS. nosniff also covers all of /media/*.
+	header {
+		Strict-Transport-Security "max-age=31536000; includeSubDomains"
+		X-Content-Type-Options "nosniff"
+		X-Frame-Options "DENY"
+		Referrer-Policy "strict-origin-when-cross-origin"
+	}
 
-    # Media previews and thumbnails
-    @previews path /media/previews/* /media/thumbnails/*
-    header @previews Cache-Control "public, max-age=3600"
+	# SvelteKit frontend — static assets with long-lived cache (content-hashed filenames)
+	@hashed_assets path_regexp hashed /_app/immutable/.*\.[a-f0-9]{8,}\.(js|css|woff2)$
+	header @hashed_assets Cache-Control "public, max-age=31536000, immutable"
 
-    # Original media files (private — only host can download)
-    @originals path /media/originals/*
-    header @originals Cache-Control "private, max-age=86400"
+	# Preview/thumbnail images. These are now served by the app through a
+	# visibility-checked alias (/api/v1/upload/{id}/{preview,thumbnail}) so moderation
+	# can revoke access; direct /media/previews|thumbnails is 404-blocked at the app.
+	# Privately cacheable for a short window (the app sets the same header; this is the
+	# edge carve-out from the blanket no-store below). Kept short so a moderated image
+	# stops being served to a direct-URL holder promptly.
+	@media_api path /api/v1/upload/*/preview /api/v1/upload/*/thumbnail
+	header @media_api Cache-Control "private, max-age=300"
 
-    # API — never cache
-    @api path /api/*
-    header @api Cache-Control "no-store"
+	# API — never cache, EXCEPT the gated image routes above.
+	@api {
+		path /api/*
+		not path /api/v1/upload/*/preview /api/v1/upload/*/thumbnail
+	}
+	header @api Cache-Control "no-store"
 
-    # Route API and media requests to the Rust backend
-    reverse_proxy /api/* app:3000
-    reverse_proxy /media/* app:3000
+	# Route API and media requests to the Rust backend
+	reverse_proxy /api/* app:3000
+	reverse_proxy /media/* app:3000
 
-    # Everything else goes to SvelteKit frontend
-    reverse_proxy frontend:3001
+	# Everything else goes to SvelteKit frontend
+	reverse_proxy frontend:3001
 }

FOLLOWUPS.md

@@ -0,0 +1,137 @@
+# Follow-ups
+
+Tracked work that was deferred during the multi-round UI/UX review pass.
+Each item has a clear acceptance criterion so a future pass can land it
+without re-deriving the context.
+
+## A11y — assistive-tech containment inside modals
+
+**Problem.** Open modals (LightboxModal, ConfirmSheet, Modal, OnboardingGuide,
+join PIN modal, account data-mode sheet, export HTML guide) trap keyboard Tab
+via `focusTrap`, but VoiceOver rotor / TalkBack arrow-key navigation can still
+escape into the page content behind the dialog. Screen-reader users hear the
+wrong context.
+
+**Why deferred.** A naive sibling-walk that sets `inert` on direct children of
+the modal's parent silences the global `<Toaster>` (`aria-live="polite"` region
+mounted in `+layout.svelte`) and the `<BottomNav>` while a modal is open —
+breaking toast announcements and the visible nav state. SvelteKit has no
+built-in portal mechanism, so dialogs render inside the route tree alongside
+the Toaster.
+
+**Acceptance criterion.** With any modal open:
+- VoiceOver rotor (iOS Safari) and TalkBack swipe navigation (Android Chrome)
+  cannot leave the dialog subtree.
+- Toasts that fire while a modal is open are still announced.
+- Nested modals (e.g. ConfirmSheet opened from inside ContextSheet) maintain
+  correct containment when the inner closes.
+
+**Sketch of an approach.** One of:
+1. **Portal pattern.** Render dialogs into a dedicated `<div id="modal-root">`
+   that's a sibling of the main app root in `app.html`. `focusTrap` then sets
+   `inert` on the main root, leaving the modal root and the toast region (also
+   moved to its own portal root) untouched.
+2. **Opt-out marker.** Walk siblings and inert them, but skip any node carrying
+   a `data-modal-passthrough` attribute. Mark `<Toaster>` with it. Document
+   the contract.
+3. **Stack-aware containment.** Maintain a module-level stack of open dialog
+   nodes; the topmost owns the inert state, popped dialogs restore the
+   previous layer. Avoids the nested-modal restoration bug.
+
+Approach 1 is the cleanest long-term but the highest blast radius. Approach 2
+is the smallest patch.
+
+**Files to touch.**
+- [frontend/src/lib/actions/focus-trap.ts](frontend/src/lib/actions/focus-trap.ts) — add inert logic
+- [frontend/src/lib/components/Toaster.svelte](frontend/src/lib/components/Toaster.svelte) — add passthrough marker (if approach 2) or move to a portal (if approach 1)
+- [frontend/src/app.html](frontend/src/app.html) — add `<div id="modal-root">` (if approach 1)
+
+## Feed — DOM-windowing virtualization (IMPLEMENTED — residual validation owed)
+
+**Status.** Implemented in [frontend/src/lib/components/VirtualFeed.svelte](frontend/src/lib/components/VirtualFeed.svelte)
+using `@tanstack/svelte-virtual`'s `createWindowVirtualizer`. Both the **list**
+view (dynamic `measureElement` heights, keyed by upload id with `anchorTo:'start'`
+so an SSE prepend doesn't yank a scrolled-down reader) and the **grid** view
+(three measured square tiles per row) now keep only the on-screen window (+overscan)
+in the DOM instead of one node per upload. The window virtualizer scrolls the
+document, so the sticky header, pull-to-refresh, the infinite-scroll sentinel and
+the bottom nav are untouched. The earlier `content-visibility` band-aid was removed
+from `FeedListCard` (it interferes with real-height measurement), and the old
+`FeedGrid.svelte` was deleted (its sole consumer migrated to `VirtualFeed`).
+
+**Verified.** `svelte-check` 0 errors, production build clean. The integration
+follows the library's documented window-virtualizer contract (confirmed against
+`virtual-core` source: item `start` includes `scrollMargin`, `getTotalSize()`
+excludes it; the SSR path is guarded by `getScrollElement()` returning null).
+
+**Residual validation owed (needs the running app — could not be done headless).**
+- Manual scroll testing on a ~1000-item event: confirm no jank, correct scrollbar
+  size, and that an SSE `new-upload` / `feed-delta` prepend while scrolled down does
+  not jump the viewport (the `anchorTo:'start'` + id-key path).
+- `scrollMargin` re-measure when grid filter chips change the header height (handled
+  reactively via the `uploads`-length-driven effect, but unverified visually).
+- A new e2e spec that scrolls far down, likes an item via SSE, and asserts scroll
+  position is retained — the existing suite only asserts a single card is visible,
+  so it cannot catch a scroll regression.
+
+**Files.**
+- [frontend/src/lib/components/VirtualFeed.svelte](frontend/src/lib/components/VirtualFeed.svelte) — new windowing component (list + grid)
+- [frontend/src/routes/feed/+page.svelte](frontend/src/routes/feed/+page.svelte) — renders `VirtualFeed` for both views
+- [frontend/src/lib/components/FeedListCard.svelte](frontend/src/lib/components/FeedListCard.svelte) — `content-visibility` removed
+
+**Known limitations (surfaced in the post-commit review, left as-is — low impact).**
+- **Grid prepend reflows tiles.** Grid rows are index-keyed and pack 3 tiles each, so
+  an SSE `new-upload` shifts every tile by one position; `anchorTo:'start'` can only
+  anchor a scrolled grid reader when the prepend crosses a 3-item boundary (it adds a
+  *row*). List view is unaffected (one id-keyed row per upload). A fix would key rows
+  by the first tile's id and accept partial-row churn; not worth it for the rarer
+  "new upload while browsing the grid" case.
+- **Filtered grid can auto-load the whole feed.** When a grid filter matches few items
+  the `VirtualFeed` is short, so the infinite-scroll sentinel sits in-viewport and
+  `loadMore()` fires until `nextCursor` is null — pulling all pages to widen the
+  client-side search. This is pre-existing (the old `FeedGrid` had the same shape, and
+  the empty-filter copy even says "scrolle weiter"), not a virtualization regression.
+  If undesired, gate auto-load to list view or to actual user scroll.
+
+## Feed — comment deletion leaves a stale live count
+
+**Problem.** `add_comment` now broadcasts a fresh `comment_count` so feed clients patch
+the card in place, but `delete_comment` and `host_delete_comment`
+([backend/src/handlers/social.rs](backend/src/handlers/social.rs),
+[host.rs](backend/src/handlers/host.rs)) soft-delete without broadcasting any
+count/event. So a deletion leaves the count too high on every client until a full
+refetch (pull-to-refresh or an unrelated `upload-processed` merge). `toggle_like`
+already broadcasts on both add and remove, so likes are fine — the gap is
+comments-on-delete. Pre-existing, but the in-place-patch scheme makes it observable.
+
+**Fix.** Emit a `new-comment` (or a `comment-deleted`) event carrying the refreshed
+`comment_count` from both delete paths, the same best-effort way `add_comment` does;
+the frontend `patchCount(..., 'comment_count')` handler already consumes it.
+
+**Note — count ordering.** The broadcast count is read just after the (auto-committed)
+mutation, not inside it, so under concurrent likes/comments on one upload the SSE
+messages are last-write-wins. The frontend *replaces* (not increments) the count, so
+steady state is correct and self-healing; only document this if strict per-event
+ordering is ever required (then compute the count in-tx with a monotonic sequence).
+
+## Feed — per-image exact CLS reservation
+
+**Problem.** `FeedListCard` now reserves a default `aspect-[4/5]` box for photos so
+the card doesn't collapse to height 0 and reflow as images stream in (matching
+`Skeleton`). But no image dimensions are stored anywhere (not on `FeedUpload`, the
+`upload` table, or any migration), so the box is a uniform guess that crops to fit —
+the original is one tap away in the lightbox.
+
+**Acceptance criterion.** Extract image width/height during the compression worker,
+store them on `upload`, expose them on `FeedUpload`, and have the card reserve the
+*true* aspect ratio (no crop, zero shift).
+
+**Files to touch.**
+- backend: `services/compression.rs`, `models/upload.rs`, `handlers/feed.rs`, a migration
+- [frontend/src/lib/types.ts](frontend/src/lib/types.ts), [FeedListCard.svelte](frontend/src/lib/components/FeedListCard.svelte)
+
+## Smaller nits, optional
+
+- **Auto-submit on retried 4th digit.** [recover/+page.svelte](frontend/src/routes/recover/+page.svelte), [join/+page.svelte](frontend/src/routes/join/+page.svelte) — after a wrong PIN, deleting one digit and retyping triggers an immediate submit. Backend's 3-attempts/15-min lockout makes this safe; could feel hair-trigger after a typo. Consider gating the second auto-submit per input session behind an explicit button press.
+- **Onboarding pip tap target on the vertical axis.** [OnboardingGuide.svelte](frontend/src/lib/components/OnboardingGuide.svelte) — current `p-2.5` yields ~26 px height, meets WCAG 2.2 AA (≥24 px) but below iOS HIG / Material's 44 / 48 dp recommendation. Bumping to `p-3` is the easy improvement; further increases start crowding the row.
+- **Migrate bespoke focus-trapped dialogs to `<Modal>`.** Join PIN modal, OnboardingGuide, LightboxModal, HTML guide, account data-mode sheet — all currently roll their own shell with `focusTrap`. They're correct, just not using the canonical primitive. Migrate when `<Modal>` gains features (e.g. the inert work above) you'd want everywhere.

PROJECT.md

@@ -42,7 +42,7 @@ A guest scans the QR code on their way in, types their name, and is immediately
 Mobile-first Progressive Web App (PWA) — accessible via browser, no app store required.
 
 ### Status
-Idea / Planning phase. Greenfield personal project.
+Implementation in progress (~v0.16). Core flows + new features all wired: auth, feed, upload, host/admin dashboards, ZIP + HTML-viewer export, SSE with delta-fetch on reconnect, toggleable rate limits + quotas with live per-user estimate, host PIN reset with one-time modal, data-mode (Saver/Original), Datenschutzhinweis, mobile gestures (long-press context sheet, double-tap to like), and the live Diashow with pluggable transitions. Open items: low-disk alert, event banner UI, chunked resumable upload for very large videos. See [FEATURES.md](docs/FEATURES.md) for the capability matrix.
 
 ---
 
@@ -208,9 +208,9 @@ Personal / private use. One event at a time. Up to ~100 users uploading ~1,000 f
 │  Axum HTTP Server (Rust — Single Binary)                        │
 │                                                                 │
 │  ┌─────────────┐  ┌──────────────┐  ┌────────────────────────┐ │
-│  │  REST API   │  │  SSE Engine  │  │  Static File Server    │ │
-│  │  /api/v1/*  │  │  /api/v1/   │  │  (SvelteKit build      │ │
-│  │             │  │  stream      │  │   output, embedded)    │ │
+│  │  REST API   │  │  SSE Engine  │  │  Media Static Server   │ │
+│  │  /api/v1/*  │  │  /api/v1/   │  │  /media/* (originals,  │ │
+│  │             │  │  stream      │  │  previews, thumbnails) │ │
 │  └──────┬──────┘  └──────┬───────┘  └────────────────────────┘ │
 │         │                │                                      │
 │  ┌──────▼──────────────────────┐  ┌──────────────────────────┐ │
@@ -245,36 +245,14 @@ Personal / private use. One event at a time. Up to ~100 users uploading ~1,000 f
 
 ### Docker Compose Stack
 
+Four services: Postgres, the Rust API (`app`), the SvelteKit Node server (`frontend`), and Caddy. Caddy routes `/api/*` and `/media/*` to the Rust binary and everything else to the SvelteKit server. See [docker-compose.yml](docker-compose.yml) for the authoritative definition.
+
 ```yaml
 services:
-  app:
-    build: ./backend          # Multi-stage Rust Dockerfile
-    env_file: .env
-    depends_on: [db]
-    volumes:
-      - media_data:/media
-    restart: unless-stopped
-
-  db:
-    image: postgres:16-alpine
-    env_file: .env
-    volumes:
-      - postgres_data:/var/lib/postgresql/data
-    restart: unless-stopped
-
-  caddy:
-    image: caddy:2-alpine
-    ports: ["80:80", "443:443"]
-    volumes:
-      - ./Caddyfile:/etc/caddy/Caddyfile:ro
-      - caddy_data:/data
-    depends_on: [app]
-    restart: unless-stopped
-
-volumes:
-  postgres_data:
-  media_data:
-  caddy_data:
+  db:        # postgres:16-alpine, persisted in postgres_data volume
+  app:       # ./backend — Rust API on :3000, mounts media_data:/media
+  frontend:  # ./frontend — SvelteKit (adapter-node) on :3001
+  caddy:     # caddy:2-alpine — terminates TLS on :80/:443, proxies app + frontend
 ```
 
 ### Caddyfile
@@ -345,11 +323,14 @@ COMPRESSION_WORKER_CONCURRENCY=2
 |-------|---------|---------|
 | `new-upload` | `{ id, preview_url, uploader, caption, created_at }` | Upload processing complete |
 | `new-comment` | `{ id, upload_id, body, uploader, created_at }` | Comment posted |
-| `new-like` | `{ upload_id, like_count }` | Like toggled |
+| `like-update` | `{ upload_id, like_count }` | Like toggled |
 | `upload-deleted` | `{ upload_id }` | Upload deleted |
 | `event-closed` | `{}` | Host locks uploads |
 | `event-opened` | `{}` | Host unlocks uploads |
 | `export-available` | `{ types: ["zip","html"] }` | Export generation complete |
+| `upload-processed` | `{ upload_id, preview_url, thumbnail_url }` | Server-side compression / preview generation finished |
+| `upload-error` | `{ upload_id, message }` | Compression / preview generation failed |
+| `export-progress` | `{ type, progress_pct }` | Periodic progress update from an export job |
 
 **Client SSE lifecycle:** `visibilitychange: hidden` → close connection · `visible` → reconnect + delta-fetch via `GET /api/v1/feed/delta?since=`
 
@@ -372,7 +353,7 @@ COMPRESSION_WORKER_CONCURRENCY=2
 | Real-Time | Axum SSE + `tokio::sync::broadcast` | Native, lightweight, perfect for fan-out at this scale |
 | ZIP Export | `async-zip` crate | Streaming ZIP generation without buffering the full archive in RAM |
 | HTML Export | `minijinja` (Rust templating) | Generates `Memories.html` as a single self-contained file |
-| Rate Limiting | `tower-governor` | Token-bucket per IP / per user; config from DB; hot-reloadable |
+| Rate Limiting | Custom in-memory sliding-window limiter ([services/rate_limiter.rs](backend/src/services/rate_limiter.rs)) | Per IP / per user; limits read from `config` DB table on each request; hot-reloadable without restart |
 | Reverse Proxy | Caddy 2 | Automatic HTTPS via Let's Encrypt; zero certificate management |
 | Containerisation | Docker + Docker Compose | Full stack in one file; `.env` for all config; single-command deploy |
 | Infrastructure | Hetzner CX33 (4 vCPU, 8 GB RAM, 80 GB SSD, 20 TB traffic) | Well-sized; 20 TB/month means post-event bulk downloads are no issue |
@@ -417,9 +398,9 @@ No paid third-party services required.
 
 | Role | Permissions |
 |------|------------|
-| Guest | Upload (within quota), caption/hashtag, like, comment, delete own content, view feed, download export (after release) |
-| Host | All guest permissions + ban/unban users (with upload visibility prompt), delete any content, promote guests to Host, lock/unlock uploads, release gallery export |
-| Admin | All Host permissions + configure storage/file/rate limits, quota tolerance, view disk usage, manage app config, trigger export generation |
+| Guest | Upload (within quota), caption/hashtag, like, comment, delete own content, view feed, download export (after release), pick data mode, read privacy note |
+| Host | All guest permissions + ban/unban users (with upload visibility prompt), delete any content, promote guests to Host, demote *other* Hosts to guest (never self), reset guest PINs (planned), lock/unlock uploads, release gallery export |
+| Admin | All Host permissions + reset any non-admin PIN, configure storage/file/rate limits with on/off toggles, edit quota tolerance and per-area quota toggles, edit the Datenschutzhinweis, view disk usage, manage app config, trigger export generation |
 | Banned Guest | View feed only — cannot upload, like, comment, or export |
 
 ### Compliance
@@ -473,37 +454,33 @@ Full-quality originals only. File naming: `{date}_{time}_{username}_{original_fi
 
 ### Export Type 2: HTML Offline Viewer (`Memories.zip`)
 
+The HTML export is a **pre-built SvelteKit static app** (`adapter-static`, `ssr=false`) shipped together with the event data. It is a non-interactive, read-only clone of the live feed — same components, same Tailwind tokens, same look — minus auth, upload, comment, and any dashboards. Full design rationale in [docs/CONCEPT_HTML_VIEWER.md](docs/CONCEPT_HTML_VIEWER.md).
+
 ```
 Memories/
-  Memories.html   ← single entry point (all CSS + JS inlined; no external deps)
-  README.txt      ← plain-text setup guide (in German, as the UI language)
-  Photos/  ...
-  Videos/  ...
+  index.html            ← entry point; open this in any browser
+  _app/
+    immutable/...       ← hashed JS/CSS bundles (viewer SPA)
+  data.json             ← event metadata, posts, comments, likes, hashtags
+  media/
+    {id}_thumb.jpg      ← grid thumbnails (≈400 px wide)
+    {id}_full.jpg/.mp4  ← full-size media for the lightbox
 ```
 
-**Fully self-contained / true offline:** `Memories.html` is a single file with all CSS and JS inlined as `<style>` and `<script>` tags — no external stylesheets, no CDN scripts, no network requests. All images and videos are referenced via **relative paths** to the sibling `Photos/` and `Videos/` folders — not base64-embedded (that would make the HTML file unworkably large). The ZIP must be unzipped first; relative paths resolve correctly from any location on disk.
+**How it works:** open `index.html` in any modern browser. The viewer hydrates client-side, `fetch('./data.json')` loads the event snapshot, all media references are relative paths into `media/`. No network calls, no service required. The ZIP must be unzipped first; the viewer does not run from inside an archive.
 
-**`Memories.html` features:** responsive photo/video grid, fullscreen lightbox, client-side hashtag filter chips, comments + like counts per upload, uploader name + timestamp, warm keepsake album aesthetic — all in self-contained vanilla JS + CSS.
+**Viewer feature parity with the live app:**
+- List view (chronological) and 3-column grid view with the same toggle as the live app
+- Lightbox with swipe navigation
+- Hashtag filter chips and grid-view search/autocomplete
+- Like counts and comment lists shown as a static snapshot from export time
+- All UI strings in German
 
-**`README.txt`** (in German, as the app's UI language):
-```
-Willkommen in der Event-Galerie!
+**Build flow:** The viewer lives at [frontend/export-viewer/](frontend/export-viewer/) and is built ahead of time into [backend/static/export-viewer/](backend/static/export-viewer/) (committed to the repo). The export job embeds those assets via `include_dir!`, generates `data.json` from the database, processes thumbnails/full-sized variants, and streams the ZIP.
 
-So geht's:
-1. Entpacke diese ZIP-Datei
-   (Windows: Rechtsklick > "Alle extrahieren"; Mac: Doppelklick;
-    Handy: Dateimanager-App verwenden).
-2. Öffne die Datei "Memories.html" in deinem Browser
-   (z. B. Chrome, Safari oder Firefox).
-3. Stöbere durch alle Fotos und Videos.
-   Du kannst nach Hashtags filtern — klicke einfach auf einen Hashtag.
-4. Eine Internetverbindung ist nicht nötig.
-   Alles ist lokal auf deinem Gerät gespeichert.
+**Source files (ZIP archive export, see below)** still contain the unmodified originals — the viewer is the polished read-only experience, the ZIP is the raw archive.
 
-Viel Freude mit den Erinnerungen!
-```
-
-For video-heavy events the ZIP can be several GB. The in-app download guide warns guests: *"Am besten im WLAN herunterladen."* ("Best downloaded on Wi-Fi.")
+For video-heavy events the viewer ZIP can be several GB. The in-app download guide warns guests: *"Am besten im WLAN herunterladen."* ("Best downloaded on Wi-Fi.")
 
 ---
 
@@ -625,7 +602,9 @@ CREATE TABLE "user" (
     recovery_pin_hash   TEXT        NOT NULL,           -- bcrypt(PIN)
     total_upload_bytes  BIGINT      NOT NULL DEFAULT 0, -- running sum for quota checks
     created_at          TIMESTAMPTZ NOT NULL DEFAULT NOW()
-    -- No UNIQUE(event_id, display_name) — PIN disambiguates name collisions
+    -- Case-insensitive UNIQUE on (event_id, LOWER(display_name)) added by migration 007.
+    -- Name collisions are rejected on join; the user is prompted to recover with their PIN
+    -- (or to pick a different name).
 );
 
 -- ─────────────────────────────────────────
@@ -730,12 +709,23 @@ CREATE TABLE config (
 INSERT INTO config (key, value) VALUES
     ('max_image_size_mb',       '20'),
     ('max_video_size_mb',       '500'),
-    ('upload_rate_per_hour',    '10'),
+    ('upload_rate_per_hour',    '100'),  -- raised from 10 in migration 015 (guests upload bursts of 10-20)
     ('feed_rate_per_min',       '60'),
     ('export_rate_per_day',     '3'),
     ('quota_tolerance',         '0.75'),
     ('estimated_guest_count',   '100'),
-    ('compression_concurrency', '2')
+    ('compression_concurrency', '2'),
+    -- Planned (see docs/FEATURES.md §2.6 and §2.7):
+    --   on/off switches for rate limits and quotas, and the privacy note text
+    ('rate_limits_enabled',     'true'),
+    ('upload_rate_enabled',     'true'),
+    ('feed_rate_enabled',       'true'),
+    ('export_rate_enabled',     'true'),
+    ('join_rate_enabled',       'true'),
+    ('quota_enabled',           'true'),
+    ('storage_quota_enabled',   'true'),
+    ('upload_count_quota_enabled', 'true'),
+    ('privacy_note',            '')      -- free text, whitespace + newlines preserved, no HTML
 ON CONFLICT (key) DO NOTHING;
 ```
 
@@ -1173,7 +1163,11 @@ The `/media` volume contains originals, previews, thumbnails, generated exports,
 
 | Decision | Chosen | Rationale |
 |----------|--------|-----------|
-| Recovery mechanism | 4-digit PIN, stored in `localStorage` + "My Account" page | Simple for non-technical guests; no email required |
+| Recovery mechanism | 4-digit PIN, stored in `localStorage` + "My Account" page; Host/Admin can issue a fresh PIN via the user list when a guest loses it entirely | Simple for non-technical guests; no email required; Host-mediated reset preserves the no-email identity model |
+| Host demotion authority | Hosts can demote other Hosts (never themselves); Admin can demote anyone non-admin | Avoids requiring an Admin for every staffing change at the event |
+| Privacy note | Free-text, plain (no HTML), admin-edited, rendered preformatted in My Account | Many events need a per-event privacy statement; preformatted text avoids any markup-injection risk |
+| Data mode | Per-device `localStorage` setting (Saver / Original), default Saver | A guest can be on cellular on one device and Wi-Fi on another; per-device is the right scope |
+| Rate-limit & quota toggles | On/off switches plus numeric values in the `config` table | Lets the Admin disable enforcement for testing or trusted events without redeploying |
 | Admin dashboard path | `/admin` (standard route) | Correct auth checks are the security; obscure paths add no meaningful protection |
 | ZIP contents | Full-quality originals only (Photos + Videos folders) | Clean and simple; no metadata JSON |
 | HTML export assets | Fully offline (relative paths, CSS/JS inlined) | True offline experience; no external dependencies |
@@ -1215,7 +1209,7 @@ The `/media` volume contains originals, previews, thumbnails, generated exports,
 | `uuid` | UUID v7 (time-sortable) |
 | `serde` / `serde_json` | Serialisation |
 | `tower` / `tower-http` | Middleware stack (CORS, compression, static files, request tracing) |
-| `tower-governor` | Token-bucket rate limiting (per IP and per user) |
+| (custom limiter, no crate) | Token-bucket / sliding window built in-tree at [services/rate_limiter.rs](backend/src/services/rate_limiter.rs) |
 | `tokio::sync::Semaphore` | Bounded worker pool for compression tasks |
 | `async-zip` | Streaming ZIP export (no in-memory buffer) |
 | `minijinja` | HTML export template rendering (`Memories.html`) |

README.md

@@ -77,6 +77,7 @@ eventsnap/
 │   ├── svelte.config.js
 │   └── Dockerfile
 ├── docker-compose.yml
+├── docker-compose.dev.yml   # opt-in dev overlay (publishes Postgres on the host)
 ├── Caddyfile
 └── .env.example
 ```
@@ -93,9 +94,9 @@ eventsnap/
 ### Deploy on a fresh VPS
 
 ```bash
-# 1. Clone the repository
-git clone https://git.mc02.dev/fabi/EventSnap.git
-cd EventSnap
+# 1. Clone the repository (into a lowercase dir, matching the paths used below)
+git clone https://git.mc02.dev/fabi/EventSnap.git eventsnap
+cd eventsnap
 
 # 2. Configure environment
 cp .env.example .env
@@ -107,6 +108,13 @@ docker compose up -d
 
 Caddy automatically obtains a Let's Encrypt certificate on first start. The app is live at `https://DOMAIN` within ~30 seconds.
 
+> **If the site never comes up:** with `APP_ENV=production` the backend **refuses to boot** while `JWT_SECRET`/`ADMIN_PASSWORD_HASH` still hold the `.env.example` placeholders (this is deliberate — a publicly-known signing key is worse than downtime). Caddy then waits on the unhealthy `app` container and never serves. Check `docker compose logs app` — a "Refusing to start … placeholder …" line means you skipped step 2. Rotate the secrets (see below) and restart.
+
+> **Production note:** `docker compose up -d` does **not** expose the database — Postgres is reachable only on the internal Docker network. For local development where you need host access to Postgres, opt into the dev overlay explicitly:
+> ```bash
+> docker compose -f docker-compose.yml -f docker-compose.dev.yml up
+> ```
+
 ### Generate required secrets
 
 ```bash
@@ -144,18 +152,18 @@ See [.env.example](.env.example) for the full list with descriptions and default
 ┌───▼────┐      ┌─────▼──────┐
 │  app   │      │  frontend  │
 │ :3000  │      │   :3001    │
-│ (Rust) │      │ (SvelteKit)│
+│ (Rust) │      │(SvelteKit) │
 └───┬────┘      └────────────┘
     │
 ┌───▼────┐
 │   db   │
 │ :5432  │
-│(Postgres│
+│(Postgres)│
 └────────┘
 ```
 
 - `/api/*` and `/media/*` → Rust backend
-- Everything else → SvelteKit frontend
+- Everything else → SvelteKit frontend (`adapter-node`)
 - Named volumes: `postgres_data`, `media_data`, `caddy_data`
 
 ---
@@ -174,21 +182,86 @@ The `/media` volume holds originals, previews, thumbnails, exports, and DB backu
 
 ---
 
+## Running the backend test suite
+
+```bash
+cd backend
+
+# The DB-backed integration tests (backend/tests/) need a live Postgres. `#[sqlx::test]` creates a
+# throwaway database per test and runs backend/migrations/ into it — it does NOT touch this one's data.
+docker run -d --name eventsnap-test-pg -p 55433:5432 \
+  -e POSTGRES_PASSWORD=postgres -e POSTGRES_DB=eventsnap postgres:16-alpine
+
+export DATABASE_URL=postgres://postgres:postgres@localhost:55433/eventsnap
+cargo test        # 44 unit + 12 DB-backed
+cargo clippy --all-targets -- -D warnings
+```
+
+**`cargo test` requires `DATABASE_URL`** — without it the integration tests panic rather than skip.
+That is deliberate. The riskiest code in this repo is SQL (the export epoch state machine, the
+atomic quota increment, the `FOR SHARE` upload lock), and for a long time *not one line of it* was
+executed by `cargo test` — every backend test was a pure-function test, so the tests clustered
+tightly around the code that could not break and stopped exactly where it started to. Tests that
+silently skip when the database is absent recreate that hole; they were meant to be a gate.
+
+## Running the E2E test suite
+
+Playwright-based end-to-end tests live in [`e2e/`](e2e/). They spin up an isolated docker-compose stack (Postgres on `:55432`, Caddy on `:3101`) and exercise the SvelteKit frontend against the real Rust backend with rate limits disabled.
+
+```bash
+cd e2e
+npm install
+npm run install:browsers      # one-time
+npm run stack:up              # bring up the test stack
+npm run test:e2e              # full Phase 1 suite on chromium-desktop
+npm run test:e2e:smoke        # cross-UA matrix (chromium, samsung-internet, webkit, firefox, …)
+npm run stack:down            # tear it down
+```
+
+See [`e2e/README.md`](e2e/README.md) for the full UA matrix, Samsung Internet escalation tiers, and the Phase 2/3 roadmap.
+
+CI runs this on every PR — see [`.github/workflows/e2e.yml`](.github/workflows/e2e.yml) (desktop **and**
+mobile projects), plus [`checks.yml`](.github/workflows/checks.yml) for `cargo test`/clippy, the frontend
+unit tests, svelte-check and the e2e typecheck, and [`audit.yml`](.github/workflows/audit.yml) for
+dependency advisories.
+
+**Playwright runs with `retries: 0`, including in CI.** This repo's real bugs are races, and from the
+outside a race is indistinguishable from a flake — so a retry silently resolves that ambiguity in
+favour of "flake" every time. A flake here is a bug report; treat it as one.
+
+---
+
 ## Development Roadmap
 
+Done:
 - [x] Project blueprint & architecture
 - [x] Monorepo scaffold (`backend/`, `frontend/`, Docker Compose)
-- [ ] DB schema + SQLx migrations
-- [ ] Auth flow (join, JWT, PIN recovery)
-- [ ] Upload pipeline (multipart → compression worker → SSE broadcast)
-- [ ] Client upload queue (IndexedDB, progress, retry)
-- [ ] Gallery feed (grid, SSE, hashtag filters)
-- [ ] Camera capture (`getUserMedia`)
-- [ ] Host Dashboard
-- [ ] Admin Dashboard
-- [ ] Export engine (ZIP + offline HTML)
-- [ ] Rate limiting middleware
-- [ ] End-to-end test event (10+ real devices)
+- [x] DB schema + SQLx migrations (8 migrations through compression status + case-insensitive unique names)
+- [x] Auth flow (join, JWT, 4-digit PIN with bcrypt + 3-attempt/15-min lockout, admin login)
+- [x] Upload pipeline (multipart → compression worker via `tokio::sync::Semaphore` → SSE broadcast)
+- [x] Client upload queue (IndexedDB, progress, retry, rate-limit auto-resume)
+- [x] Gallery feed (list + grid toggle, SSE live updates, hashtag chips, in-memory search + autocomplete)
+- [x] Camera capture (`getUserMedia` with front/back toggle, photo + `MediaRecorder` video)
+- [x] Host Dashboard (event lock, gallery release, ban modal with hide-uploads choice, promote/demote, user search)
+- [x] Admin Dashboard with inner tabs (Stats, Config, Export, Nutzer)
+- [x] Export engine: streaming ZIP + SvelteKit-static HTML viewer (see [docs/CONCEPT_HTML_VIEWER.md](docs/CONCEPT_HTML_VIEWER.md))
+- [x] Custom rate limiter (per-endpoint, hot-reloadable from `config` table)
+- [x] Mobile-first redesign (bottom nav + FAB, see [docs/CONCEPT_MOBILE_UI.md](docs/CONCEPT_MOBILE_UI.md))
+
+Open:
+- [ ] Dynamic per-user storage quota enforcement (formula in [PROJECT.md §12](PROJECT.md); only tracking exists today)
+- [ ] Own-upload deletion UI in the lightbox (backend route exists)
+- [ ] SSE delta-fetch on foreground reconnect (scaffolded in [sse.ts](frontend/src/lib/sse.ts), not wired)
+- [ ] Live diashow / slideshow mode — see [docs/CONCEPT_DIASHOW.md](docs/CONCEPT_DIASHOW.md)
+- [ ] Individual file download button per post
+- [ ] Low-disk alert (< 10 GB free)
+- [ ] Event banner / cover image
+- [ ] Chunked resumable upload for files > 100 MB
+- [ ] Shared Tailwind config between main app and export-viewer
+- [ ] End-to-end test event (10+ real devices on cellular)
+
+See [docs/FEATURES.md](docs/FEATURES.md) for the up-to-date capability matrix by role.
+Speculative / v2+ ideas live in [docs/IDEAS.md](docs/IDEAS.md).
 
 ---
 

TEST_GUIDE.md

@@ -16,67 +16,53 @@ Please test each step in order and report any errors (console errors, wrong text
 9. ✅ Expected: Overlay disappears
 
 ### Step 3 — Feed & navigation
-10. ✅ Expected: Feed shows "Noch keine Fotos." empty state with an upload button
-11. ✅ Expected: Top-right has an **upload button** (blue) and a **person icon** link
+10. ✅ Expected: Feed shows the empty state ("Noch keine Fotos." or similar) with a hint to upload
+11. ✅ Expected: A **persistent bottom nav** is visible with three slots — 🏠 **Feed** on the left, an elevated 📷+ **FAB** in the center, 👤 **Account** on the right
 
 ### Step 4 — My Account page
-12. Click the **person icon** in the top-right
+12. Tap the **👤 Account** tab in the bottom nav
 13. ✅ Expected: `/account` page shows your name (`Max`), a blue "Gast" badge, session expiry date, and your PIN displayed large in an amber box
-14. Click **Kopieren** — check clipboard contains your PIN
+14. Tap **Kopieren** — check the clipboard contains your PIN
 15. ✅ Expected: Button briefly shows "Kopiert!"
-16. Click **Zur Galerie** to go back to the feed
+16. Tap the 🏠 **Feed** tab to go back
 
 ### Step 5 — Upload
-17. Click **Hochladen** — this takes you to `/upload`
-18. Try uploading a photo from your device library
-19. ✅ Expected: Photo appears in queue with a progress bar, then completes
-20. Go back to `/feed` — ✅ Expected: your photo appears in the feed grid
+17. Tap the central **📷+ FAB** in the bottom nav
+18. ✅ Expected: A bottom sheet slides up offering **Kamera** and **Galerie** options
+19. Tap **Galerie** → pick a photo from your device library
+20. ✅ Expected: Preview screen (`/upload`) shows the staged file with an optional caption / hashtag editor
+21. Tap **Hochladen**
+22. ✅ Expected: You return to the feed immediately; the FAB shows a small badge while uploading; the photo appears in the feed once processing completes
 
 ### Step 6 — Onboarding guide not shown again
-21. Reload the page at `/feed`
-22. ✅ Expected: The onboarding overlay does **not** appear (already dismissed)
+23. Reload the page at `/feed`
+24. ✅ Expected: The onboarding overlay does **not** appear (already dismissed)
 
 ### Step 7 — Recover (open a private/incognito window)
-23. Open a new **private/incognito** window at **http://localhost:5173/recover**
-24. Enter the same name (`Max`) and the PIN you copied
-25. ✅ Expected: You're redirected to the feed with the same account
+25. Open a new **private/incognito** window at **http://localhost:5173/recover**
+26. Enter the same name (`Max`) and the PIN you copied
+27. ✅ Expected: You're redirected to the feed with the same account
 
 ### Step 8 — Upload rate-limit auto-retry
-26. Upload more than 20 photos in one hour to trigger the rate limit
-27. ✅ Expected: When the limit is hit, remaining items stay **Wartend** (not error)
-28. ✅ Expected: An amber banner appears in the queue: "Upload-Limit erreicht. Wird in Xs automatisch fortgesetzt."
-29. ✅ Expected: The countdown ticks down and uploads resume automatically when it reaches 0
+28. Upload more than the per-hour limit of photos in quick succession to trigger the rate limit
+29. ✅ Expected: When the limit is hit, remaining items stay **Wartend** (not error)
+30. ✅ Expected: An amber banner appears in the queue: "Upload-Limit erreicht. Wird in Xs automatisch fortgesetzt."
+31. ✅ Expected: The countdown ticks down and uploads resume automatically when it reaches 0
 
 ### Step 9 — Name uniqueness (case-insensitive)
-30. In a private/incognito window go to **http://localhost:5173/join**
-31. Enter `max` or `MAX` — the same name already taken in Step 1 (different case)
-32. ✅ Expected: Instead of creating a new account, an amber warning appears: „Max ist bereits vergeben." with name tips
-33. ✅ Expected: A PIN input and **Anmelden** button appear, plus an **Anderen Namen wählen** button
-34. Enter your PIN from Step 1 and click **Anmelden**
-35. ✅ Expected: You're signed in to the existing `Max` account and redirected to the feed
-36. Alternatively, click **Anderen Namen wählen** — ✅ Expected: the name input reappears with `max` pre-filled so you can edit it
+32. In a private/incognito window go to **http://localhost:5173/join**
+33. Enter `max` or `MAX` — the same name already taken in Step 1 (different case)
+34. ✅ Expected: Instead of creating a new account, an amber warning appears: „Max ist bereits vergeben." with name tips
+35. ✅ Expected: A PIN input and **Anmelden** button appear, plus an **Anderen Namen wählen** button
+36. Enter your PIN from Step 1 and click **Anmelden**
+37. ✅ Expected: You're signed in to the existing `Max` account and redirected to the feed
+38. Alternatively, click **Anderen Namen wählen** — ✅ Expected: the name input reappears with `max` pre-filled so you can edit it
 
 ---
 
 ## Admin & Host Features
 
-For these steps you need an admin session. Log in via the admin API (or use the curl command below):
-
-```bash
-curl -s -X POST http://localhost:3000/api/v1/admin/login \
-  -H 'Content-Type: application/json' \
-  -d '{"password":"admin123"}' | python3 -m json.tool
-```
-
-Copy the `jwt` value — you'll need it for the Authorization header in curl commands.
-To use the admin dashboard in the browser, paste the token into DevTools console:
-
-```js
-localStorage.setItem('token', '<jwt>');
-localStorage.setItem('role', 'admin');
-```
-
-Then navigate to **http://localhost:5173/admin**.
+For these steps you need an admin session. Go to **http://localhost:5173/admin/login** and enter the admin password (`admin123` for the dev environment). You'll be redirected to the dashboard automatically.
 
 ### Step 10 — Admin Dashboard: Stats & Config
 1. Go to **http://localhost:5173/admin**

backend/Cargo.lock

@@ -513,6 +513,17 @@ dependencies = [
  "shlex",
 ]
 
+[[package]]
+name = "cfb"
+version = "0.7.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d38f2da7a0a2c4ccf0065be06397cc26a81f4e528be095826eee9d4adbb8c60f"
+dependencies = [
+ "byteorder",
+ "fnv",
+ "uuid",
+]
+
 [[package]]
 name = "cfg-if"
 version = "1.0.4"
@@ -896,8 +907,9 @@ dependencies = [
  "dotenvy",
  "futures",
  "image",
+ "include_dir",
+ "infer",
  "jsonwebtoken",
- "minijinja",
  "oxipng",
  "rand 0.9.2",
  "serde",
@@ -1004,6 +1016,12 @@ dependencies = [
  "spin",
 ]
 
+[[package]]
+name = "fnv"
+version = "1.0.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
+
 [[package]]
 name = "foldhash"
 version = "0.1.5"
@@ -1577,6 +1595,25 @@ version = "1.12.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e7c5cedc30da3a610cac6b4ba17597bdf7152cf974e8aab3afb3d54455e371c8"
 
+[[package]]
+name = "include_dir"
+version = "0.7.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "923d117408f1e49d914f1a379a309cffe4f18c05cf4e3d12e613a15fc81bd0dd"
+dependencies = [
+ "include_dir_macros",
+]
+
+[[package]]
+name = "include_dir_macros"
+version = "0.7.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7cab85a7ed0bd5f0e76d93846e0147172bed2e2d3f859bcc33a8d9699cad1a75"
+dependencies = [
+ "proc-macro2",
+ "quote",
+]
+
 [[package]]
 name = "indexmap"
 version = "2.13.0"
@@ -1590,6 +1627,15 @@ dependencies = [
  "serde_core",
 ]
 
+[[package]]
+name = "infer"
+version = "0.15.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cb33622da908807a06f9513c19b3c1ad50fab3e4137d82a78107d502075aa199"
+dependencies = [
+ "cfb",
+]
+
 [[package]]
 name = "inout"
 version = "0.1.4"
@@ -1832,12 +1878,6 @@ version = "2.8.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f8ca58f447f06ed17d5fc4043ce1b10dd205e060fb3ce5b979b8ed8e59ff3f79"
 
-[[package]]
-name = "memo-map"
-version = "0.3.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "38d1115007560874e373613744c6fba374c17688327a71c1476d1a5954cc857b"
-
 [[package]]
 name = "mime"
 version = "0.3.17"
@@ -1854,16 +1894,6 @@ dependencies = [
  "unicase",
 ]
 
-[[package]]
-name = "minijinja"
-version = "2.18.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "328251e58ad8e415be6198888fc207502727dc77945806421ab34f35bf012e7d"
-dependencies = [
- "memo-map",
- "serde",
-]
-
 [[package]]
 name = "miniz_oxide"
 version = "0.8.9"

backend/Cargo.toml

@@ -29,7 +29,8 @@ sysinfo = "0.32"
 image = "0.25"
 oxipng = "9"
 async_zip = { version = "0.0.17", features = ["tokio", "deflate"] }
-minijinja = "2"
+include_dir = "0.7"
+infer = "0.15"
 
 [profile.release]
 opt-level = 3

backend/Dockerfile

@@ -1,5 +1,5 @@
 # --- Build stage ---
-FROM rust:1.87-alpine AS builder
+FROM rust:1.88-alpine AS builder
 
 RUN apk add --no-cache musl-dev pkgconfig openssl-dev
 
@@ -11,6 +11,8 @@ RUN mkdir src && echo "fn main(){}" > src/main.rs && \
     rm -rf src
 
 COPY src ./src
+COPY static ./static
+COPY migrations ./migrations
 RUN touch src/main.rs && cargo build --release
 
 # --- Runtime stage ---
@@ -18,8 +20,18 @@ FROM alpine:3.21
 
 RUN apk add --no-cache ca-certificates ffmpeg
 
+# Run as a non-root user. Pre-create and chown the media + export mount paths so
+# the fresh named volumes inherit the non-root ownership (Docker seeds an empty
+# named volume from the image directory, preserving its uid/gid) and uploads +
+# export archives can be written. Exports live OUTSIDE /media on purpose so the
+# public media ServeDir can't reach them.
+RUN addgroup -S app && adduser -S app -G app
+
 WORKDIR /app
 COPY --from=builder /app/target/release/eventsnap-backend ./
 
+RUN mkdir -p /media /exports && chown -R app:app /app /media /exports
+USER app
+
 EXPOSE 3000
 CMD ["./eventsnap-backend"]

backend/migrations/008_compression_status.down.sql

@@ -0,0 +1,2 @@
+-- Remove compression_status field
+ALTER TABLE upload DROP COLUMN compression_status;

backend/migrations/008_compression_status.up.sql

@@ -0,0 +1,6 @@
+-- Add compression_status to track media processing state
+ALTER TABLE upload ADD COLUMN compression_status TEXT NOT NULL DEFAULT 'pending';
+
+-- Values: 'pending', 'processing', 'done', 'failed'
+-- Add comment to document the field
+COMMENT ON COLUMN upload.compression_status IS 'Tracks media compression/preview generation: pending -> processing -> (done or failed)';

backend/migrations/009_feature_toggles.down.sql

@@ -0,0 +1,11 @@
+DELETE FROM config WHERE key IN (
+    'rate_limits_enabled',
+    'upload_rate_enabled',
+    'feed_rate_enabled',
+    'export_rate_enabled',
+    'join_rate_enabled',
+    'quota_enabled',
+    'storage_quota_enabled',
+    'upload_count_quota_enabled',
+    'privacy_note'
+);

backend/migrations/009_feature_toggles.up.sql

@@ -0,0 +1,16 @@
+-- Feature toggles for rate limits and quotas, plus the admin-configurable
+-- Datenschutzhinweis. Everything lives in the `config` table — no schema change.
+INSERT INTO config (key, value) VALUES
+    -- Rate limits (master + per-endpoint)
+    ('rate_limits_enabled',        'true'),
+    ('upload_rate_enabled',        'true'),
+    ('feed_rate_enabled',          'true'),
+    ('export_rate_enabled',        'true'),
+    ('join_rate_enabled',          'true'),
+    -- Quotas (master + per-area)
+    ('quota_enabled',              'true'),
+    ('storage_quota_enabled',      'true'),
+    ('upload_count_quota_enabled', 'true'),
+    -- Free-text privacy note shown to guests in My Account. Plain text — no HTML.
+    ('privacy_note',               '')
+ON CONFLICT (key) DO NOTHING;

backend/migrations/010_review_indexes.down.sql

@@ -0,0 +1,3 @@
+DROP INDEX IF EXISTS idx_comment_hashtag_hashtag;
+DROP INDEX IF EXISTS idx_comment_user;
+DROP INDEX IF EXISTS idx_upload_event_created_id;

backend/migrations/010_review_indexes.up.sql

@@ -0,0 +1,17 @@
+-- Composite feed index with an id tiebreaker so keyset pagination
+-- (ORDER BY created_at DESC, id DESC) stays index-covered and stable when
+-- multiple uploads share a created_at timestamp.
+CREATE INDEX idx_upload_event_created_id
+    ON upload(event_id, created_at DESC, id DESC)
+    WHERE deleted_at IS NULL;
+
+-- A user's own comments (moderation, "who commented"). The sibling
+-- idx_upload_user already exists for uploads; comment(user_id) was missing.
+CREATE INDEX idx_comment_user
+    ON comment(user_id)
+    WHERE deleted_at IS NULL;
+
+-- Hashtag filtering over comments — mirrors idx_upload_hashtag_hashtag, which
+-- only covered upload_hashtag.
+CREATE INDEX idx_comment_hashtag_hashtag
+    ON comment_hashtag(hashtag_id);

backend/migrations/011_ban_hides_and_pin_reset.down.sql

@@ -0,0 +1,25 @@
+DROP TABLE IF EXISTS pin_reset_request;
+
+-- Restore v_feed without the is_banned filter.
+CREATE OR REPLACE VIEW v_feed AS
+SELECT
+    u.id,
+    u.event_id,
+    u.user_id,
+    usr.display_name             AS uploader_name,
+    usr.is_banned,
+    usr.uploads_hidden,
+    u.preview_path,
+    u.thumbnail_path,
+    u.mime_type,
+    u.caption,
+    u.created_at,
+    COUNT(DISTINCT l.user_id)    AS like_count,
+    COUNT(DISTINCT c.id)         AS comment_count
+FROM upload u
+JOIN "user" usr ON u.user_id = usr.id
+LEFT JOIN "like" l  ON l.upload_id = u.id
+LEFT JOIN comment c ON c.upload_id = u.id AND c.deleted_at IS NULL
+WHERE u.deleted_at IS NULL
+  AND usr.uploads_hidden = FALSE
+GROUP BY u.id, usr.display_name, usr.is_banned, usr.uploads_hidden;

backend/migrations/011_ban_hides_and_pin_reset.up.sql

@@ -0,0 +1,39 @@
+-- Ban now ALWAYS hides: exclude banned uploaders from the feed view. Previously ban and
+-- hide were decoupled (a host could ban without hiding), leaving a banned user's photos on
+-- the feed and baked into the export. `find_visible_media` and the export query get the
+-- same `is_banned = FALSE` filter in code.
+CREATE OR REPLACE VIEW v_feed AS
+SELECT
+    u.id,
+    u.event_id,
+    u.user_id,
+    usr.display_name             AS uploader_name,
+    usr.is_banned,
+    usr.uploads_hidden,
+    u.preview_path,
+    u.thumbnail_path,
+    u.mime_type,
+    u.caption,
+    u.created_at,
+    COUNT(DISTINCT l.user_id)    AS like_count,
+    COUNT(DISTINCT c.id)         AS comment_count
+FROM upload u
+JOIN "user" usr ON u.user_id = usr.id
+LEFT JOIN "like" l  ON l.upload_id = u.id
+LEFT JOIN comment c ON c.upload_id = u.id AND c.deleted_at IS NULL
+WHERE u.deleted_at IS NULL
+  AND usr.uploads_hidden = FALSE
+  AND usr.is_banned = FALSE
+GROUP BY u.id, usr.display_name, usr.is_banned, usr.uploads_hidden;
+
+-- pin_reset_request: a guest who forgot their PIN (localStorage was the only copy) can ask
+-- a host to reset it in-app, instead of being permanently orphaned. One pending request per
+-- user; cleared when the host resets the PIN or dismisses it, and cascades if the user/event
+-- is removed.
+CREATE TABLE pin_reset_request (
+    id          UUID        PRIMARY KEY DEFAULT gen_random_uuid(),
+    event_id    UUID        NOT NULL REFERENCES event(id) ON DELETE CASCADE,
+    user_id     UUID        NOT NULL REFERENCES "user"(id) ON DELETE CASCADE,
+    created_at  TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+    UNIQUE (user_id)
+);

backend/migrations/012_export_release_seq.down.sql

@@ -0,0 +1,2 @@
+ALTER TABLE export_job
+    DROP COLUMN IF EXISTS release_seq;

backend/migrations/012_export_release_seq.up.sql

@@ -0,0 +1,11 @@
+-- H1: export generation guard. A reopen (which clears `export_released_at`) followed by a
+-- re-release could spawn a fresh export worker while a worker from the PRIOR release was
+-- still streaming a now-stale snapshot to `Gallery.zip`. The stale worker's finalize then
+-- re-set `export_*_ready = TRUE` on an archive that predated the reopen — a silent,
+-- permanent data loss (new uploads missing from a "ready" keepsake).
+--
+-- `release_seq` is a per-(event,type) generation counter bumped on every (re)release.
+-- A worker captures the seq it claimed and only finalizes / flips the ready flag while
+-- that seq is still current; a superseded worker discards its output instead.
+ALTER TABLE export_job
+    ADD COLUMN release_seq BIGINT NOT NULL DEFAULT 0;

backend/migrations/013_uploads_hidden_at.down.sql

@@ -0,0 +1,2 @@
+ALTER TABLE "user"
+    DROP COLUMN IF EXISTS uploads_hidden_at;

backend/migrations/013_uploads_hidden_at.up.sql

@@ -0,0 +1,12 @@
+-- Ban-replay on reconnect. A ban is not a soft-delete (it sets `uploads_hidden`/`is_banned`,
+-- never `upload.deleted_at`), and live eviction rode only on the ephemeral `user-hidden` SSE
+-- broadcast — which has no reconnect-replay. So a client (especially the unattended diashow
+-- projector) that missed that broadcast kept cycling the banned user's already-loaded slides.
+--
+-- `uploads_hidden_at` stamps WHEN a user's uploads became hidden, so the reconnect delta can
+-- return the set of users hidden since the client's cursor and the client can evict them.
+ALTER TABLE "user"
+    ADD COLUMN uploads_hidden_at TIMESTAMPTZ;
+
+-- Backfill already-hidden users so a reconnect right after this migration still evicts them.
+UPDATE "user" SET uploads_hidden_at = NOW() WHERE uploads_hidden = TRUE;

backend/migrations/014_export_epoch.down.sql

@@ -0,0 +1,28 @@
+-- Restore the stored ready flags and the per-job counter.
+DROP VIEW IF EXISTS export_current;
+
+ALTER TABLE event ADD COLUMN export_zip_ready  BOOLEAN NOT NULL DEFAULT FALSE;
+ALTER TABLE event ADD COLUMN export_html_ready BOOLEAN NOT NULL DEFAULT FALSE;
+
+-- Re-derive the flags from what the epoch model considers downloadable, so the old code sees
+-- exactly the state it would have written itself.
+UPDATE event e
+   SET export_zip_ready = EXISTS (
+         SELECT 1 FROM export_job j
+          WHERE j.event_id = e.id AND j.type = 'zip'
+            AND j.status = 'done' AND j.epoch = e.export_epoch
+       ),
+       export_html_ready = EXISTS (
+         SELECT 1 FROM export_job j
+          WHERE j.event_id = e.id AND j.type = 'html'
+            AND j.status = 'done' AND j.epoch = e.export_epoch
+       )
+ WHERE e.export_released_at IS NOT NULL;
+
+ALTER TABLE export_job RENAME COLUMN epoch TO release_seq;
+
+-- The old code treats release_seq as a non-negative per-job counter; the -1 retirement sentinel
+-- has no meaning there, so floor it back to 0.
+UPDATE export_job SET release_seq = 0 WHERE release_seq < 0;
+
+ALTER TABLE event DROP COLUMN export_epoch;

backend/migrations/014_export_epoch.up.sql

@@ -0,0 +1,86 @@
+-- Export generation, unified.
+--
+-- ⚠ ROLLBACK RUNBOOK. This is the repo's first DESTRUCTIVE migration (it DROPs two columns). The
+-- previous image will NOT boot against this schema: sqlx runs migrations before serving and errors
+-- with VersionMissing on an unknown version, so you get a crash loop, not a degraded service.
+-- To roll back to the previous release you must run 014_export_epoch.down.sql BY HAND FIRST, then
+-- deploy the old image. The down migration re-derives the old ready flags from the epoch state, so
+-- no keepsake is lost.
+--
+-- Before this migration, "which generation of the keepsake is current?" had no single answer.
+-- It was assembled at runtime from three separately-written pieces of state across two tables:
+--   * event.export_released_at   — a release marker with NO identity (you cannot ask *which* release)
+--   * export_job.release_seq     — a per-row counter, in a different table, bumped in a different statement
+--   * event.export_{zip,html}_ready — a CACHED COPY of the derivation of the other two
+-- Keeping those three in agreement took ~10 hand-written guards, and every guard was a repair of the
+-- same missing invariant. Three consecutive review rounds each found another path that slipped through.
+--
+-- This replaces all of it with ONE authority:
+--
+--   event.export_epoch — a monotonic counter bumped in the SAME UPDATE as any change to
+--                        export_released_at (release and reopen are its only writers).
+--   export_job.epoch   — a COPY of event.export_epoch taken when the job was enqueued.
+--                        NEVER independently incremented.
+--
+-- The single invariant, which replaces the entire proof:
+--
+--   An export is downloadable IFF
+--       event.export_released_at IS NOT NULL
+--   AND export_job.epoch = event.export_epoch
+--   AND export_job.status = 'done'
+--
+-- Readiness is therefore DERIVED, never stored — so it cannot drift, and no worker can set it.
+-- A worker holding a dead epoch is INERT BY CONSTRUCTION: anything it writes is invisible to every
+-- reader, with no guard involved. Losing a race can no longer corrupt state; it can only waste work.
+--
+-- Crucially, epoch equality is checked on the SAME ROW the worker updates (export_job), never via a
+-- cross-table EXISTS. That matters: under READ COMMITTED, when an UPDATE blocks on a row lock and the
+-- blocker commits, Postgres re-evaluates the WHERE against the *updated target row* but answers
+-- subqueries on OTHER tables from the statement's ORIGINAL snapshot. The old cross-row guard
+-- (`EXISTS (SELECT 1 FROM event WHERE ... export_released_at IS NOT NULL)`) was unsound for exactly
+-- that reason — it could admit a claim against an already-reopened event while RETURNING the
+-- post-bump seq. A same-row `epoch = $n` predicate is re-evaluated correctly by EPQ.
+
+ALTER TABLE event ADD COLUMN export_epoch BIGINT NOT NULL DEFAULT 0;
+
+-- A currently-released event's live generation becomes epoch 1.
+UPDATE event SET export_epoch = 1 WHERE export_released_at IS NOT NULL;
+
+-- The per-job counter becomes a plain copy of the event epoch it was enqueued for.
+ALTER TABLE export_job RENAME COLUMN release_seq TO epoch;
+
+-- Carry the OLD notion of "downloadable" across exactly: a job the old model considered ready
+-- (released + done + its ready flag) adopts the event's live epoch. Everything else is retired to
+-- -1, which can never equal a non-negative event.export_epoch, so it is inert forever.
+UPDATE export_job j
+   SET epoch = CASE
+         WHEN e.export_released_at IS NOT NULL
+          AND j.status = 'done'
+          AND ((j.type = 'zip'  AND e.export_zip_ready)
+            OR (j.type = 'html' AND e.export_html_ready))
+         THEN e.export_epoch
+         ELSE -1
+       END
+  FROM event e
+ WHERE e.id = j.event_id;
+
+-- The duplicated derivation. Gone — along with the whole class of bugs where a stale worker
+-- resurrected it, or where it disagreed with the job row it was supposed to summarise.
+ALTER TABLE event DROP COLUMN export_zip_ready;
+ALTER TABLE event DROP COLUMN export_html_ready;
+
+-- The ONE definition of "the current generation". Every reader goes through this, so readiness and
+-- the file path are resolved in a SINGLE read — closing the download-path TOCTOU where the ready
+-- flag was checked in one query and file_path fetched in another (a reopen between the two served a
+-- keepsake that should have 404'd).
+CREATE VIEW export_current AS
+SELECT j.event_id,
+       j.type,
+       j.status,
+       j.progress_pct,
+       j.file_path,
+       j.epoch
+  FROM export_job j
+  JOIN event e ON e.id = j.event_id
+ WHERE e.export_released_at IS NOT NULL   -- implied by the epoch match; kept as an assertion
+   AND j.epoch = e.export_epoch;

backend/migrations/015_raise_upload_rate.down.sql

@@ -0,0 +1,3 @@
+-- Revert the default upload rate to 10/hour for installs still on the raised
+-- default (preserves any explicit admin override at another value).
+UPDATE config SET value = '10' WHERE key = 'upload_rate_per_hour' AND value = '100';

backend/migrations/015_raise_upload_rate.up.sql

@@ -0,0 +1,10 @@
+-- Raise the default per-guest upload rate from 10/hour to 100/hour.
+--
+-- Rationale: guests routinely upload a burst of 10-20 photos at once (phone
+-- multi-select). At the old default of 10/hour a real guest's first burst was
+-- throttled — surfaced by the 2026-07-18 load test. 100/hour comfortably covers
+-- several bursts across an event while still bounding abuse.
+--
+-- Only bump installs still on the old default; an admin who deliberately set a
+-- different value keeps it (migration 005 seeded 10; this UPDATE is scoped to '10').
+UPDATE config SET value = '100' WHERE key = 'upload_rate_per_hour' AND value = '10';

backend/migrations/016_display_derivative.down.sql

@@ -0,0 +1,28 @@
+-- Drop the view (frees the column dependency), remove the column, then restore the
+-- pre-016 view definition (matches migration 011).
+DROP VIEW IF EXISTS v_feed;
+ALTER TABLE upload DROP COLUMN display_path;
+
+CREATE VIEW v_feed AS
+SELECT
+    u.id,
+    u.event_id,
+    u.user_id,
+    usr.display_name             AS uploader_name,
+    usr.is_banned,
+    usr.uploads_hidden,
+    u.preview_path,
+    u.thumbnail_path,
+    u.mime_type,
+    u.caption,
+    u.created_at,
+    COUNT(DISTINCT l.user_id)    AS like_count,
+    COUNT(DISTINCT c.id)         AS comment_count
+FROM upload u
+JOIN "user" usr ON u.user_id = usr.id
+LEFT JOIN "like" l  ON l.upload_id = u.id
+LEFT JOIN comment c ON c.upload_id = u.id AND c.deleted_at IS NULL
+WHERE u.deleted_at IS NULL
+  AND usr.uploads_hidden = FALSE
+  AND usr.is_banned = FALSE
+GROUP BY u.id, usr.display_name, usr.is_banned, usr.uploads_hidden;

backend/migrations/016_display_derivative.up.sql

@@ -0,0 +1,34 @@
+-- Display derivative: a big-screen-quality image (~2048px long edge) for the diashow.
+-- The 800px `preview_path` is sized for phone feeds (data saver); upscaled on a projector
+-- it looks soft. The diashow uses `display_path` instead — bounded in size (safe to decode
+-- on weak kiosk hardware) yet sharp on 1080p/4K. NULL until the compression worker (or the
+-- one-time backfill) generates it; consumers fall back to the original when absent.
+ALTER TABLE upload ADD COLUMN display_path TEXT;
+
+-- Recreate (not CREATE OR REPLACE, which only allows appending columns at the end) so the
+-- new column can sit alongside preview_path/thumbnail_path.
+DROP VIEW IF EXISTS v_feed;
+CREATE VIEW v_feed AS
+SELECT
+    u.id,
+    u.event_id,
+    u.user_id,
+    usr.display_name             AS uploader_name,
+    usr.is_banned,
+    usr.uploads_hidden,
+    u.preview_path,
+    u.thumbnail_path,
+    u.display_path,
+    u.mime_type,
+    u.caption,
+    u.created_at,
+    COUNT(DISTINCT l.user_id)    AS like_count,
+    COUNT(DISTINCT c.id)         AS comment_count
+FROM upload u
+JOIN "user" usr ON u.user_id = usr.id
+LEFT JOIN "like" l  ON l.upload_id = u.id
+LEFT JOIN comment c ON c.upload_id = u.id AND c.deleted_at IS NULL
+WHERE u.deleted_at IS NULL
+  AND usr.uploads_hidden = FALSE
+  AND usr.is_banned = FALSE
+GROUP BY u.id, usr.display_name, usr.is_banned, usr.uploads_hidden;

backend/migrations/README.md

@@ -0,0 +1,28 @@
+# Migrations
+
+SQLx-managed Postgres migrations. Each `NNN_topic.up.sql` has a matching
+`NNN_topic.down.sql`. Run by `sqlx::migrate!()` at app start.
+
+## Rules
+
+1. **Never edit a shipped migration.** If a column needs to change or a fix needs to
+   land, write a new migration. Production has already applied the old one and SQLx
+   tracks each by checksum — editing in place will fail to apply on existing databases.
+2. **Always pair `.up.sql` with a `.down.sql`.** Reverts may not be perfect (data
+   loss is sometimes unavoidable) but the file must exist and do the best it can.
+3. **Prefer additive changes.** New columns, new tables, new keys in `config`. Drop /
+   rename only when there is no alternative.
+4. **No business logic in migrations.** Schema + seeds only. Anything that needs Rust
+   code goes in a one-off binary, not a migration file.
+5. **One concern per migration.** Easier to revert. Easier to read in `git log`.
+
+## Numbering
+
+Zero-padded three digits, monotonically increasing. The next free number lives at the
+bottom of the directory listing — pick that.
+
+## Seed-only migrations
+
+When you only need to add `config` keys (feature flags, defaults), use
+`INSERT … ON CONFLICT DO NOTHING` so existing operator overrides survive. See
+`009_feature_toggles.up.sql` for the canonical shape.

backend/scripts/rehearse-014.sh

@@ -0,0 +1,195 @@
+#!/usr/bin/env bash
+#
+# Dress rehearsal for migration 014 (export_epoch) — the repo's first DESTRUCTIVE migration.
+#
+# 014 DROPs event.export_zip_ready / export_html_ready and rewrites export_job.release_seq into an
+# epoch. The dangerous part is not the DDL, it's the BACKFILL: it has to carry the old notion of
+# "downloadable" across exactly. Get it wrong and a released event's keepsake silently 404s for
+# every guest, on a dataset you cannot re-create.
+#
+# This script proves the backfill preserves downloadability, against a scratch copy of a REAL dump:
+#
+#   ./rehearse-014.sh /path/to/prod-dump.sql     # rehearse against production data
+#   ./rehearse-014.sh                            # synthetic: build a pre-014 DB covering every case
+#
+# It asserts:
+#   1. up:   {(event,type) downloadable BEFORE} == {(event,type) downloadable AFTER}
+#   2. down: the old ready flags come back identical  (so a rollback is actually a rollback)
+#   3. up again: still identical                      (so a roll-forward after a rollback is safe)
+#
+# It NEVER touches your real database. Everything runs in a throwaway container.
+#
+# NOTE ON ROLLBACK (see the runbook header in 014_export_epoch.up.sql): sqlx runs migrations before
+# serving and errors with VersionMissing on an unknown version, so the PREVIOUS image will not boot
+# against the 014 schema — it crash-loops. Rolling back means running 014_export_epoch.down.sql BY
+# HAND FIRST, then deploying the old image. Assertion 2 is what makes that safe. Rehearse it before
+# you need it, not while the party is happening.
+
+set -euo pipefail
+
+DUMP="${1:-}"
+MIG_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/../migrations" && pwd)"
+CONTAINER="eventsnap-rehearse-014"
+PGPASSWORD="rehearse"
+DB="eventsnap_rehearsal"
+
+cleanup() { docker rm -f "$CONTAINER" > /dev/null 2>&1 || true; }
+trap cleanup EXIT
+cleanup
+
+echo "▸ Starting a throwaway postgres:16 (nothing here touches your real DB)…"
+docker run --rm -d --name "$CONTAINER" \
+  -e POSTGRES_PASSWORD="$PGPASSWORD" -e POSTGRES_DB="$DB" \
+  postgres:16-alpine > /dev/null
+
+psql() { docker exec -i -e PGPASSWORD="$PGPASSWORD" "$CONTAINER" psql -U postgres -d "$DB" -v ON_ERROR_STOP=1 "$@"; }
+
+for _ in $(seq 1 30); do
+  if docker exec "$CONTAINER" pg_isready -U postgres > /dev/null 2>&1; then break; fi
+  sleep 1
+done
+
+if [[ -n "$DUMP" ]]; then
+  echo "▸ Restoring $DUMP …"
+  psql -q < "$DUMP"
+else
+  echo "▸ No dump given — building a synthetic pre-014 DB (migrations 001→013)…"
+  for f in "$MIG_DIR"/0[0-1][0-9]_*.up.sql; do
+    [[ "$(basename "$f")" == 014_* ]] && continue
+    psql -q < "$f"
+  done
+
+  # Every state the backfill has to classify. If 014 mishandles ANY of these, assertion 1 fails.
+  # Every state the backfill has to classify. export_job is UNIQUE (event_id, type), so each event
+  # has at most one job per type — the cases are distinguished by event, not by piling up rows.
+  #   A: released, both flags, both done  → both stay downloadable
+  #   B: released, ZIP flag only          → ZIP downloadable, HTML not (a half-built keepsake)
+  #   C: released, jobs done, NO flags    → NOT downloadable (a superseded worker's finished job:
+  #                                          the exact state the old ready-flag model used to leak)
+  #   D: never released, jobs done        → NOT downloadable (reopened, or built before release)
+  #   E: released, ZIP flag set but the job is FAILED/RUNNING → NOT downloadable (flag/job disagree,
+  #                                          which the old two-source model made representable at all)
+  echo "▸ Seeding: released+both, zip-only, done-but-stale, unreleased, flag/job-disagreement…"
+  psql -q <<'SQL'
+INSERT INTO event (id, slug, name, export_released_at, export_zip_ready, export_html_ready) VALUES
+  ('aaaaaaaa-0000-0000-0000-000000000001','ev-a','A', now(), TRUE,  TRUE),
+  ('aaaaaaaa-0000-0000-0000-000000000002','ev-b','B', now(), TRUE,  FALSE),
+  ('aaaaaaaa-0000-0000-0000-000000000003','ev-c','C', now(), FALSE, FALSE),
+  ('aaaaaaaa-0000-0000-0000-000000000004','ev-d','D', NULL,  FALSE, FALSE),
+  ('aaaaaaaa-0000-0000-0000-000000000005','ev-e','E', now(), TRUE,  TRUE);
+
+INSERT INTO export_job (event_id, type, status, progress_pct, file_path, release_seq)
+SELECT e.id, t::export_type, 'done'::export_status, 100, '/x/' || e.slug || '.' || t, 0
+  FROM event e CROSS JOIN (VALUES ('zip'),('html')) AS v(t);
+
+-- E: the flags claim ready, the jobs say otherwise. Old model required BOTH, so neither is
+-- downloadable — the migration must not "trust the flag" and resurrect them.
+UPDATE export_job SET status = 'failed',  file_path = NULL, progress_pct = 40
+ WHERE event_id = 'aaaaaaaa-0000-0000-0000-000000000005' AND type = 'zip';
+UPDATE export_job SET status = 'running', file_path = NULL, progress_pct = 70
+ WHERE event_id = 'aaaaaaaa-0000-0000-0000-000000000005' AND type = 'html';
+SQL
+fi
+
+echo "▸ Snapshotting what is downloadable under the OLD model…"
+psql -q <<'SQL'
+CREATE TABLE _rehearsal_before AS
+SELECT j.event_id, j.type
+  FROM export_job j JOIN event e ON e.id = j.event_id
+ WHERE e.export_released_at IS NOT NULL
+   AND j.status = 'done'
+   AND ((j.type = 'zip' AND e.export_zip_ready) OR (j.type = 'html' AND e.export_html_ready));
+
+-- For assertion 2: the exact flags a rollback has to reproduce.
+CREATE TABLE _rehearsal_flags AS
+SELECT id, export_zip_ready, export_html_ready FROM event;
+SQL
+
+before=$(psql -tAc "SELECT count(*) FROM _rehearsal_before")
+echo "  → $before (event,type) pair(s) downloadable before."
+
+# The assertion. A symmetric difference of zero means the backfill carried the old notion of
+# "downloadable" across EXACTLY: nothing gained (a stale keepsake resurrected), nothing lost (a
+# guest's keepsake silently 404s).
+assert_up_preserves() {
+  local label="$1" diff
+  diff=$(psql -tAc "
+    SELECT count(*) FROM (
+      (SELECT event_id, type FROM _rehearsal_before
+        EXCEPT SELECT event_id, type FROM export_current WHERE status = 'done')
+      UNION ALL
+      (SELECT event_id, type FROM export_current WHERE status = 'done'
+        EXCEPT SELECT event_id, type FROM _rehearsal_before)
+    ) d")
+  if [[ "$diff" != "0" ]]; then
+    echo "✗ FAIL ($label): $diff (event,type) pair(s) changed downloadability across the migration."
+    psql -c "
+      (SELECT 'LOST — guests can no longer download this' AS problem, event_id, type FROM _rehearsal_before
+        EXCEPT ALL SELECT 'LOST — guests can no longer download this', event_id, type FROM export_current WHERE status='done')
+      UNION ALL
+      (SELECT 'GAINED — a stale keepsake became downloadable', event_id, type FROM export_current WHERE status='done'
+        EXCEPT ALL SELECT 'GAINED — a stale keepsake became downloadable', event_id, type FROM _rehearsal_before)"
+    exit 1
+  fi
+  echo "✓ $label: downloadability preserved exactly ($before pair(s))."
+}
+
+echo "▸ Applying 014 (up)…"
+psql -q < "$MIG_DIR/014_export_epoch.up.sql"
+assert_up_preserves "up"
+
+echo "▸ Applying 014 (down) — the rollback path…"
+psql -q < "$MIG_DIR/014_export_epoch.down.sql"
+
+# The down migration is an inverse UP TO DRIFT, not a bit-exact inverse — deliberately.
+#
+# The old model let a ready flag disagree with its job row (flag TRUE over a running/failed job):
+# the flag was a CACHED copy of a derivation, and keeping a cache in agreement with its source
+# across concurrent workers is precisely what it kept failing at. The down migration re-derives the
+# flags from the epoch state, so such a pair comes back FALSE. That HEALS the drift rather than
+# faithfully restoring corruption, and it costs nothing: a flag over a non-done job had no file to
+# serve (file_path is NULL until the worker finishes), so the old code 404'd on it anyway.
+#
+# What a rollback must NOT do is drop a flag for a keepsake that was genuinely downloadable
+# (flag set AND the job actually done). That is the assertion.
+lost=$(psql -tAc "
+  SELECT count(*) FROM _rehearsal_before b
+   WHERE NOT EXISTS (
+     SELECT 1 FROM event e
+      WHERE e.id = b.event_id
+        AND ((b.type = 'zip' AND e.export_zip_ready) OR (b.type = 'html' AND e.export_html_ready)))")
+if [[ "$lost" != "0" ]]; then
+  echo "✗ FAIL (down): $lost downloadable keepsake(s) lost their ready flag — the rollback is LOSSY."
+  psql -c "
+    SELECT b.event_id, b.type, 'was downloadable, flag not restored' AS problem
+      FROM _rehearsal_before b
+     WHERE NOT EXISTS (
+       SELECT 1 FROM event e
+        WHERE e.id = b.event_id
+          AND ((b.type = 'zip' AND e.export_zip_ready) OR (b.type = 'html' AND e.export_html_ready)))"
+  exit 1
+fi
+echo "✓ down: every downloadable keepsake kept its flag — the old image sees a working state."
+
+healed=$(psql -tAc "
+  SELECT count(*) FROM event e JOIN _rehearsal_flags f ON f.id = e.id
+   WHERE (e.export_zip_ready, e.export_html_ready) IS DISTINCT FROM (f.export_zip_ready, f.export_html_ready)")
+if [[ "$healed" != "0" ]]; then
+  echo "  ℹ $healed event(s) came back with a CLEARED flag that had drifted from its job row."
+  echo "    Not a loss — those had no file to serve. The rollback normalises them; the old code"
+  echo "    will rebuild on the next release. Listing them so the behaviour is not a surprise:"
+  psql -c "SELECT e.id, f.export_zip_ready AS was_zip, e.export_zip_ready AS now_zip,
+                  f.export_html_ready AS was_html, e.export_html_ready AS now_html
+             FROM event e JOIN _rehearsal_flags f ON f.id = e.id
+            WHERE (e.export_zip_ready, e.export_html_ready)
+                  IS DISTINCT FROM (f.export_zip_ready, f.export_html_ready)"
+fi
+
+echo "▸ Re-applying 014 (up) — roll forward after a rollback…"
+psql -q < "$MIG_DIR/014_export_epoch.up.sql"
+assert_up_preserves "up (after rollback)"
+
+echo
+echo "✓ Rehearsal passed. 014 is safe to deploy against this dataset."
+[[ -z "$DUMP" ]] && echo "  (synthetic data — re-run with a real pg_dump before you deploy for real)"
+exit 0

backend/src/auth/handlers.rs

@@ -1,8 +1,8 @@
 use std::time::Duration;
 
+use axum::Json;
 use axum::extract::State;
 use axum::http::{HeaderMap, StatusCode};
-use axum::Json;
 use chrono::Utc;
 use rand::Rng;
 use serde::{Deserialize, Serialize};
@@ -14,6 +14,7 @@ use crate::error::AppError;
 use crate::models::event::Event;
 use crate::models::session::Session;
 use crate::models::user::{User, UserRole};
+use crate::services::config;
 use crate::services::rate_limiter::client_ip;
 use crate::state::AppState;
 
@@ -36,7 +37,14 @@ pub async fn join(
     Json(body): Json<JoinRequest>,
 ) -> Result<(StatusCode, Json<JoinResponse>), AppError> {
     let ip = client_ip(&headers, "unknown");
-    if !state.rate_limiter.check(format!("join:{ip}"), 5, Duration::from_secs(60)) {
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    let join_rate_on = config::get_bool(&state.config_cache, "join_rate_enabled", true).await;
+    if rate_limits_on
+        && join_rate_on
+        && !state
+            .rate_limiter
+            .check(format!("join:{ip}"), 5, Duration::from_secs(60))
+    {
         return Err(AppError::TooManyRequests(
             "Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(),
             None,
@@ -44,11 +52,19 @@ pub async fn join(
     }
 
     let display_name = body.display_name.trim();
-    if display_name.is_empty() || display_name.len() > 50 {
+    let name_chars = display_name.chars().count();
+    if name_chars == 0 || name_chars > 50 {
         return Err(AppError::BadRequest(
             "Name muss zwischen 1 und 50 Zeichen lang sein.".into(),
         ));
     }
+    // Postgres rejects 0x00 in TEXT columns with a 500. Catch it here so callers
+    // see a clean 400 instead of an internal error.
+    if display_name.contains('\0') {
+        return Err(AppError::BadRequest(
+            "Name enthält ungültige Zeichen.".into(),
+        ));
+    }
 
     let event = Event::find_or_create(
         &state.pool,
@@ -67,10 +83,21 @@ pub async fn join(
 
     // Generate a 4-digit PIN
     let pin: String = format!("{:04}", rand::rng().random_range(0..10000u32));
-    let pin_hash =
-        bcrypt::hash(&pin, 12).map_err(|e| AppError::Internal(anyhow::anyhow!(e)))?;
+    let pin_hash = bcrypt::hash(&pin, 12).map_err(|e| AppError::Internal(anyhow::anyhow!(e)))?;
 
-    let user = User::create(&state.pool, event.id, display_name, &pin_hash).await?;
+    // The pre-check above is racy: two simultaneous joins with the same name can both
+    // pass it, and the DB's unique index then rejects the loser. Map that unique
+    // violation to the same clean 409 the pre-check returns, not a generic 500.
+    let user = match User::create(&state.pool, event.id, display_name, &pin_hash).await {
+        Ok(u) => u,
+        Err(sqlx::Error::Database(db)) if db.is_unique_violation() => {
+            return Err(AppError::Conflict(format!(
+                "Der Name \"{}\" ist bereits vergeben.",
+                display_name
+            )));
+        }
+        Err(e) => return Err(e.into()),
+    };
 
     let token = jwt::create_token(
         user.id,
@@ -108,27 +135,69 @@ pub struct RecoverResponse {
     pub user_id: Uuid,
 }
 
+/// A real cost-12 bcrypt hash of a fixed dummy value, computed once and cached. Used
+/// to run a constant-time-ish verify on the "unknown display name" branch of
+/// [`recover`], so that branch costs the same as a real (user-exists) verify and can't
+/// be told apart by timing.
+fn dummy_pin_hash() -> &'static str {
+    static HASH: std::sync::OnceLock<String> = std::sync::OnceLock::new();
+    HASH.get_or_init(|| {
+        bcrypt::hash("eventsnap-dummy-not-a-real-pin", 12)
+            .expect("bcrypt hashing of a static dummy value cannot fail")
+    })
+}
+
 pub async fn recover(
     State(state): State<AppState>,
+    headers: HeaderMap,
     Json(body): Json<RecoverRequest>,
 ) -> Result<Json<RecoverResponse>, AppError> {
     let display_name = body.display_name.trim();
 
+    // Per-IP+name throttle BEFORE the per-user 3-strike counter. Without this
+    // an attacker who knows a display name (they're visible on the feed) can
+    // burn through 3 wrong PINs and lock the victim for 15 minutes — repeated
+    // every 15 minutes, indefinitely. 5 attempts per 15 minutes per (IP, name)
+    // softens that into a real cost.
+    let ip = client_ip(&headers, "unknown");
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    let recover_rate_on = config::get_bool(&state.config_cache, "recover_rate_enabled", true).await;
+    if rate_limits_on && recover_rate_on {
+        let name_key = display_name.to_lowercase();
+        if !state.rate_limiter.check(
+            format!("recover:{ip}:{name_key}"),
+            5,
+            Duration::from_secs(15 * 60),
+        ) {
+            return Err(AppError::TooManyRequests(
+                "Zu viele Versuche. Bitte warte kurz und versuche es erneut.".into(),
+                None,
+            ));
+        }
+    }
+
     let event = Event::find_by_slug(&state.pool, &state.config.event_slug)
         .await?
         .ok_or_else(|| AppError::NotFound("Event nicht gefunden.".into()))?;
 
-    let users =
-        User::find_by_event_and_name(&state.pool, event.id, display_name).await?;
+    let users = User::find_by_event_and_name(&state.pool, event.id, display_name).await?;
 
     if users.is_empty() {
-        return Err(AppError::NotFound(
-            "Kein Benutzer mit diesem Namen gefunden.".into(),
-        ));
+        // No user with this name. Run a throwaway bcrypt verify so this branch takes
+        // the same time as the user-exists path, and return the SAME error as a wrong
+        // PIN — so "no such name" and "wrong PIN" are indistinguishable by response or
+        // timing. Display names are already public on the feed, but this still closes
+        // the /recover enumeration + timing oracle.
+        let _ = bcrypt::verify(&body.pin, dummy_pin_hash());
+        return Err(AppError::Unauthorized("PIN ist falsch.".into()));
     }
 
     for user in &users {
-        // Check PIN lockout
+        // Check PIN lockout. If the lockout has expired, also reset the failed-attempt
+        // counter so the user gets a fresh 3-strike window — otherwise the counter
+        // stays at 3+ and every subsequent wrong PIN immediately re-locks them, even
+        // after waiting out the cooldown. Without this reset, a once-locked account
+        // is effectively permanently fragile.
         if let Some(locked_until) = user.pin_locked_until {
             if Utc::now() < locked_until {
                 return Err(AppError::TooManyRequests(
@@ -136,10 +205,11 @@ pub async fn recover(
                     None,
                 ));
             }
+            // Lockout window expired — wipe the counter and the timestamp.
+            User::reset_pin_attempts(&state.pool, user.id).await?;
         }
 
-        let pin_matches = bcrypt::verify(&body.pin, &user.recovery_pin_hash)
-            .unwrap_or(false);
+        let pin_matches = bcrypt::verify(&body.pin, &user.recovery_pin_hash).unwrap_or(false);
 
         if pin_matches {
             // Reset failed attempts on success
@@ -166,9 +236,22 @@ pub async fn recover(
 
         // Wrong PIN — increment failure count
         let attempts = User::increment_failed_pin(&state.pool, user.id).await?;
+        tracing::warn!(
+            user_id = %user.id,
+            event_id = %event.id,
+            ip = %ip,
+            attempts,
+            "recover: wrong PIN"
+        );
         if attempts >= 3 {
             let lockout = Utc::now() + chrono::Duration::minutes(15);
             User::lock_pin(&state.pool, user.id, lockout).await?;
+            tracing::warn!(
+                user_id = %user.id,
+                event_id = %event.id,
+                ip = %ip,
+                "recover: account locked for 15 minutes"
+            );
         }
     }
 
@@ -183,10 +266,15 @@ pub struct AdminLoginRequest {
 #[derive(Serialize)]
 pub struct AdminLoginResponse {
     pub jwt: String,
+    /// The admin's user id + display name, so the client can populate a real identity
+    /// (own-post affordances, a name on the Account page) instead of a blank session.
+    pub user_id: Uuid,
+    pub display_name: String,
 }
 
 pub async fn admin_login(
     State(state): State<AppState>,
+    headers: HeaderMap,
     Json(body): Json<AdminLoginRequest>,
 ) -> Result<Json<AdminLoginResponse>, AppError> {
     if state.config.admin_password_hash.is_empty() {
@@ -195,10 +283,30 @@ pub async fn admin_login(
         ));
     }
 
-    let valid = bcrypt::verify(&body.password, &state.config.admin_password_hash)
-        .unwrap_or(false);
+    // Throttle password attempts. The admin password is bcrypt-hashed (slow to
+    // verify) but with no IP-level limit a determined attacker can still mount
+    // a long-running guess campaign. 5 attempts / minute / IP is plenty for
+    // honest typos.
+    let ip = client_ip(&headers, "unknown");
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    let admin_rate_on =
+        config::get_bool(&state.config_cache, "admin_login_rate_enabled", true).await;
+    if rate_limits_on
+        && admin_rate_on
+        && !state
+            .rate_limiter
+            .check(format!("admin_login:{ip}"), 5, Duration::from_secs(60))
+    {
+        return Err(AppError::TooManyRequests(
+            "Zu viele Anmeldeversuche. Bitte warte kurz und versuche es erneut.".into(),
+            None,
+        ));
+    }
+
+    let valid = bcrypt::verify(&body.password, &state.config.admin_password_hash).unwrap_or(false);
 
     if !valid {
+        tracing::warn!(ip = %ip, "admin_login: wrong password");
         return Err(AppError::Unauthorized("Falsches Passwort.".into()));
     }
 
@@ -215,9 +323,14 @@ pub async fn admin_login(
     let admin_user = if let Some(u) = users.into_iter().find(|u| u.role == UserRole::Admin) {
         u
     } else {
-        // Create admin user with a dummy PIN (admin authenticates via password)
-        let dummy_hash = bcrypt::hash("0000", 4)
-            .map_err(|e| AppError::Internal(anyhow::anyhow!(e)))?;
+        // Admin authenticates via password, but the schema still requires a PIN
+        // hash. Generate a random unguessable PIN so the recovery path remains
+        // unusable as an escalation route even if the role flag ever got cleared.
+        let dummy_pin: String = (0..32)
+            .map(|_| rand::rng().random_range(b'a'..=b'z') as char)
+            .collect();
+        let dummy_hash =
+            bcrypt::hash(&dummy_pin, 4).map_err(|e| AppError::Internal(anyhow::anyhow!(e)))?;
         let user = User::create(&state.pool, event.id, admin_name, &dummy_hash).await?;
         sqlx::query("UPDATE \"user\" SET role = 'admin' WHERE id = $1")
             .bind(user.id)
@@ -228,6 +341,8 @@ pub async fn admin_login(
             .ok_or_else(|| AppError::Internal(anyhow::anyhow!("admin user creation failed")))?
     };
 
+    tracing::info!(user_id = %admin_user.id, event_id = %event.id, ip = %ip, "admin_login: success");
+
     let token = jwt::create_token(
         admin_user.id,
         event.id,
@@ -241,13 +356,88 @@ pub async fn admin_login(
     let expires_at = Utc::now() + chrono::Duration::days(1);
     Session::create(&state.pool, admin_user.id, &token_hash, expires_at).await?;
 
-    Ok(Json(AdminLoginResponse { jwt: token }))
+    Ok(Json(AdminLoginResponse {
+        jwt: token,
+        user_id: admin_user.id,
+        display_name: admin_user.display_name,
+    }))
 }
 
-pub async fn logout(
-    State(state): State<AppState>,
-    auth: AuthUser,
-) -> Result<StatusCode, AppError> {
+pub async fn logout(State(state): State<AppState>, auth: AuthUser) -> Result<StatusCode, AppError> {
     Session::delete_by_token_hash(&state.pool, &auth.token_hash).await?;
     Ok(StatusCode::NO_CONTENT)
 }
+
+/// "Sign out everywhere" — revoke every session for the caller, not just the current one.
+/// A single leaked/kept-alive device (a shared phone, a laptop recovered via PIN) can then
+/// be cut from any of the user's devices.
+pub async fn logout_all(
+    State(state): State<AppState>,
+    auth: AuthUser,
+) -> Result<StatusCode, AppError> {
+    Session::delete_all_for_user(&state.pool, auth.user_id).await?;
+    Ok(StatusCode::NO_CONTENT)
+}
+
+#[derive(Deserialize)]
+pub struct PinResetRequestBody {
+    pub display_name: String,
+}
+
+/// A guest who forgot their PIN asks a host to reset it in-app. Unauthenticated (they
+/// can't log in without the PIN) and rate-limited. Always returns 204 regardless of
+/// whether the name exists, so it can't enumerate display names beyond what the public
+/// feed already exposes.
+pub async fn request_pin_reset(
+    State(state): State<AppState>,
+    headers: HeaderMap,
+    Json(body): Json<PinResetRequestBody>,
+) -> Result<StatusCode, AppError> {
+    let display_name = body.display_name.trim();
+    let ip = client_ip(&headers, "unknown");
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    if rate_limits_on {
+        let name_key = display_name.to_lowercase();
+        if !state.rate_limiter.check(
+            format!("pin_reset_req:{ip}:{name_key}"),
+            3,
+            Duration::from_secs(15 * 60),
+        ) {
+            return Err(AppError::TooManyRequests(
+                "Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(),
+                None,
+            ));
+        }
+    }
+    if display_name.is_empty() {
+        return Ok(StatusCode::NO_CONTENT);
+    }
+
+    // Single statement so the existing-name and unknown-name paths do IDENTICAL work
+    // (same event+user index scans, an INSERT that matches 0 rows for an unknown name) —
+    // no timing oracle despite the always-204 contract. Admins recover via password, so
+    // they're excluded from the join and never get a reset request queued.
+    let _ = sqlx::query(
+        "INSERT INTO pin_reset_request (event_id, user_id)
+         SELECT e.id, u.id
+         FROM event e
+         JOIN \"user\" u
+           ON u.event_id = e.id
+          AND lower(u.display_name) = lower($2)
+          AND u.role <> 'admin'
+         WHERE e.slug = $1
+         ON CONFLICT (user_id) DO NOTHING",
+    )
+    .bind(&state.config.event_slug)
+    .bind(display_name)
+    .execute(&state.pool)
+    .await;
+
+    // Broadcast unconditionally (carries no per-name info) so an online host's request
+    // badge updates without revealing whether the name existed.
+    let _ = state
+        .sse_tx
+        .send(crate::state::SseEvent::new("pin-reset-requested", "{}"));
+
+    Ok(StatusCode::NO_CONTENT)
+}

backend/src/auth/jwt.rs

@@ -13,6 +13,13 @@ pub struct Claims {
     pub role: UserRole,
     pub exp: i64,
     pub iat: i64,
+    /// Random per-token identifier. Without it, two `create_token` calls in the
+    /// same wall-clock second for the same (sub, role, event) produce identical
+    /// JWT bytes — and identical sha256(token) hashes — which then collide on
+    /// the `session.token_hash` UNIQUE constraint. The jti is ignored by the
+    /// verifier but breaks the collision.
+    #[serde(default)]
+    pub jti: Uuid,
 }
 
 pub fn create_token(
@@ -29,6 +36,7 @@ pub fn create_token(
         role,
         iat: now.timestamp(),
         exp: (now + Duration::days(expiry_days)).timestamp(),
+        jti: Uuid::new_v4(),
     };
     jsonwebtoken::encode(
         &Header::default(),
@@ -38,10 +46,20 @@ pub fn create_token(
 }
 
 pub fn verify_token(token: &str, secret: &str) -> Result<Claims, jsonwebtoken::errors::Error> {
+    // We deliberately do NOT enforce the JWT's own `exp`. The authoritative session
+    // lifetime lives in the `session` row (`expires_at > NOW()`), which SLIDES forward on
+    // every authenticated request (see `Session::touch_and_renew`). Enforcing the JWT's
+    // fixed +Nd `exp` here would hard-log-out an actively-used client on day N+1 even
+    // though its session was renewed — the "30-day cliff" from the review. With server-
+    // side sliding sessions, the token is a signature-checked bearer credential and its
+    // lifetime is revocable (logout / expiry / ban all delete the row), which is strictly
+    // stronger than a stateless non-revocable exp.
+    let mut validation = Validation::default();
+    validation.validate_exp = false;
     let data = jsonwebtoken::decode::<Claims>(
         token,
         &DecodingKey::from_secret(secret.as_bytes()),
-        &Validation::default(),
+        &validation,
     )?;
     Ok(data.claims)
 }

backend/src/auth/middleware.rs

@@ -1,4 +1,4 @@
-use axum::extract::{FromRequestParts, State};
+use axum::extract::FromRequestParts;
 use axum::http::request::Parts;
 use uuid::Uuid;
 
@@ -13,6 +13,10 @@ pub struct AuthUser {
     pub user_id: Uuid,
     pub event_id: Uuid,
     pub role: UserRole,
+    /// Live ban flag. Banned users keep *read* access (per USER_JOURNEYS §10), so
+    /// the base extractor does NOT reject them — write handlers and the
+    /// Require{Host,Admin} extractors enforce the ban instead.
+    pub is_banned: bool,
     pub token_hash: String,
 }
 
@@ -33,27 +37,51 @@ impl FromRequestParts<AppState> for AuthUser {
             .strip_prefix("Bearer ")
             .ok_or_else(|| AppError::Unauthorized("Ungültiges Token-Format.".into()))?;
 
-        let claims = jwt::verify_token(token, &state.config.jwt_secret)
+        // Verify the JWT's signature. Expiry is deliberately NOT enforced here (see
+        // `jwt::verify_token`) — the authoritative, sliding session lifetime lives in the
+        // `session` row read below. We also don't trust the token's role/ban claims; the
+        // live user row is authoritative, so the decoded claims aren't needed beyond this.
+        jwt::verify_token(token, &state.config.jwt_secret)
             .map_err(|_| AppError::Unauthorized("Token ungültig oder abgelaufen.".into()))?;
 
         let token_hash = jwt::hash_token(token);
 
-        let session = Session::find_by_token_hash(&state.pool, &token_hash)
+        // Single round-trip: resolve the session token to its *live* user row. A
+        // role/ban stored in the token would survive a demote/ban for the full session
+        // lifetime (up to 30d), so we always re-read the user (a demoted host loses host
+        // powers immediately). We do NOT reject banned users here — they retain read
+        // access by design; writes and host/admin actions enforce the ban downstream.
+        let user = Session::find_user_by_token_hash(&state.pool, &token_hash)
             .await
             .map_err(|e| AppError::Internal(e.into()))?
-            .ok_or_else(|| AppError::Unauthorized("Sitzung nicht gefunden oder abgelaufen.".into()))?;
+            .ok_or_else(|| {
+                AppError::Unauthorized("Sitzung nicht gefunden oder abgelaufen.".into())
+            })?;
 
-        // Update last_seen_at in the background (fire-and-forget)
+        // Touch last_seen_at AND slide the session's expiry forward (fire-and-forget), so
+        // an active client's session renews instead of hitting the fixed 30-day cliff.
+        // Admin sessions keep their tighter 1-day window (they renew on activity but still
+        // lapse a day after the admin goes idle). Failures are non-fatal but worth
+        // surfacing — silent swallowing hides DB connection pressure that would otherwise
+        // be the first symptom of a real problem.
         let pool = state.pool.clone();
-        let session_id = session.id;
+        let touch_hash = token_hash.clone();
+        let expiry_days = if user.role == UserRole::Admin {
+            1
+        } else {
+            state.config.session_expiry_days
+        };
         tokio::spawn(async move {
-            let _ = Session::touch(&pool, session_id).await;
+            if let Err(e) = Session::touch_and_renew(&pool, &touch_hash, expiry_days).await {
+                tracing::warn!(error = ?e, "session touch/renew failed");
+            }
         });
 
         Ok(Self {
-            user_id: claims.sub,
-            event_id: claims.event_id,
-            role: claims.role,
+            user_id: user.id,
+            event_id: user.event_id,
+            role: user.role,
+            is_banned: user.is_banned,
             token_hash,
         })
     }
@@ -70,6 +98,9 @@ impl FromRequestParts<AppState> for RequireHost {
         state: &AppState,
     ) -> Result<Self, Self::Rejection> {
         let auth = AuthUser::from_request_parts(parts, state).await?;
+        if auth.is_banned {
+            return Err(AppError::Forbidden("Du bist gesperrt.".into()));
+        }
         match auth.role {
             UserRole::Host | UserRole::Admin => Ok(Self(auth)),
             _ => Err(AppError::Forbidden("Nur für Hosts und Admins.".into())),
@@ -88,6 +119,9 @@ impl FromRequestParts<AppState> for RequireAdmin {
         state: &AppState,
     ) -> Result<Self, Self::Rejection> {
         let auth = AuthUser::from_request_parts(parts, state).await?;
+        if auth.is_banned {
+            return Err(AppError::Forbidden("Du bist gesperrt.".into()));
+        }
         match auth.role {
             UserRole::Admin => Ok(Self(auth)),
             _ => Err(AppError::Forbidden("Nur für Admins.".into())),

backend/src/config.rs

@@ -1,6 +1,51 @@
 use std::path::PathBuf;
 
-use anyhow::{Context, Result};
+use anyhow::{Context, Result, anyhow};
+
+/// Well-known dev JWT secret shipped in `.env.example`. If APP_ENV=production
+/// we refuse to start with this value; otherwise we warn loudly.
+const DEV_JWT_SECRET_SENTINEL: &str = "dev_secret_do_not_use_in_production_32byteslong_aaaa";
+
+/// A secret is "placeholder-ish" if it's the shipped dev sentinel or still carries
+/// the tell-tale scaffolding substrings from `.env.example`. Length alone is not
+/// enough — the shipped `change_me_...` placeholder is >32 chars.
+fn looks_placeholder(s: &str) -> bool {
+    let lower = s.to_ascii_lowercase();
+    s == DEV_JWT_SECRET_SENTINEL
+        || lower.contains("change_me")
+        || lower.contains("dev_secret")
+        || lower.contains("placeholder")
+}
+
+/// Enforce secret hygiene. In production every guard is hard-fail: a booting app
+/// with a publicly-known signing key is worse than one that refuses to start.
+/// Outside production the dev sentinel is tolerated (warned) so local dev is frictionless.
+fn validate_secrets(is_prod: bool, jwt_secret: &str, admin_password_hash: &str) -> Result<()> {
+    if is_prod {
+        if looks_placeholder(jwt_secret) {
+            return Err(anyhow!(
+                "Refusing to start in production with a placeholder JWT_SECRET — \
+                 rotate it (openssl rand -hex 64)."
+            ));
+        }
+        if jwt_secret.len() < 32 {
+            return Err(anyhow!("JWT_SECRET must be at least 32 characters."));
+        }
+        if admin_password_hash.is_empty() || looks_placeholder(admin_password_hash) {
+            return Err(anyhow!(
+                "Refusing to start in production without a real ADMIN_PASSWORD_HASH — \
+                 generate one (htpasswd -bnBC 12 '' <password> | tr -d ':\\n')."
+            ));
+        }
+    } else if jwt_secret == DEV_JWT_SECRET_SENTINEL {
+        tracing::warn!(
+            "JWT_SECRET is the dev sentinel — fine for local development, NEVER ship this."
+        );
+    } else if jwt_secret.len() < 32 {
+        return Err(anyhow!("JWT_SECRET must be at least 32 characters."));
+    }
+    Ok(())
+}
 
 #[derive(Clone, Debug)]
 pub struct AppConfig {
@@ -11,33 +56,148 @@ pub struct AppConfig {
     pub event_name: String,
     pub event_slug: String,
     pub media_path: PathBuf,
+    /// Where export archives are written. MUST be outside `media_path` — that
+    /// directory is served by a public `ServeDir`, so a predictable archive name
+    /// under it would leak the whole gallery to anonymous visitors.
+    pub export_path: PathBuf,
     pub app_port: u16,
+    /// Number of concurrent media compression workers (read once at boot).
+    pub compression_concurrency: usize,
+    /// Master switch for the comment feature (env `COMMENTS_ENABLED`, default true).
+    /// When false the backend rejects new comments and the frontend hides the whole
+    /// comment UI. Existing comments stay in the DB (hidden), so flipping it back
+    /// restores them. Boot-time immutable, like `compression_concurrency`.
+    pub comments_enabled: bool,
+    /// Default colour theme, used as the fallback when the DB config keys are unset.
+    /// Runtime overrides live in the `config` table (admin UI); these env vars only
+    /// seed the initial default. `preset` is an id the frontend knows (e.g.
+    /// "champagne-gold", "rose", … or "custom"); the two seeds are `#rrggbb` brand +
+    /// accent colours the whole palette is derived from.
+    pub default_theme_preset: String,
+    pub default_theme_primary: String,
+    pub default_theme_accent: String,
 }
 
+/// The shipped default brand/accent seed (champagne gold — matches the hand-tuned
+/// ramp in tailwind-theme.css). Kept here so an unset env still yields the current look.
+const DEFAULT_THEME_SEED: &str = "#8a6a2b";
+
 impl AppConfig {
     pub fn from_env() -> Result<Self> {
+        let app_env = std::env::var("APP_ENV").unwrap_or_else(|_| "development".to_string());
+        let is_prod = app_env.eq_ignore_ascii_case("production");
+
+        let jwt_secret = std::env::var("JWT_SECRET").context("JWT_SECRET must be set")?;
+        let admin_password_hash = std::env::var("ADMIN_PASSWORD_HASH").unwrap_or_default();
+
+        validate_secrets(is_prod, &jwt_secret, &admin_password_hash)?;
+
         Ok(Self {
-            database_url: std::env::var("DATABASE_URL")
-                .context("DATABASE_URL must be set")?,
-            jwt_secret: std::env::var("JWT_SECRET")
-                .context("JWT_SECRET must be set")?,
+            database_url: std::env::var("DATABASE_URL").context("DATABASE_URL must be set")?,
+            jwt_secret,
             session_expiry_days: std::env::var("SESSION_EXPIRY_DAYS")
                 .unwrap_or_else(|_| "30".to_string())
                 .parse()
                 .context("SESSION_EXPIRY_DAYS must be a number")?,
-            admin_password_hash: std::env::var("ADMIN_PASSWORD_HASH")
-                .unwrap_or_default(),
-            event_name: std::env::var("EVENT_NAME")
-                .unwrap_or_else(|_| "EventSnap".to_string()),
-            event_slug: std::env::var("EVENT_SLUG")
-                .context("EVENT_SLUG must be set")?,
+            admin_password_hash,
+            event_name: std::env::var("EVENT_NAME").unwrap_or_else(|_| "EventSnap".to_string()),
+            event_slug: std::env::var("EVENT_SLUG").context("EVENT_SLUG must be set")?,
             media_path: PathBuf::from(
                 std::env::var("MEDIA_PATH").unwrap_or_else(|_| "/media".to_string()),
             ),
+            export_path: PathBuf::from(
+                std::env::var("EXPORT_PATH").unwrap_or_else(|_| "/exports".to_string()),
+            ),
             app_port: std::env::var("APP_PORT")
                 .unwrap_or_else(|_| "3000".to_string())
                 .parse()
                 .context("APP_PORT must be a number")?,
+            compression_concurrency: std::env::var("COMPRESSION_WORKER_CONCURRENCY")
+                .ok()
+                .and_then(|v| v.parse().ok())
+                .filter(|&n| n >= 1)
+                .unwrap_or(2),
+            comments_enabled: std::env::var("COMMENTS_ENABLED")
+                .map(|v| {
+                    !matches!(
+                        v.trim().to_ascii_lowercase().as_str(),
+                        "false" | "0" | "no" | "off"
+                    )
+                })
+                .unwrap_or(true),
+            default_theme_preset: std::env::var("THEME_PRESET")
+                .unwrap_or_else(|_| "champagne-gold".to_string()),
+            default_theme_primary: std::env::var("THEME_PRIMARY")
+                .unwrap_or_else(|_| DEFAULT_THEME_SEED.to_string()),
+            default_theme_accent: std::env::var("THEME_ACCENT")
+                .unwrap_or_else(|_| DEFAULT_THEME_SEED.to_string()),
         })
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    const REAL_SECRET: &str = "a1b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6a7b8c9d0e1f2a3b4c5d6e7f8a9b0c1d2";
+    const REAL_HASH: &str = "$2y$12$abcdefghijklmnopqrstuv.wxyzABCDEFGHIJKLMNOPQRSTUVWXYZ012";
+
+    #[test]
+    fn prod_rejects_shipped_placeholder_secret() {
+        // The exact string shipped in `.env` — >32 chars, so it must be caught by
+        // the substring guard, not the length check.
+        let err = validate_secrets(true, "change_me_to_a_random_64_byte_hex_string", REAL_HASH);
+        assert!(
+            err.is_err(),
+            "placeholder JWT_SECRET must be rejected in prod"
+        );
+    }
+
+    #[test]
+    fn prod_rejects_dev_sentinel_and_short_secret() {
+        assert!(validate_secrets(true, DEV_JWT_SECRET_SENTINEL, REAL_HASH).is_err());
+        assert!(validate_secrets(true, "tooshort", REAL_HASH).is_err());
+    }
+
+    #[test]
+    fn prod_rejects_missing_or_placeholder_admin_hash() {
+        assert!(validate_secrets(true, REAL_SECRET, "").is_err());
+        assert!(validate_secrets(true, REAL_SECRET, "$2y$12$placeholder_replace_me").is_err());
+    }
+
+    #[test]
+    fn prod_accepts_real_secrets() {
+        assert!(validate_secrets(true, REAL_SECRET, REAL_HASH).is_ok());
+    }
+
+    #[test]
+    fn non_prod_tolerates_dev_sentinel() {
+        assert!(validate_secrets(false, DEV_JWT_SECRET_SENTINEL, "").is_ok());
+    }
+
+    #[test]
+    fn non_prod_still_rejects_short_non_sentinel_secret() {
+        assert!(validate_secrets(false, "tooshort", "").is_err());
+    }
+
+    #[test]
+    fn placeholder_detection_is_case_insensitive() {
+        // looks_placeholder lowercases before matching — an upper/mixed-case
+        // placeholder must still be rejected in prod.
+        assert!(
+            validate_secrets(true, "CHANGE_ME_TO_A_RANDOM_64_BYTE_HEX_STRING", REAL_HASH).is_err()
+        );
+        assert!(validate_secrets(true, REAL_SECRET, "$2Y$12$PLACEHOLDER_replace_me").is_err());
+    }
+
+    #[test]
+    fn prod_len_boundary_at_32() {
+        // Exactly 32 non-placeholder chars is the minimum accepted; 31 is rejected.
+        const LEN_32: &str = "abcdefghijklmnopqrstuvwxyz012345";
+        const LEN_31: &str = "abcdefghijklmnopqrstuvwxyz01234";
+        assert_eq!(LEN_32.len(), 32);
+        assert_eq!(LEN_31.len(), 31);
+        assert!(validate_secrets(true, LEN_32, REAL_HASH).is_ok());
+        assert!(validate_secrets(true, LEN_31, REAL_HASH).is_err());
+    }
+}

backend/src/db.rs

@@ -1,10 +1,17 @@
 use anyhow::{Context, Result};
-use sqlx::postgres::PgPoolOptions;
 use sqlx::PgPool;
+use sqlx::postgres::PgPoolOptions;
+
+const DEFAULT_MAX_CONNECTIONS: u32 = 10;
 
 pub async fn create_pool(database_url: &str) -> Result<PgPool> {
+    let max_connections = std::env::var("DATABASE_MAX_CONNECTIONS")
+        .ok()
+        .and_then(|s| s.parse::<u32>().ok())
+        .unwrap_or(DEFAULT_MAX_CONNECTIONS);
+
     let pool = PgPoolOptions::new()
-        .max_connections(10)
+        .max_connections(max_connections)
         .connect(database_url)
         .await
         .context("failed to connect to database")?;
@@ -14,6 +21,6 @@ pub async fn create_pool(database_url: &str) -> Result<PgPool> {
         .await
         .context("failed to run database migrations")?;
 
-    tracing::info!("database connected and migrations applied");
+    tracing::info!(max_connections, "database connected and migrations applied");
     Ok(pool)
 }

backend/src/error.rs

@@ -6,10 +6,19 @@ pub enum AppError {
     BadRequest(String),
     Unauthorized(String),
     Forbidden(String),
+    /// Uploads are temporarily locked (event closed / gallery released). Distinct from
+    /// `Forbidden` so the client can tell this REVERSIBLE 403 apart from a permanent one
+    /// (banned user, quota): the queued blob is kept and retried if the host reopens,
+    /// instead of being purged like a genuinely-terminal rejection.
+    UploadsLocked(String),
     NotFound(String),
     Conflict(String),
     /// Second field: optional retry-after seconds to include in the response.
     TooManyRequests(String, Option<u64>),
+    /// Per-user storage quota exhausted. Distinct from `TooManyRequests` (rate limit) so
+    /// the client can treat it as *terminal* (413, no retry) instead of backing off and
+    /// retrying a permanently-failing upload forever.
+    QuotaExceeded(String),
     Internal(anyhow::Error),
 }
 
@@ -19,9 +28,11 @@ impl AppError {
             Self::BadRequest(_) => (StatusCode::BAD_REQUEST, "bad_request"),
             Self::Unauthorized(_) => (StatusCode::UNAUTHORIZED, "unauthorized"),
             Self::Forbidden(_) => (StatusCode::FORBIDDEN, "forbidden"),
+            Self::UploadsLocked(_) => (StatusCode::FORBIDDEN, "uploads_locked"),
             Self::NotFound(_) => (StatusCode::NOT_FOUND, "not_found"),
             Self::Conflict(_) => (StatusCode::CONFLICT, "conflict"),
             Self::TooManyRequests(..) => (StatusCode::TOO_MANY_REQUESTS, "too_many_requests"),
+            Self::QuotaExceeded(_) => (StatusCode::PAYLOAD_TOO_LARGE, "quota_exceeded"),
             Self::Internal(_) => (StatusCode::INTERNAL_SERVER_ERROR, "internal_error"),
         }
     }
@@ -31,9 +42,11 @@ impl AppError {
             Self::BadRequest(msg)
             | Self::Unauthorized(msg)
             | Self::Forbidden(msg)
+            | Self::UploadsLocked(msg)
             | Self::NotFound(msg)
             | Self::Conflict(msg) => msg.clone(),
             Self::TooManyRequests(msg, _) => msg.clone(),
+            Self::QuotaExceeded(msg) => msg.clone(),
             Self::Internal(err) => {
                 tracing::error!("internal error: {err:#}");
                 "Ein interner Fehler ist aufgetreten.".to_string()
@@ -62,10 +75,11 @@ impl IntoResponse for AppError {
         }
 
         let mut resp = (status, axum::Json(body)).into_response();
-        if let Some(secs) = retry_after_secs {
-            if let Ok(val) = axum::http::HeaderValue::from_str(&secs.to_string()) {
-                resp.headers_mut().insert(axum::http::header::RETRY_AFTER, val);
-            }
+        if let Some(secs) = retry_after_secs
+            && let Ok(val) = axum::http::HeaderValue::from_str(&secs.to_string())
+        {
+            resp.headers_mut()
+                .insert(axum::http::header::RETRY_AFTER, val);
         }
         resp
     }

backend/src/handlers/admin.rs

@@ -1,14 +1,14 @@
 use std::collections::HashMap;
 use std::time::Duration;
 
-use axum::extract::State;
-use axum::http::{HeaderMap, StatusCode};
 use axum::Json;
+use axum::extract::{Query, State};
+use axum::http::{HeaderMap, StatusCode};
 use serde::{Deserialize, Serialize};
-use sysinfo::System;
 
 use crate::auth::middleware::RequireAdmin;
 use crate::error::AppError;
+use crate::services::config;
 use crate::services::rate_limiter::client_ip;
 use crate::state::AppState;
 
@@ -45,19 +45,17 @@ pub async fn get_stats(
         .await?
         .ok_or_else(|| AppError::NotFound("Event nicht gefunden.".into()))?;
 
-    let (user_count,): (i64,) =
-        sqlx::query_as("SELECT COUNT(*) FROM \"user\" WHERE event_id = $1")
+    let (user_count,): (i64,) = sqlx::query_as("SELECT COUNT(*) FROM \"user\" WHERE event_id = $1")
+        .bind(event.id)
+        .fetch_one(&state.pool)
+        .await?;
+
+    let (upload_count,): (i64,) =
+        sqlx::query_as("SELECT COUNT(*) FROM upload WHERE event_id = $1 AND deleted_at IS NULL")
             .bind(event.id)
             .fetch_one(&state.pool)
             .await?;
 
-    let (upload_count,): (i64,) = sqlx::query_as(
-        "SELECT COUNT(*) FROM upload WHERE event_id = $1 AND deleted_at IS NULL",
-    )
-    .bind(event.id)
-    .fetch_one(&state.pool)
-    .await?;
-
     let (comment_count,): (i64,) = sqlx::query_as(
         "SELECT COUNT(*) FROM comment c
          JOIN upload u ON u.id = c.upload_id
@@ -67,23 +65,12 @@ pub async fn get_stats(
     .fetch_one(&state.pool)
     .await?;
 
-    // Disk usage via sysinfo
-    let mut sys = System::new();
-    sys.refresh_all();
-
-    let media_path = state.config.media_path.to_string_lossy().to_string();
-    let (disk_total, disk_free) = sysinfo::Disks::new_with_refreshed_list()
-        .iter()
-        .find(|d| media_path.starts_with(d.mount_point().to_string_lossy().as_ref()))
-        .map(|d| (d.total_space(), d.available_space()))
-        .unwrap_or_else(|| {
-            // Fall back to the root disk
-            sysinfo::Disks::new_with_refreshed_list()
-                .iter()
-                .find(|d| d.mount_point().to_string_lossy() == "/")
-                .map(|d| (d.total_space(), d.available_space()))
-                .unwrap_or((0, 0))
-        });
+    // Disk usage from the shared cache (unknown mount → zeros, same as before).
+    let (disk_total, disk_free) = state
+        .disk_cache
+        .snapshot(&state.config.media_path)
+        .map(|d| (d.total, d.free))
+        .unwrap_or((0, 0));
 
     let disk_used = disk_total.saturating_sub(disk_free);
 
@@ -101,14 +88,17 @@ pub async fn get_config(
     State(state): State<AppState>,
     RequireAdmin(_auth): RequireAdmin,
 ) -> Result<Json<HashMap<String, String>>, AppError> {
-    let rows: Vec<(String, String)> =
-        sqlx::query_as("SELECT key, value FROM config ORDER BY key")
-            .fetch_all(&state.pool)
-            .await?;
+    let rows: Vec<(String, String)> = sqlx::query_as("SELECT key, value FROM config ORDER BY key")
+        .fetch_all(&state.pool)
+        .await?;
 
     Ok(Json(rows.into_iter().collect()))
 }
 
+/// Documents the wire shape of `PATCH /admin/config` (a flat `{key: value}` object).
+/// `patch_config` extracts the `HashMap` directly rather than going through this newtype, so it is
+/// never constructed in Rust — it stays as the serde-derived description of the request body.
+#[allow(dead_code)]
 #[derive(Deserialize)]
 pub struct PatchConfigRequest(pub HashMap<String, String>);
 
@@ -117,38 +107,172 @@ pub async fn patch_config(
     RequireAdmin(_auth): RequireAdmin,
     Json(body): Json<HashMap<String, String>>,
 ) -> Result<StatusCode, AppError> {
-    const ALLOWED_KEYS: &[&str] = &[
-        "max_image_size_mb",
-        "max_video_size_mb",
-        "upload_rate_per_hour",
-        "feed_rate_per_min",
-        "export_rate_per_day",
-        "quota_tolerance",
-        "estimated_guest_count",
-        "compression_concurrency",
+    // Numeric keys validated as f64; boolean keys validated as truthy strings; the
+    // privacy note is free text. Splitting these explicitly is verbose but makes the
+    // failure mode for typos obvious (`Unbekannter Schlüssel: ...`).
+    // (key, integer_only, min, max). Ranges reject values that `parse::<f64>` would
+    // accept but that silently revert to the hardcoded default at read time
+    // (get_usize/get_i64 can't parse negatives/NaN/fractionals). `compression_concurrency`
+    // is intentionally absent — it's read once at boot, so a live edit was a no-op.
+    const NUMERIC_SPECS: &[(&str, bool, f64, f64)] = &[
+        ("max_image_size_mb", true, 1.0, 1024.0),
+        ("max_video_size_mb", true, 1.0, 10240.0),
+        ("upload_rate_per_hour", true, 1.0, 100_000.0),
+        ("feed_rate_per_min", true, 1.0, 100_000.0),
+        ("export_rate_per_day", true, 1.0, 100_000.0),
+        ("quota_tolerance", false, 0.0, 1.0),
+        ("estimated_guest_count", true, 1.0, 1_000_000.0),
+    ];
+    const BOOL_KEYS: &[&str] = &[
+        "rate_limits_enabled",
+        "upload_rate_enabled",
+        "feed_rate_enabled",
+        "export_rate_enabled",
+        "join_rate_enabled",
+        // These two per-area rate toggles are HONOURED by their handlers (auth/handlers.rs reads
+        // `admin_login_rate_enabled` and `recover_rate_enabled`, both defaulting true) but were
+        // missing from this allowlist — so the switch existed in code and could never be flipped.
+        "admin_login_rate_enabled",
+        "recover_rate_enabled",
+        "quota_enabled",
+        "storage_quota_enabled",
+        "upload_count_quota_enabled",
+    ];
+    const TEXT_KEYS: &[&str] = &[
+        "privacy_note",
+        "theme_preset",
+        "theme_primary",
+        "theme_accent",
+    ];
+    const PRIVACY_NOTE_MAX_LEN: usize = 16 * 1024; // 16 KiB free text is plenty
+    // Preset ids the frontend knows how to render (mirror of PRESETS in
+    // frontend/src/lib/theme/palette.ts). "custom" means "use the theme_primary/accent
+    // seeds verbatim". Kept in sync by hand — a new preset must be added in both places.
+    const THEME_PRESETS: &[&str] = &[
+        "champagne-gold",
+        "rose",
+        "sage",
+        "dusk-blue",
+        "classic-silver",
+        "custom",
     ];
 
+    let mut privacy_note_changed = false;
+    let mut theme_changed = false;
+
+    // Validate every key first so a bad value in the batch can't leave a partial
+    // update behind — validation must fully precede any write.
     for (key, value) in &body {
-        if !ALLOWED_KEYS.contains(&key.as_str()) {
-            return Err(AppError::BadRequest(format!("Unbekannter Konfigurationsschlüssel: {key}")));
-        }
-        // Validate numeric values
-        if value.parse::<f64>().is_err() {
-            return Err(AppError::BadRequest(format!("Ungültiger Wert für {key}: muss eine Zahl sein.")));
+        let key_str = key.as_str();
+        if let Some(&(_, integer_only, min, max)) =
+            NUMERIC_SPECS.iter().find(|(k, ..)| *k == key_str)
+        {
+            let n = value.trim().parse::<f64>().ok().filter(|n| n.is_finite());
+            let n = match n {
+                Some(n) => n,
+                None => {
+                    return Err(AppError::BadRequest(format!(
+                        "Ungültiger Wert für {key}: muss eine Zahl sein."
+                    )));
+                }
+            };
+            if integer_only && n.fract() != 0.0 {
+                return Err(AppError::BadRequest(format!(
+                    "Ungültiger Wert für {key}: muss eine ganze Zahl sein."
+                )));
+            }
+            if n < min || n > max {
+                return Err(AppError::BadRequest(format!(
+                    "Wert für {key} liegt außerhalb des zulässigen Bereichs ({min}–{max})."
+                )));
+            }
+        } else if BOOL_KEYS.contains(&key_str) {
+            match value.trim().to_ascii_lowercase().as_str() {
+                "true" | "false" | "1" | "0" | "yes" | "no" | "on" | "off" => {}
+                _ => {
+                    return Err(AppError::BadRequest(format!(
+                        "Ungültiger Wert für {key}: muss true oder false sein."
+                    )));
+                }
+            }
+        } else if TEXT_KEYS.contains(&key_str) {
+            // Count characters, not bytes — the message says "Zeichen" and a
+            // multi-byte grapheme shouldn't count against the limit multiple times.
+            if value.chars().count() > PRIVACY_NOTE_MAX_LEN {
+                return Err(AppError::BadRequest(format!(
+                    "Wert für {key} ist zu lang (max. {PRIVACY_NOTE_MAX_LEN} Zeichen)."
+                )));
+            }
+            match key_str {
+                "privacy_note" => privacy_note_changed = true,
+                "theme_preset" => {
+                    if !THEME_PRESETS.contains(&value.trim()) {
+                        return Err(AppError::BadRequest(format!("Ungültiges Theme: {value}.")));
+                    }
+                    theme_changed = true;
+                }
+                "theme_primary" | "theme_accent" => {
+                    if !is_hex_color(value.trim()) {
+                        return Err(AppError::BadRequest(format!(
+                            "Ungültige Farbe für {key}: muss #rrggbb sein."
+                        )));
+                    }
+                    theme_changed = true;
+                }
+                _ => {}
+            }
+        } else {
+            return Err(AppError::BadRequest(format!(
+                "Unbekannter Konfigurationsschlüssel: {key}"
+            )));
         }
+    }
+
+    // Apply all writes in one transaction — the batch is all-or-nothing.
+    let mut tx = state.pool.begin().await?;
+    for (key, value) in &body {
         sqlx::query(
             "INSERT INTO config (key, value, updated_at) VALUES ($1, $2, NOW())
              ON CONFLICT (key) DO UPDATE SET value = EXCLUDED.value, updated_at = NOW()",
         )
         .bind(key)
         .bind(value)
-        .execute(&state.pool)
+        .execute(&mut *tx)
         .await?;
     }
+    tx.commit().await?;
+
+    // The config cache must reflect this write on the very next read (tests PATCH then
+    // immediately assert the new value takes effect). Invalidate synchronously here —
+    // the TTL is only a backstop and must not be relied on for correctness.
+    state.config_cache.invalidate();
+
+    // Notify all clients that a publicly-readable config value changed so their stores
+    // (e.g. the privacy note in My Account) refresh without a manual reload.
+    if privacy_note_changed || theme_changed {
+        let mut keys: Vec<&str> = Vec::new();
+        if privacy_note_changed {
+            keys.push("privacy_note");
+        }
+        if theme_changed {
+            keys.push("theme");
+        }
+        let _ = state.sse_tx.send(crate::state::SseEvent::new(
+            "event-updated",
+            serde_json::json!({ "keys": keys }).to_string(),
+        ));
+    }
 
     Ok(StatusCode::NO_CONTENT)
 }
 
+/// A strict `#rrggbb` hex-colour check (6 hex digits, leading `#`). Deliberately not
+/// accepting shorthand/`#rgba` so the value is safe to drop straight into CSS.
+fn is_hex_color(s: &str) -> bool {
+    let bytes = s.as_bytes();
+    bytes.len() == 7 && bytes[0] == b'#' && bytes[1..].iter().all(|b| b.is_ascii_hexdigit())
+}
+
 pub async fn get_export_jobs(
     State(state): State<AppState>,
     RequireAdmin(_auth): RequireAdmin,
@@ -172,61 +296,106 @@ pub async fn get_export_jobs(
 
 // ── Export download endpoints (authenticated guests) ─────────────────────────
 
+#[derive(Deserialize)]
+pub struct DownloadQuery {
+    pub ticket: String,
+}
+
+/// Mint a short-lived ticket for a browser-driven export download. The download
+/// is a top-level navigation so the multi-GB ZIP streams straight to disk instead
+/// of being buffered in memory by `fetch()` + `blob()` — but a navigation can't
+/// carry an `Authorization` header, so the client exchanges its Bearer token for
+/// an opaque ticket here, then hits `/export/zip?ticket=...`. Reuses the same
+/// single-use, 30s-TTL store as the SSE stream.
+pub async fn export_ticket(
+    State(state): State<AppState>,
+    auth: crate::auth::middleware::AuthUser,
+) -> Json<serde_json::Value> {
+    // NOTE: intentionally NOT gated on `is_banned`. A banned user keeps *read* access
+    // by design (USER_JOURNEYS §10.3, FEATURES: "Can still download the export once
+    // released — Spec design choice"). The export is read-only, so it stays available
+    // to them, consistent with the read-only-ban model.
+    let ticket = state.sse_tickets.issue(auth.token_hash);
+    Json(serde_json::json!({ "ticket": ticket }))
+}
+
+/// Validate a download ticket (single-use) and confirm its session still exists.
+async fn authenticate_download_ticket(state: &AppState, ticket: &str) -> Result<(), AppError> {
+    let token_hash = state
+        .sse_tickets
+        .consume(ticket)
+        .ok_or_else(|| AppError::Unauthorized("Ticket ungültig oder abgelaufen.".into()))?;
+    crate::models::session::Session::find_by_token_hash(&state.pool, &token_hash)
+        .await
+        .map_err(|e| AppError::Internal(e.into()))?
+        .ok_or_else(|| AppError::Unauthorized("Sitzung nicht gefunden.".into()))?;
+    Ok(())
+}
+
 pub async fn download_zip(
     State(state): State<AppState>,
-    _auth: crate::auth::middleware::AuthUser,
+    Query(q): Query<DownloadQuery>,
     headers: HeaderMap,
 ) -> Result<axum::response::Response, AppError> {
-    let ip = client_ip(&headers, "unknown");
-    let limit = get_config_usize(&state.pool, "export_rate_per_day", 3).await;
-    if !state.rate_limiter.check(format!("export:{ip}"), limit, Duration::from_secs(86400)) {
-        return Err(AppError::TooManyRequests("Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(), None));
-    }
+    authenticate_download_ticket(&state, &q.ticket).await?;
+    enforce_export_rate(&state, &headers).await?;
 
-    let event = crate::models::event::Event::find_by_slug(&state.pool, &state.config.event_slug)
-        .await?
-        .ok_or_else(|| AppError::NotFound("Event nicht gefunden.".into()))?;
+    let path =
+        resolve_export_file(&state, "zip", "Der ZIP-Export ist noch nicht verfügbar.").await?;
+    serve_file(path, "Gallery.zip", "application/zip").await
+}
 
-    if !event.export_zip_ready {
-        return Err(AppError::NotFound(
-            "Der ZIP-Export ist noch nicht verfügbar.".into(),
-        ));
-    }
+/// Resolve the on-disk path of the CURRENT export generation — readiness check and path lookup in
+/// ONE read, through the `export_current` view (migration 014).
+///
+/// This used to be two statements: the caller checked `event.export_zip_ready`, then this fetched
+/// `export_job.file_path`. A reopen landing between them served a keepsake that should have 404'd —
+/// the same stale-keepsake class, leaking through the read path, and it existed only because
+/// readiness was a stored copy rather than a derivation. `export_current` gives both facts from one
+/// consistent snapshot: it yields a row only when the event is released AND the job is `done` at the
+/// event's current epoch, so "is it ready" and "which file" can no longer disagree.
+async fn resolve_export_file(
+    state: &AppState,
+    export_type: &str,
+    not_ready_msg: &str,
+) -> Result<std::path::PathBuf, AppError> {
+    let file_path: Option<(Option<String>,)> = sqlx::query_as(
+        "SELECT c.file_path FROM export_current c
+         JOIN event e ON e.id = c.event_id
+         WHERE e.slug = $1 AND c.type = $2::export_type AND c.status = 'done'",
+    )
+    .bind(&state.config.event_slug)
+    .bind(export_type)
+    .fetch_optional(&state.pool)
+    .await
+    .map_err(|e| AppError::Internal(e.into()))?;
 
-    let path = state.config.media_path.join("exports").join("Gallery.zip");
+    let Some((Some(rel),)) = file_path else {
+        return Err(AppError::NotFound(not_ready_msg.into()));
+    };
+
+    // `file_path` is stored as `exports/<name>`; the base dir is already `export_path`, so
+    // join only the file name (defends against any absolute/`..` content too).
+    let name = std::path::Path::new(&rel)
+        .file_name()
+        .ok_or_else(|| AppError::NotFound("Exportdatei nicht gefunden.".into()))?;
+    let path = state.config.export_path.join(name);
     if !path.exists() {
         return Err(AppError::NotFound("Exportdatei nicht gefunden.".into()));
     }
-
-    serve_file(path, "Gallery.zip", "application/zip").await
+    Ok(path)
 }
 
 pub async fn download_html(
     State(state): State<AppState>,
-    _auth: crate::auth::middleware::AuthUser,
+    Query(q): Query<DownloadQuery>,
     headers: HeaderMap,
 ) -> Result<axum::response::Response, AppError> {
-    let ip = client_ip(&headers, "unknown");
-    let limit = get_config_usize(&state.pool, "export_rate_per_day", 3).await;
-    if !state.rate_limiter.check(format!("export:{ip}"), limit, Duration::from_secs(86400)) {
-        return Err(AppError::TooManyRequests("Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(), None));
-    }
-
-    let event = crate::models::event::Event::find_by_slug(&state.pool, &state.config.event_slug)
-        .await?
-        .ok_or_else(|| AppError::NotFound("Event nicht gefunden.".into()))?;
-
-    if !event.export_html_ready {
-        return Err(AppError::NotFound(
-            "Der HTML-Export ist noch nicht verfügbar.".into(),
-        ));
-    }
-
-    let path = state.config.media_path.join("exports").join("Memories.zip");
-    if !path.exists() {
-        return Err(AppError::NotFound("Exportdatei nicht gefunden.".into()));
-    }
+    authenticate_download_ticket(&state, &q.ticket).await?;
+    enforce_export_rate(&state, &headers).await?;
 
+    let path =
+        resolve_export_file(&state, "html", "Der HTML-Export ist noch nicht verfügbar.").await?;
     serve_file(path, "Memories.zip", "application/zip").await
 }
 
@@ -236,7 +405,7 @@ async fn serve_file(
     content_type: &str,
 ) -> Result<axum::response::Response, AppError> {
     use axum::body::Body;
-    use axum::http::{header, Response, StatusCode};
+    use axum::http::{Response, StatusCode, header};
     use tokio_util::io::ReaderStream;
 
     let file = tokio::fs::File::open(&path)
@@ -272,8 +441,19 @@ pub async fn export_status(
 
     let released = event.export_released_at.is_some();
 
+    // ONE statement: the epoch comparison happens inside the query, against a single snapshot.
+    // Binding the epoch read by a previous statement would let a release/regeneration commit in
+    // between and yield `{released: true, zip: locked, html: locked}` — a state that never existed.
+    //
+    // Only jobs at the CURRENT epoch are reported. A row left behind by a retired generation (a
+    // worker superseded mid-run) is meaningless — surfacing its frozen `running`/77% would show a
+    // progress bar that never moves for a keepsake nobody is building. It reads as "locked" (no
+    // current job), which is exactly what it is.
     let jobs: Vec<(String, String, i16)> = sqlx::query_as(
-        "SELECT type::text, status::text, progress_pct FROM export_job WHERE event_id = $1",
+        "SELECT j.type::text, j.status::text, j.progress_pct
+           FROM export_job j
+           JOIN event e ON e.id = j.event_id
+          WHERE e.id = $1 AND j.epoch = e.export_epoch",
     )
     .bind(event.id)
     .fetch_all(&state.pool)
@@ -282,9 +462,7 @@ pub async fn export_status(
     let job_status = |type_name: &str| {
         jobs.iter()
             .find(|(t, _, _)| t == type_name)
-            .map(|(_, status, pct)| {
-                serde_json::json!({ "status": status, "progress_pct": pct })
-            })
+            .map(|(_, status, pct)| serde_json::json!({ "status": status, "progress_pct": pct }))
             .unwrap_or_else(|| serde_json::json!({ "status": "locked", "progress_pct": 0 }))
     };
 
@@ -295,12 +473,25 @@ pub async fn export_status(
     })))
 }
 
-async fn get_config_usize(pool: &sqlx::PgPool, key: &str, default: usize) -> usize {
-    let row: Option<(String,)> =
-        sqlx::query_as("SELECT value FROM config WHERE key = $1")
-            .bind(key)
-            .fetch_optional(pool)
-            .await
-            .unwrap_or(None);
-    row.and_then(|r| r.0.parse().ok()).unwrap_or(default)
+/// Centralised guard for the export rate limit. Same pattern as upload/feed: master
+/// switch + per-endpoint switch + numeric value, all stored in `config` and read on
+/// each request.
+async fn enforce_export_rate(state: &AppState, headers: &HeaderMap) -> Result<(), AppError> {
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    let export_rate_on = config::get_bool(&state.config_cache, "export_rate_enabled", true).await;
+    if !(rate_limits_on && export_rate_on) {
+        return Ok(());
+    }
+    let ip = client_ip(headers, "unknown");
+    let limit = config::get_usize(&state.config_cache, "export_rate_per_day", 3).await;
+    if !state
+        .rate_limiter
+        .check(format!("export:{ip}"), limit, Duration::from_secs(86400))
+    {
+        return Err(AppError::TooManyRequests(
+            "Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(),
+            None,
+        ));
+    }
+    Ok(())
 }

backend/src/handlers/feed.rs

@@ -1,14 +1,15 @@
 use std::time::Duration;
 
+use axum::Json;
 use axum::extract::{Query, State};
 use axum::http::HeaderMap;
-use axum::Json;
 use chrono::{DateTime, Utc};
 use serde::{Deserialize, Serialize};
 use uuid::Uuid;
 
 use crate::auth::middleware::AuthUser;
 use crate::error::AppError;
+use crate::services::config;
 use crate::services::rate_limiter::client_ip;
 use crate::state::AppState;
 
@@ -26,6 +27,8 @@ pub struct FeedUpload {
     pub uploader_name: String,
     pub preview_url: Option<String>,
     pub thumbnail_url: Option<String>,
+    /// Big-screen (~2048px) variant for the diashow. Absent until the derivative exists.
+    pub display_url: Option<String>,
     pub mime_type: String,
     pub caption: Option<String>,
     pub like_count: i64,
@@ -47,6 +50,7 @@ struct FeedRow {
     uploader_name: String,
     preview_path: Option<String>,
     thumbnail_path: Option<String>,
+    display_path: Option<String>,
     mime_type: String,
     caption: Option<String>,
     like_count: i64,
@@ -61,56 +65,68 @@ pub async fn feed(
     Query(q): Query<FeedQuery>,
 ) -> Result<Json<FeedResponse>, AppError> {
     let ip = client_ip(&headers, "unknown");
-    let rate_limit = get_config_usize(&state.pool, "feed_rate_per_min", 60).await;
-    if !state.rate_limiter.check(format!("feed:{ip}"), rate_limit, Duration::from_secs(60)) {
-        return Err(AppError::TooManyRequests("Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(), None));
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    let feed_rate_on = config::get_bool(&state.config_cache, "feed_rate_enabled", true).await;
+    if rate_limits_on && feed_rate_on {
+        let rate_limit = config::get_usize(&state.config_cache, "feed_rate_per_min", 60).await;
+        if !state
+            .rate_limiter
+            .check(format!("feed:{ip}"), rate_limit, Duration::from_secs(60))
+        {
+            return Err(AppError::TooManyRequests(
+                "Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(),
+                None,
+            ));
+        }
     }
 
     let limit = q.limit.unwrap_or(20).min(100);
 
+    // Resolve the cursor to a (created_at, id) position. The pair is compared as a
+    // tuple so ties on created_at break on id — keyset pagination on created_at
+    // alone would silently drop rows sharing a timestamp across a page boundary.
+    let (cursor_time, cursor_id) = match q.cursor {
+        Some(c) => match get_cursor_pos(&state.pool, c).await {
+            Some((t, id)) => (Some(t), Some(id)),
+            None => (None, None),
+        },
+        None => (None, None),
+    };
+
     let rows = if let Some(hashtag) = &q.hashtag {
         let tag = hashtag.trim().trim_start_matches('#').to_lowercase();
         sqlx::query_as::<_, FeedRow>(
             "SELECT v.id, v.user_id, v.uploader_name, v.preview_path, v.thumbnail_path,
-                    v.mime_type, v.caption, v.like_count, v.comment_count, v.created_at
+                    v.display_path, v.mime_type, v.caption, v.like_count, v.comment_count,
+                    v.created_at
              FROM v_feed v
              JOIN upload_hashtag uh ON uh.upload_id = v.id
              JOIN hashtag h ON h.id = uh.hashtag_id AND h.tag = $1
              WHERE v.event_id = $2
-               AND ($3::timestamptz IS NULL OR v.created_at < $3)
-             ORDER BY v.created_at DESC
-             LIMIT $4",
+               AND ($3::timestamptz IS NULL OR (v.created_at, v.id) < ($3, $4))
+             ORDER BY v.created_at DESC, v.id DESC
+             LIMIT $5",
         )
         .bind(&tag)
         .bind(auth.event_id)
-        .bind(
-            if let Some(cursor) = q.cursor {
-                get_cursor_time(&state.pool, cursor).await
-            } else {
-                None
-            },
-        )
+        .bind(cursor_time)
+        .bind(cursor_id)
         .bind(limit + 1)
         .fetch_all(&state.pool)
         .await?
     } else {
         sqlx::query_as::<_, FeedRow>(
             "SELECT id, user_id, uploader_name, preview_path, thumbnail_path,
-                    mime_type, caption, like_count, comment_count, created_at
+                    display_path, mime_type, caption, like_count, comment_count, created_at
              FROM v_feed
              WHERE event_id = $1
-               AND ($2::timestamptz IS NULL OR created_at < $2)
-             ORDER BY created_at DESC
-             LIMIT $3",
+               AND ($2::timestamptz IS NULL OR (created_at, id) < ($2, $3))
+             ORDER BY created_at DESC, id DESC
+             LIMIT $4",
         )
         .bind(auth.event_id)
-        .bind(
-            if let Some(cursor) = q.cursor {
-                get_cursor_time(&state.pool, cursor).await
-            } else {
-                None
-            },
-        )
+        .bind(cursor_time)
+        .bind(cursor_id)
         .bind(limit + 1)
         .fetch_all(&state.pool)
         .await?
@@ -118,7 +134,11 @@ pub async fn feed(
 
     let has_more = rows.len() as i64 > limit;
     let rows: Vec<FeedRow> = rows.into_iter().take(limit as usize).collect();
-    let next_cursor = if has_more { rows.last().map(|r| r.id) } else { None };
+    let next_cursor = if has_more {
+        rows.last().map(|r| r.id)
+    } else {
+        None
+    };
 
     // Batch check which uploads the current user has liked
     let upload_ids: Vec<Uuid> = rows.iter().map(|r| r.id).collect();
@@ -127,8 +147,21 @@ pub async fn feed(
     let uploads = rows
         .into_iter()
         .map(|r| {
-            let preview_url = r.preview_path.map(|p| format!("/media/{p}"));
-            let thumbnail_url = r.thumbnail_path.map(|p| format!("/media/{p}"));
+            // Gated media aliases (visibility-checked, direct /media blocked). Emit the
+            // URL only when the variant actually exists — the URL is what signals the
+            // client which variant to load.
+            let preview_url = r
+                .preview_path
+                .as_ref()
+                .map(|_| format!("/api/v1/upload/{}/preview", r.id));
+            let thumbnail_url = r
+                .thumbnail_path
+                .as_ref()
+                .map(|_| format!("/api/v1/upload/{}/thumbnail", r.id));
+            let display_url = r
+                .display_path
+                .as_ref()
+                .map(|_| format!("/api/v1/upload/{}/display", r.id));
             FeedUpload {
                 liked_by_me: liked_set.contains(&r.id),
                 id: r.id,
@@ -136,6 +169,7 @@ pub async fn feed(
                 uploader_name: r.uploader_name,
                 preview_url,
                 thumbnail_url,
+                display_url,
                 mime_type: r.mime_type,
                 caption: r.caption,
                 like_count: r.like_count,
@@ -160,6 +194,22 @@ pub struct DeltaQuery {
 pub struct DeltaResponse {
     pub uploads: Vec<FeedUpload>,
     pub deleted_ids: Vec<Uuid>,
+    /// Users whose uploads became hidden (banned / uploads_hidden) since `since`. A ban is
+    /// not a soft-delete, so it never appears in `deleted_ids`; without this, a client that
+    /// missed the ephemeral `user-hidden` SSE (a reconnecting projector, most acutely) would
+    /// keep displaying the banned user's already-loaded slides. The client evicts every
+    /// upload from these users on receipt.
+    pub hidden_user_ids: Vec<Uuid>,
+    /// True when the upload query hit `DELTA_LIMIT`: the response carries only the
+    /// newest slice of the gap, so the client must fall back to a full feed refresh
+    /// rather than merging (the older missed uploads are absent and unrecoverable
+    /// via a later delta, which advances `since` past them).
+    pub truncated: bool,
+    /// The server's clock at the moment this delta was computed. The client advances its
+    /// reconnect cursor from THIS, never `new Date()` — a browser clock even seconds fast
+    /// would otherwise silently skip uploads whose server `created_at` falls in the skew
+    /// window (they'd never reappear without a hard refresh).
+    pub server_time: DateTime<Utc>,
 }
 
 pub async fn feed_delta(
@@ -167,21 +217,78 @@ pub async fn feed_delta(
     auth: AuthUser,
     Query(q): Query<DeltaQuery>,
 ) -> Result<Json<DeltaResponse>, AppError> {
+    // Rate-limit the delta the same way as the paginated feed. Without this, ~100 clients
+    // reconnecting at once (post-outage, or a flapping network) each fire an unbounded
+    // delta fetch — a reconnect stampede. Keyed per-user so one client can't starve others
+    // behind a shared NAT.
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    let feed_rate_on = config::get_bool(&state.config_cache, "feed_rate_enabled", true).await;
+    if rate_limits_on && feed_rate_on {
+        let rate_limit = config::get_usize(&state.config_cache, "feed_rate_per_min", 60).await;
+        if !state.rate_limiter.check(
+            format!("feed_delta:{}", auth.user_id),
+            rate_limit,
+            Duration::from_secs(60),
+        ) {
+            return Err(AppError::TooManyRequests(
+                "Zu viele Anfragen. Bitte warte kurz und versuche es erneut.".into(),
+                None,
+            ));
+        }
+    }
+
+    // Anchor the next cursor to the DB clock, captured *before* the queries so an upload
+    // committed during this handler is re-fetched next time rather than skipped (a
+    // duplicate id merges idempotently on the client; a miss is unrecoverable).
+    let server_time: DateTime<Utc> = sqlx::query_scalar("SELECT NOW()")
+        .fetch_one(&state.pool)
+        .await?;
+
+    // Bounded like the paginated feed: a stale `since` could otherwise pull the
+    // entire event's uploads in one response. If a client hits the cap it should
+    // fall back to a full feed refresh rather than another delta.
+    const DELTA_LIMIT: i64 = 200;
+    // `>= since` (not `>`): `created_at` is not unique, so a strict `>` anchored to the exact
+    // timestamp of the last-seen upload silently drops a SECOND upload committed in the same
+    // microsecond — an unrecoverable live-update miss. `>=` re-includes the boundary rows;
+    // the client merges by id (see the feed-delta handler's `seen` set), so the duplicate is
+    // harmless while the tied upload is no longer lost. The cursor still advances to the
+    // response's `server_time`, so this doesn't re-fetch on every subsequent delta.
     let rows = sqlx::query_as::<_, FeedRow>(
         "SELECT id, user_id, uploader_name, preview_path, thumbnail_path,
-                mime_type, caption, like_count, comment_count, created_at
+                display_path, mime_type, caption, like_count, comment_count, created_at
          FROM v_feed
-         WHERE event_id = $1 AND created_at > $2
-         ORDER BY created_at DESC",
+         WHERE event_id = $1 AND created_at >= $2
+         ORDER BY created_at DESC, id DESC
+         LIMIT $3",
+    )
+    .bind(auth.event_id)
+    .bind(q.since)
+    .bind(DELTA_LIMIT)
+    .fetch_all(&state.pool)
+    .await?;
+
+    // Hit the cap => this is only the newest slice of a larger gap. Signal the
+    // client to full-refresh instead of merging a partial delta.
+    let truncated = rows.len() as i64 >= DELTA_LIMIT;
+
+    // `>=` for the same tie-break reason as the uploads query above; re-signalling an
+    // already-removed id is idempotent on the client (it filters its list by these ids).
+    let deleted_ids: Vec<(Uuid,)> = sqlx::query_as(
+        "SELECT id FROM upload
+         WHERE event_id = $1 AND deleted_at IS NOT NULL AND deleted_at >= $2",
     )
     .bind(auth.event_id)
     .bind(q.since)
     .fetch_all(&state.pool)
     .await?;
 
-    let deleted_ids: Vec<(Uuid,)> = sqlx::query_as(
-        "SELECT id FROM upload
-         WHERE event_id = $1 AND deleted_at IS NOT NULL AND deleted_at > $2",
+    // Users hidden since the cursor (ban / uploads_hidden). Uncapped like `deleted_ids` and
+    // for the same reason — an eviction the client misses is unrecoverable via a later delta.
+    let hidden_user_ids: Vec<(Uuid,)> = sqlx::query_as(
+        "SELECT id FROM \"user\"
+         WHERE event_id = $1 AND uploads_hidden = TRUE
+           AND uploads_hidden_at IS NOT NULL AND uploads_hidden_at >= $2",
     )
     .bind(auth.event_id)
     .bind(q.since)
@@ -198,8 +305,18 @@ pub async fn feed_delta(
             id: r.id,
             user_id: r.user_id,
             uploader_name: r.uploader_name,
-            preview_url: r.preview_path.map(|p| format!("/media/{p}")),
-            thumbnail_url: r.thumbnail_path.map(|p| format!("/media/{p}")),
+            preview_url: r
+                .preview_path
+                .as_ref()
+                .map(|_| format!("/api/v1/upload/{}/preview", r.id)),
+            thumbnail_url: r
+                .thumbnail_path
+                .as_ref()
+                .map(|_| format!("/api/v1/upload/{}/thumbnail", r.id)),
+            display_url: r
+                .display_path
+                .as_ref()
+                .map(|_| format!("/api/v1/upload/{}/display", r.id)),
             mime_type: r.mime_type,
             caption: r.caption,
             like_count: r.like_count,
@@ -211,6 +328,9 @@ pub async fn feed_delta(
     Ok(Json(DeltaResponse {
         uploads,
         deleted_ids: deleted_ids.into_iter().map(|r| r.0).collect(),
+        hidden_user_ids: hidden_user_ids.into_iter().map(|r| r.0).collect(),
+        truncated,
+        server_time,
     }))
 }
 
@@ -224,12 +344,11 @@ pub async fn hashtags(
     State(state): State<AppState>,
     auth: AuthUser,
 ) -> Result<Json<Vec<HashtagCount>>, AppError> {
-    let rows: Vec<(String, i64)> = sqlx::query_as(
-        "SELECT tag, upload_count FROM v_hashtag_counts WHERE event_id = $1",
-    )
-    .bind(auth.event_id)
-    .fetch_all(&state.pool)
-    .await?;
+    let rows: Vec<(String, i64)> =
+        sqlx::query_as("SELECT tag, upload_count FROM v_hashtag_counts WHERE event_id = $1")
+            .bind(auth.event_id)
+            .fetch_all(&state.pool)
+            .await?;
 
     Ok(Json(
         rows.into_iter()
@@ -238,24 +357,17 @@ pub async fn hashtags(
     ))
 }
 
-async fn get_config_usize(pool: &sqlx::PgPool, key: &str, default: usize) -> usize {
-    let row: Option<(String,)> =
-        sqlx::query_as("SELECT value FROM config WHERE key = $1")
-            .bind(key)
-            .fetch_optional(pool)
-            .await
-            .unwrap_or(None);
-    row.and_then(|r| r.0.parse().ok()).unwrap_or(default)
-}
-
-async fn get_cursor_time(pool: &sqlx::PgPool, cursor_id: Uuid) -> Option<DateTime<Utc>> {
-    let row: Option<(DateTime<Utc>,)> =
-        sqlx::query_as("SELECT created_at FROM upload WHERE id = $1")
+/// Resolve a cursor id to its `(created_at, id)` position. Both are needed:
+/// `created_at` alone isn't unique, so pagination must break ties on `id` to
+/// avoid silently dropping rows that share a timestamp across a page boundary.
+async fn get_cursor_pos(pool: &sqlx::PgPool, cursor_id: Uuid) -> Option<(DateTime<Utc>, Uuid)> {
+    let row: Option<(DateTime<Utc>, Uuid)> =
+        sqlx::query_as("SELECT created_at, id FROM upload WHERE id = $1")
             .bind(cursor_id)
             .fetch_optional(pool)
             .await
             .ok()?;
-    row.map(|r| r.0)
+    row
 }
 
 async fn get_liked_set(
@@ -266,14 +378,13 @@ async fn get_liked_set(
     if upload_ids.is_empty() {
         return std::collections::HashSet::new();
     }
-    let rows: Vec<(Uuid,)> = sqlx::query_as(
-        "SELECT upload_id FROM \"like\" WHERE user_id = $1 AND upload_id = ANY($2)",
-    )
-    .bind(user_id)
-    .bind(upload_ids)
-    .fetch_all(pool)
-    .await
-    .unwrap_or_default();
+    let rows: Vec<(Uuid,)> =
+        sqlx::query_as("SELECT upload_id FROM \"like\" WHERE user_id = $1 AND upload_id = ANY($2)")
+            .bind(user_id)
+            .bind(upload_ids)
+            .fetch_all(pool)
+            .await
+            .unwrap_or_default();
 
     rows.into_iter().map(|r| r.0).collect()
 }

backend/src/handlers/host.rs

@@ -1,6 +1,6 @@
+use axum::Json;
 use axum::extract::{Path, State};
 use axum::http::StatusCode;
-use axum::Json;
 use chrono::{DateTime, Utc};
 use serde::{Deserialize, Serialize};
 use uuid::Uuid;
@@ -9,8 +9,11 @@ use crate::auth::middleware::RequireHost;
 use crate::error::AppError;
 use crate::models::comment::Comment;
 use crate::models::event::Event;
+use crate::models::session::Session;
 use crate::models::upload::Upload;
-use crate::state::AppState;
+use crate::models::user::UserRole;
+use crate::services::export::Affects;
+use crate::state::{AppState, SseEvent};
 
 // ── DTOs ─────────────────────────────────────────────────────────────────────
 
@@ -34,9 +37,24 @@ pub struct EventStatus {
     pub export_released: bool,
 }
 
-#[derive(Deserialize)]
-pub struct BanRequest {
-    pub hide_uploads: bool,
+/// Count non-banned hosts/admins in the event OTHER than `excluding` — the operators
+/// who would remain if `excluding` were demoted or banned. Used to enforce the "an event
+/// always keeps at least one operator" floor.
+async fn remaining_operators(
+    state: &AppState,
+    event_id: Uuid,
+    excluding: Uuid,
+) -> Result<i64, AppError> {
+    let count = sqlx::query_scalar::<_, i64>(
+        "SELECT COUNT(*) FROM \"user\"
+         WHERE event_id = $1 AND id != $2
+           AND role IN ('host', 'admin') AND is_banned = FALSE",
+    )
+    .bind(event_id)
+    .bind(excluding)
+    .fetch_one(&state.pool)
+    .await?;
+    Ok(count)
 }
 
 #[derive(Deserialize)]
@@ -92,11 +110,13 @@ pub async fn ban_user(
     State(state): State<AppState>,
     RequireHost(auth): RequireHost,
     Path(user_id): Path<Uuid>,
-    Json(body): Json<BanRequest>,
 ) -> Result<StatusCode, AppError> {
+    // The ban request carries no body — ban always hides (no per-request options).
     // Cannot ban yourself or another host/admin
     if user_id == auth.user_id {
-        return Err(AppError::BadRequest("Du kannst dich nicht selbst sperren.".into()));
+        return Err(AppError::BadRequest(
+            "Du kannst dich nicht selbst sperren.".into(),
+        ));
     }
     let target = sqlx::query_as::<_, (String,)>(
         "SELECT role::text FROM \"user\" WHERE id = $1 AND event_id = $2",
@@ -107,30 +127,190 @@ pub async fn ban_user(
     .await?
     .ok_or_else(|| AppError::NotFound("Benutzer nicht gefunden.".into()))?;
 
-    if target.0 == "admin" || (target.0 == "host" && auth.role != crate::models::user::UserRole::Admin) {
-        return Err(AppError::Forbidden("Du kannst diesen Benutzer nicht sperren.".into()));
+    if target.0 == "admin"
+        || (target.0 == "host" && auth.role != crate::models::user::UserRole::Admin)
+    {
+        return Err(AppError::Forbidden(
+            "Du kannst diesen Benutzer nicht sperren.".into(),
+        ));
     }
 
+    // Floor: never leave the event with zero operators. Banning removes the target from
+    // the active-operator pool, so refuse if they're the last non-banned host/admin.
+    if target.0 == "host" && remaining_operators(&state, auth.event_id, user_id).await? == 0 {
+        return Err(AppError::BadRequest(
+            "Der letzte Host kann nicht gesperrt werden.".into(),
+        ));
+    }
+
+    // Ban ALWAYS hides: a banned user's content is "gone" everywhere. The visibility
+    // views/queries now also filter on `is_banned` (defense in depth), and we set
+    // `uploads_hidden` so the existing `user-hidden` live-eviction path fires too. The old
+    // opt-out checkbox is gone — `hide_uploads` in the request is ignored.
+    //
+    // We deliberately do NOT revoke the banned user's sessions. This is a *read-only ban*
+    // by design (USER_JOURNEYS §10.3): the user keeps read access to the feed and can still
+    // download the released export — writes and host/admin actions are what the ban blocks
+    // (enforced live on the write handlers + Require{Host,Admin}). Revoking sessions would
+    // contradict that model, break the documented "banned guest can still download the
+    // keepsake" flow, and be ineffective anyway (the user could just /recover a new session).
+    //
+    // The ban and the keepsake invalidation are ONE transaction — see `host_delete_upload`.
+    let mut tx = state.pool.begin().await?;
     sqlx::query(
-        "UPDATE \"user\" SET is_banned = TRUE, uploads_hidden = $2 WHERE id = $1",
+        "UPDATE \"user\"
+         SET is_banned = TRUE, uploads_hidden = TRUE, uploads_hidden_at = NOW()
+         WHERE id = $1 AND event_id = $2",
     )
     .bind(user_id)
-    .bind(body.hide_uploads)
-    .execute(&state.pool)
+    .bind(auth.event_id)
+    .execute(&mut *tx)
     .await?;
 
+    // A ban hides the user's uploads EVERYWHERE — and the keepsake is the place that matters most,
+    // because it is the copy people keep. The export already filters `is_banned = FALSE`, so a
+    // FUTURE export excludes them; without this, an ALREADY-RELEASED archive would keep serving a
+    // banned user's photos forever. Same class as a takedown, so same treatment.
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        Affects::Both,
+    )
+    .await?;
+    tx.commit().await?;
+    if let Some(r) = regen {
+        start_regen(&state, r);
+    }
+
+    // Evict their content live from every feed + the diashow so it disappears without
+    // each viewer having to reload. (Their own SSE stream is separately dropped by the
+    // is_banned revalidation in `sse::stream`, so they stop receiving live pushes while
+    // retaining plain read access.)
+    let _ = state.sse_tx.send(SseEvent::new(
+        "user-hidden",
+        serde_json::json!({ "user_id": user_id }).to_string(),
+    ));
+
+    tracing::info!(
+        actor_user_id = %auth.user_id,
+        target_user_id = %user_id,
+        event_id = %auth.event_id,
+        "host: ban_user"
+    );
+
     Ok(StatusCode::NO_CONTENT)
 }
 
 pub async fn unban_user(
     State(state): State<AppState>,
-    RequireHost(_auth): RequireHost,
+    RequireHost(auth): RequireHost,
     Path(user_id): Path<Uuid>,
 ) -> Result<StatusCode, AppError> {
-    sqlx::query("UPDATE \"user\" SET is_banned = FALSE WHERE id = $1")
-        .bind(user_id)
-        .execute(&state.pool)
-        .await?;
+    // Mirror the ban guard: a host may only lift bans on guests, never on hosts or
+    // admins. Without this a host could override an admin's ban of another host,
+    // which is asymmetric with `ban_user` and lets a host escalate a peer back in.
+    let target = sqlx::query_as::<_, (String,)>(
+        "SELECT role::text FROM \"user\" WHERE id = $1 AND event_id = $2",
+    )
+    .bind(user_id)
+    .bind(auth.event_id)
+    .fetch_optional(&state.pool)
+    .await?
+    .ok_or_else(|| AppError::NotFound("Benutzer nicht gefunden.".into()))?;
+
+    if target.0 == "admin"
+        || (target.0 == "host" && auth.role != crate::models::user::UserRole::Admin)
+    {
+        return Err(AppError::Forbidden(
+            "Du kannst diesen Benutzer nicht entsperren.".into(),
+        ));
+    }
+
+    // Unban restores visibility too: ban set `uploads_hidden = TRUE`, so clearing only
+    // `is_banned` would leave their content invisible. Clear all three (the timestamp too,
+    // so a future ban stamps a fresh `uploads_hidden_at` the reconnect delta will replay).
+    let mut tx = state.pool.begin().await?;
+    let result = sqlx::query(
+        "UPDATE \"user\"
+         SET is_banned = FALSE, uploads_hidden = FALSE, uploads_hidden_at = NULL
+         WHERE id = $1 AND event_id = $2",
+    )
+    .bind(user_id)
+    .bind(auth.event_id)
+    .execute(&mut *tx)
+    .await?;
+    if result.rows_affected() == 0 {
+        return Err(AppError::NotFound("Benutzer nicht gefunden.".into()));
+    }
+
+    // The mirror of the ban case: an unban RESTORES their uploads to the export query, so an
+    // already-released keepsake is now missing content it should contain. Rebuild it.
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        Affects::Both,
+    )
+    .await?;
+    tx.commit().await?;
+    if let Some(r) = regen {
+        start_regen(&state, r);
+    }
+
+    tracing::info!(
+        actor_user_id = %auth.user_id,
+        target_user_id = %user_id,
+        event_id = %auth.event_id,
+        "host: unban_user"
+    );
+    Ok(StatusCode::NO_CONTENT)
+}
+
+/// Force a keepsake rebuild. The ESCAPE HATCH.
+///
+/// Without this, a failed or stranded export is terminal at runtime: `release_gallery` refuses an
+/// already-released event ("bereits freigegeben"), `recover_exports` only runs at boot, and there is
+/// no other retry path — so the host's only options were restarting the container or reopening the
+/// event (which unlocks uploads to every guest and discards the release). This is also the recovery
+/// path for a keepsake that went stale for any reason we haven't thought of.
+pub async fn rebuild_export(
+    State(state): State<AppState>,
+    RequireHost(auth): RequireHost,
+) -> Result<StatusCode, AppError> {
+    let mut tx = state.pool.begin().await?;
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        Affects::Both,
+    )
+    .await?;
+    tx.commit().await?;
+
+    let Some(r) = regen else {
+        return Err(AppError::BadRequest(
+            "Die Galerie ist nicht freigegeben — es gibt nichts neu zu erzeugen.".into(),
+        ));
+    };
+
+    // No debounce: this is an explicit, deliberate host action, not a burst.
+    for export_type in ["zip", "html"] {
+        let _ = state.sse_tx.send(SseEvent::new(
+            "export-progress",
+            serde_json::json!({ "type": export_type, "progress_pct": 0 }).to_string(),
+        ));
+    }
+    crate::services::export::spawn_export_jobs(
+        r.event_id,
+        r.event_name,
+        r.epoch,
+        state.config.comments_enabled,
+        std::time::Duration::ZERO,
+        state.pool.clone(),
+        state.config.media_path.clone(),
+        state.config.export_path.clone(),
+        state.sse_tx.clone(),
+    );
+
+    tracing::info!(actor_user_id = %auth.user_id, "host: rebuild_export");
     Ok(StatusCode::NO_CONTENT)
 }
 
@@ -141,51 +321,324 @@ pub async fn set_role(
     Json(body): Json<SetRoleRequest>,
 ) -> Result<StatusCode, AppError> {
     if user_id == auth.user_id {
-        return Err(AppError::BadRequest("Du kannst deine eigene Rolle nicht ändern.".into()));
+        return Err(AppError::BadRequest(
+            "Du kannst deine eigene Rolle nicht ändern.".into(),
+        ));
     }
     let new_role = match body.role.as_str() {
         "guest" => "guest",
         "host" => "host",
-        _ => return Err(AppError::BadRequest("Ungültige Rolle. Erlaubt: guest, host.".into())),
+        _ => {
+            return Err(AppError::BadRequest(
+                "Ungültige Rolle. Erlaubt: guest, host.".into(),
+            ));
+        }
     };
+
+    // Look up the current role so we can apply the host-vs-admin guard. A plain host may
+    // promote/demote GUESTS only; it may not change any host's or admin's role (see the
+    // guard below — this closes the demote-a-peer-host→ban/PIN-reset takeover chain, F1).
+    // Only an admin may change a host's role. Admins may do anything except change
+    // themselves (blocked above).
+    let target = sqlx::query_as::<_, (String,)>(
+        "SELECT role::text FROM \"user\" WHERE id = $1 AND event_id = $2",
+    )
+    .bind(user_id)
+    .bind(auth.event_id)
+    .fetch_optional(&state.pool)
+    .await?
+    .ok_or_else(|| AppError::NotFound("Benutzer nicht gefunden.".into()))?;
+
+    // Admins are untouchable by hosts. A plain host also may not demote another
+    // *host*: without this guard a host could demote a peer host to guest and then
+    // ban / PIN-reset (→ account-takeover via /recover) them — the ban/pin-reset peer
+    // guards key off the target's *current* role, so a prior demotion would launder
+    // past them. Only an admin may change a host's role.
+    if target.0 == "admin" || (target.0 == "host" && auth.role != UserRole::Admin) {
+        return Err(AppError::Forbidden(
+            "Du darfst die Rolle dieses Benutzers nicht ändern.".into(),
+        ));
+    }
+
+    // Floor: demoting the last non-banned host/admin to guest would leave the event with
+    // no operator. Refuse.
+    if new_role == "guest"
+        && target.0 == "host"
+        && remaining_operators(&state, auth.event_id, user_id).await? == 0
+    {
+        return Err(AppError::BadRequest(
+            "Der letzte Host kann nicht zum Gast gemacht werden.".into(),
+        ));
+    }
+
     sqlx::query("UPDATE \"user\" SET role = $2::user_role WHERE id = $1 AND event_id = $3")
         .bind(user_id)
         .bind(new_role)
         .bind(auth.event_id)
         .execute(&state.pool)
         .await?;
+    tracing::info!(
+        actor_user_id = %auth.user_id,
+        target_user_id = %user_id,
+        event_id = %auth.event_id,
+        old_role = %target.0,
+        new_role,
+        "host: set_role"
+    );
     Ok(StatusCode::NO_CONTENT)
 }
 
+#[derive(Serialize)]
+pub struct PinResetResponse {
+    /// Plaintext PIN — shown to the operator **once**. Never persisted client-side.
+    pub pin: String,
+}
+
+/// Generate a fresh PIN for another user, returning the plaintext exactly once.
+///
+/// Authorisation:
+/// - Host caller → may reset **guest** PINs only.
+/// - Admin caller → may reset **guest** and **host** PINs (never another admin).
+/// - Target ≠ caller.
+pub async fn reset_user_pin(
+    State(state): State<AppState>,
+    RequireHost(auth): RequireHost,
+    Path(user_id): Path<Uuid>,
+) -> Result<Json<PinResetResponse>, AppError> {
+    use rand::Rng;
+
+    if user_id == auth.user_id {
+        return Err(AppError::BadRequest(
+            "Du kannst deine eigene PIN nicht über diese Funktion zurücksetzen.".into(),
+        ));
+    }
+
+    let target = sqlx::query_as::<_, (String,)>(
+        "SELECT role::text FROM \"user\" WHERE id = $1 AND event_id = $2",
+    )
+    .bind(user_id)
+    .bind(auth.event_id)
+    .fetch_optional(&state.pool)
+    .await?
+    .ok_or_else(|| AppError::NotFound("Benutzer nicht gefunden.".into()))?;
+
+    match (auth.role.clone(), target.0.as_str()) {
+        (UserRole::Admin, "guest" | "host") => {}
+        (UserRole::Host, "guest") => {}
+        _ => {
+            return Err(AppError::Forbidden(
+                "Du darfst die PIN dieses Benutzers nicht zurücksetzen.".into(),
+            ));
+        }
+    }
+
+    let pin: String = format!("{:04}", rand::rng().random_range(0..10000u32));
+    let pin_hash = bcrypt::hash(&pin, 12).map_err(|e| AppError::Internal(anyhow::anyhow!(e)))?;
+
+    sqlx::query(
+        "UPDATE \"user\"
+         SET recovery_pin_hash = $1,
+             failed_pin_attempts = 0,
+             pin_locked_until = NULL
+         WHERE id = $2 AND event_id = $3",
+    )
+    .bind(&pin_hash)
+    .bind(user_id)
+    .bind(auth.event_id)
+    .execute(&state.pool)
+    .await?;
+
+    // A PIN reset means the old credential is compromised/forgotten — revoke every
+    // existing session so old devices must re-authenticate with the new PIN. This is a
+    // security-relevant revoke: if it fails, the old sessions stay valid (sessions are
+    // token- not PIN-bound), so surface the error in logs rather than swallowing it
+    // silently while reporting success to the host.
+    if let Err(e) = Session::delete_all_for_user(&state.pool, user_id).await {
+        tracing::error!(error = ?e, user_id = %user_id, "PIN reset: failed to revoke sessions");
+    }
+
+    // Resolve any pending in-app "I forgot my PIN" request for this user.
+    let _ = sqlx::query("DELETE FROM pin_reset_request WHERE user_id = $1")
+        .bind(user_id)
+        .execute(&state.pool)
+        .await;
+
+    // Notify the *recipient* device(s) if they happen to be online so they can clear
+    // their cached local PIN. They'll save the new one on the next /recover.
+    let _ = state.sse_tx.send(SseEvent::new(
+        "pin-reset",
+        serde_json::json!({ "user_id": user_id }).to_string(),
+    ));
+
+    tracing::info!(
+        actor_user_id = %auth.user_id,
+        target_user_id = %user_id,
+        event_id = %auth.event_id,
+        "host: reset_user_pin"
+    );
+
+    Ok(Json(PinResetResponse { pin }))
+}
+
+#[derive(Serialize, sqlx::FromRow)]
+pub struct PinResetRequestSummary {
+    pub id: Uuid,
+    pub user_id: Uuid,
+    pub display_name: String,
+    pub created_at: DateTime<Utc>,
+}
+
+/// List pending in-app PIN-reset requests so a host can action them (via the existing
+/// `reset_user_pin`, which also clears the request).
+pub async fn list_pin_reset_requests(
+    State(state): State<AppState>,
+    RequireHost(auth): RequireHost,
+) -> Result<Json<Vec<PinResetRequestSummary>>, AppError> {
+    let rows = sqlx::query_as::<_, PinResetRequestSummary>(
+        "SELECT r.id, r.user_id, u.display_name, r.created_at
+         FROM pin_reset_request r
+         JOIN \"user\" u ON u.id = r.user_id
+         WHERE r.event_id = $1
+         ORDER BY r.created_at ASC",
+    )
+    .bind(auth.event_id)
+    .fetch_all(&state.pool)
+    .await?;
+    Ok(Json(rows))
+}
+
+/// Dismiss a PIN-reset request without resetting (e.g. the host couldn't verify the
+/// requester's identity).
+pub async fn dismiss_pin_reset_request(
+    State(state): State<AppState>,
+    RequireHost(auth): RequireHost,
+    Path(id): Path<Uuid>,
+) -> Result<StatusCode, AppError> {
+    sqlx::query("DELETE FROM pin_reset_request WHERE id = $1 AND event_id = $2")
+        .bind(id)
+        .bind(auth.event_id)
+        .execute(&state.pool)
+        .await?;
+    Ok(StatusCode::NO_CONTENT)
+}
+
+/// Content changed AFTER the gallery was released — regenerate the keepsake.
+///
+/// Deleting a photo used to remove it from the live feed but leave it in the already-generated
+/// archive FOREVER: the old `if ready { continue }` skip guaranteed the export was never rebuilt.
+/// For a takedown ("please remove my photo") that is the one place it most needs to disappear.
+///
+/// Bumping the epoch (while staying released) retires the current generation, so the stale archive
+/// stops being downloadable the instant the delete commits, and a fresh worker rebuilds it without
+/// the removed content. The download 404s in the meantime, which is the correct answer — serving
+/// the old archive would serve the deleted photo.
+/// Start the workers for a regeneration that was armed inside a (now-committed) transaction, and
+/// tell every client the current keepsake just became undownloadable.
+///
+/// The SSE matters: bumping the epoch retires the archive INSTANTLY, so `/export/zip` starts 404ing
+/// the moment the change commits. Without a nudge, a guest sitting on `/export` keeps rendering an
+/// enabled "download" button for the whole rebuild. Both the nav badge and the export page already
+/// refetch `/export/status` on `export-progress`, so a 0% tick is the cheapest correct signal.
+pub fn start_regen(state: &AppState, regen: crate::services::export::PendingRegen) {
+    for export_type in ["zip", "html"] {
+        let _ = state.sse_tx.send(SseEvent::new(
+            "export-progress",
+            serde_json::json!({ "type": export_type, "progress_pct": 0 }).to_string(),
+        ));
+    }
+    crate::services::export::spawn_export_jobs(
+        regen.event_id,
+        regen.event_name,
+        regen.epoch,
+        state.config.comments_enabled,
+        // Debounced: a takedown pass is a burst, and each request retires the last generation. The
+        // delay lets superseded workers fail their claim and do zero work instead of each building
+        // a full archive. See export::REGEN_DEBOUNCE.
+        crate::services::export::REGEN_DEBOUNCE,
+        state.pool.clone(),
+        state.config.media_path.clone(),
+        state.config.export_path.clone(),
+        state.sse_tx.clone(),
+    );
+}
+
 pub async fn host_delete_upload(
     State(state): State<AppState>,
-    RequireHost(_auth): RequireHost,
+    RequireHost(auth): RequireHost,
     Path(upload_id): Path<Uuid>,
 ) -> Result<StatusCode, AppError> {
-    let upload = Upload::find_by_id(&state.pool, upload_id)
+    let upload = Upload::find_by_id_and_event(&state.pool, upload_id, auth.event_id)
         .await?
         .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
 
-    Upload::soft_delete(&state.pool, upload_id).await?;
+    // The delete and the keepsake invalidation are ONE transaction: if the delete committed and the
+    // invalidation didn't, the taken-down photo would stay downloadable forever and nothing would
+    // notice (the keepsake still looks complete, and the host can no longer find the upload to retry).
+    let mut tx = state.pool.begin().await?;
+    let deleted = Upload::soft_delete_in_event(&mut tx, upload_id, auth.event_id).await?;
+    if !deleted {
+        return Err(AppError::NotFound("Upload nicht gefunden.".into()));
+    }
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        Affects::Both,
+    )
+    .await?;
+    tx.commit().await?;
 
-    let _ = state.sse_tx.send(crate::state::SseEvent {
-        event_type: "upload-deleted".to_string(),
-        data: serde_json::json!({ "upload_id": upload.id }).to_string(),
-    });
+    let _ = state.sse_tx.send(SseEvent::new(
+        "upload-deleted",
+        serde_json::json!({ "upload_id": upload.id }).to_string(),
+    ));
+    if let Some(r) = regen {
+        start_regen(&state, r);
+    }
+
+    tracing::info!(
+        actor_user_id = %auth.user_id,
+        event_id = %auth.event_id,
+        upload_id = %upload.id,
+        "host: host_delete_upload"
+    );
 
     Ok(StatusCode::NO_CONTENT)
 }
 
 pub async fn host_delete_comment(
     State(state): State<AppState>,
-    RequireHost(_auth): RequireHost,
+    RequireHost(auth): RequireHost,
     Path(comment_id): Path<Uuid>,
 ) -> Result<StatusCode, AppError> {
-    Comment::find_by_id(&state.pool, comment_id)
-        .await?
-        .ok_or_else(|| AppError::NotFound("Kommentar nicht gefunden.".into()))?;
+    let mut tx = state.pool.begin().await?;
+    let deleted = Comment::soft_delete_in_event(&mut tx, comment_id, auth.event_id).await?;
+    if !deleted {
+        return Err(AppError::NotFound("Kommentar nicht gefunden.".into()));
+    }
+    // Only the HTML viewer embeds comments — the ZIP is media-only, so it is carried forward rather
+    // than rebuilt. Otherwise moderating one comment would 404 the photo download for minutes to
+    // change nothing in it.
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        Affects::ViewerOnly,
+    )
+    .await?;
+    tx.commit().await?;
 
-    Comment::soft_delete(&state.pool, comment_id).await?;
+    let _ = state.sse_tx.send(SseEvent::new(
+        "comment-deleted",
+        serde_json::json!({ "comment_id": comment_id }).to_string(),
+    ));
+    if let Some(r) = regen {
+        start_regen(&state, r);
+    }
+    tracing::info!(
+        actor_user_id = %auth.user_id,
+        event_id = %auth.event_id,
+        comment_id = %comment_id,
+        "host: host_delete_comment"
+    );
     Ok(StatusCode::NO_CONTENT)
 }
 
@@ -193,17 +646,18 @@ pub async fn close_event(
     State(state): State<AppState>,
     RequireHost(_auth): RequireHost,
 ) -> Result<StatusCode, AppError> {
-    sqlx::query(
+    let result = sqlx::query(
         "UPDATE event SET uploads_locked_at = NOW() WHERE slug = $1 AND uploads_locked_at IS NULL",
     )
     .bind(&state.config.event_slug)
     .execute(&state.pool)
     .await?;
 
-    let _ = state.sse_tx.send(crate::state::SseEvent {
-        event_type: "event-closed".to_string(),
-        data: "{}".to_string(),
-    });
+    // Only broadcast when this call actually flipped the lock — closing an
+    // already-closed event is a no-op and shouldn't spam listeners.
+    if result.rows_affected() > 0 {
+        let _ = state.sse_tx.send(SseEvent::new("event-closed", "{}"));
+    }
 
     Ok(StatusCode::NO_CONTENT)
 }
@@ -212,17 +666,29 @@ pub async fn open_event(
     State(state): State<AppState>,
     RequireHost(_auth): RequireHost,
 ) -> Result<StatusCode, AppError> {
-    sqlx::query(
-        "UPDATE event SET uploads_locked_at = NULL WHERE slug = $1",
+    // Reopening invalidates any prior release: the keepsake was snapshotted at release time, so
+    // allowing new uploads afterwards would silently diverge the live feed from the frozen export.
+    //
+    // ONE statement retires the entire export generation. Bumping `export_epoch` in the same write
+    // that clears `export_released_at` instantly invalidates every in-flight worker, every `done`
+    // row and all readiness — because readiness is DERIVED from this epoch (migration 014), not
+    // stored. There is no export_job row to touch and nothing to keep in sync, so this needs no
+    // transaction. Any worker still streaming holds the old epoch and is now inert: its
+    // epoch-guarded finalize matches nothing, and it discards its own output.
+    let result = sqlx::query(
+        "UPDATE event
+         SET uploads_locked_at  = NULL,
+             export_released_at = NULL,
+             export_epoch       = export_epoch + 1
+         WHERE slug = $1 AND (uploads_locked_at IS NOT NULL OR export_released_at IS NOT NULL)",
     )
     .bind(&state.config.event_slug)
     .execute(&state.pool)
     .await?;
 
-    let _ = state.sse_tx.send(crate::state::SseEvent {
-        event_type: "event-opened".to_string(),
-        data: "{}".to_string(),
-    });
+    if result.rows_affected() > 0 {
+        let _ = state.sse_tx.send(SseEvent::new("event-opened", "{}"));
+    }
 
     Ok(StatusCode::NO_CONTENT)
 }
@@ -231,37 +697,69 @@ pub async fn release_gallery(
     State(state): State<AppState>,
     RequireHost(_auth): RequireHost,
 ) -> Result<StatusCode, AppError> {
-    let event = Event::find_by_slug(&state.pool, &state.config.event_slug)
-        .await?
-        .ok_or_else(|| AppError::NotFound("Event nicht gefunden.".into()))?;
+    // The release claim, the epoch bump, the upload lock and BOTH job rows are written in ONE
+    // transaction. Two reasons, both of which were live bugs:
+    //
+    //  1. Cancellation. This handler used to commit `export_released_at` and only THEN await the
+    //     enqueue. Axum drops the handler future when the client disconnects (closed tab, proxy
+    //     timeout), which left the event released and uploads locked with ZERO export_job rows and
+    //     no workers: downloads 404 forever and the host cannot retry, because release_gallery
+    //     rejects an already-released event ("bereits freigegeben"). Only a restart escaped it.
+    //     Now nothing is committed until every row is written, and the workers are spawned AFTER
+    //     the commit (a detached `tokio::spawn` survives cancellation).
+    //  2. Atomicity vs. a concurrent reopen. Bumping the epoch in the same statement that sets
+    //     `export_released_at` means no worker and no reader can ever observe "released again but
+    //     the generation hasn't moved on yet" — the window every previous fix kept leaving open.
+    //
+    // Release also locks uploads in the same statement (release ⇒ lock), so the export snapshot is
+    // taken against a frozen upload set. `COALESCE` preserves an earlier explicit lock time.
+    let mut tx = state.pool.begin().await?;
 
-    if event.export_released_at.is_some() {
-        return Err(AppError::BadRequest("Galerie wurde bereits freigegeben.".into()));
-    }
+    let claimed: Option<(Uuid, String, i64)> = sqlx::query_as(
+        "UPDATE event
+         SET export_released_at = NOW(),
+             uploads_locked_at  = COALESCE(uploads_locked_at, NOW()),
+             export_epoch       = export_epoch + 1
+         WHERE slug = $1 AND export_released_at IS NULL
+         RETURNING id, name, export_epoch",
+    )
+    .bind(&state.config.event_slug)
+    .fetch_optional(&mut *tx)
+    .await?;
 
-    sqlx::query("UPDATE event SET export_released_at = NOW() WHERE slug = $1")
-        .bind(&state.config.event_slug)
-        .execute(&state.pool)
-        .await?;
+    let Some((event_id, event_name, epoch)) = claimed else {
+        // Distinguish "no such event" from "already released" for a clean error.
+        let exists = Event::find_by_slug(&state.pool, &state.config.event_slug)
+            .await?
+            .is_some();
+        return Err(if exists {
+            AppError::BadRequest("Galerie wurde bereits freigegeben.".into())
+        } else {
+            AppError::NotFound("Event nicht gefunden.".into())
+        });
+    };
 
-    // Enqueue export jobs
-    for export_type in ["zip", "html"] {
-        sqlx::query(
-            "INSERT INTO export_job (event_id, type) VALUES ($1, $2::export_type)
-             ON CONFLICT (event_id, type) DO NOTHING",
-        )
-        .bind(event.id)
-        .bind(export_type)
-        .execute(&state.pool)
-        .await?;
-    }
+    // Arm both types at THIS epoch. No "skip the type that's already ready" check any more: the
+    // epoch bump above retired every prior generation, so there is nothing to preserve and nothing
+    // to be fooled by. (That skip was how a stale ready flag used to suppress regeneration.)
+    crate::services::export::enqueue_jobs_at_epoch(&mut tx, event_id, epoch).await?;
 
-    // Spawn export workers
+    tx.commit().await?;
+
+    // Release locks uploads too — tell any open composer to flip to the locked UI live rather than
+    // discovering it via a rejected upload.
+    let _ = state.sse_tx.send(SseEvent::new("event-closed", "{}"));
+
+    // Detached — survives this handler being cancelled.
     crate::services::export::spawn_export_jobs(
-        event.id,
-        event.name,
+        event_id,
+        event_name,
+        epoch,
+        state.config.comments_enabled,
+        std::time::Duration::ZERO,
         state.pool.clone(),
         state.config.media_path.clone(),
+        state.config.export_path.clone(),
         state.sse_tx.clone(),
     );
 

backend/src/handlers/me.rs

@@ -0,0 +1,114 @@
+//! Endpoints scoped to the *current user*. Kept separate from `auth::handlers` because
+//! these aren't about acquiring / refreshing a session — they're about reading my own
+//! state once I'm already signed in.
+//!
+//! Current routes:
+//! - `GET /api/v1/me/context` — bundled profile + feature flags + privacy note. The
+//!   account page loads this once on mount instead of issuing several round trips.
+//! - `GET /api/v1/me/quota`  — live per-user storage quota estimate.
+
+use axum::Json;
+use axum::extract::State;
+use serde::Serialize;
+
+use crate::auth::middleware::AuthUser;
+use crate::error::AppError;
+use crate::handlers::upload::compute_storage_quota;
+use crate::models::user::{User, UserRole};
+use crate::services::config;
+use crate::state::AppState;
+
+#[derive(Serialize)]
+pub struct QuotaDto {
+    pub enabled: bool,
+    pub used_bytes: i64,
+    pub limit_bytes: Option<i64>,
+    pub active_uploaders: i64,
+    pub free_disk_bytes: i64,
+}
+
+pub async fn get_quota(
+    State(state): State<AppState>,
+    auth: AuthUser,
+) -> Result<Json<QuotaDto>, AppError> {
+    let user = User::find_by_id(&state.pool, auth.user_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Benutzer nicht gefunden.".into()))?;
+
+    let estimate = compute_storage_quota(&state).await;
+
+    // Raw server telemetry (free disk, concurrent uploader count) is staff-only — it
+    // must never reach a guest, even though the guest upload UI no longer renders it.
+    // A guest still gets their own `used`/`limit` so enforcement stays transparent to
+    // the code paths that consume it; only the server-wide fields are zeroed.
+    let is_staff = matches!(auth.role, UserRole::Host | UserRole::Admin);
+
+    Ok(Json(QuotaDto {
+        enabled: estimate.limit_bytes.is_some(),
+        used_bytes: user.total_upload_bytes,
+        limit_bytes: estimate.limit_bytes,
+        active_uploaders: if is_staff {
+            estimate.active_uploaders
+        } else {
+            0
+        },
+        free_disk_bytes: if is_staff {
+            estimate.free_disk_bytes
+        } else {
+            0
+        },
+    }))
+}
+
+#[derive(Serialize)]
+pub struct MeContextDto {
+    pub user_id: uuid::Uuid,
+    pub display_name: String,
+    pub role: String,
+    /// Plain-text Datenschutzhinweis set by the admin. Empty string when not configured.
+    pub privacy_note: String,
+    pub quota_enabled: bool,
+    pub storage_quota_enabled: bool,
+    /// Uploads are locked (event closed) — the composer should show a locked state live
+    /// instead of letting a guest compose an upload only to eat a 403.
+    pub uploads_locked: bool,
+    /// The gallery has been released and the export snapshotted — uploads are permanently
+    /// closed for this run (release ⇒ lock, and reopening regenerates).
+    pub gallery_released: bool,
+}
+
+pub async fn get_context(
+    State(state): State<AppState>,
+    auth: AuthUser,
+) -> Result<Json<MeContextDto>, AppError> {
+    let user = User::find_by_id(&state.pool, auth.user_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Benutzer nicht gefunden.".into()))?;
+
+    let privacy_note = config::get_str(&state.config_cache, "privacy_note", "").await;
+    let quota_enabled = config::get_bool(&state.config_cache, "quota_enabled", true).await;
+    let storage_quota_enabled =
+        config::get_bool(&state.config_cache, "storage_quota_enabled", true).await;
+
+    let event =
+        crate::models::event::Event::find_by_slug(&state.pool, &state.config.event_slug).await?;
+    let uploads_locked = event
+        .as_ref()
+        .map(|e| e.uploads_locked_at.is_some())
+        .unwrap_or(false);
+    let gallery_released = event
+        .as_ref()
+        .map(|e| e.export_released_at.is_some())
+        .unwrap_or(false);
+
+    Ok(Json(MeContextDto {
+        user_id: user.id,
+        display_name: user.display_name,
+        role: user.role.as_str().to_string(),
+        privacy_note,
+        quota_enabled,
+        storage_quota_enabled,
+        uploads_locked,
+        gallery_released,
+    }))
+}

backend/src/handlers/mod.rs

@@ -1,6 +1,9 @@
 pub mod admin;
 pub mod feed;
 pub mod host;
+pub mod me;
+pub mod public;
 pub mod social;
 pub mod sse;
+pub mod test_admin;
 pub mod upload;

backend/src/handlers/public.rs

@@ -0,0 +1,44 @@
+//! Unauthenticated, read-only endpoints safe to expose before a user has joined.
+
+use axum::Json;
+use axum::extract::State;
+use serde::Serialize;
+
+use crate::services::config;
+use crate::state::AppState;
+
+#[derive(Serialize)]
+pub struct PublicEventDto {
+    pub name: String,
+    pub slug: String,
+    /// Whether the comment feature is on (env `COMMENTS_ENABLED`). The frontend hides
+    /// the whole comment UI when false; exposed here so even the pre-auth shell knows.
+    pub comments_enabled: bool,
+    /// Active colour theme. `preset` is an id the frontend maps to a palette (or
+    /// "custom"); `primary`/`accent` are the `#rrggbb` seeds the ramps derive from.
+    /// Resolved as DB-config override → env default. Public so the theme applies on
+    /// the very first (pre-auth) paint without a flash.
+    pub theme_preset: String,
+    pub theme_primary: String,
+    pub theme_accent: String,
+}
+
+/// Public event identity + presentation config, used by the pre-auth join/recover
+/// screens (which event am I joining, what does it look like). Only non-user-scoped
+/// fields are exposed, so this is safe without a token. Identity comes straight from
+/// instance config; the theme is resolved from the runtime `config` table (admin UI)
+/// falling back to the env-seeded default.
+pub async fn get_public_event(State(state): State<AppState>) -> Json<PublicEventDto> {
+    let cache = &state.config_cache;
+    Json(PublicEventDto {
+        name: state.config.event_name.clone(),
+        slug: state.config.event_slug.clone(),
+        comments_enabled: state.config.comments_enabled,
+        theme_preset: config::get_str(cache, "theme_preset", &state.config.default_theme_preset)
+            .await,
+        theme_primary: config::get_str(cache, "theme_primary", &state.config.default_theme_primary)
+            .await,
+        theme_accent: config::get_str(cache, "theme_accent", &state.config.default_theme_accent)
+            .await,
+    })
+}

backend/src/handlers/social.rs

@@ -1,20 +1,33 @@
-use axum::extract::{Path, State};
-use axum::http::StatusCode;
 use axum::Json;
-use serde::Deserialize;
+use axum::extract::{Path, Query, State};
+use axum::http::StatusCode;
+use chrono::{DateTime, Utc};
+use serde::{Deserialize, Serialize};
 use uuid::Uuid;
 
 use crate::auth::middleware::AuthUser;
 use crate::error::AppError;
 use crate::models::comment::{Comment, CommentDto};
 use crate::models::hashtag::{self, Hashtag};
+use crate::models::upload::Upload;
 use crate::state::AppState;
 
+#[derive(Serialize)]
+pub struct LikeResponse {
+    /// The caller's like state *after* this toggle. The client sets `liked_by_me` from
+    /// this rather than blind-inverting local state — otherwise a second device (same
+    /// recovered user) drifts, since the `like-update` broadcast only carries `like_count`.
+    pub liked: bool,
+    /// Fresh like count, or `null` if the (best-effort) count query hiccuped. The client
+    /// keeps its current count when this is null rather than adopting a wrong number.
+    pub like_count: Option<i64>,
+}
+
 pub async fn toggle_like(
     State(state): State<AppState>,
     auth: AuthUser,
     Path(upload_id): Path<Uuid>,
-) -> Result<StatusCode, AppError> {
+) -> Result<Json<LikeResponse>, AppError> {
     // Check if user is banned
     let user = crate::models::user::User::find_by_id(&state.pool, auth.user_id)
         .await?
@@ -23,7 +36,17 @@ pub async fn toggle_like(
         return Err(AppError::Forbidden("Du bist gesperrt.".into()));
     }
 
-    // Try to insert; if conflict, delete (toggle)
+    // Event-scope: the upload must belong to the caller's event (404 otherwise),
+    // matching the host handlers' find_by_id_and_event pattern.
+    Upload::find_by_id_and_event(&state.pool, upload_id, auth.event_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
+
+    // NOTE: liking is intentionally allowed while the event is locked. Locking
+    // ("Event schließen") freezes *new uploads* only — likes, comments and
+    // browsing stay open (USER_JOURNEYS §9.3, FEATURES capability matrix).
+
+    // Try to insert; if conflict, delete (toggle). `liked` = the caller's state afterwards.
     let result = sqlx::query(
         "INSERT INTO \"like\" (upload_id, user_id) VALUES ($1, $2)
          ON CONFLICT (upload_id, user_id) DO NOTHING",
@@ -33,7 +56,8 @@ pub async fn toggle_like(
     .execute(&state.pool)
     .await?;
 
-    if result.rows_affected() == 0 {
+    let liked = result.rows_affected() > 0;
+    if !liked {
         // Already liked — remove
         sqlx::query("DELETE FROM \"like\" WHERE upload_id = $1 AND user_id = $2")
             .bind(upload_id)
@@ -42,21 +66,55 @@ pub async fn toggle_like(
             .await?;
     }
 
-    // Broadcast SSE
-    let _ = state.sse_tx.send(crate::state::SseEvent {
-        event_type: "like-update".to_string(),
-        data: serde_json::json!({ "upload_id": upload_id }).to_string(),
-    });
+    // Fresh count so feed clients can patch the single card in place instead of
+    // refetching page 1 (mirrors v_feed.like_count = COUNT(DISTINCT user_id)). The like
+    // itself is already committed, so a failed count must not fail the request — but we
+    // also must NOT broadcast/return a bogus 0 (that would push like_count: 0 to every
+    // client until the next event). On error we skip the broadcast and return null.
+    let like_count = sqlx::query_scalar::<_, i64>(
+        "SELECT COUNT(DISTINCT user_id) FROM \"like\" WHERE upload_id = $1",
+    )
+    .bind(upload_id)
+    .fetch_one(&state.pool)
+    .await
+    .ok();
 
-    Ok(StatusCode::NO_CONTENT)
+    if let Some(count) = like_count {
+        // Broadcast the new count so other clients patch their card. Only `like_count` is
+        // shared — each client's own `liked_by_me` only changes via its own toggle (which
+        // now reads it straight from this response).
+        let _ = state.sse_tx.send(crate::state::SseEvent {
+            event_type: "like-update".to_string(),
+            data: serde_json::json!({ "upload_id": upload_id, "like_count": count }).to_string(),
+        });
+    }
+
+    Ok(Json(LikeResponse { liked, like_count }))
 }
 
+#[derive(Deserialize, Default)]
+pub struct ListCommentsQuery {
+    /// RFC3339 timestamp — return only comments older than this. Pass the
+    /// `created_at` of the oldest currently-loaded comment to fetch the next
+    /// older page.
+    pub before: Option<DateTime<Utc>>,
+}
+
+const COMMENT_PAGE_SIZE: i64 = 50;
+
 pub async fn list_comments(
     State(state): State<AppState>,
-    _auth: AuthUser,
+    auth: AuthUser,
     Path(upload_id): Path<Uuid>,
+    Query(q): Query<ListCommentsQuery>,
 ) -> Result<Json<Vec<CommentDto>>, AppError> {
-    let comments = Comment::list_for_upload(&state.pool, upload_id).await?;
+    // Event-scope: only list comments for an upload in the caller's event.
+    Upload::find_by_id_and_event(&state.pool, upload_id, auth.event_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
+
+    let comments =
+        Comment::list_for_upload(&state.pool, upload_id, q.before, COMMENT_PAGE_SIZE).await?;
     Ok(Json(comments))
 }
 
@@ -71,6 +129,12 @@ pub async fn add_comment(
     Path(upload_id): Path<Uuid>,
     Json(body): Json<AddCommentRequest>,
 ) -> Result<(StatusCode, Json<CommentDto>), AppError> {
+    // Comments can be disabled instance-wide (env COMMENTS_ENABLED). The frontend hides
+    // the UI, but gate the API too so a stale client or direct call can't slip one in.
+    if !state.config.comments_enabled {
+        return Err(AppError::Forbidden("Kommentare sind deaktiviert.".into()));
+    }
+
     let user = crate::models::user::User::find_by_id(&state.pool, auth.user_id)
         .await?
         .ok_or_else(|| AppError::NotFound("Benutzer nicht gefunden.".into()))?;
@@ -78,33 +142,58 @@ pub async fn add_comment(
         return Err(AppError::Forbidden("Du bist gesperrt.".into()));
     }
 
+    // Event-scope: only comment on an upload that belongs to the caller's event.
+    Upload::find_by_id_and_event(&state.pool, upload_id, auth.event_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
+
+    // NOTE: commenting is intentionally allowed while the event is locked. Locking
+    // freezes *new uploads* only — likes, comments and browsing stay open
+    // (USER_JOURNEYS §9.3, FEATURES capability matrix).
+
     let text = body.body.trim();
-    if text.is_empty() || text.len() > 500 {
+    let text_chars = text.chars().count();
+    if text_chars == 0 || text_chars > 500 {
         return Err(AppError::BadRequest(
             "Kommentar muss zwischen 1 und 500 Zeichen lang sein.".into(),
         ));
     }
 
-    let comment = Comment::create(&state.pool, upload_id, auth.user_id, text).await?;
-
-    // Process hashtags in comment body
+    // Insert the comment and link its hashtags atomically, so a crash mid-loop
+    // can't leave a committed comment with only some of its tags indexed.
     let tags = hashtag::extract_hashtags(text);
+    let mut tx = state.pool.begin().await?;
+    let comment = Comment::create(&mut *tx, upload_id, auth.user_id, text).await?;
     for tag in &tags {
-        let h = Hashtag::upsert(&state.pool, auth.event_id, tag).await?;
+        let h = Hashtag::upsert(&mut *tx, auth.event_id, tag).await?;
         sqlx::query(
             "INSERT INTO comment_hashtag (comment_id, hashtag_id) VALUES ($1, $2) ON CONFLICT DO NOTHING",
         )
         .bind(comment.id)
         .bind(h.id)
-        .execute(&state.pool)
+        .execute(&mut *tx)
         .await?;
     }
+    tx.commit().await?;
 
-    // Broadcast SSE
-    let _ = state.sse_tx.send(crate::state::SseEvent {
-        event_type: "new-comment".to_string(),
-        data: serde_json::json!({ "upload_id": upload_id }).to_string(),
-    });
+    // Fresh count so feed clients can patch the single card in place instead of
+    // refetching page 1 (mirrors v_feed.comment_count = COUNT(DISTINCT c.id); COUNT(*)
+    // over the same deleted_at filter is identical since comment.id is the PK). The
+    // count + broadcast are a UI optimisation — the comment is already committed, so a
+    // failure here must not fail the request. Swallow the error and skip the broadcast.
+    if let Ok(comment_count) = sqlx::query_scalar::<_, i64>(
+        "SELECT COUNT(*) FROM comment WHERE upload_id = $1 AND deleted_at IS NULL",
+    )
+    .bind(upload_id)
+    .fetch_one(&state.pool)
+    .await
+    {
+        let _ = state.sse_tx.send(crate::state::SseEvent {
+            event_type: "new-comment".to_string(),
+            data: serde_json::json!({ "upload_id": upload_id, "comment_count": comment_count })
+                .to_string(),
+        });
+    }
 
     let dto = CommentDto {
         id: comment.id,
@@ -123,6 +212,10 @@ pub async fn delete_comment(
     auth: AuthUser,
     Path(comment_id): Path<Uuid>,
 ) -> Result<StatusCode, AppError> {
+    // Banned users keep read access but cannot mutate (USER_JOURNEYS §10).
+    if auth.is_banned {
+        return Err(AppError::Forbidden("Du bist gesperrt.".into()));
+    }
     let comment = Comment::find_by_id(&state.pool, comment_id)
         .await?
         .ok_or_else(|| AppError::NotFound("Kommentar nicht gefunden.".into()))?;
@@ -131,6 +224,27 @@ pub async fn delete_comment(
         return Err(AppError::Forbidden("Nur eigene Kommentare löschen.".into()));
     }
 
-    Comment::soft_delete(&state.pool, comment_id).await?;
+    // Event-scope: soft_delete_in_event only matches comments whose upload is in
+    // the caller's event, so a cross-event comment_id resolves to a 404 here.
+    let mut tx = state.pool.begin().await?;
+    let deleted = Comment::soft_delete_in_event(&mut tx, comment_id, auth.event_id).await?;
+    if !deleted {
+        return Err(AppError::NotFound("Kommentar nicht gefunden.".into()));
+    }
+    // Comments live only in the HTML viewer, so the ZIP is carried forward, not rebuilt.
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        crate::services::export::Affects::ViewerOnly,
+    )
+    .await?;
+    tx.commit().await?;
+    if let Some(r) = regen {
+        crate::handlers::host::start_regen(&state, r);
+    }
+    let _ = state.sse_tx.send(crate::state::SseEvent::new(
+        "comment-deleted",
+        serde_json::json!({ "comment_id": comment_id, "upload_id": comment.upload_id }).to_string(),
+    ));
     Ok(StatusCode::NO_CONTENT)
 }

backend/src/handlers/sse.rs

@@ -1,47 +1,115 @@
 use std::convert::Infallible;
 use std::time::Duration;
 
+use axum::Json;
 use axum::extract::{Query, State};
 use axum::response::sse::{Event, KeepAlive, Sse};
 use futures::stream::Stream;
-use serde::Deserialize;
-use tokio_stream::wrappers::BroadcastStream;
+use serde::{Deserialize, Serialize};
 use tokio_stream::StreamExt;
+use tokio_stream::wrappers::BroadcastStream;
+use tokio_stream::wrappers::errors::BroadcastStreamRecvError;
 
-use crate::auth::jwt;
+use crate::auth::middleware::AuthUser;
 use crate::error::AppError;
 use crate::models::session::Session;
 use crate::state::AppState;
 
 #[derive(Deserialize)]
 pub struct SseQuery {
-    pub token: String,
+    pub ticket: String,
 }
 
-/// SSE stream endpoint. Accepts JWT via query param since EventSource
-/// doesn't support custom headers.
+#[derive(Serialize)]
+pub struct StreamTicketResponse {
+    pub ticket: String,
+    /// Server clock at mint time — the client seeds its SSE reconnect cursor from this
+    /// instead of `new Date()`, so a skewed browser clock can't drop uploads. See
+    /// `DeltaResponse::server_time`.
+    pub server_time: chrono::DateTime<chrono::Utc>,
+}
+
+/// Mint a short-lived single-use SSE ticket. The browser's `EventSource` cannot
+/// send an `Authorization` header, so the alternative used to be passing the JWT
+/// as `?token=...` — which leaks the bearer token into access logs, referer
+/// headers, and browser history. The client now exchanges its Bearer token for
+/// an opaque ticket via this endpoint and passes that on the stream open.
+pub async fn issue_ticket(
+    State(state): State<AppState>,
+    auth: AuthUser,
+) -> Result<Json<StreamTicketResponse>, AppError> {
+    let ticket = state.sse_tickets.issue(auth.token_hash);
+    let server_time = sqlx::query_scalar("SELECT NOW()")
+        .fetch_one(&state.pool)
+        .await?;
+    Ok(Json(StreamTicketResponse {
+        ticket,
+        server_time,
+    }))
+}
+
+/// SSE stream endpoint. Authenticates via a single-use ticket (see
+/// [`issue_ticket`]) — never the raw JWT.
 pub async fn stream(
     State(state): State<AppState>,
     Query(q): Query<SseQuery>,
 ) -> Result<Sse<impl Stream<Item = Result<Event, Infallible>>>, AppError> {
-    // Verify token
-    let _claims = jwt::verify_token(&q.token, &state.config.jwt_secret)
-        .map_err(|_| AppError::Unauthorized("Token ungültig.".into()))?;
+    let token_hash = state
+        .sse_tickets
+        .consume(&q.ticket)
+        .ok_or_else(|| AppError::Unauthorized("Ticket ungültig oder abgelaufen.".into()))?;
 
-    let token_hash = jwt::hash_token(&q.token);
+    // NOTE: this authenticates via ticket→session, not the `AuthUser` extractor. The
+    // SSE stream is read-only, and under the read-only-ban model (USER_JOURNEYS §10)
+    // banned users retain read access — so both minting a ticket and holding a stream
+    // open are intentionally allowed for banned users; only writes are blocked.
     Session::find_by_token_hash(&state.pool, &token_hash)
         .await
         .map_err(|e| AppError::Internal(e.into()))?
         .ok_or_else(|| AppError::Unauthorized("Sitzung nicht gefunden.".into()))?;
 
     let rx = state.sse_tx.subscribe();
-    let stream = BroadcastStream::new(rx).filter_map(|msg| match msg {
+    let events = BroadcastStream::new(rx).filter_map(|msg| match msg {
         Ok(sse_event) => Some(Ok(Event::default()
             .event(sse_event.event_type)
             .data(sse_event.data))),
-        Err(_) => None,
+        // A consumer that falls behind the broadcast buffer would otherwise silently
+        // lose events, leaving its feed permanently stale. Instead of dropping the
+        // gap, tell the client to resync — the frontend responds by running a full
+        // feed-delta fetch (which reconciles both new uploads and deletions).
+        Err(BroadcastStreamRecvError::Lagged(n)) => {
+            tracing::warn!("SSE consumer lagged, dropped {n} event(s); emitting resync");
+            Some(Ok(Event::default().event("resync").data(n.to_string())))
+        }
     });
 
+    // The session is only checked once at open. Re-validate it periodically so a
+    // logged-out, expired, OR banned session's stream is closed rather than kept alive
+    // until the client happens to disconnect. Bounds a stale stream to ~60s. NOTE: a ban is
+    // deliberately read-only and does NOT revoke sessions (see `ban_user`), so the
+    // `is_banned` re-read below is LOAD-BEARING — it is the only thing that stops a banned
+    // user's live push stream. Do not remove it. Only a *definitive* gone/expired/banned
+    // state ends the stream; a transient DB error just retries next tick.
+    let pool = state.pool.clone();
+    let session_hash = token_hash.clone();
+    let session_gone = async move {
+        let mut ticker = tokio::time::interval(Duration::from_secs(60));
+        ticker.tick().await; // consume the immediate first tick
+        loop {
+            ticker.tick().await;
+            match Session::find_user_by_token_hash(&pool, &session_hash).await {
+                Ok(Some(user)) if !user.is_banned => {} // still valid — keep streaming
+                Ok(Some(_)) => break,                   // banned — cut the stream
+                Ok(None) => break,                      // logged out or expired — stop
+                Err(e) => {
+                    tracing::warn!(error = ?e, "SSE session revalidation query failed; retrying");
+                }
+            }
+        }
+    };
+
+    let stream = futures::StreamExt::take_until(events, Box::pin(session_gone));
+
     Ok(Sse::new(stream).keep_alive(
         KeepAlive::new()
             .interval(Duration::from_secs(30))

backend/src/handlers/test_admin.rs

@@ -0,0 +1,116 @@
+//! Test-only admin routes. **Compiled in always, but only registered when
+//! `EVENTSNAP_TEST_MODE=1` is set in the environment.** The route returns a hard
+//! 404 in production builds because [`crate::main`] skips registering the handler.
+//!
+//! These exist to give the Playwright E2E suite a quick "reset everything"
+//! escape hatch without forcing tests to maintain raw SQL fixtures or spin up a
+//! fresh database container per test.
+
+use axum::extract::State;
+use axum::http::StatusCode;
+
+use crate::auth::middleware::RequireAdmin;
+use crate::error::AppError;
+use crate::state::AppState;
+
+/// Truncates every event-scoped table, wipes media on disk, and reseeds the
+/// `config` table from migration defaults. Requires an admin JWT — even with
+/// `EVENTSNAP_TEST_MODE=1` it cannot be hit anonymously.
+pub async fn truncate_all(
+    State(state): State<AppState>,
+    RequireAdmin(_auth): RequireAdmin,
+) -> Result<StatusCode, AppError> {
+    // Truncate in dependency order doesn't matter with CASCADE, but listing the
+    // tables explicitly makes the blast radius obvious in code review.
+    sqlx::query(
+        r#"TRUNCATE
+            comment_hashtag,
+            upload_hashtag,
+            hashtag,
+            "like",
+            comment,
+            export_job,
+            upload,
+            session,
+            "user",
+            event,
+            config
+        RESTART IDENTITY CASCADE"#,
+    )
+    .execute(&state.pool)
+    .await?;
+
+    // Reseed config — mirrors migrations 005, 009 and 015. Kept in sync by hand
+    // because pulling SQL out of the migration files at runtime is fragile.
+    sqlx::query(
+        r#"INSERT INTO config (key, value) VALUES
+            ('max_image_size_mb',          '20'),
+            ('max_video_size_mb',          '500'),
+            ('upload_rate_per_hour',       '100'),
+            ('feed_rate_per_min',          '60'),
+            ('export_rate_per_day',        '3'),
+            ('quota_tolerance',            '0.75'),
+            ('estimated_guest_count',      '100'),
+            ('compression_concurrency',    '2'),
+            ('rate_limits_enabled',        'false'),
+            ('upload_rate_enabled',        'false'),
+            ('feed_rate_enabled',          'false'),
+            ('export_rate_enabled',        'false'),
+            ('join_rate_enabled',          'false'),
+            ('quota_enabled',              'false'),
+            ('storage_quota_enabled',      'false'),
+            ('upload_count_quota_enabled', 'false'),
+            ('privacy_note',               '')
+        ON CONFLICT (key) DO UPDATE SET value = EXCLUDED.value"#,
+    )
+    .execute(&state.pool)
+    .await?;
+
+    // Wipe media directory. Best-effort: if it doesn't exist, that's fine.
+    let _ = tokio::fs::remove_dir_all(&state.config.media_path).await;
+    let _ = tokio::fs::create_dir_all(&state.config.media_path).await;
+
+    // Wipe the export directory too. Exports moved OUT of media_path (CR2 fix), so
+    // the media wipe above no longer covers them — without this a real export in
+    // one test would leave Gallery.zip on disk and contaminate the next.
+    let _ = tokio::fs::remove_dir_all(&state.config.export_path).await;
+    let _ = tokio::fs::create_dir_all(&state.config.export_path).await;
+
+    // The rate limiter holds an in-memory HashMap; clear it so a previous test's
+    // counters don't leak into the next one.
+    state.rate_limiter.clear();
+
+    // The reseed above wrote the `config` table directly (bypassing patch_config), so
+    // the cache must be invalidated too — otherwise the first request after a truncate
+    // could serve the previous test's toggles.
+    state.config_cache.invalidate();
+
+    // The other two in-memory singletons that TRUNCATE used to leave standing.
+    //
+    // `disk_cache` holds a free-space reading for up to its TTL. TRUNCATE has just deleted every
+    // uploaded file, which materially changes free space — so without this the next test can
+    // compute a storage quota from the PREVIOUS test's disk. That was harmless only while quotas
+    // were globally disabled in e2e (they no longer are: see specs/02-upload/quota.spec.ts, which
+    // steers the per-user limit off `free_disk_bytes`), i.e. two holes were masking each other.
+    state.disk_cache.invalidate();
+
+    // `sse_tickets` maps a ticket to a session token hash. TRUNCATE deletes the sessions, so every
+    // surviving ticket is a dangling reference to a user that no longer exists.
+    state.sse_tickets.clear();
+
+    // Invalidate any in-flight/queued compression task spawned by the previous test. Without this a
+    // task still waiting on the concurrency semaphore wakes AFTER this wipe, fails to find its
+    // (now-deleted) file, and broadcasts upload-error/upload-deleted into the NEXT test's SSE
+    // stream. (Export workers are already inert across a truncate: they are epoch-guarded on the
+    // event row, and truncate gives the event a fresh random UUID, so their writes match nothing.)
+    state.compression.bump_generation();
+
+    Ok(StatusCode::NO_CONTENT)
+}
+
+/// Returns whether the truncate endpoint is enabled. Used by the e2e harness
+/// during global-setup to fail loud if the test backend was started without
+/// `EVENTSNAP_TEST_MODE=1`.
+pub fn is_test_mode() -> bool {
+    std::env::var("EVENTSNAP_TEST_MODE").as_deref() == Ok("1")
+}

backend/src/handlers/upload.rs

@@ -1,8 +1,9 @@
 use std::time::Duration;
 
+use axum::Json;
 use axum::extract::{Multipart, Path, State};
 use axum::http::StatusCode;
-use axum::Json;
+use chrono::{DateTime, Utc};
 use serde::Deserialize;
 use uuid::Uuid;
 
@@ -11,24 +12,54 @@ use crate::error::AppError;
 use crate::models::hashtag::{self, Hashtag};
 use crate::models::upload::{Upload, UploadDto};
 use crate::models::user::User;
+use crate::services::config;
 use crate::state::AppState;
 
+const MAX_CAPTION_LENGTH: usize = 2000;
+
+/// Allowlist of accepted media types, keyed by the MIME that `infer` derives from
+/// the file's magic bytes. The detected MIME (not the client-declared one) is what
+/// we trust, store, and hand to the compression pipeline — so a text-based payload
+/// (SVG/HTML/JS) can never be stored or served on-origin. Each entry maps to the
+/// server-controlled file extension we persist the original under.
+///
+/// HEIC/HEIF are deliberately excluded: the preview pipeline (`image` crate, and
+/// the bundled ffmpeg 6.1) cannot decode them, so accepting them would store files
+/// that never get a thumbnail. iOS Safari already transcodes HEIC→JPEG when a photo
+/// is selected via a file input, so this rejects only the rare HEIC-preserving
+/// upload path — with a clear error rather than a silently broken post.
+const ALLOWED_MEDIA: &[(&str, &str)] = &[
+    ("image/jpeg", "jpg"),
+    ("image/png", "png"),
+    ("image/webp", "webp"),
+    ("image/gif", "gif"),
+    ("video/mp4", "mp4"),
+    ("video/quicktime", "mov"),
+    ("video/webm", "webm"),
+];
+
 pub async fn upload(
     State(state): State<AppState>,
     auth: AuthUser,
     mut multipart: Multipart,
 ) -> Result<(StatusCode, Json<UploadDto>), AppError> {
-    // Rate limit: N uploads per hour per user
-    let upload_rate = get_config_i64(&state.pool, "upload_rate_per_hour", 10).await as usize;
-    if let Err(retry_after_secs) = state
-        .rate_limiter
-        .check_with_retry(format!("upload:{}", auth.user_id), upload_rate, Duration::from_secs(3600))
-    {
-        drain_multipart(multipart).await;
-        return Err(AppError::TooManyRequests(
-            "Du hast dein Upload-Limit für diese Stunde erreicht.".into(),
-            Some(retry_after_secs),
-        ));
+    // Rate limit: N uploads per hour per user. Gated by master + per-endpoint toggles.
+    let rate_limits_on = config::get_bool(&state.config_cache, "rate_limits_enabled", true).await;
+    let upload_rate_on = config::get_bool(&state.config_cache, "upload_rate_enabled", true).await;
+    if rate_limits_on && upload_rate_on {
+        let upload_rate =
+            config::get_i64(&state.config_cache, "upload_rate_per_hour", 100).await as usize;
+        if let Err(retry_after_secs) = state.rate_limiter.check_with_retry(
+            format!("upload:{}", auth.user_id),
+            upload_rate,
+            Duration::from_secs(3600),
+        ) {
+            drain_multipart(multipart).await;
+            return Err(AppError::TooManyRequests(
+                "Du hast dein Upload-Limit für diese Stunde erreicht.".into(),
+                Some(retry_after_secs),
+            ));
+        }
     }
 
     // Check if user is banned
@@ -46,99 +77,195 @@ pub async fn upload(
         .ok_or_else(|| AppError::NotFound("Event nicht gefunden.".into()))?;
     if event.uploads_locked_at.is_some() {
         drain_multipart(multipart).await;
-        return Err(AppError::Forbidden("Uploads sind gesperrt.".into()));
+        // Reversible: a host can reopen the event, so the client keeps the queued blob and
+        // retries on `event-opened` rather than purging it (UploadsLocked, not Forbidden).
+        return Err(AppError::UploadsLocked("Uploads sind gesperrt.".into()));
+    }
+    // Belt-and-suspenders on top of the lock (release ⇒ lock): once the gallery is
+    // released the export has been snapshotted, so a late upload could never make it into
+    // the keepsake. Reject it explicitly rather than silently diverging the live feed.
+    // Also reversible (reopen clears `export_released_at`), so likewise UploadsLocked.
+    if event.export_released_at.is_some() {
+        drain_multipart(multipart).await;
+        return Err(AppError::UploadsLocked(
+            "Galerie wurde bereits freigegeben.".into(),
+        ));
     }
 
     // Read config limits from DB
-    let max_image_mb: i64 = get_config_i64(&state.pool, "max_image_size_mb", 20).await;
-    let max_video_mb: i64 = get_config_i64(&state.pool, "max_video_size_mb", 500).await;
+    let max_image_mb: i64 = config::get_i64(&state.config_cache, "max_image_size_mb", 20).await;
+    let max_video_mb: i64 = config::get_i64(&state.config_cache, "max_video_size_mb", 500).await;
 
-    let mut file_data: Option<Vec<u8>> = None;
-    let mut file_name: Option<String> = None;
-    let mut content_type: Option<String> = None;
+    // The uploaded file is streamed straight to a temp file on disk (never buffered
+    // whole in memory — a 500 MB video used to cost 500 MB of RAM per concurrent
+    // upload). We only keep the first ≤512 bytes in memory for magic-byte sniffing.
+    // On success the temp file is renamed into place under its detected extension.
+    let upload_id = Uuid::new_v4();
+    let event_slug = &state.config.event_slug;
+    let originals_dir = state
+        .config
+        .media_path
+        .join(format!("originals/{event_slug}"));
+    let temp_abs = originals_dir.join(format!("{upload_id}.tmp"));
+
+    let mut streamed: Option<(i64, Vec<u8>)> = None; // (size, head bytes for sniffing)
     let mut caption: Option<String> = None;
     let mut hashtags_csv: Option<String> = None;
 
-    while let Some(field) = multipart.next_field().await.map_err(|e| AppError::BadRequest(e.to_string()))? {
-        let name = field.name().unwrap_or_default().to_string();
-        match name.as_str() {
-            "file" => {
-                file_name = field.file_name().map(|s| s.to_string());
-                content_type = field.content_type().map(|s| s.to_string());
-                file_data = Some(
-                    field.bytes().await
-                        .map_err(|e| AppError::BadRequest(format!("Datei konnte nicht gelesen werden: {e}")))?
-                        .to_vec(),
-                );
+    // Wrap the multipart read so any error after the temp file is created still cleans
+    // it up (a mid-stream parse failure must not leave a stray `.tmp` on disk).
+    let parse_result: Result<(), AppError> = async {
+        while let Some(field) = multipart
+            .next_field()
+            .await
+            .map_err(|e| AppError::BadRequest(e.to_string()))?
+        {
+            let name = field.name().unwrap_or_default().to_string();
+            match name.as_str() {
+                "file" => {
+                    // The client-declared Content-Type does NOT determine the stored
+                    // MIME/extension — those come from the file's magic bytes below. The
+                    // declared type only picks the streaming cap so an oversized body is
+                    // aborted early; a mislabelled type only makes the cap *stricter*
+                    // (safe), and the authoritative per-class check still runs on the
+                    // detected type.
+                    let declared = field.content_type().unwrap_or("").to_string();
+                    let cap_bytes = if declared.starts_with("video/") {
+                        (max_video_mb * 1024 * 1024) as usize
+                    } else if declared.starts_with("image/") {
+                        (max_image_mb * 1024 * 1024) as usize
+                    } else {
+                        (max_image_mb.max(max_video_mb) * 1024 * 1024) as usize
+                    };
+                    tokio::fs::create_dir_all(&originals_dir)
+                        .await
+                        .map_err(|e| AppError::Internal(e.into()))?;
+                    streamed = Some(stream_field_to_file(field, &temp_abs, cap_bytes).await?);
+                }
+                "caption" => {
+                    caption = Some(
+                        field
+                            .text()
+                            .await
+                            .map_err(|e| AppError::BadRequest(e.to_string()))?,
+                    );
+                }
+                "hashtags" => {
+                    hashtags_csv = Some(
+                        field
+                            .text()
+                            .await
+                            .map_err(|e| AppError::BadRequest(e.to_string()))?,
+                    );
+                }
+                _ => {}
             }
-            "caption" => {
-                caption = Some(
-                    field.text().await
-                        .map_err(|e| AppError::BadRequest(e.to_string()))?,
-                );
-            }
-            "hashtags" => {
-                hashtags_csv = Some(
-                    field.text().await
-                        .map_err(|e| AppError::BadRequest(e.to_string()))?,
-                );
-            }
-            _ => {}
         }
+        Ok(())
+    }
+    .await;
+
+    if let Err(e) = parse_result {
+        let _ = tokio::fs::remove_file(&temp_abs).await;
+        return Err(e);
     }
 
-    let data = file_data.ok_or_else(|| AppError::BadRequest("Keine Datei hochgeladen.".into()))?;
-    let mime = content_type.unwrap_or_else(|| "application/octet-stream".to_string());
-    let size = data.len() as i64;
+    // From here on the temp file may exist; every validation failure removes it before
+    // returning so a rejected upload never leaves bytes behind.
+    let (size, head) = match streamed {
+        Some(s) => s,
+        None => return Err(AppError::BadRequest("Keine Datei hochgeladen.".into())),
+    };
 
-    // Validate file size
+    // Validate caption length. Counted in chars (code points) to match the
+    // "Zeichen" wording in the error message — `.len()` would be bytes and
+    // reject perfectly valid German/emoji captions early.
+    if let Some(ref cap) = caption
+        && cap.chars().count() > MAX_CAPTION_LENGTH
+    {
+        let _ = tokio::fs::remove_file(&temp_abs).await;
+        return Err(AppError::BadRequest(format!(
+            "Beschreibung ist zu lang. Maximum: {} Zeichen.",
+            MAX_CAPTION_LENGTH
+        )));
+    }
+
+    // Determine the file type from its magic bytes and require it to be on the
+    // allowlist. `infer` returns None for text-based payloads (SVG/HTML/JS), so
+    // those are rejected outright — closing the stored-XSS vector. Both the MIME
+    // we persist and the on-disk extension come from the detected type, never from
+    // client-supplied values.
+    let kind = match infer::get(&head) {
+        Some(k) => k,
+        None => {
+            let _ = tokio::fs::remove_file(&temp_abs).await;
+            return Err(AppError::BadRequest(
+                "Dateityp nicht erkannt oder nicht unterstützt.".into(),
+            ));
+        }
+    };
+    let (mime, ext) = match ALLOWED_MEDIA
+        .iter()
+        .find(|(allowed, _)| *allowed == kind.mime_type())
+        .map(|(m, e)| ((*m).to_string(), *e))
+    {
+        Some(v) => v,
+        None => {
+            let _ = tokio::fs::remove_file(&temp_abs).await;
+            return Err(AppError::BadRequest(format!(
+                "Dateityp wird nicht unterstützt: {}.",
+                kind.mime_type()
+            )));
+        }
+    };
+
+    // Validate file size against the authoritative per-detected-class limit.
     let max_bytes = if mime.starts_with("video/") {
         max_video_mb * 1024 * 1024
     } else {
         max_image_mb * 1024 * 1024
     };
     if size > max_bytes {
+        let _ = tokio::fs::remove_file(&temp_abs).await;
         return Err(AppError::BadRequest(format!(
             "Datei ist zu groß. Maximum: {} MB.",
             max_bytes / (1024 * 1024)
         )));
     }
 
-    // Determine file extension
-    let ext = file_name
-        .as_deref()
-        .and_then(|n| n.rsplit('.').next())
-        .unwrap_or(if mime.starts_with("video/") { "mp4" } else { "jpg" });
+    // Per-user storage quota — dynamic formula based on available disk space and the
+    // number of active uploaders. Gated by master + per-area toggles so the admin can
+    // disable it on trusted instances.
+    let quota_on = config::get_bool(&state.config_cache, "quota_enabled", true).await;
+    let storage_quota_on =
+        config::get_bool(&state.config_cache, "storage_quota_enabled", true).await;
+    // When quota is enforced, this holds the byte ceiling so the increment UPDATE below can
+    // enforce it atomically (`WHERE total + size <= limit`). Without that guard, two
+    // concurrent uploads from the same user (e.g. phone + laptop) both pass this stale
+    // pre-check and both increment, blowing past the quota. The pre-check stays as a
+    // fast path that avoids the disk write when the user is already clearly over.
+    let mut quota_limit: Option<i64> = None;
+    if quota_on && storage_quota_on {
+        let estimate = compute_storage_quota(&state).await;
+        if let Some(limit) = estimate.limit_bytes {
+            quota_limit = Some(limit);
+            let prospective_total = user.total_upload_bytes.saturating_add(size);
+            if prospective_total > limit {
+                let _ = tokio::fs::remove_file(&temp_abs).await;
+                return Err(AppError::QuotaExceeded(
+                    "Du hast dein Upload-Limit für dieses Event erreicht.".into(),
+                ));
+            }
+        }
+    }
 
-    let upload_id = Uuid::new_v4();
-    let event_slug = &state.config.event_slug;
+    // All checks passed — atomically move the temp file to its final, extension-correct
+    // path (same directory, so the rename is cheap and atomic).
     let relative_path = format!("originals/{event_slug}/{upload_id}.{ext}");
     let absolute_path = state.config.media_path.join(&relative_path);
-
-    // Ensure directory exists and write file
-    if let Some(parent) = absolute_path.parent() {
-        tokio::fs::create_dir_all(parent).await.map_err(|e| AppError::Internal(e.into()))?;
-    }
-    tokio::fs::write(&absolute_path, &data).await.map_err(|e| AppError::Internal(e.into()))?;
-
-    // Update user's total upload bytes
-    sqlx::query("UPDATE \"user\" SET total_upload_bytes = total_upload_bytes + $2 WHERE id = $1")
-        .bind(auth.user_id)
-        .bind(size)
-        .execute(&state.pool)
-        .await?;
-
-    // Insert upload record
-    let upload = Upload::create(
-        &state.pool,
-        auth.event_id,
-        auth.user_id,
-        &relative_path,
-        &mime,
-        size,
-        caption.as_deref(),
-    )
-    .await?;
+    tokio::fs::rename(&temp_abs, &absolute_path)
+        .await
+        .map_err(|e| AppError::Internal(e.into()))?;
 
     // Process hashtags from caption and explicit CSV
     let mut tags: Vec<String> = Vec::new();
@@ -156,10 +283,98 @@ pub async fn upload(
     tags.sort();
     tags.dedup();
 
-    for tag in &tags {
-        let h = Hashtag::upsert(&state.pool, auth.event_id, tag).await?;
-        Hashtag::link_to_upload(&state.pool, upload.id, h.id).await?;
+    // Quota accounting, the upload row, and its hashtag links must be atomic: a
+    // crash between the bytes increment and the insert would permanently charge
+    // bytes with no row to reclaim them (silent quota erosion / spurious lockout).
+    let tx_result: Result<Upload, AppError> = async {
+        let mut tx = state.pool.begin().await?;
+
+        // RE-CHECK THE LOCK, UNDER A ROW LOCK, INSIDE THE COMMIT TX.
+        //
+        // The pre-flight check at the top of this handler ran BEFORE we streamed the body — which
+        // for a 500 MB video is minutes. Trusting it here is a TOCTOU that silently loses photos
+        // from the keepsake, and it is the real cause of the "stale keepsake" bug that survived
+        // three rounds of fixes inside the export state machine:
+        //
+        //   1. guest starts a big upload; the lock check passes (event open)
+        //   2. host releases the gallery → uploads lock, export workers snapshot the uploads table
+        //   3. this upload commits AFTER that snapshot → it shows up in the live feed but is
+        //      MISSING from the downloaded keepsake, permanently (nothing ever regenerates it)
+        //
+        // `FOR SHARE` conflicts with the `UPDATE event` in `release_gallery`, which serializes us
+        // against it. Either we take the lock first — and release (hence the export snapshot) is
+        // strictly ordered after our commit, so the snapshot CONTAINS this upload — or release
+        // commits first and we observe the lock here and reject. Either way the keepsake is
+        // complete. `UploadsLocked` (not Forbidden) is reversible: the client keeps the blob and
+        // resumes it when the host reopens.
+        let (locked_at, released_at): (Option<DateTime<Utc>>, Option<DateTime<Utc>>) =
+            sqlx::query_as(
+                "SELECT uploads_locked_at, export_released_at FROM event WHERE id = $1 FOR SHARE",
+            )
+            .bind(auth.event_id)
+            .fetch_one(&mut *tx)
+            .await?;
+        if locked_at.is_some() || released_at.is_some() {
+            return Err(AppError::UploadsLocked("Uploads sind gesperrt.".into()));
+        }
+
+        // Increment the user's byte total. When a quota is in force, guard it atomically
+        // (`total + size <= limit`) so two concurrent uploads can't both slip past the
+        // stale pre-check — the loser's UPDATE matches 0 rows and we abort with the same
+        // terminal quota error (the tx rolls back on drop; the on-disk file is cleaned by
+        // the error path below).
+        let inc = if let Some(limit) = quota_limit {
+            sqlx::query(
+                "UPDATE \"user\" SET total_upload_bytes = total_upload_bytes + $2
+                 WHERE id = $1 AND total_upload_bytes + $2 <= $3",
+            )
+            .bind(auth.user_id)
+            .bind(size)
+            .bind(limit)
+            .execute(&mut *tx)
+            .await?
+        } else {
+            sqlx::query(
+                "UPDATE \"user\" SET total_upload_bytes = total_upload_bytes + $2 WHERE id = $1",
+            )
+            .bind(auth.user_id)
+            .bind(size)
+            .execute(&mut *tx)
+            .await?
+        };
+        if inc.rows_affected() == 0 {
+            return Err(AppError::QuotaExceeded(
+                "Du hast dein Upload-Limit für dieses Event erreicht.".into(),
+            ));
+        }
+        let upload = Upload::create(
+            &mut *tx,
+            auth.event_id,
+            auth.user_id,
+            &relative_path,
+            &mime,
+            size,
+            caption.as_deref(),
+        )
+        .await?;
+        for tag in &tags {
+            let h = Hashtag::upsert(&mut *tx, auth.event_id, tag).await?;
+            Hashtag::link_to_upload(&mut *tx, upload.id, h.id).await?;
+        }
+        tx.commit().await?;
+        Ok(upload)
     }
+    .await;
+
+    // The file is already on disk at `absolute_path`. If the transaction failed, no DB
+    // row will ever reference it, so remove it now rather than orphan bytes on disk.
+    let upload = match tx_result {
+        Ok(u) => u,
+        Err(e) => {
+            let _ = tokio::fs::remove_file(&absolute_path).await;
+            return Err(e);
+        }
+    };
 
     // Spawn compression task
     state
@@ -182,10 +397,10 @@ pub async fn upload(
         created_at: upload.created_at,
     };
 
-    let _ = state.sse_tx.send(crate::state::SseEvent {
-        event_type: "new-upload".to_string(),
-        data: serde_json::to_string(&dto).unwrap_or_default(),
-    });
+    let _ = state.sse_tx.send(crate::state::SseEvent::new(
+        "new-upload",
+        serde_json::to_string(&dto).unwrap_or_default(),
+    ));
 
     Ok((StatusCode::CREATED, Json(dto)))
 }
@@ -202,7 +417,11 @@ pub async fn edit_upload(
     Path(upload_id): Path<Uuid>,
     Json(body): Json<EditUploadRequest>,
 ) -> Result<StatusCode, AppError> {
-    let upload = Upload::find_by_id(&state.pool, upload_id)
+    // Banned users keep read access but cannot mutate (USER_JOURNEYS §10).
+    if auth.is_banned {
+        return Err(AppError::Forbidden("Du bist gesperrt.".into()));
+    }
+    let upload = Upload::find_by_id_and_event(&state.pool, upload_id, auth.event_id)
         .await?
         .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
 
@@ -210,17 +429,38 @@ pub async fn edit_upload(
         return Err(AppError::Forbidden("Nur eigene Uploads bearbeiten.".into()));
     }
 
+    // Caption update + hashtag wipe-then-relink in one transaction, so a crash
+    // mid-relink can't leave the upload with its hashtags stripped.
+    //
+    // Editing is intentionally allowed while uploads are locked or the gallery is released — like
+    // comments and likes, the lock freezes *new uploads* only (USER_JOURNEYS §9.3). But a caption
+    // is embedded in the HTML viewer keepsake (the ZIP holds media only — see export.rs), so an
+    // edit AFTER release must regenerate the viewer, or the downloadable keepsake keeps showing the
+    // old caption forever while the live feed shows the new one. Same atomicity as delete_upload:
+    // the edit and its invalidation share one tx so a dropped handler can't leave them disagreeing.
+    // `Affects::ViewerOnly` carries the finished ZIP forward (the media didn't change); when the
+    // gallery isn't released, `invalidate_and_arm` returns None and this is a no-op.
+    let mut tx = state.pool.begin().await?;
     if let Some(ref caption) = body.caption {
-        Upload::update_caption(&state.pool, upload_id, Some(caption)).await?;
+        Upload::update_caption(&mut *tx, upload_id, Some(caption)).await?;
     }
-
     if let Some(ref hashtags) = body.hashtags {
-        Hashtag::unlink_all_from_upload(&state.pool, upload_id).await?;
+        Hashtag::unlink_all_from_upload(&mut *tx, upload_id).await?;
         for tag in hashtags {
-            let h = Hashtag::upsert(&state.pool, auth.event_id, tag).await?;
-            Hashtag::link_to_upload(&state.pool, upload_id, h.id).await?;
+            let h = Hashtag::upsert(&mut *tx, auth.event_id, tag).await?;
+            Hashtag::link_to_upload(&mut *tx, upload_id, h.id).await?;
         }
     }
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        crate::services::export::Affects::ViewerOnly,
+    )
+    .await?;
+    tx.commit().await?;
+    if let Some(r) = regen {
+        crate::handlers::host::start_regen(&state, r);
+    }
 
     Ok(StatusCode::OK)
 }
@@ -230,7 +470,11 @@ pub async fn delete_upload(
     auth: AuthUser,
     Path(upload_id): Path<Uuid>,
 ) -> Result<StatusCode, AppError> {
-    let upload = Upload::find_by_id(&state.pool, upload_id)
+    // Banned users keep read access but cannot mutate (USER_JOURNEYS §10).
+    if auth.is_banned {
+        return Err(AppError::Forbidden("Du bist gesperrt.".into()));
+    }
+    let upload = Upload::find_by_id_and_event(&state.pool, upload_id, auth.event_id)
         .await?
         .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
 
@@ -238,11 +482,96 @@ pub async fn delete_upload(
         return Err(AppError::Forbidden("Nur eigene Uploads löschen.".into()));
     }
 
-    Upload::soft_delete(&state.pool, upload_id).await?;
+    // Atomic with the keepsake invalidation: a guest removing their own photo must have it removed
+    // from the downloadable archive too, and a half-applied delete would leave it there forever.
+    let mut tx = state.pool.begin().await?;
+    Upload::soft_delete_in_event(&mut tx, upload_id, auth.event_id).await?;
+    let regen = crate::services::export::invalidate_and_arm(
+        &mut tx,
+        &state.config.event_slug,
+        crate::services::export::Affects::Both,
+    )
+    .await?;
+    tx.commit().await?;
+    if let Some(r) = regen {
+        crate::handlers::host::start_regen(&state, r);
+    }
+
+    // Evict the card live on every other feed + the projector diashow — otherwise
+    // a self-deleted post lingers until each viewer manually reloads. Same event
+    // the host-delete path already emits and the frontend already handles.
+    let _ = state.sse_tx.send(crate::state::SseEvent::new(
+        "upload-deleted",
+        serde_json::json!({ "upload_id": upload_id }).to_string(),
+    ));
 
     Ok(StatusCode::NO_CONTENT)
 }
 
+/// Number of leading bytes retained in memory for magic-byte (`infer`) sniffing. Every
+/// allowed type's signature sits well within this; 512 is comfortably generous.
+const HEAD_SNIFF_BYTES: usize = 512;
+
+/// Stream a multipart field straight to `dest`, aborting with a 400 the moment it
+/// exceeds `max_bytes`. Only the first [`HEAD_SNIFF_BYTES`] bytes are kept in memory
+/// (for type detection); the rest goes chunk-by-chunk to disk, so peak memory is a
+/// single chunk rather than the whole file. Returns `(total_size, head_bytes)`. On any
+/// error the partial temp file is removed so no stray `.tmp` is left behind.
+async fn stream_field_to_file(
+    mut field: axum::extract::multipart::Field<'_>,
+    dest: &std::path::Path,
+    max_bytes: usize,
+) -> Result<(i64, Vec<u8>), AppError> {
+    use tokio::io::AsyncWriteExt;
+
+    let mut file = tokio::fs::File::create(dest)
+        .await
+        .map_err(|e| AppError::Internal(e.into()))?;
+    let mut total: usize = 0;
+    let mut head: Vec<u8> = Vec::with_capacity(HEAD_SNIFF_BYTES);
+
+    loop {
+        let chunk = match field.chunk().await {
+            Ok(Some(c)) => c,
+            Ok(None) => break,
+            Err(e) => {
+                let _ = file.shutdown().await;
+                let _ = tokio::fs::remove_file(dest).await;
+                return Err(AppError::BadRequest(format!(
+                    "Datei konnte nicht gelesen werden: {e}"
+                )));
+            }
+        };
+
+        total = total.saturating_add(chunk.len());
+        if total > max_bytes {
+            let _ = file.shutdown().await;
+            let _ = tokio::fs::remove_file(dest).await;
+            return Err(AppError::BadRequest(format!(
+                "Datei ist zu groß. Maximum: {} MB.",
+                max_bytes / (1024 * 1024)
+            )));
+        }
+
+        if head.len() < HEAD_SNIFF_BYTES {
+            let need = HEAD_SNIFF_BYTES - head.len();
+            head.extend_from_slice(&chunk[..need.min(chunk.len())]);
+        }
+
+        if let Err(e) = file.write_all(&chunk).await {
+            let _ = tokio::fs::remove_file(dest).await;
+            return Err(AppError::Internal(e.into()));
+        }
+    }
+
+    if let Err(e) = file.flush().await {
+        let _ = tokio::fs::remove_file(dest).await;
+        return Err(AppError::Internal(e.into()));
+    }
+
+    Ok((total as i64, head))
+}
+
 /// Drain a multipart body so the HTTP connection stays clean when returning an early error.
 /// Without draining, the client may still be sending the body after we've sent our response,
 /// which can corrupt the keep-alive connection for subsequent requests.
@@ -252,12 +581,248 @@ async fn drain_multipart(mut mp: Multipart) {
     }
 }
 
-async fn get_config_i64(pool: &sqlx::PgPool, key: &str, default: i64) -> i64 {
-    let row: Option<(String,)> =
-        sqlx::query_as("SELECT value FROM config WHERE key = $1")
-            .bind(key)
-            .fetch_optional(pool)
-            .await
-            .unwrap_or(None);
-    row.and_then(|r| r.0.parse().ok()).unwrap_or(default)
+/// Snapshot of the dynamic per-user quota used both by the upload pre-check and the
+/// `GET /me/quota` endpoint. `limit_bytes = None` means quota enforcement is currently
+/// off (the frontend hides the widget in that case).
+pub struct QuotaEstimate {
+    pub limit_bytes: Option<i64>,
+    pub active_uploaders: i64,
+    pub free_disk_bytes: i64,
+    /// The tolerance factor the limit above was computed with. Carried on the snapshot so the
+    /// number is self-describing; no caller reads it back today.
+    #[allow(dead_code)]
+    pub tolerance: f64,
+}
+
+/// Pure per-user quota formula: `floor((free_disk * tolerance) / max(active, 1))`.
+/// Extracted from `compute_storage_quota` so it's unit-testable without a DB or disk.
+fn quota_limit_bytes(free_disk: i64, tolerance: f64, active_uploaders: i64) -> i64 {
+    let active = active_uploaders.max(1);
+    ((free_disk as f64 * tolerance) / active as f64).floor() as i64
+}
+
+/// Computes the per-user storage quota using
+/// `floor((free_disk * tolerance) / max(active_uploaders, 1))`. Returns `limit_bytes =
+/// None` whenever the storage quota is currently disabled — callers should skip the
+/// check (upload handler) or hide the UI (quota endpoint).
+pub async fn compute_storage_quota(state: &AppState) -> QuotaEstimate {
+    let quota_on = config::get_bool(&state.config_cache, "quota_enabled", true).await;
+    let storage_quota_on =
+        config::get_bool(&state.config_cache, "storage_quota_enabled", true).await;
+    let tolerance = config::get_f64(&state.config_cache, "quota_tolerance", 0.75).await;
+
+    let (active_count,): (i64,) =
+        sqlx::query_as("SELECT COUNT(DISTINCT user_id) FROM upload WHERE deleted_at IS NULL")
+            .fetch_one(&state.pool)
+            .await
+            .unwrap_or((0,));
+    let active = active_count.max(1);
+
+    // Cached disk reading. `None` means we couldn't resolve the media filesystem.
+    let disk = state.disk_cache.snapshot(&state.config.media_path);
+    let free_disk = disk.map(|d| d.free as i64).unwrap_or(0);
+
+    let limit_bytes = if quota_on && storage_quota_on {
+        match disk {
+            Some(d) => Some(quota_limit_bytes(d.free as i64, tolerance, active)),
+            // Fail OPEN, not closed: if the disk can't be read we don't know the real
+            // free space, and enforcing a 0-byte limit would reject every upload with a
+            // spurious "quota reached". Skip enforcement this round and warn instead.
+            None => {
+                tracing::warn!(
+                    "disk snapshot unavailable; skipping storage-quota enforcement this round"
+                );
+                None
+            }
+        }
+    } else {
+        None
+    };
+
+    QuotaEstimate {
+        limit_bytes,
+        active_uploaders: active,
+        free_disk_bytes: free_disk,
+        tolerance,
+    }
+}
+
+/// Stream a media file from disk into an HTTP response with a fixed set of security
+/// headers. Every media response (original, preview, thumbnail) goes through here so
+/// they consistently carry `X-Content-Type-Options: nosniff` (defense-in-depth against
+/// content-type confusion, even if the edge proxy is bypassed) plus an explicit
+/// `Content-Disposition` and `Cache-Control`.
+async fn stream_media_file(
+    absolute: &std::path::Path,
+    content_type: String,
+    disposition: &str,
+    cache_control: &str,
+) -> Result<axum::response::Response, AppError> {
+    use axum::body::Body;
+    use axum::http::{Response, StatusCode, header};
+    use tokio_util::io::ReaderStream;
+
+    if !absolute.exists() {
+        return Err(AppError::NotFound("Datei nicht gefunden.".into()));
+    }
+
+    let file = tokio::fs::File::open(absolute)
+        .await
+        .map_err(|e| AppError::Internal(e.into()))?;
+    let metadata = file
+        .metadata()
+        .await
+        .map_err(|e| AppError::Internal(e.into()))?;
+    let stream = ReaderStream::new(file);
+
+    Response::builder()
+        .status(StatusCode::OK)
+        .header(header::CONTENT_TYPE, content_type)
+        .header(header::CONTENT_DISPOSITION, disposition)
+        .header(header::CONTENT_LENGTH, metadata.len())
+        .header(header::CACHE_CONTROL, cache_control)
+        .header(header::X_CONTENT_TYPE_OPTIONS, "nosniff")
+        .body(Body::from_stream(stream))
+        .map_err(|e| AppError::Internal(e.into()))
+}
+
+/// Streaming download of the original file behind an upload. Used by:
+///   - the per-post "Original anzeigen" context action (`window.open`)
+///   - `<img src>` / `<video src>` in the feed, lightbox, and diashow when the user is in
+///     Data Mode = Original
+///
+/// **Auth model:** the route is intentionally unauthenticated so it works from
+/// `<img src>` / `window.open`. The URL contains the upload's unguessable UUID. Unlike
+/// raw `/media` files, this alias is the *only* way to fetch an original: direct
+/// `/media/originals/**` access is blocked in the router, and this handler filters out
+/// soft-deleted and ban-hidden uploads (via `find_visible_media`) so moderation actually
+/// removes access to content. Preview and thumbnail variants are gated the same way (see
+/// [`get_preview`] / [`get_thumbnail`]).
+pub async fn get_original(
+    State(state): State<AppState>,
+    Path(upload_id): Path<Uuid>,
+) -> Result<axum::response::Response, AppError> {
+    let media = Upload::find_visible_media(&state.pool, upload_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
+
+    let absolute = state.config.media_path.join(&media.original_path);
+    let filename = absolute
+        .file_name()
+        .and_then(|n| n.to_str())
+        .unwrap_or("original");
+    let disposition = format!("attachment; filename=\"{filename}\"");
+
+    // Full-res original: force download, never cache at the edge.
+    stream_media_file(&absolute, media.mime_type, &disposition, "no-store").await
+}
+
+/// Streaming access to an upload's compressed **preview** image. Gated exactly like
+/// [`get_original`]: `find_visible_media` drops soft-deleted / ban-hidden uploads, so a
+/// deleted or moderated post's preview 404s here even though the file may still be on
+/// disk. Direct `/media/previews/**` is blocked in the router, making this the only path
+/// to a preview.
+///
+/// Served inline (it's an `<img src>`) and privately cacheable for a short window. The
+/// window is intentionally short (5 min) so a moderated image stops being served to a
+/// direct-URL holder promptly; the live feed already evicts the card instantly via the
+/// `upload-deleted` / `user-hidden` SSE events, so this only bounds the raw-URL edge case.
+pub async fn get_preview(
+    State(state): State<AppState>,
+    Path(upload_id): Path<Uuid>,
+) -> Result<axum::response::Response, AppError> {
+    let media = Upload::find_visible_media(&state.pool, upload_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
+    let rel = media
+        .preview_path
+        .ok_or_else(|| AppError::NotFound("Vorschau nicht verfügbar.".into()))?;
+    let absolute = state.config.media_path.join(&rel);
+    stream_media_file(
+        &absolute,
+        "image/jpeg".to_string(),
+        "inline",
+        "private, max-age=300",
+    )
+    .await
+}
+
+/// Streaming access to an upload's big-screen **display** derivative (~2048px), used by the
+/// diashow. Gated identically to [`get_preview`]. 404s when the derivative doesn't exist yet
+/// (still compressing, or an old upload the backfill hasn't reached) — the diashow then falls
+/// back to the original.
+pub async fn get_display(
+    State(state): State<AppState>,
+    Path(upload_id): Path<Uuid>,
+) -> Result<axum::response::Response, AppError> {
+    let media = Upload::find_visible_media(&state.pool, upload_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
+    let rel = media
+        .display_path
+        .ok_or_else(|| AppError::NotFound("Anzeige nicht verfügbar.".into()))?;
+    let absolute = state.config.media_path.join(&rel);
+    stream_media_file(
+        &absolute,
+        "image/jpeg".to_string(),
+        "inline",
+        "private, max-age=300",
+    )
+    .await
+}
+
+/// Streaming access to an upload's **thumbnail** (video poster). Gated identically to
+/// [`get_preview`].
+pub async fn get_thumbnail(
+    State(state): State<AppState>,
+    Path(upload_id): Path<Uuid>,
+) -> Result<axum::response::Response, AppError> {
+    let media = Upload::find_visible_media(&state.pool, upload_id)
+        .await?
+        .ok_or_else(|| AppError::NotFound("Upload nicht gefunden.".into()))?;
+    let rel = media
+        .thumbnail_path
+        .ok_or_else(|| AppError::NotFound("Thumbnail nicht verfügbar.".into()))?;
+    let absolute = state.config.media_path.join(&rel);
+    stream_media_file(
+        &absolute,
+        "image/jpeg".to_string(),
+        "inline",
+        "private, max-age=300",
+    )
+    .await
+}
+
+#[cfg(test)]
+mod tests {
+    use super::quota_limit_bytes;
+
+    #[test]
+    fn divides_free_space_by_uploaders_with_tolerance() {
+        // 1000 * 0.75 / 3 = 250
+        assert_eq!(quota_limit_bytes(1000, 0.75, 3), 250);
+    }
+
+    #[test]
+    fn floors_fractional_results() {
+        // 1000 * 0.75 / 7 = 107.14… → 107
+        assert_eq!(quota_limit_bytes(1000, 0.75, 7), 107);
+    }
+
+    #[test]
+    fn active_uploaders_below_one_is_clamped_to_one() {
+        // Guards against divide-by-zero when no one has uploaded yet.
+        assert_eq!(quota_limit_bytes(1000, 1.0, 0), 1000);
+        assert_eq!(quota_limit_bytes(1000, 1.0, -5), 1000);
+    }
+
+    #[test]
+    fn zero_free_disk_yields_zero() {
+        assert_eq!(quota_limit_bytes(0, 0.75, 3), 0);
+    }
+
+    #[test]
+    fn full_tolerance_is_identity_for_a_single_uploader() {
+        assert_eq!(quota_limit_bytes(500, 1.0, 1), 500);
+    }
 }

backend/src/main.rs

@@ -1,7 +1,7 @@
 use anyhow::Result;
+use axum::Router;
 use axum::extract::DefaultBodyLimit;
 use axum::routing::{delete, get, patch, post};
-use axum::Router;
 use tower_http::services::ServeDir;
 use tower_http::trace::TraceLayer;
 use tracing_subscriber::{layer::SubscriberExt, util::SubscriberInitExt};
@@ -18,63 +18,187 @@ mod state;
 use config::AppConfig;
 use state::AppState;
 
+/// Hard HTTP body cap for the upload endpoint (576 MiB). Backstop against
+/// memory-exhaustion; precise per-class size limits are enforced in the handler.
+const MAX_UPLOAD_BYTES: usize = 576 * 1024 * 1024;
+
 #[tokio::main]
 async fn main() -> Result<()> {
     dotenvy::dotenv().ok();
 
     tracing_subscriber::registry()
-        .with(tracing_subscriber::EnvFilter::try_from_default_env().unwrap_or_else(|_| {
-            "eventsnap_backend=debug,tower_http=debug".into()
-        }))
+        .with(
+            tracing_subscriber::EnvFilter::try_from_default_env()
+                .unwrap_or_else(|_| "eventsnap_backend=debug,tower_http=debug".into()),
+        )
         .with(tracing_subscriber::fmt::layer())
         .init();
 
     let config = AppConfig::from_env()?;
     let pool = db::create_pool(&config.database_url).await?;
-    let state = AppState::new(pool, config.clone());
+
+    // Reset any rows left mid-flight by a previous (possibly crashed) instance —
+    // stuck `compression_status='processing'` uploads and `status='running'` export
+    // jobs. Must run before the server starts taking requests so clients never see
+    // the half-state.
+    services::maintenance::startup_recovery(&pool).await;
+
+    let state = AppState::new(pool.clone(), config.clone());
+
+    // Backfill the big-screen display derivative for image uploads processed before it
+    // existed (v0.17.x). Fire-and-forget behind the compression semaphore; each upload
+    // falls back to its original in the diashow until its display is generated.
+    state.compression.backfill_missing_display().await;
+
+    // Re-spawn exports for events that were released but whose keepsake never finished
+    // (crash mid-export). Needs the media/export paths + SSE sender, so it runs here
+    // rather than inside `startup_recovery`. Fire-and-forget: the workers run in the
+    // background; the HTTP server can start accepting requests meanwhile.
+    services::export::recover_exports(
+        pool.clone(),
+        config.media_path.clone(),
+        config.export_path.clone(),
+        config.comments_enabled,
+        state.sse_tx.clone(),
+    )
+    .await;
+
+    // Hourly background hygiene: prune expired sessions, evict cold rate-limiter
+    // keys. Keeps the DB and process from growing unboundedly over multi-day events.
+    services::maintenance::spawn_periodic_tasks(
+        pool,
+        state.rate_limiter.clone(),
+        state.sse_tickets.clone(),
+    );
 
     // Ensure media directories exist
     tokio::fs::create_dir_all(&config.media_path).await.ok();
 
     let api = Router::new()
         // Auth
+        .route("/api/v1/event", get(handlers::public::get_public_event))
         .route("/api/v1/join", post(auth::handlers::join))
         .route("/api/v1/recover", post(auth::handlers::recover))
+        // Forgotten-PIN escape hatch: ask a host to reset it (unauthenticated, throttled).
+        .route(
+            "/api/v1/recover/request",
+            post(auth::handlers::request_pin_reset),
+        )
         .route("/api/v1/admin/login", post(auth::handlers::admin_login))
         .route("/api/v1/session", delete(auth::handlers::logout))
-        // Upload — body limit disabled; size validation is done inside the handler
-        .route("/api/v1/upload", post(handlers::upload::upload)
-            .route_layer(DefaultBodyLimit::disable()))
+        // "Sign out everywhere" — revoke all of the caller's sessions.
+        .route("/api/v1/sessions", delete(auth::handlers::logout_all))
+        // Upload — HTTP-level body cap as an OOM backstop. The handler still enforces
+        // the precise per-class limits from DB config (max_image/video_size_mb); this
+        // layer just stops a multi-GB body from being buffered into memory before that
+        // check runs. Sized generously above the default 500 MB video limit + multipart
+        // overhead — if an admin raises max_video_size_mb above this, bump MAX_UPLOAD_BYTES.
+        .route(
+            "/api/v1/upload",
+            post(handlers::upload::upload).route_layer(DefaultBodyLimit::max(MAX_UPLOAD_BYTES)),
+        )
         .route(
             "/api/v1/upload/{id}",
             patch(handlers::upload::edit_upload).delete(handlers::upload::delete_upload),
         )
+        .route(
+            "/api/v1/upload/{id}/original",
+            get(handlers::upload::get_original),
+        )
+        // Preview/thumbnail variants are gated the same way as originals (visibility
+        // check + direct /media block below) so moderation actually revokes access to
+        // the displayed images, not just the full-res download.
+        .route(
+            "/api/v1/upload/{id}/preview",
+            get(handlers::upload::get_preview),
+        )
+        .route(
+            "/api/v1/upload/{id}/display",
+            get(handlers::upload::get_display),
+        )
+        .route(
+            "/api/v1/upload/{id}/thumbnail",
+            get(handlers::upload::get_thumbnail),
+        )
+        // Current-user endpoints (live quota estimate, profile + privacy note bundle)
+        .route("/api/v1/me/context", get(handlers::me::get_context))
+        .route("/api/v1/me/quota", get(handlers::me::get_quota))
         // Feed
         .route("/api/v1/feed", get(handlers::feed::feed))
         .route("/api/v1/feed/delta", get(handlers::feed::feed_delta))
         .route("/api/v1/hashtags", get(handlers::feed::hashtags))
         // Social
-        .route("/api/v1/upload/{id}/like", post(handlers::social::toggle_like))
+        .route(
+            "/api/v1/upload/{id}/like",
+            post(handlers::social::toggle_like),
+        )
         .route(
             "/api/v1/upload/{id}/comments",
             get(handlers::social::list_comments).post(handlers::social::add_comment),
         )
-        .route("/api/v1/comment/{id}", delete(handlers::social::delete_comment))
+        .route(
+            "/api/v1/comment/{id}",
+            delete(handlers::social::delete_comment),
+        )
         // SSE
         .route("/api/v1/stream", get(handlers::sse::stream))
+        .route("/api/v1/stream/ticket", post(handlers::sse::issue_ticket))
         // Host Dashboard
         .route("/api/v1/host/event", get(handlers::host::get_event_status))
-        .route("/api/v1/host/event/close", post(handlers::host::close_event))
+        .route(
+            "/api/v1/host/event/close",
+            post(handlers::host::close_event),
+        )
         .route("/api/v1/host/event/open", post(handlers::host::open_event))
-        .route("/api/v1/host/gallery/release", post(handlers::host::release_gallery))
+        .route(
+            "/api/v1/host/gallery/release",
+            post(handlers::host::release_gallery),
+        )
+        // Escape hatch: force a keepsake rebuild. Without it a failed export is terminal at runtime
+        // (release_gallery refuses an already-released event; recovery only runs at boot).
+        .route(
+            "/api/v1/host/export/rebuild",
+            post(handlers::host::rebuild_export),
+        )
         .route("/api/v1/host/users", get(handlers::host::list_users))
-        .route("/api/v1/host/users/{id}/ban", post(handlers::host::ban_user))
-        .route("/api/v1/host/users/{id}/unban", post(handlers::host::unban_user))
-        .route("/api/v1/host/users/{id}/role", patch(handlers::host::set_role))
-        .route("/api/v1/host/upload/{id}", delete(handlers::host::host_delete_upload))
-        .route("/api/v1/host/comment/{id}", delete(handlers::host::host_delete_comment))
+        .route(
+            "/api/v1/host/users/{id}/ban",
+            post(handlers::host::ban_user),
+        )
+        .route(
+            "/api/v1/host/users/{id}/unban",
+            post(handlers::host::unban_user),
+        )
+        .route(
+            "/api/v1/host/users/{id}/role",
+            patch(handlers::host::set_role),
+        )
+        .route(
+            "/api/v1/host/users/{id}/pin-reset",
+            post(handlers::host::reset_user_pin),
+        )
+        .route(
+            "/api/v1/host/pin-reset-requests",
+            get(handlers::host::list_pin_reset_requests),
+        )
+        .route(
+            "/api/v1/host/pin-reset-requests/{id}",
+            delete(handlers::host::dismiss_pin_reset_request),
+        )
+        .route(
+            "/api/v1/host/upload/{id}",
+            delete(handlers::host::host_delete_upload),
+        )
+        .route(
+            "/api/v1/host/comment/{id}",
+            delete(handlers::host::host_delete_comment),
+        )
         // Export (all authenticated users)
         .route("/api/v1/export/status", get(handlers::admin::export_status))
+        .route(
+            "/api/v1/export/ticket",
+            post(handlers::admin::export_ticket),
+        )
         .route("/api/v1/export/zip", get(handlers::admin::download_zip))
         .route("/api/v1/export/html", get(handlers::admin::download_html))
         // Admin Dashboard
@@ -83,7 +207,27 @@ async fn main() -> Result<()> {
             "/api/v1/admin/config",
             get(handlers::admin::get_config).patch(handlers::admin::patch_config),
         )
-        .route("/api/v1/admin/export/jobs", get(handlers::admin::get_export_jobs));
+        .route(
+            "/api/v1/admin/export/jobs",
+            get(handlers::admin::get_export_jobs),
+        );
+
+    // Test-only route: a hard reset for the Playwright E2E harness. The handler
+    // is compiled in always, but the route is only attached when
+    // `EVENTSNAP_TEST_MODE=1`. In production the call returns 404 — the route
+    // simply isn't there.
+    let api = if handlers::test_admin::is_test_mode() {
+        tracing::warn!(
+            "EVENTSNAP_TEST_MODE=1 — registering /api/v1/admin/__truncate. \
+             DO NOT enable this in production."
+        );
+        api.route(
+            "/api/v1/admin/__truncate",
+            post(handlers::test_admin::truncate_all),
+        )
+    } else {
+        api
+    };
 
     // Serve media files from disk
     let media_service = ServeDir::new(&config.media_path);
@@ -91,13 +235,99 @@ async fn main() -> Result<()> {
     let router = Router::new()
         .route("/health", get(|| async { "ok" }))
         .merge(api)
+        // Block direct HTTP access to ALL media subtrees. The files live under
+        // `media_path` (so the compression worker and export can read them off disk) but
+        // must NOT be pullable straight from `/media/**` — that bypasses the visibility
+        // checks (soft-delete + ban-hide) in the gated handlers. Every legitimate fetch
+        // goes through `/api/v1/upload/{id}/{original,preview,thumbnail}`, which filter
+        // via `find_visible_media`. The more specific nests take precedence over the
+        // `/media` ServeDir below (which, with all three subtrees blocked, now serves
+        // nothing — kept as a backstop).
+        .nest_service(
+            "/media/originals",
+            get(|| async { axum::http::StatusCode::NOT_FOUND }),
+        )
+        .nest_service(
+            "/media/previews",
+            get(|| async { axum::http::StatusCode::NOT_FOUND }),
+        )
+        .nest_service(
+            "/media/displays",
+            get(|| async { axum::http::StatusCode::NOT_FOUND }),
+        )
+        .nest_service(
+            "/media/thumbnails",
+            get(|| async { axum::http::StatusCode::NOT_FOUND }),
+        )
         .nest_service("/media", media_service)
         .layer(TraceLayer::new_for_http())
         .with_state(state);
 
     let listener = tokio::net::TcpListener::bind(("0.0.0.0", config.app_port)).await?;
     tracing::info!("listening on {}", listener.local_addr()?);
-    axum::serve(listener, router).await?;
+    axum::serve(listener, router)
+        .with_graceful_shutdown(shutdown_signal())
+        .await?;
 
     Ok(())
 }
+
+/// Hard cap on how long we wait for in-flight connections to drain after a shutdown
+/// signal. Uploads (streamed to disk) finish in well under this; the cap exists because
+/// long-lived SSE streams never end on their own and would otherwise keep the graceful
+/// drain — and thus the process — pending until the orchestrator force-kills it.
+const SHUTDOWN_GRACE: std::time::Duration = std::time::Duration::from_secs(10);
+
+/// Resolves on SIGINT (Ctrl-C) or SIGTERM (container stop / deploy). Letting
+/// `axum::serve` drain in-flight requests on this signal means a redeploy no longer
+/// truncates uploads mid-flight; any background compression/export half-states that a
+/// hard kill would leave are already reconciled by `startup_recovery` on the next boot.
+///
+/// Once the signal fires we also arm a detached backstop that force-exits after
+/// [`SHUTDOWN_GRACE`]. Without it, open SSE streams (which have no natural end) would
+/// hold the graceful drain open indefinitely; the backstop bounds shutdown regardless
+/// of the orchestrator's own kill timeout. If the drain completes first, `main` returns
+/// and the process exits before the timer ever fires.
+async fn shutdown_signal() {
+    let ctrl_c = async {
+        let _ = tokio::signal::ctrl_c().await;
+    };
+
+    #[cfg(unix)]
+    let terminate = async {
+        match tokio::signal::unix::signal(tokio::signal::unix::SignalKind::terminate()) {
+            Ok(mut sig) => {
+                sig.recv().await;
+            }
+            Err(e) => {
+                tracing::warn!(error = ?e, "failed to install SIGTERM handler");
+                // Never resolve — fall back to ctrl_c only.
+                std::future::pending::<()>().await;
+            }
+        }
+    };
+
+    #[cfg(not(unix))]
+    let terminate = std::future::pending::<()>();
+
+    tokio::select! {
+        _ = ctrl_c => {}
+        _ = terminate => {}
+    }
+
+    tracing::info!(
+        "shutdown signal received, draining in-flight requests (max {}s)",
+        SHUTDOWN_GRACE.as_secs()
+    );
+
+    // Backstop: if the graceful drain is still blocked after the grace window (almost
+    // always because SSE streams are still open), exit anyway so deploys aren't stalled.
+    tokio::spawn(async {
+        tokio::time::sleep(SHUTDOWN_GRACE).await;
+        tracing::warn!(
+            "graceful drain exceeded {}s (likely open SSE streams); forcing exit",
+            SHUTDOWN_GRACE.as_secs()
+        );
+        std::process::exit(0);
+    });
+}

backend/src/models/comment.rs

@@ -3,6 +3,10 @@ use serde::Serialize;
 use sqlx::PgPool;
 use uuid::Uuid;
 
+// Row shape for `comment`: every field is populated by sqlx from `SELECT *` / `RETURNING *`.
+// `deleted_at` is not read in Rust today (the soft-delete filter lives in SQL), but it is part of
+// the row and stays here so the struct keeps mirroring the table.
+#[allow(dead_code)]
 #[derive(Debug, sqlx::FromRow)]
 pub struct Comment {
     pub id: Uuid,
@@ -24,49 +28,87 @@ pub struct CommentDto {
 }
 
 impl Comment {
-    pub async fn create(
-        pool: &PgPool,
+    /// Takes any executor so the caller can insert the comment and link its
+    /// hashtags inside a single transaction.
+    pub async fn create<'e, E>(
+        executor: E,
         upload_id: Uuid,
         user_id: Uuid,
         body: &str,
-    ) -> Result<Self, sqlx::Error> {
+    ) -> Result<Self, sqlx::Error>
+    where
+        E: sqlx::PgExecutor<'e>,
+    {
         sqlx::query_as::<_, Self>(
             "INSERT INTO comment (upload_id, user_id, body) VALUES ($1, $2, $3) RETURNING *",
         )
         .bind(upload_id)
         .bind(user_id)
         .bind(body)
-        .fetch_one(pool)
+        .fetch_one(executor)
         .await
     }
 
-    pub async fn list_for_upload(pool: &PgPool, upload_id: Uuid) -> Result<Vec<CommentDto>, sqlx::Error> {
+    /// Paginated comment listing — returns up to `limit` rows in chronological
+    /// order (oldest first). If `before` is set, only comments older than that
+    /// timestamp are returned, enabling backward cursor pagination ("load
+    /// earlier"). Without the LIMIT a hot post with thousands of comments could
+    /// OOM the server on a single GET.
+    pub async fn list_for_upload(
+        pool: &PgPool,
+        upload_id: Uuid,
+        before: Option<DateTime<Utc>>,
+        limit: i64,
+    ) -> Result<Vec<CommentDto>, sqlx::Error> {
+        // Two-step: pick the newest `limit` rows older than `before`, then flip
+        // them back into ascending order so the caller can render top-to-bottom.
         sqlx::query_as::<_, CommentDto>(
-            "SELECT c.id, c.upload_id, c.user_id, u.display_name AS uploader_name, c.body, c.created_at
-             FROM comment c
-             JOIN \"user\" u ON u.id = c.user_id
-             WHERE c.upload_id = $1 AND c.deleted_at IS NULL
-             ORDER BY c.created_at ASC",
+            "SELECT * FROM (
+               SELECT c.id, c.upload_id, c.user_id, u.display_name AS uploader_name,
+                      c.body, c.created_at
+               FROM comment c
+               JOIN \"user\" u ON u.id = c.user_id
+               WHERE c.upload_id = $1 AND c.deleted_at IS NULL
+                 AND ($2::timestamptz IS NULL OR c.created_at < $2)
+               ORDER BY c.created_at DESC
+               LIMIT $3
+             ) page
+             ORDER BY created_at ASC",
         )
         .bind(upload_id)
+        .bind(before)
+        .bind(limit)
         .fetch_all(pool)
         .await
     }
 
     pub async fn find_by_id(pool: &PgPool, id: Uuid) -> Result<Option<Self>, sqlx::Error> {
-        sqlx::query_as::<_, Self>(
-            "SELECT * FROM comment WHERE id = $1 AND deleted_at IS NULL",
-        )
-        .bind(id)
-        .fetch_optional(pool)
-        .await
+        sqlx::query_as::<_, Self>("SELECT * FROM comment WHERE id = $1 AND deleted_at IS NULL")
+            .bind(id)
+            .fetch_optional(pool)
+            .await
     }
 
-    pub async fn soft_delete(pool: &PgPool, id: Uuid) -> Result<(), sqlx::Error> {
-        sqlx::query("UPDATE comment SET deleted_at = NOW() WHERE id = $1")
-            .bind(id)
-            .execute(pool)
-            .await?;
-        Ok(())
+    /// Event-scoped soft delete. Returns `false` if the comment doesn't exist or belongs to a
+    /// different event.
+    /// Executor-generic so the delete and the keepsake regeneration can share one transaction
+    /// (see `Upload::soft_delete_in_event` for why that must be atomic).
+    pub async fn soft_delete_in_event(
+        conn: &mut sqlx::PgConnection,
+        id: Uuid,
+        event_id: Uuid,
+    ) -> Result<bool, sqlx::Error> {
+        let result = sqlx::query(
+            "UPDATE comment
+             SET deleted_at = NOW()
+             WHERE id = $1
+               AND deleted_at IS NULL
+               AND upload_id IN (SELECT id FROM upload WHERE event_id = $2)",
+        )
+        .bind(id)
+        .bind(event_id)
+        .execute(conn)
+        .await?;
+        Ok(result.rows_affected() > 0)
     }
 }

backend/src/models/event.rs

@@ -2,6 +2,11 @@ use chrono::{DateTime, Utc};
 use sqlx::PgPool;
 use uuid::Uuid;
 
+// Row shape for `event`: every field is populated by sqlx from `SELECT *` / `RETURNING *`. Several
+// (`slug`, `cover_image_path`, `export_epoch`, `created_at`) are not read through this struct today
+// — callers that need them query the column directly — but they are part of the row and stay here so
+// the struct keeps mirroring the table.
+#[allow(dead_code)]
 #[derive(Debug, sqlx::FromRow)]
 pub struct Event {
     pub id: Uuid,
@@ -11,8 +16,11 @@ pub struct Event {
     pub is_active: bool,
     pub uploads_locked_at: Option<DateTime<Utc>>,
     pub export_released_at: Option<DateTime<Utc>>,
-    pub export_zip_ready: bool,
-    pub export_html_ready: bool,
+    /// Monotonic generation counter for the keepsake. Bumped in the SAME UPDATE as any change to
+    /// `export_released_at` (release and reopen are its only writers). An export is downloadable
+    /// iff a `done` `export_job` row carries this exact epoch — readiness is derived from that,
+    /// never stored, so it cannot drift and no worker can resurrect it. See migration 014.
+    pub export_epoch: i64,
     pub created_at: DateTime<Utc>,
 }
 
@@ -25,13 +33,11 @@ impl Event {
     }
 
     pub async fn create(pool: &PgPool, slug: &str, name: &str) -> Result<Self, sqlx::Error> {
-        sqlx::query_as::<_, Self>(
-            "INSERT INTO event (slug, name) VALUES ($1, $2) RETURNING *",
-        )
-        .bind(slug)
-        .bind(name)
-        .fetch_one(pool)
-        .await
+        sqlx::query_as::<_, Self>("INSERT INTO event (slug, name) VALUES ($1, $2) RETURNING *")
+            .bind(slug)
+            .bind(name)
+            .fetch_one(pool)
+            .await
     }
 
     pub async fn find_or_create(

backend/src/models/hashtag.rs

@@ -1,6 +1,8 @@
-use sqlx::PgPool;
 use uuid::Uuid;
 
+// Row shape for `hashtag`, populated by sqlx from `RETURNING *` in `upsert`. Callers only use
+// `id` today; `event_id`/`tag` are the rest of the row and stay part of the struct.
+#[allow(dead_code)]
 #[derive(Debug, sqlx::FromRow)]
 pub struct Hashtag {
     pub id: Uuid,
@@ -10,7 +12,13 @@ pub struct Hashtag {
 
 impl Hashtag {
     /// Upsert a hashtag (insert if not exists, return existing if it does).
-    pub async fn upsert(pool: &PgPool, event_id: Uuid, tag: &str) -> Result<Self, sqlx::Error> {
+    ///
+    /// Takes any executor so callers can run it inside a transaction (atomic
+    /// upload/comment writes) or standalone against the pool.
+    pub async fn upsert<'e, E>(executor: E, event_id: Uuid, tag: &str) -> Result<Self, sqlx::Error>
+    where
+        E: sqlx::PgExecutor<'e>,
+    {
         let normalized = tag.trim().trim_start_matches('#').to_lowercase();
         sqlx::query_as::<_, Self>(
             "INSERT INTO hashtag (event_id, tag) VALUES ($1, $2)
@@ -19,59 +27,124 @@ impl Hashtag {
         )
         .bind(event_id)
         .bind(&normalized)
-        .fetch_one(pool)
+        .fetch_one(executor)
         .await
     }
 
-    pub async fn link_to_upload(
-        pool: &PgPool,
+    pub async fn link_to_upload<'e, E>(
+        executor: E,
         upload_id: Uuid,
         hashtag_id: Uuid,
-    ) -> Result<(), sqlx::Error> {
+    ) -> Result<(), sqlx::Error>
+    where
+        E: sqlx::PgExecutor<'e>,
+    {
         sqlx::query(
             "INSERT INTO upload_hashtag (upload_id, hashtag_id) VALUES ($1, $2)
              ON CONFLICT DO NOTHING",
         )
         .bind(upload_id)
         .bind(hashtag_id)
-        .execute(pool)
+        .execute(executor)
         .await?;
         Ok(())
     }
 
-    pub async fn unlink_all_from_upload(
-        pool: &PgPool,
+    pub async fn unlink_all_from_upload<'e, E>(
+        executor: E,
         upload_id: Uuid,
-    ) -> Result<(), sqlx::Error> {
+    ) -> Result<(), sqlx::Error>
+    where
+        E: sqlx::PgExecutor<'e>,
+    {
         sqlx::query("DELETE FROM upload_hashtag WHERE upload_id = $1")
             .bind(upload_id)
-            .execute(pool)
+            .execute(executor)
             .await?;
         Ok(())
     }
-
-    pub async fn tags_for_upload(
-        pool: &PgPool,
-        upload_id: Uuid,
-    ) -> Result<Vec<String>, sqlx::Error> {
-        let rows: Vec<(String,)> = sqlx::query_as(
-            "SELECT h.tag FROM hashtag h
-             JOIN upload_hashtag uh ON uh.hashtag_id = h.id
-             WHERE uh.upload_id = $1
-             ORDER BY h.tag",
-        )
-        .bind(upload_id)
-        .fetch_all(pool)
-        .await?;
-        Ok(rows.into_iter().map(|r| r.0).collect())
-    }
 }
 
-/// Extract #hashtags from text (caption or body).
+/// Extract `#hashtags` from text (caption or body). Tags are restricted to
+/// ASCII letters, digits, and underscore — emoji / punctuation / accented
+/// characters are rejected. This is deliberately strict: hashtags are an index
+/// (used for filtering and SQL JOINs), and tags like `#🎉` or `#!?` accumulate
+/// noise without helping anyone find content.
 pub fn extract_hashtags(text: &str) -> Vec<String> {
+    const MAX_TAG_LEN: usize = 40;
     text.split_whitespace()
-        .filter(|w| w.starts_with('#') && w.len() > 1)
-        .map(|w| w.trim_start_matches('#').to_lowercase())
-        .filter(|t| !t.is_empty())
+        .filter_map(|w| w.strip_prefix('#'))
+        .map(|t| {
+            t.chars()
+                .take_while(|c| c.is_ascii_alphanumeric() || *c == '_')
+                .collect::<String>()
+                .to_lowercase()
+        })
+        .filter(|t| !t.is_empty() && t.chars().count() <= MAX_TAG_LEN)
         .collect()
 }
+
+#[cfg(test)]
+mod tests {
+    use super::extract_hashtags;
+
+    #[test]
+    fn ascii_word_chars_extracted() {
+        assert_eq!(
+            extract_hashtags("hello #wedding #Day_2 #cake!"),
+            vec!["wedding", "day_2", "cake"]
+        );
+    }
+
+    #[test]
+    fn emojis_and_punctuation_excluded() {
+        // The 🎉 tag drops out entirely (no ASCII chars after #), the next tag
+        // stops at the !? and yields only "fun".
+        assert_eq!(extract_hashtags("#🎉 #!? #fun!"), vec!["fun"]);
+    }
+
+    #[test]
+    fn empty_or_bare_hash_skipped() {
+        assert_eq!(extract_hashtags("# #"), Vec::<String>::new());
+    }
+
+    #[test]
+    fn tag_stops_at_first_non_word_char() {
+        // A tag runs until the first char that isn't ascii-alphanumeric or '_'.
+        assert_eq!(extract_hashtags("#foo#bar"), vec!["foo"]);
+        assert_eq!(extract_hashtags("#foo-bar"), vec!["foo"]);
+        assert_eq!(extract_hashtags("##tag"), Vec::<String>::new()); // '#' after the strip is non-word
+    }
+
+    #[test]
+    fn tag_length_is_capped_at_40_chars() {
+        let ok = "a".repeat(40);
+        assert_eq!(extract_hashtags(&format!("#{ok}")), vec![ok.clone()]);
+        // 41+ chars → dropped entirely (not truncated).
+        let too_long = "a".repeat(41);
+        assert_eq!(
+            extract_hashtags(&format!("#{too_long}")),
+            Vec::<String>::new()
+        );
+    }
+
+    #[test]
+    fn duplicate_tags_are_returned_verbatim_not_deduplicated() {
+        // Dedup is the DB's job (Hashtag::upsert ON CONFLICT); extraction returns each
+        // occurrence so callers can count/link them independently. Case folds to lower.
+        assert_eq!(
+            extract_hashtags("#fun #Fun #fun!"),
+            vec!["fun", "fun", "fun"]
+        );
+    }
+
+    #[test]
+    fn non_ascii_word_chars_truncate_the_tag() {
+        // KNOWN LIMITATION for a German app: `is_ascii_alphanumeric` excludes umlauts
+        // and ß, so a tag truncates at the first non-ASCII letter. Pinned here so a
+        // future Unicode-aware change is a deliberate, test-visible decision.
+        assert_eq!(extract_hashtags("#Grüße"), vec!["gr"]);
+        assert_eq!(extract_hashtags("#Straße"), vec!["stra"]);
+        assert_eq!(extract_hashtags("#café"), vec!["caf"]);
+    }
+}

backend/src/models/session.rs

@@ -2,6 +2,9 @@ use chrono::{DateTime, Utc};
 use sqlx::PgPool;
 use uuid::Uuid;
 
+// Row shape for `session`, populated by sqlx from `RETURNING *`. Session validation is done in SQL
+// (expiry/last-seen predicates), so no field is read in Rust — the struct is the row's shape.
+#[allow(dead_code)]
 #[derive(Debug, sqlx::FromRow)]
 pub struct Session {
     pub id: Uuid,
@@ -43,22 +46,65 @@ impl Session {
         .await
     }
 
-    pub async fn touch(pool: &PgPool, id: Uuid) -> Result<(), sqlx::Error> {
-        sqlx::query("UPDATE session SET last_seen_at = NOW() WHERE id = $1")
-            .bind(id)
-            .execute(pool)
-            .await?;
+    /// Resolve a session token straight to its live user row in one round-trip.
+    ///
+    /// The auth extractor runs on every authenticated request and used to do two
+    /// sequential queries (session lookup, then user lookup); this collapses them into
+    /// a single `session JOIN "user"`. The `expires_at` guard mirrors
+    /// [`Self::find_by_token_hash`], and the user row is read live (role/ban are never
+    /// trusted from the JWT). Returns `None` when the session is missing/expired.
+    pub async fn find_user_by_token_hash(
+        pool: &PgPool,
+        token_hash: &str,
+    ) -> Result<Option<crate::models::user::User>, sqlx::Error> {
+        sqlx::query_as::<_, crate::models::user::User>(
+            "SELECT u.*
+             FROM session s
+             JOIN \"user\" u ON u.id = s.user_id
+             WHERE s.token_hash = $1 AND s.expires_at > NOW()",
+        )
+        .bind(token_hash)
+        .fetch_optional(pool)
+        .await
+    }
+
+    /// Touch `last_seen_at` AND slide `expires_at` forward by `expiry_days` from now, so
+    /// an actively-used session never hits the fixed 30-day cliff (it renews on every
+    /// authenticated request). An idle session still expires `expiry_days` after its last
+    /// activity. Keyed by token hash so the auth extractor needs no prior id lookup.
+    pub async fn touch_and_renew(
+        pool: &PgPool,
+        token_hash: &str,
+        expiry_days: i64,
+    ) -> Result<(), sqlx::Error> {
+        sqlx::query(
+            "UPDATE session
+             SET last_seen_at = NOW(),
+                 expires_at = NOW() + ($2 || ' days')::interval
+             WHERE token_hash = $1",
+        )
+        .bind(token_hash)
+        .bind(expiry_days.to_string())
+        .execute(pool)
+        .await?;
         Ok(())
     }
 
-    pub async fn delete_by_token_hash(
-        pool: &PgPool,
-        token_hash: &str,
-    ) -> Result<(), sqlx::Error> {
+    pub async fn delete_by_token_hash(pool: &PgPool, token_hash: &str) -> Result<(), sqlx::Error> {
         sqlx::query("DELETE FROM session WHERE token_hash = $1")
             .bind(token_hash)
             .execute(pool)
             .await?;
         Ok(())
     }
+
+    /// Revoke every session for a user. Backs "sign out everywhere" and the forced
+    /// re-auth after a host PIN-reset or a ban. Returns the number of sessions cleared.
+    pub async fn delete_all_for_user(pool: &PgPool, user_id: Uuid) -> Result<u64, sqlx::Error> {
+        let r = sqlx::query("DELETE FROM session WHERE user_id = $1")
+            .bind(user_id)
+            .execute(pool)
+            .await?;
+        Ok(r.rows_affected())
+    }
 }

backend/src/models/upload.rs

@@ -3,6 +3,10 @@ use serde::Serialize;
 use sqlx::PgPool;
 use uuid::Uuid;
 
+// Row shape for `upload`: every field is populated by sqlx from `RETURNING *` in `create`. Callers
+// mostly use `id` and hand the rest to the compression/feed queries, so most fields are never read
+// through this struct — they stay here so it keeps mirroring the table.
+#[allow(dead_code)]
 #[derive(Debug, sqlx::FromRow)]
 pub struct Upload {
     pub id: Uuid,
@@ -14,6 +18,7 @@ pub struct Upload {
     pub mime_type: String,
     pub original_size_bytes: i64,
     pub caption: Option<String>,
+    pub compression_status: String,
     pub created_at: DateTime<Utc>,
     pub deleted_at: Option<DateTime<Utc>>,
 }
@@ -34,16 +39,32 @@ pub struct UploadDto {
     pub created_at: DateTime<Utc>,
 }
 
+/// Minimal projection of an upload's on-disk file paths, used by the visibility-gated
+/// media aliases so they don't hydrate the entire `Upload` row per request.
+#[derive(Debug, sqlx::FromRow)]
+pub struct VisibleMedia {
+    pub original_path: String,
+    pub preview_path: Option<String>,
+    pub thumbnail_path: Option<String>,
+    pub display_path: Option<String>,
+    pub mime_type: String,
+}
+
 impl Upload {
-    pub async fn create(
-        pool: &PgPool,
+    /// Takes any executor so the caller can run it inside a transaction (atomic
+    /// quota + insert) or standalone against the pool.
+    pub async fn create<'e, E>(
+        executor: E,
         event_id: Uuid,
         user_id: Uuid,
         original_path: &str,
         mime_type: &str,
         original_size_bytes: i64,
         caption: Option<&str>,
-    ) -> Result<Self, sqlx::Error> {
+    ) -> Result<Self, sqlx::Error>
+    where
+        E: sqlx::PgExecutor<'e>,
+    {
         sqlx::query_as::<_, Self>(
             "INSERT INTO upload (event_id, user_id, original_path, mime_type, original_size_bytes, caption)
              VALUES ($1, $2, $3, $4, $5, $6)
@@ -55,19 +76,52 @@ impl Upload {
         .bind(mime_type)
         .bind(original_size_bytes)
         .bind(caption)
-        .fetch_one(pool)
+        .fetch_one(executor)
         .await
     }
 
-    pub async fn find_by_id(pool: &PgPool, id: Uuid) -> Result<Option<Self>, sqlx::Error> {
-        sqlx::query_as::<_, Self>(
-            "SELECT * FROM upload WHERE id = $1 AND deleted_at IS NULL",
+    /// Lean lookup for the public media aliases (`get_original`/`get_preview`/
+    /// `get_thumbnail`): returns ONLY the file paths + mime for a visible upload —
+    /// excluding soft-deleted rows, hidden owners (`uploads_hidden`), and banned owners
+    /// (`is_banned`) — the same filter `v_feed` applies. So moderation that removes a post
+    /// from the feed also stops its original/preview/thumbnail from being pulled by UUID.
+    ///
+    /// Selects four columns instead of the whole `Upload` row: this runs once per image
+    /// per cache-miss on the media hot path, so we avoid hydrating fields the response
+    /// never uses.
+    pub async fn find_visible_media(
+        pool: &PgPool,
+        id: Uuid,
+    ) -> Result<Option<VisibleMedia>, sqlx::Error> {
+        sqlx::query_as::<_, VisibleMedia>(
+            "SELECT up.original_path, up.preview_path, up.thumbnail_path, up.display_path, up.mime_type
+             FROM upload up
+             JOIN \"user\" u ON u.id = up.user_id
+             WHERE up.id = $1 AND up.deleted_at IS NULL
+               AND u.uploads_hidden = false AND u.is_banned = false",
         )
         .bind(id)
         .fetch_optional(pool)
         .await
     }
 
+    /// Event-scoped lookup used by host endpoints so a host of event A cannot
+    /// reach uploads belonging to event B.
+    pub async fn find_by_id_and_event(
+        pool: &PgPool,
+        id: Uuid,
+        event_id: Uuid,
+    ) -> Result<Option<Self>, sqlx::Error> {
+        sqlx::query_as::<_, Self>(
+            "SELECT * FROM upload
+             WHERE id = $1 AND event_id = $2 AND deleted_at IS NULL",
+        )
+        .bind(id)
+        .bind(event_id)
+        .fetch_optional(pool)
+        .await
+    }
+
     pub async fn set_preview_path(
         pool: &PgPool,
         id: Uuid,
@@ -81,6 +135,19 @@ impl Upload {
         Ok(())
     }
 
+    pub async fn set_display_path(
+        pool: &PgPool,
+        id: Uuid,
+        display_path: &str,
+    ) -> Result<(), sqlx::Error> {
+        sqlx::query("UPDATE upload SET display_path = $2 WHERE id = $1")
+            .bind(id)
+            .bind(display_path)
+            .execute(pool)
+            .await?;
+        Ok(())
+    }
+
     pub async fn set_thumbnail_path(
         pool: &PgPool,
         id: Uuid,
@@ -94,22 +161,104 @@ impl Upload {
         Ok(())
     }
 
+    /// Soft-deletes the upload and decrements the uploader's `total_upload_bytes`.
+    /// Done in a single transaction so a crash between the two writes can't leave
+    /// the quota counter pointing at bytes the user has already deleted (which would
+    /// silently lock them out of future uploads).
+    ///
+    /// No-op if the row is already deleted — protects against a double-tap on the
+    /// delete action double-decrementing the counter.
     pub async fn soft_delete(pool: &PgPool, id: Uuid) -> Result<(), sqlx::Error> {
-        sqlx::query("UPDATE upload SET deleted_at = NOW() WHERE id = $1")
+        let mut tx = pool.begin().await?;
+        let row: Option<(Uuid, i64)> = sqlx::query_as(
+            "UPDATE upload
+             SET deleted_at = NOW()
+             WHERE id = $1 AND deleted_at IS NULL
+             RETURNING user_id, original_size_bytes",
+        )
+        .bind(id)
+        .fetch_optional(&mut *tx)
+        .await?;
+        if let Some((user_id, bytes)) = row {
+            sqlx::query(
+                "UPDATE \"user\"
+                 SET total_upload_bytes = GREATEST(0, total_upload_bytes - $2)
+                 WHERE id = $1",
+            )
+            .bind(user_id)
+            .bind(bytes)
+            .execute(&mut *tx)
+            .await?;
+        }
+        tx.commit().await?;
+        Ok(())
+    }
+
+    /// Event-scoped variant of [`Self::soft_delete`]. Returns `false` if no row
+    /// matched (already deleted, wrong event, or unknown id) so host handlers
+    /// can return a clean 404 instead of silently no-op'ing.
+    /// Executor-generic so a caller can run the delete and the keepsake regeneration in ONE
+    /// transaction. They must be atomic: if the delete commits and the regeneration doesn't (a
+    /// dropped handler future, a failed second tx), the taken-down photo stays in the downloadable
+    /// archive forever, and recovery can't tell — the keepsake still looks complete at the current
+    /// epoch, and the host can no longer even find the upload to retry.
+    pub async fn soft_delete_in_event(
+        conn: &mut sqlx::PgConnection,
+        id: Uuid,
+        event_id: Uuid,
+    ) -> Result<bool, sqlx::Error> {
+        let tx = conn;
+        let row: Option<(Uuid, i64)> = sqlx::query_as(
+            "UPDATE upload
+             SET deleted_at = NOW()
+             WHERE id = $1 AND event_id = $2 AND deleted_at IS NULL
+             RETURNING user_id, original_size_bytes",
+        )
+        .bind(id)
+        .bind(event_id)
+        .fetch_optional(&mut *tx)
+        .await?;
+        let deleted = if let Some((user_id, bytes)) = row {
+            sqlx::query(
+                "UPDATE \"user\"
+                 SET total_upload_bytes = GREATEST(0, total_upload_bytes - $2)
+                 WHERE id = $1",
+            )
+            .bind(user_id)
+            .bind(bytes)
+            .execute(&mut *tx)
+            .await?;
+            true
+        } else {
+            false
+        };
+        Ok(deleted)
+    }
+
+    pub async fn update_caption<'e, E>(
+        executor: E,
+        id: Uuid,
+        caption: Option<&str>,
+    ) -> Result<(), sqlx::Error>
+    where
+        E: sqlx::PgExecutor<'e>,
+    {
+        sqlx::query("UPDATE upload SET caption = $2 WHERE id = $1")
             .bind(id)
-            .execute(pool)
+            .bind(caption)
+            .execute(executor)
             .await?;
         Ok(())
     }
 
-    pub async fn update_caption(
+    pub async fn set_compression_status(
         pool: &PgPool,
         id: Uuid,
-        caption: Option<&str>,
+        status: &str,
     ) -> Result<(), sqlx::Error> {
-        sqlx::query("UPDATE upload SET caption = $2 WHERE id = $1")
+        sqlx::query("UPDATE upload SET compression_status = $2 WHERE id = $1")
             .bind(id)
-            .bind(caption)
+            .bind(status)
             .execute(pool)
             .await?;
         Ok(())

backend/src/models/user.rs

@@ -12,6 +12,20 @@ pub enum UserRole {
     Admin,
 }
 
+impl UserRole {
+    pub fn as_str(&self) -> &'static str {
+        match self {
+            UserRole::Guest => "guest",
+            UserRole::Host => "host",
+            UserRole::Admin => "admin",
+        }
+    }
+}
+
+// Row shape for `user`: every field is populated by sqlx from `SELECT *` / `RETURNING *`.
+// `uploads_hidden`, `failed_pin_attempts` and `created_at` are enforced/updated in SQL rather than
+// read in Rust, but they are part of the row and stay here so the struct keeps mirroring the table.
+#[allow(dead_code)]
 #[derive(Debug, sqlx::FromRow)]
 pub struct User {
     pub id: Uuid,
@@ -95,14 +109,16 @@ impl User {
         Ok(row.0)
     }
 
-    pub async fn lock_pin(pool: &PgPool, id: Uuid, until: DateTime<Utc>) -> Result<(), sqlx::Error> {
-        sqlx::query(
-            "UPDATE \"user\" SET pin_locked_until = $2 WHERE id = $1",
-        )
-        .bind(id)
-        .bind(until)
-        .execute(pool)
-        .await?;
+    pub async fn lock_pin(
+        pool: &PgPool,
+        id: Uuid,
+        until: DateTime<Utc>,
+    ) -> Result<(), sqlx::Error> {
+        sqlx::query("UPDATE \"user\" SET pin_locked_until = $2 WHERE id = $1")
+            .bind(id)
+            .bind(until)
+            .execute(pool)
+            .await?;
         Ok(())
     }
 

backend/src/services/compression.rs

@@ -1,36 +1,102 @@
 use std::path::{Path, PathBuf};
 use std::sync::Arc;
+use std::sync::atomic::{AtomicU64, Ordering};
 
 use anyhow::{Context, Result};
 use sqlx::PgPool;
-use tokio::sync::Semaphore;
+use tokio::sync::{Semaphore, broadcast};
 use uuid::Uuid;
 
 use crate::models::upload::Upload;
+use crate::state::SseEvent;
 
 #[derive(Clone)]
 pub struct CompressionWorker {
     semaphore: Arc<Semaphore>,
     pool: PgPool,
     media_path: PathBuf,
+    sse_tx: broadcast::Sender<SseEvent>,
+    /// Bumped whenever the underlying data is reset out from under in-flight work (only the e2e
+    /// TRUNCATE does this today). A task captures the value at spawn and abandons itself if it has
+    /// changed by the time it runs — see `process`.
+    generation: Arc<AtomicU64>,
 }
 
 impl CompressionWorker {
-    pub fn new(pool: PgPool, media_path: PathBuf, concurrency: usize) -> Self {
+    pub fn new(
+        pool: PgPool,
+        media_path: PathBuf,
+        concurrency: usize,
+        sse_tx: broadcast::Sender<SseEvent>,
+    ) -> Self {
         Self {
             semaphore: Arc::new(Semaphore::new(concurrency)),
             pool,
             media_path,
+            sse_tx,
+            generation: Arc::new(AtomicU64::new(0)),
         }
     }
 
+    /// Invalidate all in-flight and queued compression work. Called by the e2e TRUNCATE endpoint:
+    /// truncating deletes the upload rows and wipes `media/`, so a worker that was queued on the
+    /// semaphore when the wipe happened would otherwise wake in the NEXT test, fail to find its
+    /// file, and broadcast `upload-error` / `upload-deleted` into that test's live SSE stream —
+    /// corrupting any test that asserts on toasts or feed contents. Bumping the generation makes
+    /// those stale tasks return silently instead. A no-op in production (never called there).
+    pub fn bump_generation(&self) {
+        self.generation.fetch_add(1, Ordering::SeqCst);
+    }
+
     /// Spawn a background task to process an uploaded file.
     pub fn process(&self, upload_id: Uuid, original_path: String, mime_type: String) {
         let worker = self.clone();
+        let born_at = worker.generation.load(Ordering::SeqCst);
         tokio::spawn(async move {
             let _permit = worker.semaphore.acquire().await;
-            if let Err(e) = worker.do_process(upload_id, &original_path, &mime_type).await {
-                tracing::error!("compression failed for upload {upload_id}: {e:#}");
+            // The data this task was queued against may have been reset while it waited for a permit
+            // (e2e TRUNCATE). If so, its file and row are gone; doing anything — including
+            // broadcasting a failure — would leak into an unrelated test. Abandon quietly.
+            if worker.generation.load(Ordering::SeqCst) != born_at {
+                return;
+            }
+            match worker
+                .do_process(upload_id, &original_path, &mime_type)
+                .await
+            {
+                Ok(_) => {
+                    tracing::info!("compression completed for upload {upload_id}");
+                    let _ = worker.sse_tx.send(SseEvent {
+                        event_type: "upload-processed".to_string(),
+                        data: serde_json::json!({ "upload_id": upload_id }).to_string(),
+                    });
+                }
+                Err(e) => {
+                    tracing::error!("compression failed for upload {upload_id}: {e:#}");
+                    // Auto-cleanup: a failed transcode would otherwise leave a
+                    // permanently broken feed card, silently charge the uploader's
+                    // quota, and orphan the original on disk. Refund + soft-delete
+                    // (one tx, so v_feed excludes it), remove the orphan file, then
+                    // tell the uploader (upload-error toast) and evict the card
+                    // everywhere (upload-deleted, already handled by the feed).
+                    let _ = Upload::set_compression_status(&worker.pool, upload_id, "failed").await;
+                    if let Err(del) = Upload::soft_delete(&worker.pool, upload_id).await {
+                        tracing::warn!(error = ?del, %upload_id, "failed to soft-delete after compression failure");
+                    }
+                    let orphan = worker.media_path.join(&original_path);
+                    if let Err(rm) = tokio::fs::remove_file(&orphan).await {
+                        tracing::warn!(error = ?rm, path = %orphan.display(), "failed to remove orphaned original");
+                    }
+                    let _ = worker.sse_tx.send(SseEvent {
+                        event_type: "upload-error".to_string(),
+                        data: serde_json::json!({ "upload_id": upload_id, "error": e.to_string() })
+                            .to_string(),
+                    });
+                    let _ = worker.sse_tx.send(SseEvent {
+                        event_type: "upload-deleted".to_string(),
+                        data: serde_json::json!({ "upload_id": upload_id }).to_string(),
+                    });
+                }
             }
         });
     }
@@ -41,45 +107,96 @@ impl CompressionWorker {
         original_path: &str,
         mime_type: &str,
     ) -> Result<()> {
+        Upload::set_compression_status(&self.pool, upload_id, "processing").await?;
+
         let original = self.media_path.join(original_path);
 
         if mime_type.starts_with("image/") {
-            let preview_rel = self.generate_image_preview(upload_id, &original, mime_type).await?;
+            let (preview_rel, display_rel) = self
+                .generate_image_derivatives(upload_id, &original, mime_type)
+                .await?;
             Upload::set_preview_path(&self.pool, upload_id, &preview_rel).await?;
-            tracing::info!("preview generated for upload {upload_id}");
+            Upload::set_display_path(&self.pool, upload_id, &display_rel).await?;
+            tracing::info!("preview + display generated for upload {upload_id}");
         } else if mime_type.starts_with("video/") {
             let thumb_rel = self.generate_video_thumbnail(upload_id, &original).await?;
             Upload::set_thumbnail_path(&self.pool, upload_id, &thumb_rel).await?;
             tracing::info!("thumbnail generated for upload {upload_id}");
         }
 
+        Upload::set_compression_status(&self.pool, upload_id, "done").await?;
         Ok(())
     }
 
-    async fn generate_image_preview(
+    /// Longest edge of the big-screen "display" derivative used by the diashow. Sized to be
+    /// sharp on 1080p/4K while staying bounded (a ~2048px JPEG decodes to ~16 MB — trivial
+    /// for any kiosk, unlike a raw multi-thousand-pixel original).
+    const DISPLAY_MAX_EDGE: u32 = 2048;
+    /// Longest edge of the phone-feed "preview" (data-saver default).
+    const PREVIEW_MAX_EDGE: u32 = 800;
+
+    /// Decode the image ONCE and emit both derivatives — the 800px `preview` (phone feed)
+    /// and the 2048px `display` (diashow). Returns `(preview_rel, display_rel)`.
+    async fn generate_image_derivatives(
         &self,
         upload_id: Uuid,
         original: &Path,
         mime_type: &str,
-    ) -> Result<String> {
+    ) -> Result<(String, String)> {
         let previews_dir = self.media_path.join("previews");
+        let displays_dir = self.media_path.join("displays");
         tokio::fs::create_dir_all(&previews_dir).await?;
+        tokio::fs::create_dir_all(&displays_dir).await?;
 
-        let preview_filename = format!("{upload_id}.jpg");
-        let preview_path = previews_dir.join(&preview_filename);
+        let filename = format!("{upload_id}.jpg");
+        let preview_path = previews_dir.join(&filename);
+        let display_path = displays_dir.join(&filename);
         let original = original.to_path_buf();
-        let preview_path_clone = preview_path.clone();
         let mime_owned = mime_type.to_string();
+        let preview_max = Self::PREVIEW_MAX_EDGE;
+        let display_max = Self::DISPLAY_MAX_EDGE;
 
         // Run blocking image operations in a spawn_blocking task
         tokio::task::spawn_blocking(move || -> Result<()> {
-            let img = image::open(&original)
-                .context("failed to open image")?;
+            // Reject decompression bombs *before* fully decoding: the upload body
+            // cap bounds the file size on disk, but a small file can still decode to
+            // enormous dimensions (e.g. a ~1 MB image expanding to 50k×50k px →
+            // gigabytes), OOM-ing the box during decode/resize. 12000×12000 covers
+            // any real phone photo; max_alloc hard-caps the decode allocation.
+            let mut reader = image::ImageReader::open(&original)
+                .context("failed to open image")?
+                .with_guessed_format()
+                .context("failed to read image header")?;
+            let mut limits = image::Limits::default();
+            limits.max_image_width = Some(12_000);
+            limits.max_image_height = Some(12_000);
+            limits.max_alloc = Some(256 * 1024 * 1024);
+            reader.limits(limits);
+            let img = reader.decode().context("failed to decode image")?;
 
-            // Resize to max 800px wide, preserving aspect ratio
-            let preview = img.resize(800, 800, image::imageops::FilterType::Lanczos3);
-            preview.save_with_format(&preview_path_clone, image::ImageFormat::Jpeg)
-                .context("failed to save preview")?;
+            // Preview: max 800px, preserving aspect ratio (data-saver feed).
+            img.resize(
+                preview_max,
+                preview_max,
+                image::imageops::FilterType::Lanczos3,
+            )
+            .save_with_format(&preview_path, image::ImageFormat::Jpeg)
+            .context("failed to save preview")?;
+
+            // Display: max 2048px for the diashow. Only DOWNSCALE — never upscale a smaller
+            // original (that adds bytes with no quality gain); re-encode it as JPEG as-is.
+            let display = if img.width() > display_max || img.height() > display_max {
+                img.resize(
+                    display_max,
+                    display_max,
+                    image::imageops::FilterType::Lanczos3,
+                )
+            } else {
+                img
+            };
+            display
+                .save_with_format(&display_path, image::ImageFormat::Jpeg)
+                .context("failed to save display")?;
 
             // If the original is PNG, try lossless compression in-place
             if mime_owned == "image/png" {
@@ -98,21 +215,75 @@ impl CompressionWorker {
         })
         .await??;
 
-        Ok(format!("previews/{preview_filename}"))
+        Ok((
+            format!("previews/{filename}"),
+            format!("displays/{filename}"),
+        ))
     }
 
-    async fn generate_video_thumbnail(
-        &self,
-        upload_id: Uuid,
-        original: &Path,
-    ) -> Result<String> {
+    /// One-time backfill: existing image uploads processed before the display derivative
+    /// existed have a preview but no `display_path`. Regenerate both derivatives for them
+    /// (decode is cheap and idempotent) and set the path. Unlike the failure path in
+    /// `process`, a backfill error is logged and skipped — it must NEVER soft-delete an
+    /// upload that already has a working preview. Fire-and-forget from startup.
+    pub async fn backfill_missing_display(&self) {
+        let rows = sqlx::query_as::<_, (Uuid, String, String)>(
+            "SELECT id, original_path, mime_type FROM upload
+             WHERE display_path IS NULL AND preview_path IS NOT NULL
+               AND deleted_at IS NULL AND mime_type LIKE 'image/%'",
+        )
+        .fetch_all(&self.pool)
+        .await;
+        let rows = match rows {
+            Ok(r) => r,
+            Err(e) => {
+                tracing::warn!(error = ?e, "display backfill query failed");
+                return;
+            }
+        };
+        if rows.is_empty() {
+            return;
+        }
+        tracing::info!(
+            "backfilling display derivative for {} upload(s)",
+            rows.len()
+        );
+        for (id, original_path, mime_type) in rows {
+            let worker = self.clone();
+            tokio::spawn(async move {
+                let _permit = worker.semaphore.acquire().await;
+                let original = worker.media_path.join(&original_path);
+                match worker
+                    .generate_image_derivatives(id, &original, &mime_type)
+                    .await
+                {
+                    Ok((preview_rel, display_rel)) => {
+                        let _ = Upload::set_preview_path(&worker.pool, id, &preview_rel).await;
+                        let _ = Upload::set_display_path(&worker.pool, id, &display_rel).await;
+                        tracing::info!("display backfilled for upload {id}");
+                    }
+                    Err(e) => {
+                        // Leave the existing preview intact; the diashow falls back to the
+                        // original for this upload until a later successful pass.
+                        tracing::warn!(error = ?e, %id, "display backfill failed; leaving as-is");
+                    }
+                }
+            });
+        }
+    }
+
+    async fn generate_video_thumbnail(&self, upload_id: Uuid, original: &Path) -> Result<String> {
         let thumbs_dir = self.media_path.join("thumbnails");
         tokio::fs::create_dir_all(&thumbs_dir).await?;
 
         let thumb_filename = format!("{upload_id}.jpg");
         let thumb_path = thumbs_dir.join(&thumb_filename);
 
-        let output = tokio::process::Command::new("ffmpeg")
+        // Hard timeout — a malformed video can hang `ffmpeg` indefinitely. Without a
+        // cap, the held compression-worker semaphore permit is never released and the
+        // pool eventually deadlocks (no further uploads ever processed). 120s is well
+        // above the time to extract one frame from any sane input.
+        let mut child = tokio::process::Command::new("ffmpeg")
             .args([
                 "-i",
                 original.to_str().unwrap_or_default(),
@@ -125,13 +296,29 @@ impl CompressionWorker {
                 "-y",
                 thumb_path.to_str().unwrap_or_default(),
             ])
-            .output()
-            .await
-            .context("failed to run ffmpeg")?;
+            .stdout(std::process::Stdio::piped())
+            .stderr(std::process::Stdio::piped())
+            .kill_on_drop(true)
+            .spawn()
+            .context("failed to spawn ffmpeg")?;
 
-        if !output.status.success() {
-            let stderr = String::from_utf8_lossy(&output.stderr);
-            anyhow::bail!("ffmpeg failed: {stderr}");
+        let status =
+            match tokio::time::timeout(std::time::Duration::from_secs(120), child.wait()).await {
+                Ok(res) => res.context("ffmpeg wait failed")?,
+                Err(_) => {
+                    let _ = child.kill().await;
+                    anyhow::bail!("ffmpeg timeout after 120s");
+                }
+            };
+
+        if !status.success() {
+            // Best-effort: drain stderr for the log.
+            let mut stderr = Vec::new();
+            if let Some(mut handle) = child.stderr.take() {
+                use tokio::io::AsyncReadExt;
+                let _ = handle.read_to_end(&mut stderr).await;
+            }
+            anyhow::bail!("ffmpeg failed: {}", String::from_utf8_lossy(&stderr));
         }
 
         Ok(format!("thumbnails/{thumb_filename}"))

backend/src/services/config.rs

@@ -0,0 +1,150 @@
+//! Reads of the runtime-tunable `config` table, fronted by an in-memory cache.
+//!
+//! Each handler used to keep a small local copy of these helpers; consolidating them
+//! here means one place to add a parser, one place to mock for tests, and one place to
+//! find when a key changes. New keys do not require code changes — they're picked up
+//! the next time the cache reloads.
+//!
+//! ## Why a cache
+//!
+//! The `config` table is effectively static during an event, yet it was the busiest
+//! query in the system: every request re-read each key with its own `SELECT`
+//! (an upload touched it ~8 times). Against the small connection pool that was the
+//! throughput ceiling. [`ConfigCache`] loads the whole table once and serves reads
+//! from memory.
+//!
+//! ## Consistency contract
+//!
+//! Correctness comes from **synchronous invalidation on every write**, not from the
+//! TTL. The two runtime write paths — the admin `PATCH /admin/config` handler and the
+//! test-mode truncate/reseed — both call [`ConfigCache::invalidate`] after committing,
+//! so the *next* read reloads from the DB and sees the new value immediately. The
+//! [`RELOAD_TTL`] is only a safety net for out-of-band changes (e.g. a migration or a
+//! manual DB edit); it is deliberately short but never the primary mechanism.
+//!
+//! Values are read with a default fallback so the app still starts if a key is missing
+//! (e.g. during a migration window). Production seeds keys via migrations 005 and 009.
+
+use std::collections::HashMap;
+use std::sync::{Arc, RwLock};
+use std::time::{Duration, Instant};
+
+use sqlx::PgPool;
+
+/// How long a loaded snapshot is trusted before the next read reloads it. This is a
+/// backstop for out-of-band DB changes only — every in-process write invalidates the
+/// cache synchronously, so tests that PATCH-then-assert never depend on this expiring.
+const RELOAD_TTL: Duration = Duration::from_secs(30);
+
+struct Snapshot {
+    values: HashMap<String, String>,
+    loaded_at: Instant,
+}
+
+/// In-memory cache of the entire `config` table. Cheap to `clone` (shares the pool and
+/// the `Arc`), so it lives in `AppState` and every handler reads through it.
+#[derive(Clone)]
+pub struct ConfigCache {
+    pool: PgPool,
+    inner: Arc<RwLock<Option<Snapshot>>>,
+}
+
+impl ConfigCache {
+    pub fn new(pool: PgPool) -> Self {
+        Self {
+            pool,
+            inner: Arc::new(RwLock::new(None)),
+        }
+    }
+
+    /// Drop the cached snapshot so the next read reloads the whole table from the DB.
+    /// Call this after any write to the `config` table (admin PATCH, test reseed).
+    pub fn invalidate(&self) {
+        *self.inner.write().unwrap() = None;
+    }
+
+    /// Return the fresh snapshot if one is loaded and still within [`RELOAD_TTL`].
+    fn fresh_snapshot(&self) -> Option<HashMap<String, String>> {
+        let guard = self.inner.read().unwrap();
+        match guard.as_ref() {
+            Some(snap) if snap.loaded_at.elapsed() < RELOAD_TTL => Some(snap.values.clone()),
+            _ => None,
+        }
+    }
+
+    /// Read one key, loading the whole table into the cache on a miss/expiry. On a DB
+    /// error we return `None` (callers fall back to their default) without poisoning
+    /// the cache.
+    async fn get_raw(&self, key: &str) -> Option<String> {
+        if let Some(values) = self.fresh_snapshot() {
+            return values.get(key).cloned();
+        }
+
+        // Cache miss or stale — reload the entire table in one query.
+        let rows: Vec<(String, String)> = match sqlx::query_as::<_, (String, String)>(
+            "SELECT key, value FROM config",
+        )
+        .fetch_all(&self.pool)
+        .await
+        {
+            Ok(rows) => rows,
+            Err(e) => {
+                tracing::warn!(error = ?e, "config reload failed; using defaults for this read");
+                return None;
+            }
+        };
+
+        let values: HashMap<String, String> = rows.into_iter().collect();
+        let result = values.get(key).cloned();
+        *self.inner.write().unwrap() = Some(Snapshot {
+            values,
+            loaded_at: Instant::now(),
+        });
+        result
+    }
+}
+
+pub async fn get_str(cache: &ConfigCache, key: &str, default: &str) -> String {
+    cache
+        .get_raw(key)
+        .await
+        .unwrap_or_else(|| default.to_string())
+}
+
+pub async fn get_i64(cache: &ConfigCache, key: &str, default: i64) -> i64 {
+    cache
+        .get_raw(key)
+        .await
+        .and_then(|v| v.parse().ok())
+        .unwrap_or(default)
+}
+
+pub async fn get_usize(cache: &ConfigCache, key: &str, default: usize) -> usize {
+    cache
+        .get_raw(key)
+        .await
+        .and_then(|v| v.parse().ok())
+        .unwrap_or(default)
+}
+
+pub async fn get_f64(cache: &ConfigCache, key: &str, default: f64) -> f64 {
+    cache
+        .get_raw(key)
+        .await
+        .and_then(|v| v.parse().ok())
+        .unwrap_or(default)
+}
+
+/// Parses common truthy spellings used by both the migration seeds and the admin form.
+/// Accepts `true/false`, `1/0`, `yes/no`, `on/off` — case-insensitive. Anything else
+/// returns `default`.
+pub async fn get_bool(cache: &ConfigCache, key: &str, default: bool) -> bool {
+    let Some(raw) = cache.get_raw(key).await else {
+        return default;
+    };
+    match raw.trim().to_ascii_lowercase().as_str() {
+        "true" | "1" | "yes" | "on" => true,
+        "false" | "0" | "no" | "off" => false,
+        _ => default,
+    }
+}

backend/src/services/disk.rs

@@ -0,0 +1,158 @@
+//! Cached view of the filesystem backing the media directory.
+//!
+//! Free/total disk space is needed on two hot paths — the per-user storage quota
+//! (checked on every upload *and* every `GET /me/quota` poll) and the admin stats
+//! endpoint. Reading it means `sysinfo::Disks::new_with_refreshed_list()`, which stats
+//! every mounted filesystem; doing that per request is wasteful for a number that
+//! barely moves. [`DiskCache`] refreshes it at most once per [`TTL`] and serves the
+//! rest from memory.
+
+use std::path::{Path, PathBuf};
+use std::sync::{Arc, RwLock};
+use std::time::{Duration, Instant};
+
+/// How long a disk reading is trusted before the next call re-stats the filesystem.
+const TTL: Duration = Duration::from_secs(15);
+
+#[derive(Clone, Copy)]
+pub struct DiskInfo {
+    pub total: u64,
+    pub free: u64,
+}
+
+/// Cheap-to-clone cache of the media filesystem's total/free bytes. Lives in
+/// `AppState`.
+#[derive(Clone)]
+pub struct DiskCache {
+    inner: Arc<RwLock<Option<(PathBuf, DiskInfo, Instant)>>>,
+}
+
+impl DiskCache {
+    pub fn new() -> Self {
+        Self {
+            inner: Arc::new(RwLock::new(None)),
+        }
+    }
+
+    /// Drop the cached reading so the next `snapshot()` re-measures the filesystem.
+    ///
+    /// Used by the e2e TRUNCATE endpoint. Truncating deletes every uploaded file, which materially
+    /// changes free space — but the cached reading survives for up to the TTL, so the next test can
+    /// compute a quota from the PREVIOUS test's disk. That matters now that the quota tests steer
+    /// the per-user limit off `free_disk_bytes`: a stale reading makes the limit wrong and the test
+    /// flaky, for reasons that have nothing to do with the code under test.
+    pub fn invalidate(&self) {
+        *self.inner.write().unwrap() = None;
+    }
+
+    /// Cached `(total, free)` bytes for the filesystem that holds `media_path`.
+    ///
+    /// Returns `None` when the mount can't be resolved — callers MUST treat that as
+    /// "unknown", never "zero free". (The quota path in particular fails *open* on
+    /// `None`: enforcing a 0-byte limit would lock every user out of uploading.)
+    /// Cached free-space reading for `path`.
+    ///
+    /// The cache is keyed BY PATH. It used to hold a single slot and ignore its argument on a hit,
+    /// so it would happily return the media volume's numbers for any other path within the TTL.
+    /// That was invisible only because every caller happened to pass `media_path` — the first
+    /// caller to ask about a different volume (e.g. the exports volume, which is a separate mount)
+    /// would have silently got the wrong filesystem's free space.
+    pub fn snapshot(&self, path: &Path) -> Option<DiskInfo> {
+        if let Some((cached_path, info, at)) = self.inner.read().unwrap().as_ref()
+            && cached_path == path
+            && at.elapsed() < TTL
+        {
+            return Some(*info);
+        }
+        let info = read_disk_for_path(path)?;
+        *self.inner.write().unwrap() = Some((path.to_path_buf(), info, Instant::now()));
+        Some(info)
+    }
+}
+
+impl Default for DiskCache {
+    fn default() -> Self {
+        Self::new()
+    }
+}
+
+/// Resolve the filesystem backing `media_path` and read its total/free bytes.
+///
+/// Snapshots the mount table via `sysinfo`, then delegates the selection to the pure
+/// [`select_disk`] so the (fiddly, edge-case-prone) matching logic is unit-testable
+/// without touching the real filesystem.
+fn read_disk_for_path(media_path: &Path) -> Option<DiskInfo> {
+    let disks = sysinfo::Disks::new_with_refreshed_list();
+    let mounts: Vec<(String, u64, u64)> = disks
+        .iter()
+        .map(|d| {
+            (
+                d.mount_point().to_string_lossy().to_string(),
+                d.total_space(),
+                d.available_space(),
+            )
+        })
+        .collect();
+    select_disk(&mounts, &media_path.to_string_lossy())
+}
+
+/// Pick the filesystem for `media_path` from a `(mount_point, total, free)` table.
+///
+/// Chooses the **longest** mount point that is a prefix of `media_path` (the most
+/// specific filesystem) rather than the first match — otherwise the root `/` mount,
+/// which prefixes every absolute path, could shadow a dedicated `/media` volume. Falls
+/// back to `/` when nothing prefixes the path (e.g. a relative media path), and to
+/// `None` when even that is absent — the caller treats `None` as "unknown" and fails
+/// open on quota.
+fn select_disk(mounts: &[(String, u64, u64)], media_path: &str) -> Option<DiskInfo> {
+    mounts
+        .iter()
+        .filter(|(mp, _, _)| media_path.starts_with(mp.as_str()))
+        .max_by_key(|(mp, _, _)| mp.len())
+        .or_else(|| mounts.iter().find(|(mp, _, _)| mp == "/"))
+        .map(|(_, total, free)| DiskInfo {
+            total: *total,
+            free: *free,
+        })
+}
+
+#[cfg(test)]
+mod tests {
+    use super::select_disk;
+
+    #[test]
+    fn picks_longest_matching_mount() {
+        // Both "/" and "/media" prefix the path; the dedicated volume must win.
+        let mounts = vec![("/".to_string(), 100, 40), ("/media".to_string(), 200, 150)];
+        let d = select_disk(&mounts, "/media/originals/x.jpg").unwrap();
+        assert_eq!((d.total, d.free), (200, 150));
+    }
+
+    #[test]
+    fn falls_back_to_root_when_no_specific_mount_matches() {
+        let mounts = vec![("/".to_string(), 100, 40), ("/media".to_string(), 200, 150)];
+        // "/var/lib" is only prefixed by "/".
+        let d = select_disk(&mounts, "/var/lib/data").unwrap();
+        assert_eq!((d.total, d.free), (100, 40));
+    }
+
+    #[test]
+    fn relative_path_uses_root_fallback() {
+        let mounts = vec![("/".to_string(), 100, 40)];
+        // A relative path prefixes nothing, so the explicit "/" fallback applies.
+        let d = select_disk(&mounts, "media/originals").unwrap();
+        assert_eq!((d.total, d.free), (100, 40));
+    }
+
+    #[test]
+    fn none_when_no_mount_matches_and_no_root() {
+        // No "/" present and nothing prefixes the relative path → unknown (fail-open).
+        let mounts = vec![("/data".to_string(), 100, 40)];
+        assert!(select_disk(&mounts, "relative/path").is_none());
+    }
+
+    #[test]
+    fn none_on_empty_mount_table() {
+        assert!(select_disk(&[], "/media/x").is_none());
+    }
+}

backend/src/services/maintenance.rs

@@ -0,0 +1,115 @@
+//! Startup recovery + periodic background hygiene.
+//!
+//! Two responsibilities:
+//!
+//! 1. **Startup sweep** — when the server boots, fix rows left in an "in-progress"
+//!    state by the previous (possibly crashed) instance. Compression and export jobs
+//!    each leave a status row when they begin; if the process is killed mid-run, that
+//!    row stays `'processing'` / `'running'` forever, blocking re-tries and leaving
+//!    users staring at a spinner. Resetting them on startup recovers gracefully.
+//!
+//! 2. **Periodic tasks** — pruning that should happen "every hour" rather than per
+//!    request: expired sessions (otherwise the table grows unboundedly), and the
+//!    rate-limiter's in-memory windows (so keys for IPs that left long ago don't
+//!    accumulate).
+
+use std::time::Duration;
+
+use sqlx::PgPool;
+
+use crate::services::rate_limiter::RateLimiter;
+use crate::services::sse_tickets::SseTicketStore;
+
+/// Reset rows left in flight by a previous crashed instance. Run once on startup,
+/// before the HTTP server starts taking requests, so users never observe the
+/// half-state.
+pub async fn startup_recovery(pool: &PgPool) {
+    // Uploads whose preview generation was interrupted. Marking them 'failed' is
+    // safer than re-queueing — the original file is still on disk, the user can
+    // delete + re-upload if they care, and we avoid double-processing risk.
+    match sqlx::query(
+        "UPDATE upload SET compression_status = 'failed'
+         WHERE compression_status = 'processing'",
+    )
+    .execute(pool)
+    .await
+    {
+        Ok(r) if r.rows_affected() > 0 => {
+            tracing::warn!(
+                "startup recovery: reset {} stuck upload(s) from 'processing' to 'failed'",
+                r.rows_affected()
+            );
+        }
+        Ok(_) => {}
+        Err(e) => tracing::error!("startup recovery: failed to sweep uploads: {e:#}"),
+    }
+
+    // Export jobs interrupted mid-run are marked 'failed' here so they aren't left
+    // 'running' forever. The host CANNOT re-trigger a released export (release_gallery
+    // rejects an already-released event), so `export::recover_exports` re-spawns these
+    // failed-but-released jobs from `main` once `AppState` exists (it needs the media
+    // paths + SSE sender this fn doesn't have).
+    //
+    // SINGLE-INSTANCE ASSUMPTION: this sweep is unscoped — it reaps every `running` row, not just
+    // this process's. That is correct for the supported deployment (one app container; see
+    // docker-compose.yml), where no other process can own a job at boot. If EventSnap is ever run
+    // with more than one replica, a booting replica would mark a peer's live workers `failed`.
+    // This is NOT merely wasteful — do not assume the epoch model contains it. Recovery re-arms the
+    // job at the SAME epoch, so the reaped-but-still-alive worker and the fresh one would BOTH hold
+    // that epoch; the guards carry no owner/lease token, so they share the same artifact paths and
+    // either can win the other's `finalize_job`. That can corrupt or delete the keepsake. Running
+    // more than one replica therefore requires an owner/lease/heartbeat on export_job first. That
+    // machinery is not worth building for a single-container app — so this is a documented
+    // CONSTRAINT, not a hidden assumption: do not scale this service horizontally as-is.
+    match sqlx::query(
+        "UPDATE export_job
+         SET status = 'failed',
+             error_message = COALESCE(error_message, 'Server-Neustart während des Exports')
+         WHERE status = 'running'",
+    )
+    .execute(pool)
+    .await
+    {
+        Ok(r) if r.rows_affected() > 0 => {
+            tracing::warn!(
+                "startup recovery: reset {} stuck export job(s) from 'running' to 'failed'",
+                r.rows_affected()
+            );
+        }
+        Ok(_) => {}
+        Err(e) => tracing::error!("startup recovery: failed to sweep export_job: {e:#}"),
+    }
+}
+
+/// Spawns a background task that periodically:
+/// - deletes session rows whose `expires_at` is more than a day in the past
+/// - prunes the in-memory rate-limiter HashMap of empty windows
+/// - drops expired SSE tickets (30s TTL but the map keeps the slot until pruned)
+///
+/// Cadence is 1h — fine for both jobs at our scale.
+pub fn spawn_periodic_tasks(pool: PgPool, rate_limiter: RateLimiter, sse_tickets: SseTicketStore) {
+    tokio::spawn(async move {
+        let mut tick = tokio::time::interval(Duration::from_secs(3600));
+        // Fire the first tick immediately, then hourly.
+        tick.tick().await;
+        loop {
+            tick.tick().await;
+            cleanup_sessions(&pool).await;
+            rate_limiter.prune();
+            sse_tickets.prune();
+        }
+    });
+}
+
+async fn cleanup_sessions(pool: &PgPool) {
+    match sqlx::query("DELETE FROM session WHERE expires_at < NOW() - INTERVAL '1 day'")
+        .execute(pool)
+        .await
+    {
+        Ok(r) if r.rows_affected() > 0 => {
+            tracing::info!("cleaned up {} expired session(s)", r.rows_affected());
+        }
+        Ok(_) => {}
+        Err(e) => tracing::warn!("session cleanup failed: {e:#}"),
+    }
+}

backend/src/services/mod.rs

@@ -1,3 +1,7 @@
 pub mod compression;
+pub mod config;
+pub mod disk;
 pub mod export;
+pub mod maintenance;
 pub mod rate_limiter;
+pub mod sse_tickets;

backend/src/services/rate_limiter.rs

@@ -24,7 +24,12 @@ impl RateLimiter {
 
     /// Returns `Ok(())` if allowed, `Err(retry_after_secs)` if rate-limited.
     /// `retry_after_secs` is how long until the oldest slot in the window expires.
-    pub fn check_with_retry(&self, key: impl Into<String>, max: usize, window: Duration) -> Result<(), u64> {
+    pub fn check_with_retry(
+        &self,
+        key: impl Into<String>,
+        max: usize,
+        window: Duration,
+    ) -> Result<(), u64> {
         let now = Instant::now();
         let key = key.into();
         let mut map = self.windows.lock().unwrap();
@@ -41,15 +46,248 @@ impl RateLimiter {
             Err(remaining.as_secs().max(1))
         }
     }
+
+    /// Wipe every tracked window. Used by the test-mode truncate route so a previous
+    /// test's accumulated counters don't bleed into the next test's rate-limit checks.
+    pub fn clear(&self) {
+        self.windows.lock().unwrap().clear();
+    }
+
+    /// Drop keys whose windows are empty after expiring old timestamps. Called from a
+    /// background task (see [`crate::services::maintenance`]) so that long-lived
+    /// processes don't accumulate one HashMap entry per IP that ever connected.
+    ///
+    /// Uses a conservative 24h ceiling — anything older than that is gone regardless
+    /// of which endpoint's window it was tracked under (the longest window today is
+    /// 24h for export downloads). If we ever add longer windows, raise this constant.
+    pub fn prune(&self) {
+        let now = Instant::now();
+        let ceiling = Duration::from_secs(24 * 60 * 60);
+        let mut map = self.windows.lock().unwrap();
+        let before = map.len();
+        map.retain(|_, ts| {
+            ts.retain(|&t| now.duration_since(t) < ceiling);
+            !ts.is_empty()
+        });
+        let dropped = before.saturating_sub(map.len());
+        if dropped > 0 {
+            tracing::debug!("rate limiter pruned {dropped} idle keys");
+        }
+    }
 }
 
-/// Extract the client IP from X-Forwarded-For (Caddy sets this) or fall back
-/// to a provided socket address string.
+/// Extract the client IP from X-Forwarded-For or fall back to a provided socket
+/// address string.
+///
+/// We take the **right-most** entry, not the left-most. Caddy is the sole ingress
+/// and the app port is only `expose`d (never published), so the last hop Caddy
+/// appends is the real client. A client can prepend arbitrary spoofed values to
+/// the left of XFF to dodge throttles — those are ignored here. This assumes
+/// exactly one trusted proxy (Caddy); revisit if that changes.
 pub fn client_ip(headers: &axum::http::HeaderMap, fallback: &str) -> String {
     headers
         .get("x-forwarded-for")
         .and_then(|v| v.to_str().ok())
-        .and_then(|s| s.split(',').next())
+        .and_then(|s| s.rsplit(',').next())
         .map(|s| s.trim().to_owned())
+        .filter(|s| !s.is_empty())
         .unwrap_or_else(|| fallback.to_owned())
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use axum::http::HeaderMap;
+
+    const MIN: Duration = Duration::from_secs(60);
+
+    #[test]
+    fn allows_up_to_max_then_blocks() {
+        let rl = RateLimiter::new();
+        assert!(rl.check("k", 3, MIN));
+        assert!(rl.check("k", 3, MIN));
+        assert!(rl.check("k", 3, MIN));
+        assert!(!rl.check("k", 3, MIN), "the 4th request must be blocked");
+    }
+
+    #[test]
+    fn keys_are_independent() {
+        let rl = RateLimiter::new();
+        assert!(rl.check("a", 1, MIN));
+        assert!(!rl.check("a", 1, MIN));
+        assert!(rl.check("b", 1, MIN), "a different key has its own window");
+    }
+
+    #[test]
+    fn window_slides_and_allows_again_after_expiry() {
+        let rl = RateLimiter::new();
+        let w = Duration::from_millis(40);
+        assert!(rl.check("k", 1, w));
+        assert!(!rl.check("k", 1, w));
+        std::thread::sleep(Duration::from_millis(55));
+        assert!(
+            rl.check("k", 1, w),
+            "the slot should expire once the window passes"
+        );
+    }
+
+    /// `retry_after` is not a "some number in range" — it is the time until the oldest slot
+    /// in the window frees up, and it is surfaced to clients as the backoff they sleep for
+    /// (see `upload-queue.ts`). Asserting only `(1..=60)` spans the entire reachable domain
+    /// of a 60s window, so a hardcoded `Err(1)` would satisfy it while telling every client
+    /// to hammer the server a second later. Pin the actual value.
+    #[test]
+    fn retry_after_is_the_remaining_window() {
+        let rl = RateLimiter::new();
+
+        // The slot was consumed just now, so essentially the whole window remains.
+        // `as_secs()` truncates the sub-second remainder, so a 30s window reports 29.
+        let w30 = Duration::from_secs(30);
+        assert!(rl.check_with_retry("a", 1, w30).is_ok());
+        let a = rl.check_with_retry("a", 1, w30).unwrap_err();
+        assert_eq!(a, 29, "retry_after must be the remaining window, got {a}");
+
+        // A different window must yield a different retry_after: no single constant can
+        // satisfy both this and the assertion above.
+        let w10 = Duration::from_secs(10);
+        assert!(rl.check_with_retry("b", 1, w10).is_ok());
+        let b = rl.check_with_retry("b", 1, w10).unwrap_err();
+        assert_eq!(b, 9, "retry_after must scale with the window, got {b}");
+    }
+
+    #[test]
+    fn retry_after_counts_down_as_the_window_elapses() {
+        let rl = RateLimiter::new();
+        let w = Duration::from_secs(30);
+        assert!(rl.check_with_retry("k", 1, w).is_ok());
+        let first = rl.check_with_retry("k", 1, w).unwrap_err();
+
+        std::thread::sleep(Duration::from_millis(1200));
+        let second = rl.check_with_retry("k", 1, w).unwrap_err();
+
+        // A client that waits 1.2s must be told to wait ~1.2s less — otherwise the advertised
+        // backoff is a constant, not a deadline.
+        let shaved = first - second;
+        assert!(
+            (1..=2).contains(&shaved),
+            "1.2s of waiting must shorten the advertised backoff by ~1s (got {first} then {second})"
+        );
+    }
+
+    #[test]
+    fn retry_after_floors_at_one_second() {
+        let rl = RateLimiter::new();
+        let w = Duration::from_millis(800);
+        assert!(rl.check_with_retry("k", 1, w).is_ok());
+        let retry = rl.check_with_retry("k", 1, w).unwrap_err();
+        // The sub-second remainder truncates to 0; clients must never be told "retry in 0s"
+        // (that's a busy-loop). The `.max(1)` floor is what prevents it.
+        assert_eq!(
+            retry, 1,
+            "a sub-second remainder must floor to 1, got {retry}"
+        );
+    }
+
+    #[test]
+    fn clear_resets_every_window() {
+        let rl = RateLimiter::new();
+        assert!(rl.check("k", 1, MIN));
+        assert!(!rl.check("k", 1, MIN));
+        rl.clear();
+        assert!(rl.check("k", 1, MIN), "clear() must free the window");
+    }
+
+    /// `prune()` is a memory-leak guard: without it a long-lived process keeps one HashMap
+    /// entry per IP that ever connected. Nothing in the public API observes the map size, so
+    /// the only way to catch a no-op body (`fn prune(&self) {}`) is to look at the map — the
+    /// tests module can see the private field.
+    #[test]
+    fn prune_drops_keys_whose_windows_have_fully_expired() {
+        let rl = RateLimiter::new();
+
+        // A key whose only timestamp is older than the 24h ceiling. We can't sleep for a day,
+        // so backdate the Instant directly.
+        let ancient = Instant::now()
+            .checked_sub(Duration::from_secs(25 * 60 * 60))
+            .expect("backdating an Instant by 25h");
+        rl.windows
+            .lock()
+            .unwrap()
+            .insert("stale".to_string(), vec![ancient]);
+
+        // ...alongside a key that is still inside its window.
+        assert!(rl.check("live", 5, MIN));
+        assert_eq!(rl.windows.lock().unwrap().len(), 2);
+
+        rl.prune();
+
+        let map = rl.windows.lock().unwrap();
+        assert!(
+            !map.contains_key("stale"),
+            "prune() must drop keys whose timestamps have all expired"
+        );
+        assert!(
+            map.contains_key("live"),
+            "prune() must keep keys that still have live timestamps"
+        );
+        assert_eq!(map.len(), 1, "exactly one key should survive the prune");
+    }
+
+    #[test]
+    fn prune_does_not_reset_a_live_window() {
+        // The counterpart to the test above: pruning must reclaim memory, never quota. If
+        // prune() dropped live keys, every background sweep would hand attackers a fresh
+        // budget.
+        let rl = RateLimiter::new();
+        assert!(rl.check("k", 1, MIN));
+        assert!(!rl.check("k", 1, MIN));
+
+        rl.prune();
+
+        assert!(
+            !rl.check("k", 1, MIN),
+            "prune() must not clear a window that is still active"
+        );
+    }
+
+    #[test]
+    fn client_ip_takes_rightmost_forwarded_for_entry() {
+        // The right-most entry is the hop our trusted proxy (Caddy) appended.
+        let mut h = HeaderMap::new();
+        h.insert("x-forwarded-for", "10.0.0.1, 203.0.113.7".parse().unwrap());
+        assert_eq!(client_ip(&h, "fallback"), "203.0.113.7");
+    }
+
+    #[test]
+    fn client_ip_ignores_spoofed_leftmost_entry() {
+        // A client prepending a fake IP to dodge throttles must not win.
+        let mut h = HeaderMap::new();
+        h.insert(
+            "x-forwarded-for",
+            "1.2.3.4, 9.9.9.9, 203.0.113.7".parse().unwrap(),
+        );
+        assert_eq!(client_ip(&h, "fallback"), "203.0.113.7");
+    }
+
+    #[test]
+    fn client_ip_trims_surrounding_whitespace() {
+        let mut h = HeaderMap::new();
+        h.insert("x-forwarded-for", "  198.51.100.5  ".parse().unwrap());
+        assert_eq!(client_ip(&h, "fb"), "198.51.100.5");
+    }
+
+    #[test]
+    fn client_ip_falls_back_when_header_absent() {
+        assert_eq!(client_ip(&HeaderMap::new(), "127.0.0.1"), "127.0.0.1");
+    }
+
+    #[test]
+    fn client_ip_falls_back_on_trailing_comma_empty_entry() {
+        // A trailing comma leaves an empty right-most segment after trimming; the
+        // `.filter(!is_empty)` must reject it and fall through to the fallback
+        // rather than returning "" (which would collapse callers into one bucket).
+        let mut h = HeaderMap::new();
+        h.insert("x-forwarded-for", "203.0.113.7, ".parse().unwrap());
+        assert_eq!(client_ip(&h, "127.0.0.1"), "127.0.0.1");
+    }
+}

backend/src/services/sse_tickets.rs

@@ -0,0 +1,144 @@
+use std::collections::HashMap;
+use std::sync::{Arc, Mutex};
+use std::time::{Duration, Instant};
+
+use rand::Rng;
+
+/// Short-lived single-use tickets that let `EventSource` clients open the SSE
+/// stream without putting the JWT in the URL (where it would leak via access
+/// logs / referer / browser history).
+///
+/// Flow: client `POST /api/v1/stream/ticket` with `Authorization: Bearer <jwt>`,
+/// server returns an opaque ticket, client passes it as `?ticket=...` on the
+/// stream open. Tickets are consumed on use and expire after `TTL`.
+const TTL: Duration = Duration::from_secs(30);
+
+#[derive(Clone)]
+pub struct SseTicketStore {
+    inner: Arc<Mutex<HashMap<String, Entry>>>,
+}
+
+#[derive(Clone)]
+struct Entry {
+    token_hash: String,
+    issued_at: Instant,
+}
+
+impl SseTicketStore {
+    pub fn new() -> Self {
+        Self {
+            inner: Arc::new(Mutex::new(HashMap::new())),
+        }
+    }
+
+    /// Drop every outstanding ticket. Used by the e2e TRUNCATE endpoint: tickets are bound to a
+    /// session token hash, and TRUNCATE deletes the sessions out from under them, so anything left
+    /// here is a dangling reference to a user that no longer exists.
+    pub fn clear(&self) {
+        self.inner.lock().unwrap().clear();
+    }
+
+    /// Mint a new ticket bound to the caller's session (identified by token hash).
+    pub fn issue(&self, token_hash: String) -> String {
+        let ticket = random_ticket();
+        let mut map = self.inner.lock().unwrap();
+        map.insert(
+            ticket.clone(),
+            Entry {
+                token_hash,
+                issued_at: Instant::now(),
+            },
+        );
+        ticket
+    }
+
+    /// Consume a ticket. Returns `Some(token_hash)` if the ticket exists and is
+    /// not expired. Single-use: the ticket is removed regardless of whether it
+    /// was still fresh, so a replay can't slip through after expiry.
+    pub fn consume(&self, ticket: &str) -> Option<String> {
+        let mut map = self.inner.lock().unwrap();
+        let entry = map.remove(ticket)?;
+        if entry.issued_at.elapsed() > TTL {
+            return None;
+        }
+        Some(entry.token_hash)
+    }
+
+    /// Drop expired entries — called from the background maintenance task so a
+    /// long-running process doesn't accumulate stale tickets.
+    pub fn prune(&self) {
+        let mut map = self.inner.lock().unwrap();
+        map.retain(|_, e| e.issued_at.elapsed() <= TTL);
+    }
+}
+
+fn random_ticket() -> String {
+    // 192 bits of randomness, base32-ish hex. Plenty of entropy and URL-safe.
+    let mut rng = rand::rng();
+    let mut bytes = [0u8; 24];
+    rng.fill(&mut bytes);
+    bytes.iter().map(|b| format!("{b:02x}")).collect()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn issue_then_consume_returns_the_hash_exactly_once() {
+        let store = SseTicketStore::new();
+        let ticket = store.issue("hash-1".into());
+        assert_eq!(store.consume(&ticket).as_deref(), Some("hash-1"));
+        // Single-use: a replay of the same ticket is rejected.
+        assert_eq!(
+            store.consume(&ticket),
+            None,
+            "a consumed ticket must not be reusable"
+        );
+    }
+
+    #[test]
+    fn unknown_ticket_consumes_to_none() {
+        let store = SseTicketStore::new();
+        assert_eq!(store.consume("never-issued"), None);
+    }
+
+    #[test]
+    fn issued_tickets_are_unique_and_hex() {
+        let store = SseTicketStore::new();
+        let a = store.issue("h".into());
+        let b = store.issue("h".into());
+        assert_ne!(a, b, "each ticket must be unique");
+        assert_eq!(a.len(), 48, "24 random bytes → 48 hex chars");
+        assert!(a.chars().all(|c| c.is_ascii_hexdigit()));
+    }
+
+    #[test]
+    fn fresh_ticket_survives_prune() {
+        let store = SseTicketStore::new();
+        let ticket = store.issue("h".into());
+        store.prune(); // not expired → kept
+        assert_eq!(store.consume(&ticket).as_deref(), Some("h"));
+    }
+
+    #[test]
+    fn expired_ticket_consumes_to_none() {
+        // Construct an entry that is already past the TTL and confirm consume() rejects it.
+        let store = SseTicketStore::new();
+        let stale = "stale-ticket".to_string();
+        store.inner.lock().unwrap().insert(
+            stale.clone(),
+            Entry {
+                token_hash: "h".into(),
+                issued_at: Instant::now()
+                    .checked_sub(TTL + Duration::from_secs(1))
+                    .expect("host uptime should exceed the ticket TTL"),
+            },
+        );
+        assert_eq!(
+            store.consume(&stale),
+            None,
+            "an expired ticket must not authenticate"
+        );
+    }
+}

backend/src/state.rs

@@ -3,7 +3,10 @@ use tokio::sync::broadcast;
 
 use crate::config::AppConfig;
 use crate::services::compression::CompressionWorker;
+use crate::services::config::ConfigCache;
+use crate::services::disk::DiskCache;
 use crate::services::rate_limiter::RateLimiter;
+use crate::services::sse_tickets::SseTicketStore;
 
 #[derive(Clone, Debug)]
 pub struct SseEvent {
@@ -11,6 +14,17 @@ pub struct SseEvent {
     pub data: String,
 }
 
+impl SseEvent {
+    /// Standardised constructor. Prefer this over building the struct inline so the
+    /// event-type strings stay consistent across handlers.
+    pub fn new(event_type: impl Into<String>, data: impl Into<String>) -> Self {
+        Self {
+            event_type: event_type.into(),
+            data: data.into(),
+        }
+    }
+}
+
 #[derive(Clone)]
 pub struct AppState {
     pub pool: PgPool,
@@ -18,19 +32,37 @@ pub struct AppState {
     pub sse_tx: broadcast::Sender<SseEvent>,
     pub compression: CompressionWorker,
     pub rate_limiter: RateLimiter,
+    pub sse_tickets: SseTicketStore,
+    /// In-memory cache in front of the `config` table. Reads go through here; the
+    /// admin PATCH handler and the test reseed invalidate it after committing.
+    pub config_cache: ConfigCache,
+    /// Cached total/free bytes for the media filesystem (quota + admin stats).
+    pub disk_cache: DiskCache,
 }
 
 impl AppState {
     pub fn new(pool: PgPool, config: AppConfig) -> Self {
-        let (sse_tx, _) = broadcast::channel(256);
-        let compression =
-            CompressionWorker::new(pool.clone(), config.media_path.clone(), 2);
+        // Broadcast buffer for live SSE fan-out. Sized to absorb a burst (e.g. many
+        // uploads landing at once during a busy moment) before a slow consumer lags and
+        // has to `resync`. The resync path is a correctness backstop, not the happy path —
+        // a roomier buffer keeps ~1000 concurrent clients from all resyncing at once.
+        let (sse_tx, _) = broadcast::channel(1024);
+        let compression = CompressionWorker::new(
+            pool.clone(),
+            config.media_path.clone(),
+            config.compression_concurrency,
+            sse_tx.clone(),
+        );
+        let config_cache = ConfigCache::new(pool.clone());
         Self {
             pool,
             config,
             sse_tx,
             compression,
             rate_limiter: RateLimiter::new(),
+            sse_tickets: SseTicketStore::new(),
+            config_cache,
+            disk_cache: DiskCache::new(),
         }
     }
 }

backend/tests/common/mod.rs

@@ -0,0 +1,257 @@
+//! Shared fixtures for the DB-backed integration tests.
+//!
+//! Every helper here executes SQL that is **character-for-character identical** to what `src/`
+//! actually runs (see the `// SRC:` markers). That is the whole point: a paraphrased query is a
+//! query nobody runs, and a test that passes against a paraphrase proves nothing about production.
+
+#![allow(dead_code)] // each integration-test crate uses a different subset of these helpers
+
+use sqlx::PgPool;
+use uuid::Uuid;
+
+/// Insert a bare, unreleased event (epoch 0, uploads open).
+pub async fn seed_event(pool: &PgPool, slug: &str) -> Uuid {
+    sqlx::query_scalar("INSERT INTO event (slug, name) VALUES ($1, $2) RETURNING id")
+        .bind(slug)
+        .bind("Hochzeit")
+        .fetch_one(pool)
+        .await
+        .expect("seed event")
+}
+
+/// Insert a guest with a zeroed byte total.
+pub async fn seed_user(pool: &PgPool, event_id: Uuid, name: &str) -> Uuid {
+    sqlx::query_scalar(
+        "INSERT INTO \"user\" (event_id, display_name, recovery_pin_hash)
+         VALUES ($1, $2, 'x') RETURNING id",
+    )
+    .bind(event_id)
+    .bind(name)
+    .fetch_one(pool)
+    .await
+    .expect("seed user")
+}
+
+/// SRC: `handlers/host.rs::release_gallery` — the claim + epoch bump, verbatim.
+/// Returns the POST-increment epoch, exactly as the handler consumes it.
+pub async fn release_gallery(pool: &PgPool, slug: &str) -> Option<i64> {
+    let claimed: Option<(Uuid, String, i64)> = sqlx::query_as(
+        "UPDATE event
+         SET export_released_at = NOW(),
+             uploads_locked_at  = COALESCE(uploads_locked_at, NOW()),
+             export_epoch       = export_epoch + 1
+         WHERE slug = $1 AND export_released_at IS NULL
+         RETURNING id, name, export_epoch",
+    )
+    .bind(slug)
+    .fetch_optional(pool)
+    .await
+    .expect("release_gallery");
+
+    if let Some((event_id, _, epoch)) = claimed {
+        // The handler arms both jobs in the SAME transaction; for a single-connection fixture the
+        // sequencing is equivalent.
+        let mut conn = pool.acquire().await.expect("acquire");
+        enqueue_types_at_epoch(&mut conn, event_id, epoch, &["zip", "html"]).await;
+        Some(epoch)
+    } else {
+        None
+    }
+}
+
+/// SRC: `handlers/host.rs::open_event` — the one statement that retires an entire generation.
+/// Returns rows affected.
+pub async fn open_event(pool: &PgPool, slug: &str) -> u64 {
+    sqlx::query(
+        "UPDATE event
+         SET uploads_locked_at  = NULL,
+             export_released_at = NULL,
+             export_epoch       = export_epoch + 1
+         WHERE slug = $1 AND (uploads_locked_at IS NOT NULL OR export_released_at IS NOT NULL)",
+    )
+    .bind(slug)
+    .execute(pool)
+    .await
+    .expect("open_event")
+    .rows_affected()
+}
+
+/// SRC: `services/export.rs::enqueue_types_at_epoch` — verbatim upsert.
+pub async fn enqueue_types_at_epoch(
+    conn: &mut sqlx::PgConnection,
+    event_id: Uuid,
+    epoch: i64,
+    types: &[&str],
+) {
+    for export_type in types {
+        sqlx::query(
+            "INSERT INTO export_job (event_id, type, status, progress_pct, epoch)
+             VALUES ($1, $2::export_type, 'pending', 0, $3)
+             ON CONFLICT (event_id, type) DO UPDATE
+                 SET status = 'pending', progress_pct = 0, file_path = NULL,
+                     error_message = NULL, completed_at = NULL,
+                     epoch = EXCLUDED.epoch
+             WHERE export_job.status <> 'done' OR export_job.epoch <> EXCLUDED.epoch",
+        )
+        .bind(event_id)
+        .bind(export_type)
+        .bind(epoch)
+        .execute(&mut *conn)
+        .await
+        .expect("enqueue_types_at_epoch");
+    }
+}
+
+/// SRC: `services/export.rs::claim_job` — verbatim. `true` = we won the generation.
+pub async fn claim_job(pool: &PgPool, event_id: Uuid, export_type: &str, epoch: i64) -> bool {
+    sqlx::query(
+        "UPDATE export_job SET status = 'running'
+         WHERE event_id = $1 AND type = $2::export_type
+           AND epoch = $3 AND status = 'pending'",
+    )
+    .bind(event_id)
+    .bind(export_type)
+    .bind(epoch)
+    .execute(pool)
+    .await
+    .expect("claim_job")
+    .rows_affected()
+        > 0
+}
+
+/// SRC: `services/export.rs::finalize_job` — verbatim. This IS the publish step.
+pub async fn finalize_job(
+    pool: &PgPool,
+    event_id: Uuid,
+    export_type: &str,
+    epoch: i64,
+    file_path: &str,
+) -> bool {
+    sqlx::query(
+        "UPDATE export_job
+         SET status = 'done', progress_pct = 100, file_path = $3, completed_at = NOW()
+         WHERE event_id = $1 AND type = $2::export_type
+           AND epoch = $4 AND status = 'running'",
+    )
+    .bind(event_id)
+    .bind(export_type)
+    .bind(file_path)
+    .bind(epoch)
+    .execute(pool)
+    .await
+    .expect("finalize_job")
+    .rows_affected()
+        > 0
+}
+
+/// SRC: `services/export.rs::update_progress` — verbatim. Doubles as the worker's liveness check:
+/// `false` means "your generation was retired, stop working".
+pub async fn update_progress(
+    pool: &PgPool,
+    event_id: Uuid,
+    export_type: &str,
+    epoch: i64,
+    pct: i16,
+) -> bool {
+    sqlx::query(
+        "UPDATE export_job SET progress_pct = $3
+         WHERE event_id = $1 AND type = $2::export_type
+           AND epoch = $4 AND status = 'running'",
+    )
+    .bind(event_id)
+    .bind(export_type)
+    .bind(pct)
+    .bind(epoch)
+    .execute(pool)
+    .await
+    .expect("update_progress")
+    .rows_affected()
+        > 0
+}
+
+/// SRC: `services/export.rs::invalidate_and_arm` — the ViewerOnly ZIP carry-forward, verbatim.
+/// Returns `rows_affected() == 1`, which is what the production code branches on.
+pub async fn carry_zip_forward(pool: &PgPool, event_id: Uuid, epoch: i64) -> bool {
+    sqlx::query(
+        "UPDATE export_job SET epoch = $2
+         WHERE event_id = $1 AND type = 'zip'::export_type
+           AND status = 'done' AND epoch = $2 - 1",
+    )
+    .bind(event_id)
+    .bind(epoch)
+    .execute(pool)
+    .await
+    .expect("carry_zip_forward")
+    .rows_affected()
+        == 1
+}
+
+/// SRC: `services/export.rs::invalidate_and_arm` — the epoch bump, verbatim.
+pub async fn bump_epoch(pool: &PgPool, slug: &str) -> Option<(Uuid, String, i64)> {
+    sqlx::query_as(
+        "UPDATE event SET export_epoch = export_epoch + 1
+         WHERE slug = $1 AND export_released_at IS NOT NULL
+         RETURNING id, name, export_epoch",
+    )
+    .bind(slug)
+    .fetch_optional(pool)
+    .await
+    .expect("bump_epoch")
+}
+
+/// The event's authoritative epoch.
+pub async fn event_epoch(pool: &PgPool, event_id: Uuid) -> i64 {
+    sqlx::query_scalar("SELECT export_epoch FROM event WHERE id = $1")
+        .bind(event_id)
+        .fetch_one(pool)
+        .await
+        .expect("event_epoch")
+}
+
+/// The raw job row, bypassing `export_current` — what the WORKER sees.
+pub async fn job_row(
+    pool: &PgPool,
+    event_id: Uuid,
+    export_type: &str,
+) -> Option<(String, i64, Option<String>)> {
+    sqlx::query_as(
+        "SELECT status::text, epoch, file_path FROM export_job
+          WHERE event_id = $1 AND type = $2::export_type",
+    )
+    .bind(event_id)
+    .bind(export_type)
+    .fetch_optional(pool)
+    .await
+    .expect("job_row")
+}
+
+/// Does `export_current` expose this job at all? (The view itself, without the `status` filter —
+/// it is what `handlers/admin.rs::export_status` reports to the host UI.)
+pub async fn in_export_current(pool: &PgPool, event_id: Uuid, export_type: &str) -> bool {
+    sqlx::query_scalar::<_, i64>(
+        "SELECT COUNT(*) FROM export_current
+          WHERE event_id = $1 AND type = $2::export_type",
+    )
+    .bind(event_id)
+    .bind(export_type)
+    .fetch_one(pool)
+    .await
+    .expect("in_export_current")
+        > 0
+}
+
+/// THE download predicate. SRC: `handlers/admin.rs::download_export` reads exactly this shape —
+/// `SELECT c.file_path FROM export_current c WHERE ... AND c.status = 'done'`. If this returns
+/// `Some`, a guest can download the keepsake; if `None`, they get a 404.
+pub async fn downloadable(pool: &PgPool, event_id: Uuid, export_type: &str) -> Option<String> {
+    sqlx::query_scalar(
+        "SELECT c.file_path FROM export_current c
+          WHERE c.event_id = $1 AND c.type = $2::export_type AND c.status = 'done'",
+    )
+    .bind(event_id)
+    .bind(export_type)
+    .fetch_optional(pool)
+    .await
+    .expect("downloadable")
+    .flatten()
+}

backend/tests/export_epoch.rs

@@ -0,0 +1,435 @@
+//! DB-backed integration tests for the export epoch state machine (migration 014).
+//!
+//! These run against a REAL Postgres: `#[sqlx::test]` creates a throwaway database per test and
+//! runs `backend/migrations/` into it, so the schema, the enums, the `UNIQUE (event_id, type)`
+//! constraint and the `export_current` view are the production ones — not a mock.
+//!
+//! THE INVARIANT, from migration 014:
+//!
+//!     An export is downloadable IFF
+//!         event.export_released_at IS NOT NULL
+//!     AND export_job.epoch = event.export_epoch
+//!     AND export_job.status = 'done'
+//!
+//! Readiness is DERIVED (the `export_current` view), never stored. Every test below pins one leg of
+//! that invariant with the exact SQL `src/` executes (see `tests/common/mod.rs`).
+
+mod common;
+
+use common::*;
+use sqlx::PgPool;
+
+// ─────────────────────────────────────────────────────────────────────────────
+// 1. Epoch monotonicity
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Release, reopen and re-release each bump `export_epoch`, and `RETURNING export_epoch` hands the
+/// caller the POST-increment value.
+///
+/// PREVENTS: a worker born with the PRE-increment epoch. It would be inert from the instant it
+/// started — every one of its writes is `epoch`-guarded, so `claim_job`/`finalize_job` would match
+/// nothing, the job row would sit at `pending` 0% forever with no live worker, and the host's
+/// download button would spin and then 404. The keepsake would never be built at all.
+#[sqlx::test]
+async fn release_returns_post_increment_epoch(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    assert_eq!(
+        event_epoch(&pool, event_id).await,
+        0,
+        "a fresh event starts at epoch 0"
+    );
+
+    let released = release_gallery(&pool, "wedding")
+        .await
+        .expect("release claims the event");
+    assert_eq!(
+        released, 1,
+        "RETURNING must give the epoch AFTER the +1, not before"
+    );
+    assert_eq!(
+        event_epoch(&pool, event_id).await,
+        released,
+        "worker's epoch == event's epoch"
+    );
+
+    // The jobs armed by the release carry exactly that epoch — this is what makes the worker's
+    // guarded writes match.
+    for t in ["zip", "html"] {
+        let (status, epoch, _) = job_row(&pool, event_id, t).await.expect("job armed");
+        assert_eq!(status, "pending");
+        assert_eq!(
+            epoch, released,
+            "{t} job must be armed at the epoch the worker was born with"
+        );
+    }
+}
+
+/// Epoch is strictly monotonic across the whole release/reopen/re-release cycle, and a second
+/// release attempt while already released is rejected WITHOUT bumping.
+///
+/// PREVENTS: epoch reuse. If a reopen could return the event to an epoch some old `done` row still
+/// carries, a retired keepsake — one that a guest asked to be taken down from — would silently
+/// become downloadable again.
+#[sqlx::test]
+async fn epoch_is_strictly_monotonic_across_reopen(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+
+    assert_eq!(release_gallery(&pool, "wedding").await, Some(1));
+
+    // A duplicate release is a no-op (`WHERE export_released_at IS NULL`) and must NOT bump.
+    assert_eq!(
+        release_gallery(&pool, "wedding").await,
+        None,
+        "already released"
+    );
+    assert_eq!(
+        event_epoch(&pool, event_id).await,
+        1,
+        "a rejected release must not move the epoch"
+    );
+
+    // Reopen retires the generation with ONE write.
+    assert_eq!(open_event(&pool, "wedding").await, 1);
+    assert_eq!(event_epoch(&pool, event_id).await, 2, "reopen bumps");
+
+    // And re-releasing bumps again — never back to 1.
+    assert_eq!(
+        release_gallery(&pool, "wedding").await,
+        Some(3),
+        "re-release bumps again"
+    );
+    assert_eq!(event_epoch(&pool, event_id).await, 3);
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+// 2. A retired-epoch worker is inert
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// A worker holding a retired epoch cannot write anything anybody can see: once the rows have been
+/// re-armed at a newer epoch, its `update_progress` and `finalize_job` both match 0 rows, and
+/// `export_current` never exposes its output.
+///
+/// PREVENTS: the classic lost race — a slow worker from BEFORE a takedown finishing afterwards and
+/// publishing an archive that still contains the photo a guest asked to have removed. "Please take
+/// my photo out" is the one request that most needs to reach the keepsake, and the keepsake is the
+/// artifact people keep forever.
+#[sqlx::test]
+async fn retired_epoch_worker_writes_are_no_ops(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    let old_epoch = release_gallery(&pool, "wedding").await.unwrap();
+
+    // Worker A is born at epoch 1 and claims the ZIP.
+    assert!(
+        claim_job(&pool, event_id, "zip", old_epoch).await,
+        "worker A wins its claim"
+    );
+    assert!(
+        update_progress(&pool, event_id, "zip", old_epoch, 40).await,
+        "still live at 40%"
+    );
+
+    // ── A takedown lands mid-export: `invalidate_and_arm(Affects::Both)` bumps and re-arms. ──
+    let (_, _, new_epoch) = bump_epoch(&pool, "wedding")
+        .await
+        .expect("bump on a released event");
+    assert_eq!(new_epoch, old_epoch + 1);
+    let mut conn = pool.acquire().await.unwrap();
+    enqueue_types_at_epoch(&mut conn, event_id, new_epoch, &["zip", "html"]).await;
+    drop(conn);
+
+    // Worker A is now INERT BY CONSTRUCTION. Every write is guarded on its own birth epoch.
+    assert!(
+        !update_progress(&pool, event_id, "zip", old_epoch, 90).await,
+        "the liveness check must report `false` so worker A stops grinding through the gallery"
+    );
+    assert!(
+        !finalize_job(&pool, event_id, "zip", old_epoch, "exports/Gallery.1.zip").await,
+        "worker A's finalize MUST affect 0 rows — this is the write that would have published a \
+         keepsake still containing the taken-down photo"
+    );
+
+    // The re-armed row is untouched by the loser: still pending at the LIVE epoch, waiting for the
+    // fresh worker. (If worker A had won, this row would read `done` at epoch 1.)
+    let (status, epoch, file_path) = job_row(&pool, event_id, "zip").await.unwrap();
+    assert_eq!((status.as_str(), epoch), ("pending", new_epoch));
+    assert_eq!(
+        file_path, None,
+        "the loser's file_path must never be recorded"
+    );
+
+    // And nothing is downloadable — not the stale archive, not anything.
+    assert_eq!(downloadable(&pool, event_id, "zip").await, None);
+}
+
+/// The documented, deliberate nuance in `claim_job`: after a bare `open_event` (which writes
+/// NOTHING to `export_job` — that is the point of the design), a worker at the old epoch still WINS
+/// its claim and can still write `done`. That is wasted work, not incorrectness: retirement is
+/// enforced at READ time. `export_current` must refuse to expose the row.
+///
+/// PREVENTS: someone "optimising" `claim_job` into a cross-table `EXISTS (SELECT ... FROM event)`
+/// guard — the exact unsound guard migration 014 removed (under READ COMMITTED, a blocked UPDATE
+/// re-evaluates same-row predicates but answers other-table subqueries from a stale snapshot).
+/// This test pins the read-time enforcement so the write-time guard is never re-added.
+#[sqlx::test]
+async fn reopen_retires_at_read_time_not_write_time(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    let epoch = release_gallery(&pool, "wedding").await.unwrap();
+
+    assert!(claim_job(&pool, event_id, "zip", epoch).await);
+
+    // Host reopens uploads. No export_job row is touched.
+    assert_eq!(open_event(&pool, "wedding").await, 1);
+
+    // The in-flight worker's row-local writes still match — it was never told to stop.
+    assert!(
+        finalize_job(&pool, event_id, "zip", epoch, "exports/Gallery.1.zip").await,
+        "documented: the claim/finalize is guarded on the JOB row's epoch, not the event's"
+    );
+    let (status, _, _) = job_row(&pool, event_id, "zip").await.unwrap();
+    assert_eq!(status, "done", "the row really does say done");
+
+    // …and yet it is invisible. `export_current` requires the event to be released AND the epochs to
+    // match; the reopen broke both. A worker at a dead epoch writes a row nobody can see.
+    assert!(!in_export_current(&pool, event_id, "zip").await);
+    assert_eq!(
+        downloadable(&pool, event_id, "zip").await,
+        None,
+        "a reopened event must serve NO keepsake, however finished the job row looks"
+    );
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+// 3. `export_current` exactness (table-driven)
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// The view is the ONE definition of "downloadable". Released + `done` + matching epoch ⇒ present;
+/// break ANY single leg ⇒ absent. Nothing else may make it appear or disappear.
+///
+/// PREVENTS, leg by leg:
+///  * `released` — serving a keepsake for an event whose uploads are still open, i.e. an archive
+///    missing every photo taken after the snapshot.
+///  * `done` — handing out a half-written ZIP (a corrupt keepsake, downloaded once, kept forever).
+///  * `epoch` — the retired-generation download: the 404-forever keepsake, or worse, the archive
+///    still containing content that was taken down.
+#[sqlx::test]
+async fn export_current_is_exactly_the_invariant(pool: PgPool) {
+    // (name, released?, status, job epoch offset from the event epoch, expected visible)
+    let cases: &[(&str, bool, &str, i64, bool)] = &[
+        ("released + done + current epoch", true, "done", 0, true),
+        (
+            "NOT released (done, epoch matches)",
+            false,
+            "done",
+            0,
+            false,
+        ),
+        ("NOT done: pending", true, "pending", 0, false),
+        ("NOT done: running", true, "running", 0, false),
+        ("NOT done: failed", true, "failed", 0, false),
+        ("stale epoch (done, released)", true, "done", -1, false),
+        ("future epoch (done, released)", true, "done", 1, false),
+        (
+            "migration-014 retired sentinel epoch -1",
+            true,
+            "done",
+            -2,
+            false,
+        ),
+    ];
+
+    for (i, (name, released, status, offset, expect_visible)) in cases.iter().enumerate() {
+        let slug = format!("case{i}");
+        let event_id = seed_event(&pool, &slug).await;
+
+        // Get the event to a known epoch (1) either by releasing it, or — for the unreleased case —
+        // by releasing and reopening, which leaves it unreleased at a non-zero epoch.
+        let event_epoch_now = if *released {
+            release_gallery(&pool, &slug).await.unwrap()
+        } else {
+            release_gallery(&pool, &slug).await.unwrap();
+            open_event(&pool, &slug).await;
+            event_epoch(&pool, event_id).await
+        };
+
+        // Plant a single ZIP job row in the exact state under test. `-2` encodes "the sentinel the
+        // migration stamps on retired rows", which must never equal a non-negative event epoch.
+        // (Clear the rows the release armed first — `UNIQUE (event_id, type)`.)
+        sqlx::query("DELETE FROM export_job WHERE event_id = $1")
+            .bind(event_id)
+            .execute(&pool)
+            .await
+            .expect("clear armed jobs");
+
+        let job_epoch = if *offset == -2 {
+            -1
+        } else {
+            event_epoch_now + offset
+        };
+        sqlx::query(
+            "INSERT INTO export_job (event_id, type, status, progress_pct, epoch, file_path)
+             VALUES ($1, 'zip', $2::export_status, 100, $3, 'exports/Gallery.zip')",
+        )
+        .bind(event_id)
+        .bind(*status)
+        .bind(job_epoch)
+        .execute(&pool)
+        .await
+        .expect("plant job row");
+
+        let visible = downloadable(&pool, event_id, "zip").await.is_some();
+        assert_eq!(
+            visible, *expect_visible,
+            "export_current exactness violated for case: {name} \
+             (released={released}, status={status}, job_epoch={job_epoch}, event_epoch={event_epoch_now})"
+        );
+    }
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+// 4. The ViewerOnly ZIP carry-forward
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Branch A — the ZIP is `done` at the outgoing epoch: the carry-forward re-stamps it to the new
+/// epoch (rows_affected = 1), so only the HTML viewer is rebuilt and the finished ZIP stays
+/// downloadable throughout.
+///
+/// PREVENTS: rebuilding a multi-GB archive because someone deleted a comment. The ZIP holds media,
+/// not comments — a needless rebuild would 404 the photo download for minutes to change nothing
+/// inside it.
+#[sqlx::test]
+async fn viewer_only_carries_a_done_zip_forward(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    let e1 = release_gallery(&pool, "wedding").await.unwrap();
+
+    // Both halves finish at epoch 1 — the keepsake is live.
+    for t in ["zip", "html"] {
+        assert!(claim_job(&pool, event_id, t, e1).await);
+        assert!(finalize_job(&pool, event_id, t, e1, &format!("exports/{t}.{e1}.zip")).await);
+    }
+    let zip_file = downloadable(&pool, event_id, "zip")
+        .await
+        .expect("zip is live");
+
+    // ── A comment is moderated: invalidate_and_arm(Affects::ViewerOnly). ──
+    let (_, _, e2) = bump_epoch(&pool, "wedding").await.unwrap();
+    let carried = carry_zip_forward(&pool, event_id, e2).await;
+    assert!(
+        carried,
+        "a `done` ZIP at epoch-1 MUST be carried forward (rows_affected == 1)"
+    );
+
+    // Only the viewer is re-armed…
+    let mut conn = pool.acquire().await.unwrap();
+    enqueue_types_at_epoch(&mut conn, event_id, e2, &["html"]).await;
+    drop(conn);
+
+    // …and the ZIP is STILL DOWNLOADABLE, at the new epoch, pointing at the same, unrenamed file.
+    let (status, epoch, _) = job_row(&pool, event_id, "zip").await.unwrap();
+    assert_eq!(
+        (status.as_str(), epoch),
+        ("done", e2),
+        "the ZIP row rode the epoch bump"
+    );
+    assert_eq!(
+        downloadable(&pool, event_id, "zip").await,
+        Some(zip_file),
+        "the carried archive must never stop being served — same file, new epoch"
+    );
+
+    // The viewer, meanwhile, is correctly retired and pending a rebuild.
+    assert_eq!(job_row(&pool, event_id, "html").await.unwrap().0, "pending");
+    assert_eq!(downloadable(&pool, event_id, "html").await, None);
+}
+
+/// Branch B — THE BUG WE JUST FIXED. If the ZIP is still `pending`/`running` when the comment is
+/// moderated (which is MINUTES for a real multi-GB gallery, and deleting a comment right after
+/// release is an utterly ordinary thing to do), the carry-forward matches NOTHING
+/// (rows_affected = 0) — so the caller must NOT assume it carried, and must re-arm the ZIP too.
+///
+/// PREVENTS: the stranded ZIP. Blindly re-arming only the viewer would leave the ZIP row at the
+/// retired epoch; the in-flight worker then finishes and writes `done` at an epoch `export_current`
+/// no longer matches, nothing ever re-arms it, and `GET /export/zip` 404s FOREVER — a keepsake the
+/// couple paid for that simply never appears, short of a reboot.
+#[sqlx::test]
+async fn viewer_only_carry_forward_matches_nothing_when_zip_unfinished(pool: PgPool) {
+    for zip_state in ["pending", "running"] {
+        let slug = format!("wedding-{zip_state}");
+        let event_id = seed_event(&pool, &slug).await;
+        let e1 = release_gallery(&pool, &slug).await.unwrap();
+
+        // The ZIP worker is still going; only the viewer has finished.
+        if zip_state == "running" {
+            assert!(claim_job(&pool, event_id, "zip", e1).await);
+        }
+        assert!(claim_job(&pool, event_id, "html", e1).await);
+        assert!(finalize_job(&pool, event_id, "html", e1, "exports/Memories.1.zip").await);
+
+        // ── The comment is moderated. ──
+        let (_, _, e2) = bump_epoch(&pool, &slug).await.unwrap();
+        let carried = carry_zip_forward(&pool, event_id, e2).await;
+
+        assert!(
+            !carried,
+            "a {zip_state} ZIP has nothing to carry forward — the UPDATE must affect 0 rows \
+             (its `status = 'done'` predicate is the whole precondition)"
+        );
+
+        // The carry-forward's OWN result decides. It didn't match ⇒ rebuild the ZIP as well.
+        let types: &[&str] = if carried { &["html"] } else { &["zip", "html"] };
+        let mut conn = pool.acquire().await.unwrap();
+        enqueue_types_at_epoch(&mut conn, event_id, e2, types).await;
+        drop(conn);
+
+        // THE ASSERTION THAT WOULD HAVE CAUGHT THE BUG: the ZIP must not be stranded at the dead
+        // epoch. It is re-armed at the live one, so a fresh worker will actually build it.
+        let (status, epoch, _) = job_row(&pool, event_id, "zip").await.unwrap();
+        assert_eq!(
+            (status.as_str(), epoch),
+            ("pending", e2),
+            "the unfinished ZIP MUST be re-armed at the new epoch, not left stranded at {e1}"
+        );
+
+        // Even if the old in-flight worker now "finishes", it is inert and cannot resurrect itself.
+        assert!(!finalize_job(&pool, event_id, "zip", e1, "exports/Gallery.1.zip").await);
+        assert_eq!(downloadable(&pool, event_id, "zip").await, None);
+    }
+}
+
+/// The re-arm upsert must never clobber the archive it just carried forward.
+///
+/// `enqueue_types_at_epoch`'s `WHERE export_job.status <> 'done' OR export_job.epoch <> EXCLUDED.epoch`
+/// is the "startup recovery must not clobber a good half" rule, expressed as the readiness predicate
+/// itself. PREVENTS: boot recovery resetting a perfectly good, downloadable ZIP back to `pending`
+/// and making the keepsake 404 while it needlessly rebuilds.
+#[sqlx::test]
+async fn enqueue_preserves_a_done_half_at_the_current_epoch(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    let e1 = release_gallery(&pool, "wedding").await.unwrap();
+
+    // The ZIP finished; the HTML worker was killed mid-flight (crash) and sits at `running`.
+    assert!(claim_job(&pool, event_id, "zip", e1).await);
+    assert!(finalize_job(&pool, event_id, "zip", e1, "exports/Gallery.1.zip").await);
+    assert!(claim_job(&pool, event_id, "html", e1).await);
+
+    // Boot recovery re-arms both types at the SAME epoch.
+    let mut conn = pool.acquire().await.unwrap();
+    enqueue_types_at_epoch(&mut conn, event_id, e1, &["zip", "html"]).await;
+    drop(conn);
+
+    // The good half survives untouched…
+    let (status, epoch, file_path) = job_row(&pool, event_id, "zip").await.unwrap();
+    assert_eq!(
+        (status.as_str(), epoch),
+        ("done", e1),
+        "a done half at the live epoch is preserved"
+    );
+    assert_eq!(
+        file_path.as_deref(),
+        Some("exports/Gallery.1.zip"),
+        "file_path not nulled"
+    );
+    assert!(downloadable(&pool, event_id, "zip").await.is_some());
+
+    // …and only the missing half is re-armed.
+    assert_eq!(job_row(&pool, event_id, "html").await.unwrap().0, "pending");
+}

backend/tests/upload_concurrency.rs

@@ -0,0 +1,297 @@
+//! DB-backed integration tests for the two concurrency guards in the upload commit path
+//! (`handlers/upload.rs`). Both are SQL — a `FOR SHARE` row lock and an atomic compare-and-increment
+//! — and both are load-bearing for things a user can actually lose: a wedding photo, or the disk.
+//!
+//! `#[sqlx::test]` gives each test a fresh database with the real migrations applied.
+
+mod common;
+
+use std::sync::Arc;
+use std::sync::atomic::{AtomicBool, Ordering};
+use std::time::Duration;
+
+use chrono::{DateTime, Utc};
+use common::*;
+use sqlx::PgPool;
+use uuid::Uuid;
+
+/// SRC: `handlers/upload.rs:313-322` — the guarded quota increment, verbatim.
+/// Returns `rows_affected()`; the handler aborts the whole upload tx when this is 0.
+async fn quota_inc(exec: impl sqlx::PgExecutor<'_>, user_id: Uuid, size: i64, limit: i64) -> u64 {
+    sqlx::query(
+        "UPDATE \"user\" SET total_upload_bytes = total_upload_bytes + $2
+         WHERE id = $1 AND total_upload_bytes + $2 <= $3",
+    )
+    .bind(user_id)
+    .bind(size)
+    .bind(limit)
+    .execute(exec)
+    .await
+    .expect("quota_inc")
+    .rows_affected()
+}
+
+async fn total_bytes(pool: &PgPool, user_id: Uuid) -> i64 {
+    sqlx::query_scalar("SELECT total_upload_bytes FROM \"user\" WHERE id = $1")
+        .bind(user_id)
+        .fetch_one(pool)
+        .await
+        .expect("total_bytes")
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+// 5. The atomic quota increment
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Two attempts sized off ONE stale snapshot, each of which "fits" on its own, cannot both commit.
+/// The predicate re-reads `total_upload_bytes` inside the UPDATE, so the second matches 0 rows.
+///
+/// PREVENTS: one guest filling the disk. The handler's pre-flight quota check runs BEFORE the body is
+/// streamed — minutes earlier, for a 500 MB video. If the commit trusted that snapshot, a guest could
+/// start N uploads that each individually fit under the limit and land all N, blowing straight through
+/// the quota and (in a 1 GB container) taking the event down for everyone.
+#[sqlx::test]
+async fn quota_two_attempts_from_one_stale_snapshot_cannot_both_commit(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    let user_id = seed_user(&pool, event_id, "Gierige Gudrun").await;
+
+    const LIMIT: i64 = 100;
+    const SIZE: i64 = 60;
+
+    // THE STALE SNAPSHOT: the pre-flight check both uploads were admitted on.
+    let snapshot = total_bytes(&pool, user_id).await;
+    assert_eq!(snapshot, 0);
+    // Each upload, judged against that snapshot alone, fits: 0 + 60 <= 100. Twice.
+    assert!(snapshot + SIZE <= LIMIT);
+
+    assert_eq!(
+        quota_inc(&pool, user_id, SIZE, LIMIT).await,
+        1,
+        "the first upload commits"
+    );
+    assert_eq!(
+        quota_inc(&pool, user_id, SIZE, LIMIT).await,
+        0,
+        "the second MUST affect 0 rows — it was admitted on a snapshot that is now a lie \
+         (60 + 60 = 120 > 100). rows_affected() == 0 is what makes the handler abort."
+    );
+
+    assert_eq!(
+        total_bytes(&pool, user_id).await,
+        SIZE,
+        "never 120 — the quota held"
+    );
+}
+
+/// The same, but genuinely CONCURRENT: two transactions that both read `total = 0`, then both try to
+/// commit 60 bytes against a 100-byte limit. The second UPDATE blocks on the first's row lock and —
+/// because every predicate is on the ROW BEING UPDATED — Postgres re-evaluates it against the
+/// post-commit row (EPQ) rather than the statement's original snapshot. It matches nothing.
+///
+/// PREVENTS: exactly the same disk-filling overrun, on the path it actually happens — two uploads
+/// in flight at once, which is the normal case at a party.
+#[sqlx::test]
+async fn quota_guard_is_atomic_under_concurrent_transactions(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    let user_id = seed_user(&pool, event_id, "Gierige Gudrun").await;
+
+    const LIMIT: i64 = 100;
+    const SIZE: i64 = 60;
+
+    let mut tx1 = pool.begin().await.unwrap();
+    let mut tx2 = pool.begin().await.unwrap();
+
+    // Both transactions read the same snapshot and both would pass a naive `total + size <= limit`
+    // check done in Rust.
+    for tx in [&mut tx1, &mut tx2] {
+        let seen: i64 = sqlx::query_scalar("SELECT total_upload_bytes FROM \"user\" WHERE id = $1")
+            .bind(user_id)
+            .fetch_one(&mut **tx)
+            .await
+            .unwrap();
+        assert_eq!(seen, 0, "both see an empty quota");
+    }
+
+    // tx1 takes the row lock and commits.
+    assert_eq!(quota_inc(&mut *tx1, user_id, SIZE, LIMIT).await, 1);
+    tx1.commit().await.unwrap();
+
+    // tx2's UPDATE was written against the stale snapshot but is evaluated against the row as it
+    // now stands.
+    assert_eq!(
+        quota_inc(&mut *tx2, user_id, SIZE, LIMIT).await,
+        0,
+        "the loser MUST see 0 rows affected — this is the entire quota guarantee"
+    );
+    tx2.rollback().await.unwrap();
+
+    assert_eq!(total_bytes(&pool, user_id).await, SIZE);
+
+    // And an upload that legitimately fits in what's left still succeeds — the guard rejects
+    // overruns, not everything.
+    assert_eq!(
+        quota_inc(&pool, user_id, 40, LIMIT).await,
+        1,
+        "0 + 60 + 40 == 100, exactly at the limit"
+    );
+    assert_eq!(total_bytes(&pool, user_id).await, LIMIT);
+    assert_eq!(
+        quota_inc(&pool, user_id, 1, LIMIT).await,
+        0,
+        "and one byte more is refused"
+    );
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+// 6. The `FOR SHARE` upload lock vs. the release
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// SRC: `handlers/upload.rs:297-303` — the in-transaction re-check under a row lock, verbatim.
+async fn lock_and_read_event(
+    tx: &mut sqlx::PgConnection,
+    event_id: Uuid,
+) -> (Option<DateTime<Utc>>, Option<DateTime<Utc>>) {
+    sqlx::query_as(
+        "SELECT uploads_locked_at, export_released_at FROM event WHERE id = $1 FOR SHARE",
+    )
+    .bind(event_id)
+    .fetch_one(tx)
+    .await
+    .expect("FOR SHARE re-check")
+}
+
+/// THE GUARD AGAINST SILENT, PERMANENT PHOTO LOSS.
+///
+/// An upload holding `FOR SHARE` on the event row must BLOCK the `UPDATE event SET
+/// export_released_at = NOW()` in `release_gallery` until it commits. Either the upload commits first
+/// — and the release (hence the export snapshot) is strictly ordered after it, so the keepsake
+/// CONTAINS the photo — or the release commits first and the upload observes the lock and rejects
+/// (reversibly: the client keeps the blob and resumes after a reopen).
+///
+/// PREVENTS: the lost wedding photo. Without this serialization: a guest starts a 500 MB video, the
+/// pre-flight lock check passes, the host releases the gallery, the export workers snapshot the
+/// uploads table, and THEN the upload commits. The photo appears in the live feed but is missing from
+/// the downloaded keepsake, forever — nothing ever regenerates it and nobody ever notices.
+#[sqlx::test]
+async fn for_share_upload_lock_serializes_against_release(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+    let user_id = seed_user(&pool, event_id, "Fotograf Fritz").await;
+
+    // ── The guest's upload transaction takes the share lock. ──
+    let mut upload_tx = pool.begin().await.unwrap();
+    let (locked, released) = lock_and_read_event(&mut upload_tx, event_id).await;
+    assert!(
+        locked.is_none() && released.is_none(),
+        "uploads are open, so we proceed to commit"
+    );
+
+    // ── Concurrently, the host hits "Galerie freigeben". ──
+    let release_done = Arc::new(AtomicBool::new(false));
+    let release_task = {
+        let pool = pool.clone();
+        let release_done = release_done.clone();
+        tokio::spawn(async move {
+            sqlx::query(
+                "UPDATE event
+                 SET export_released_at = NOW(),
+                     uploads_locked_at  = COALESCE(uploads_locked_at, NOW()),
+                     export_epoch       = export_epoch + 1
+                 WHERE id = $1 AND export_released_at IS NULL",
+            )
+            .bind(event_id)
+            .execute(&pool)
+            .await
+            .expect("release");
+            release_done.store(true, Ordering::SeqCst);
+        })
+    };
+
+    // The release MUST be stuck behind our `FOR SHARE` row lock. (`FOR SHARE` conflicts with the
+    // `FOR UPDATE` lock the UPDATE needs, so Postgres makes it wait — this is not a timing race,
+    // it is a lock-conflict guarantee; the sleep only gives it every chance to wrongly proceed.)
+    tokio::time::sleep(Duration::from_millis(750)).await;
+    assert!(
+        !release_done.load(Ordering::SeqCst),
+        "the release MUST block while an upload holds FOR SHARE — if it can slip past, the export \
+         snapshot is taken while a photo is still committing and that photo is lost forever"
+    );
+
+    // The photo commits. It is now unambiguously part of the upload set.
+    let upload_id: Uuid = sqlx::query_scalar(
+        "INSERT INTO upload (event_id, user_id, original_path, mime_type, original_size_bytes)
+         VALUES ($1, $2, 'originals/wedding/x.jpg', 'image/jpeg', 1234) RETURNING id",
+    )
+    .bind(event_id)
+    .bind(user_id)
+    .fetch_one(&mut *upload_tx)
+    .await
+    .unwrap();
+    upload_tx.commit().await.unwrap();
+
+    // Only now can the release proceed.
+    tokio::time::timeout(Duration::from_secs(5), release_task)
+        .await
+        .expect("the release must unblock once the upload commits")
+        .unwrap();
+
+    // THE PAYOFF: the export snapshot — the very query the ZIP worker runs — sees the photo. Order
+    // enforced by the lock: upload commit < release < snapshot.
+    let snapshot: Vec<Uuid> = sqlx::query_scalar(
+        "SELECT u.id FROM upload u
+           JOIN \"user\" usr ON usr.id = u.user_id
+          WHERE u.event_id = $1 AND u.deleted_at IS NULL
+            AND usr.uploads_hidden = FALSE AND usr.is_banned = FALSE",
+    )
+    .bind(event_id)
+    .fetch_all(&pool)
+    .await
+    .unwrap();
+    assert_eq!(
+        snapshot,
+        vec![upload_id],
+        "the released keepsake CONTAINS the in-flight photo"
+    );
+}
+
+/// The other side of the same lock: once the release has COMMITTED, the next upload's `FOR SHARE`
+/// re-read sees `export_released_at` set and the handler rejects it with `UploadsLocked`.
+///
+/// PREVENTS: the same lost photo, on the losing side of the race — a photo committing AFTER the
+/// export snapshot would be in the live feed but missing from the keepsake. Rejecting is the correct
+/// outcome, and it is reversible: `UploadsLocked` (not Forbidden) tells the client to keep the blob
+/// and resume when the host reopens.
+#[sqlx::test]
+async fn upload_after_release_commits_sees_the_lock_and_is_rejected(pool: PgPool) {
+    let event_id = seed_event(&pool, "wedding").await;
+
+    // Before the release, the re-check passes.
+    let mut tx = pool.begin().await.unwrap();
+    let (locked, released) = lock_and_read_event(&mut tx, event_id).await;
+    assert!(locked.is_none() && released.is_none());
+    tx.rollback().await.unwrap();
+
+    assert_eq!(release_gallery(&pool, "wedding").await, Some(1));
+
+    // After it, the identical re-check sees the release and the handler bails out.
+    let mut tx = pool.begin().await.unwrap();
+    let (locked, released) = lock_and_read_event(&mut tx, event_id).await;
+    assert!(
+        released.is_some(),
+        "the FOR SHARE re-read MUST observe the committed release"
+    );
+    assert!(
+        locked.is_some(),
+        "release locks uploads in the same statement (release ⇒ lock)"
+    );
+    tx.rollback().await.unwrap();
+
+    // And a reopen makes it uploadable again — the rejection was reversible, not terminal.
+    assert_eq!(open_event(&pool, "wedding").await, 1);
+    let mut tx = pool.begin().await.unwrap();
+    let (locked, released) = lock_and_read_event(&mut tx, event_id).await;
+    assert!(
+        locked.is_none() && released.is_none(),
+        "the guest can resume their upload"
+    );
+    tx.rollback().await.unwrap();
+}

docker-compose.dev.yml

@@ -0,0 +1,41 @@
+# Dev-only overlay. NOT loaded automatically (unlike docker-compose.override.yml).
+# Opt in explicitly for local development when you need host access to Postgres:
+#   docker compose -f docker-compose.yml -f docker-compose.dev.yml up
+# Never use this overlay in production — it publishes the database port on the host.
+services:
+  db:
+    ports:
+      - "5432:5432"
+  app:
+    # Relax the production secret guard for local dev — the dev sentinel JWT_SECRET
+    # is tolerated (warned) rather than rejected.
+    environment:
+      APP_ENV: development
+      # `.env` sets DATABASE_URL to @localhost for the run-backend-natively workflow
+      # (the db port is published above for that). When the app runs IN a container,
+      # localhost is the app itself — point it at the `db` service instead. Creds are
+      # interpolated from .env so nothing is hardcoded.
+      DATABASE_URL: postgres://${POSTGRES_USER}:${POSTGRES_PASSWORD}@db:5432/${POSTGRES_DB}
+      # `.env` sets MEDIA_PATH to a HOST path (/home/fabi/EventSnap/media) for the
+      # run-backend-natively workflow. In a container that path doesn't exist and the
+      # app user can't create it from `/`, so every upload 500s with EACCES. The media
+      # volume is mounted at /media (see docker-compose.yml) — point the app there.
+      MEDIA_PATH: /media
+      # Recent Docker Compose interpolates env_file values, so the `$` segments of the
+      # bcrypt ADMIN_PASSWORD_HASH in .env get eaten (the salt reads as an unset var) —
+      # every admin login then 401s. Re-supply it here with `$` doubled to `$$` so Compose
+      # passes the literal hash. NOTE: production (docker-compose.yml + .env) has the SAME
+      # bug — escape the hash as `$$` in .env, or set it via `environment:` there too.
+      ADMIN_PASSWORD_HASH: "$$2b$$12$$PAteqCNpsbm6d0HTJcywfOaUovjAU.iNVlsL7EDYaRC/z4P/xv7ye"
+      # Smoke-testing the comment kill-switch: boot-time flag, so it needs a restart
+      # (not an admin-UI toggle). Backend rejects new comments (403) and the frontend
+      # hides the whole comment UI. Flip back to true (or drop this line) to restore.
+      COMMENTS_ENABLED: "false"
+
+  caddy:
+    # The prod caddy service has no env_file, so the Caddyfile's `{$DOMAIN}` expands
+    # to empty and the site block collapses into a malformed global block. Supply it
+    # for local dev (from .env → localhost, which Caddy serves with a local self-signed
+    # cert). NOTE: the prod compose likely needs DOMAIN wired to caddy too.
+    environment:
+      DOMAIN: ${DOMAIN}

docker-compose.override.yml

@@ -1,4 +0,0 @@
-services:
-  db:
-    ports:
-      - "5432:5432"

docker-compose.yml

@@ -14,6 +14,10 @@ services:
       interval: 5s
       timeout: 5s
       retries: 10
+    deploy:
+      resources:
+        limits:
+          memory: 512M
 
   app:
     build:
@@ -21,13 +25,42 @@ services:
       dockerfile: Dockerfile
     restart: unless-stopped
     env_file: .env
+    environment:
+      # Activates the production secret guard in config.rs — refuses to boot with
+      # placeholder JWT_SECRET / ADMIN_PASSWORD_HASH.
+      APP_ENV: production
+      # The media volume is mounted at /media (below), so the app MUST write there.
+      # Pin it here rather than trusting .env: if MEDIA_PATH in .env points elsewhere
+      # (e.g. a host path used for running the backend natively) the container can't
+      # create it and every upload 500s with EACCES. `environment` overrides `env_file`,
+      # so this is authoritative for the container.
+      MEDIA_PATH: /media
     depends_on:
       db:
         condition: service_healthy
     volumes:
       - media_data:/media
+      # Export archives live OUTSIDE /media so the public media ServeDir can't
+      # serve them — downloads go only through the ticket-gated handler.
+      - exports_data:/exports
     expose:
       - "3000"
+    healthcheck:
+      # Use 127.0.0.1, NOT localhost: the app binds IPv4 (0.0.0.0) but `localhost`
+      # resolves to ::1 (IPv6) first inside the container, so a localhost probe gets
+      # "connection refused" and the container never turns healthy — which would leave
+      # Caddy (gated on `condition: service_healthy` below) blocked forever on boot.
+      test: ["CMD-SHELL", "wget -q -O- http://127.0.0.1:3000/health || exit 1"]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+      start_period: 20s
+    deploy:
+      resources:
+        limits:
+          # Bounds a runaway ffmpeg transcode (large uploads, 2 workers) so it can't
+          # OOM the single box and take down Postgres.
+          memory: 1G
 
   frontend:
     build:
@@ -35,14 +68,36 @@ services:
       dockerfile: Dockerfile
     restart: unless-stopped
     env_file: .env
+    environment:
+      # adapter-node behind Caddy TLS needs the public origin for CSRF checks on
+      # POST form actions — without it they fail only in production.
+      ORIGIN: "https://${DOMAIN}"
     depends_on:
       - app
     expose:
       - "3001"
+    healthcheck:
+      # 127.0.0.1, not localhost — see the app healthcheck note above (IPv4 bind vs
+      # ::1 resolution would leave this container permanently unhealthy).
+      test: ["CMD-SHELL", "wget -q -O- http://127.0.0.1:3001/ >/dev/null 2>&1 || exit 1"]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+      start_period: 15s
+    deploy:
+      resources:
+        limits:
+          memory: 256M
 
   caddy:
     image: caddy:2-alpine
     restart: unless-stopped
+    environment:
+      # The Caddyfile's site address is `{$DOMAIN}`, read from THIS container's env.
+      # Without it, `{$DOMAIN}` expands to empty, the site block collapses, and Caddy
+      # serves nothing / fails to obtain a TLS cert. `env_file` alone wouldn't help —
+      # Caddy needs it in `environment`, and this keeps the Caddyfile the single source.
+      DOMAIN: ${DOMAIN}
     ports:
       - "80:80"
       - "443:443"
@@ -50,10 +105,17 @@ services:
       - ./Caddyfile:/etc/caddy/Caddyfile:ro
       - caddy_data:/data
     depends_on:
-      - app
-      - frontend
+      app:
+        condition: service_healthy
+      frontend:
+        condition: service_healthy
+    deploy:
+      resources:
+        limits:
+          memory: 256M
 
 volumes:
   postgres_data:
   media_data:
+  exports_data:
   caddy_data:

docs/CONCEPT_DIASHOW.md

@@ -0,0 +1,196 @@
+# Live Diashow Concept
+
+> **Status: SHIPPED.** Implementation lives at
+> [frontend/src/lib/diashow/](../frontend/src/lib/diashow/) and
+> [frontend/src/routes/diashow/+page.svelte](../frontend/src/routes/diashow/+page.svelte).
+> Treat this doc as the design reference; code is the source of truth.
+
+## Goal
+
+A fullscreen, auto-advancing slideshow that any user can start from their device. Suitable
+for a venue projector or TV running off a single phone/laptop. Two behaviours combine in one
+view:
+
+1. **Live drain** — when a new post is uploaded mid-event, it appears on the next slide
+   transition.
+2. **Shuffle fallback** — between new uploads (and they will be rare in quiet stretches), the
+   diashow rotates through everything posted so far, in shuffled order.
+
+The user does **not** need to be Host or Admin. Any guest can start the diashow on their own
+device. There is no global "the room's diashow" — each device runs its own (though they will
+look very similar if started at the same time).
+
+---
+
+## Behavioural model
+
+Two FIFO queues live in the client:
+
+| Queue          | Source                                    | Drain priority |
+|----------------|-------------------------------------------|----------------|
+| `liveQueue`    | SSE events for new processed uploads      | First          |
+| `shuffleQueue` | Snapshot of all known uploads, shuffled   | When live empty |
+
+### Slide-advance algorithm
+
+```ts
+function nextSlide(): Slide | null {
+  // 1. Drain live posts first (FIFO).
+  if (liveQueue.length) return liveQueue.shift()!;
+
+  // 2. Refill shuffle queue from `allKnown` if drained.
+  if (!shuffleQueue.length) {
+    shuffleQueue = shuffle(
+      [...allKnown.values()].filter(s => !recentlyShown.has(s.id))
+    );
+  }
+  return shuffleQueue.shift() ?? null;
+}
+```
+
+A small ring buffer `recentlyShown` (last ~5 IDs) prevents the same picture coming back
+within seconds when the shuffle queue is rebuilt.
+
+### Live insertion
+
+```ts
+sseClient.on('upload-processed', (msg) => {
+  if (allKnown.has(msg.upload_id)) return;
+  const slide = await fetchUpload(msg.upload_id); // or use payload directly
+  allKnown.set(slide.id, slide);
+  liveQueue.push(slide);
+});
+```
+
+Listen on **`upload-processed`**, not `new-upload` — the preview/thumbnail must exist before
+we try to display the slide.
+
+### Deletion / hiding
+
+```ts
+sseClient.on('upload-deleted', ({ upload_id }) => {
+  allKnown.delete(upload_id);
+  liveQueue   = liveQueue.filter(s => s.id !== upload_id);
+  shuffleQueue = shuffleQueue.filter(s => s.id !== upload_id);
+  if (currentSlide?.id === upload_id) advanceImmediately();
+});
+```
+
+Hidden uploads (banned user with `uploads_hidden=true`) need either a new SSE event or to
+piggyback `upload-deleted` for diashow purposes. Simplest path: emit `upload-deleted` for
+hidden posts to all subscribers (the live feed already filters them via `v_feed`, so this is
+a backwards-compatible signal).
+
+---
+
+## Initial pool
+
+On start:
+
+1. Call `GET /api/v1/feed?limit=200` (or paginate-and-drain in the background while the
+   diashow runs).
+2. Push every returned upload into `allKnown`.
+3. Build the first `shuffleQueue` from it.
+4. Open the SSE stream and route `upload-processed` / `upload-deleted` into the queues.
+
+If the event is empty, show a friendly placeholder:
+*"Noch keine Beiträge — neue erscheinen automatisch."*
+
+---
+
+## Frontend surface
+
+### Entry point
+
+A small **Diashow / "Präsentation starten"** action visible:
+
+- In the feed header (icon next to the list/grid toggle) on tablet/desktop layouts.
+- In the Account page on mobile (less prominent — diashow is primarily a venue-screen
+  feature).
+
+Tapping it navigates to the `/diashow` route (full-screen, bottom nav hidden).
+
+### Route: `/diashow`
+
+- Fullscreen request via `element.requestFullscreen()` after first user gesture.
+- **Screen Wake Lock**: `navigator.wakeLock.request('screen')` to keep the screen on during
+  long shows. Renew on `visibilitychange` if needed.
+- Default dwell: **6 seconds** per slide. Configurable via overlay control: 3 / 6 / 10 s.
+- Tap or `Escape` reveals an overlay with: pause/resume, dwell selector, **transition
+  picker**, exit.
+- Transitions: crossfade (≈400 ms) by default; Ken Burns, zoom, slide, etc. available as
+  pluggable components — see "Pluggable transitions" below.
+- Videos: autoplay muted, fit-to-screen, advance on `ended` or after `max(dwell, 12 s)`,
+  whichever first.
+- Preload the next slide's media into a hidden `<img>`/`<video>` to avoid flashing.
+- **Media source respects the user's data mode**
+  (see [FEATURES.md §2.5](FEATURES.md)). In Saver mode the diashow loads `preview_url`;
+  in Original mode it loads the original. The data-usage warning is shown once when the
+  mode is toggled in My Account — the diashow itself stays silent.
+
+### Pluggable transitions
+
+Each transition is a **drop-in Svelte component** under
+`frontend/src/lib/diashow/transitions/` (path finalised at implementation time). The
+interface is intentionally tiny:
+
+```ts
+// pseudocode — the real shape lands with the feature
+export interface SlideTransition {
+  id: string;                  // 'crossfade', 'kenburns', ...
+  label: string;               // shown in the dwell/transition picker
+  durationMs: number;          // default; can be overridden per-event
+  // The actual motion is implemented by mounting the component with `from` / `to` slides.
+}
+```
+
+A small registry maps `id → component`; the settings popover renders that registry as a
+dropdown. **Adding a new animation is one new file plus one line in the registry — no
+other changes required.** This is the maintainability target called out in
+[FEATURES.md §2.9](FEATURES.md) and [IDEAS.md](IDEAS.md) ("Animation pack").
+
+The same pattern is a candidate for whole-event "themes" later — a bundle of (transition
++ dwell + optional background-music defaults).
+
+### Edge cases
+
+| Case                                       | Behaviour                                              |
+|--------------------------------------------|--------------------------------------------------------|
+| Empty event                                | Placeholder card; live SSE will trigger the first show |
+| All known uploads are still compressing    | Same placeholder — wait for `upload-processed`         |
+| Network drop / SSE reconnect               | EventSource auto-reconnects; queues survive            |
+| Current slide gets deleted                 | Advance immediately                                    |
+| Event is closed (no new uploads possible)  | Diashow keeps running on shuffle queue indefinitely    |
+| Banned user's content (`uploads_hidden`)   | Removed via `upload-deleted` signal (see Deletion)     |
+
+---
+
+## Backend changes
+
+**Essentially none.** The diashow reuses:
+
+- `GET /api/v1/feed` (initial pool)
+- `GET /api/v1/stream` SSE (`upload-processed`, `upload-deleted`)
+
+Optional small additions:
+
+1. Emit `upload-deleted` (or a new `upload-hidden`) when a host bans a user with
+   `hide_uploads=true`, so that diashow clients can scrub the relevant slides without
+   reloading.
+2. Consider raising the cap on `GET /api/v1/feed?limit=` for diashow clients (or paginate
+   the initial pool in the background — preferred, no API change needed).
+
+---
+
+## Future extensions (not in scope for v1)
+
+The big ones live in [IDEAS.md](IDEAS.md) under "Diashow extensions" — most notably the
+**global / synchronised diashow** where multiple screens share one server-side cursor.
+Short list of others kept here for context:
+
+- **Curated highlights mode** — only show uploads tagged with a specific hashtag, or
+  Host-pinned "Story" uploads (depends on the story-highlights feature).
+- **Audio bed** — host can pick a background track; mute videos so they don't fight the
+  music.
+- **Slide caption / uploader chyron** — small lower-third with the uploader's name and
+  caption. Out by default to keep the visual clean.

docs/CONCEPT_HTML_VIEWER.md

@@ -0,0 +1,288 @@
+# HTML Viewer Export Concept
+
+> **Status: IMPLEMENTED.** Viewer source: [frontend/export-viewer/](../frontend/export-viewer/).
+> Pre-built output committed to [backend/static/export-viewer/](../backend/static/export-viewer/).
+> Backend export pipeline: [backend/src/services/export.rs](../backend/src/services/export.rs).
+>
+> Outstanding follow-ups:
+> - The export-viewer's `tailwind.config.js` does not yet extend the main app's config. Visual
+>   drift risk — see "Shared Tailwind Config" section below.
+> - Service-worker (offline PWA caching) is still "Future" — fine for v1 since the ZIP is
+>   already fully offline by virtue of relative paths.
+
+## Overview
+
+The HTML Viewer export produces a **self-contained offline ZIP** that is a read-only clone
+of the live EventSnap feed. Opening `index.html` in any modern browser shows the full event
+gallery — list view, grid view, search, filter, lightbox — with no internet connection or
+server required.
+
+It **replaces the current HTML export job type**. The old HTML export produced a raw
+minijinja-rendered template; the new viewer supersedes it entirely. The existing `html`
+job variant in the backend is repurposed to run this flow instead of being kept alongside
+it.
+
+It is distinct from the ZIP archive export (which is raw media files). The viewer is a
+polished, navigable web app bundled with the event's content.
+
+---
+
+## User-Facing Behavior
+
+- Download `event_name_viewer.zip`, unzip, open `index.html`
+- Full list view (chronological, newest first) and grid view with search/filter
+- Likes, comments, and reaction counts shown (static snapshot from export time)
+- Read-only: no uploads, no auth, no dashboards
+- Works offline, no CDN or external resources
+
+---
+
+## Architecture
+
+### Separate Static SvelteKit App
+
+A new mini SvelteKit project lives at `frontend/export-viewer/` within the same monorepo.
+It uses `adapter-static` and is kept completely independent of the main app.
+
+**Why separate rather than a shared route:**
+- The viewer must be distributable as a standalone static bundle; the main app uses
+  `adapter-node` and cannot be mixed
+- Keeping it separate avoids auth, store, and routing dependencies leaking in
+- Simpler to reason about: the viewer has exactly two concerns (list view, grid view)
+
+**Why same repo:**
+- Shares Tailwind config and design tokens → visual parity with the main app
+- Single `pnpm` workspace, no separate CI needed
+- Backend can reference the pre-built output by relative path
+
+### Pre-Built Output Committed to Repo
+
+The viewer is built once and its output committed to `backend/static/export-viewer/`.
+The backend export job **does not run a Node build** at runtime — it just copies the
+pre-built assets and injects event data alongside them.
+
+When the viewer source changes, a developer rebuilds it locally (`pnpm build` in
+`frontend/export-viewer/`) and commits the updated `backend/static/export-viewer/` output.
+
+---
+
+## ZIP Structure
+
+```
+event_name_viewer.zip/
+├── index.html                  ← entry point; open this in any browser
+├── _app/
+│   └── immutable/
+│       ├── viewer.[hash].js    ← all Svelte/app logic, single bundle
+│       └── viewer.[hash].css   ← all styles including Tailwind output
+├── data.json                   ← injected by backend at export time
+└── media/
+    ├── abc123_thumb.jpg        ← ~400 px wide, used in grid cells
+    ├── abc123_full.jpg         ← original or capped (see Media Strategy)
+    ├── def456_thumb.jpg
+    └── def456.mp4              ← videos included as-is
+```
+
+No external font CDN, no Google Fonts, no remote scripts. All assets are local.
+
+---
+
+## data.json Schema
+
+Generated by the backend export job. The viewer fetches this file on startup via
+`fetch('./data.json')` (relative path, works from filesystem).
+
+```json
+{
+  "event": {
+    "name": "Sommerfest 2025",
+    "exported_at": "2025-07-15T20:00:00Z"
+  },
+  "posts": [
+    {
+      "id": "abc123",
+      "uploader": "MaxMustermann",
+      "caption": "Tolle Stimmung! #party #spaß",
+      "tags": ["party", "spaß"],
+      "timestamp": "2025-07-15T18:30:00Z",
+      "likes": 12,
+      "comments": [
+        {
+          "author": "AnnaSchulz",
+          "text": "So schön!",
+          "timestamp": "2025-07-15T18:35:00Z"
+        }
+      ],
+      "media": [
+        {
+          "type": "image",
+          "thumb": "media/abc123_thumb.jpg",
+          "full": "media/abc123_full.jpg",
+          "width": 1920,
+          "height": 1080
+        }
+      ]
+    }
+  ]
+}
+```
+
+All post data (likes, comments, tags) reflects the state at export time. No live updates.
+
+---
+
+## Media Strategy
+
+### Images
+
+| Variant | Purpose | Max dimension | Format |
+|---------|---------|---------------|--------|
+| `_thumb` | Grid cells, list post thumbnail | 400 px wide | JPEG q75 |
+| `_full` | Lightbox / full-screen view | Original, or 2000 px cap if >5 MB | JPEG q85 |
+
+The backend applies compression only when the original exceeds a threshold (e.g. >5 MB for
+images). Below that threshold the original is used as `_full` unchanged.
+
+The full-resolution original is always available via the separate ZIP archive export.
+
+### Videos
+
+Included as-is (no server-side transcoding). The viewer uses a standard `<video>` element.
+The `_thumb` variant for videos is a JPEG frame extracted at the 1-second mark.
+
+---
+
+## SvelteKit SSG Configuration
+
+```
+// frontend/export-viewer/src/routes/+layout.ts
+export const prerender = true;
+export const ssr = false;
+```
+
+`ssr = false` produces a pure client-side SPA: SvelteKit emits a minimal `index.html` shell
+and the JavaScript bundle hydrates it entirely in the browser. This is correct for a ZIP
+distribution where no server exists to handle SSR.
+
+```
+// frontend/export-viewer/svelte.config.js
+import adapter from '@sveltejs/adapter-static';
+
+export default {
+  kit: {
+    adapter: adapter({
+      fallback: 'index.html',
+      pages: '../../backend/static/export-viewer',
+      assets: '../../backend/static/export-viewer',
+    })
+  }
+};
+```
+
+The build output is written directly into `backend/static/export-viewer/` so the backend
+can reference it without a copy step.
+
+### Shared Tailwind Config
+
+```
+// frontend/export-viewer/tailwind.config.js
+import baseConfig from '../tailwind.config.js';
+export default { ...baseConfig, content: ['./src/**/*.{svelte,ts}'] };
+```
+
+Imports the main app's Tailwind config to guarantee visual parity. Only the `content` glob
+is overridden.
+
+---
+
+## Viewer Feature Set
+
+| Feature | Included | Notes |
+|---------|----------|-------|
+| List view (chronological, newest first) | ✓ | Full-width cards, same layout as live app |
+| Grid view (3-column) | ✓ | Square cells, video duration badge |
+| List/grid toggle | ✓ | Same toggle icons as live app |
+| Search bar (grid view only) | ✓ | Appears only in grid view |
+| Tag filter chips | ✓ | Built from tags in data.json |
+| Uploader filter | ✓ | Dropdown from uploaders in data.json |
+| Autocomplete suggestions | ✓ | From data.json — no network requests |
+| Lightbox (tap to expand) | ✓ | Swipe left/right navigates filtered set |
+| Like counts (static) | ✓ | Snapshot from export time |
+| Comment list (static) | ✓ | Expandable under each post |
+| Like/comment actions | ✗ | Read-only export |
+| Upload button / FAB | ✗ | |
+| Account / Host / Admin | ✗ | |
+| Authentication | ✗ | No JWT, no PIN |
+| Service Worker (offline cache) | Future | Could be added later for PWA behavior |
+
+---
+
+## Backend Export Job Flow
+
+The `html` job variant is repurposed. The old minijinja template rendering path is removed
+and replaced entirely by the steps below.
+
+```
+1. Query all posts, media, reactions, and comments for the event from the DB
+2. Copy pre-built viewer assets:
+     backend/static/export-viewer/ → tmp/{job_id}/
+3. Generate data.json:
+     - Build the JSON structure from queried data
+     - Write to tmp/{job_id}/data.json
+4. Process and copy media:
+     For each media file:
+       a. Copy original; if image >5 MB, also produce compressed _full variant
+       b. Generate _thumb (resize to 400 px wide via image library)
+       c. For video, extract JPEG frame for _thumb
+       d. Write to tmp/{job_id}/media/
+5. Create ZIP:
+     zip -r event_name_viewer.zip tmp/{job_id}/
+6. Store ZIP path, mark job as complete
+7. Clean up tmp/{job_id}/
+```
+
+The backend needs an image processing dependency (e.g. `image` crate in Rust) for thumbnail
+generation and compression. Video frame extraction requires `ffmpeg` available in the
+deployment environment (already used for video handling if applicable, otherwise add to
+docker-compose).
+
+---
+
+## Monorepo Structure After Implementation
+
+```
+EventSnap/
+├── backend/
+│   ├── static/
+│   │   └── export-viewer/        ← pre-built viewer output (committed)
+│   │       ├── index.html
+│   │       └── _app/...
+│   └── src/
+│       └── handlers/
+│           └── export.rs          ← export job assembles ZIP
+├── frontend/
+│   ├── export-viewer/             ← new mini SvelteKit project
+│   │   ├── package.json
+│   │   ├── svelte.config.js       ← adapter-static, output → backend/static/export-viewer
+│   │   ├── tailwind.config.js     ← extends ../tailwind.config.js
+│   │   └── src/
+│   │       └── routes/
+│   │           ├── +layout.ts     ← prerender=true, ssr=false
+│   │           └── +page.svelte   ← list/grid feed, lightbox, search
+│   └── src/                       ← existing main app (unchanged)
+└── docs/
+    ├── CONCEPT_MOBILE_UI.md
+    └── CONCEPT_HTML_VIEWER.md
+```
+
+---
+
+## Open Questions for Implementation
+
+1. **Image processing library**: The `image` crate handles JPEG resize/compress; is it
+   already a backend dependency, or does it need to be added?
+2. **Video thumbnail extraction**: Is `ffmpeg` available in the Docker environment?
+   If not, a fallback (no video thumb, use a placeholder) is needed.
+3. **Viewer rebuild workflow**: Add a `make build-viewer` or `pnpm --filter export-viewer build`
+   step to the developer workflow docs and CI so the committed output stays in sync.
+4. **ZIP file naming**: `{event_slug}_viewer_{date}.zip` or a fixed name?

docs/CONCEPT_MOBILE_UI.md

@@ -0,0 +1,471 @@
+# Mobile-First UI/UX Redesign Concept
+
+> **Status: IMPLEMENTED (v0.15).** This document captures the design intent. The redesign
+> has shipped — see [BottomNav.svelte](../frontend/src/lib/components/BottomNav.svelte),
+> [UploadSheet.svelte](../frontend/src/lib/components/UploadSheet.svelte),
+> [CameraCapture.svelte](../frontend/src/lib/components/CameraCapture.svelte),
+> [feed/+page.svelte](../frontend/src/routes/feed/+page.svelte),
+> [account/+page.svelte](../frontend/src/routes/account/+page.svelte),
+> [host/+page.svelte](../frontend/src/routes/host/+page.svelte),
+> [admin/+page.svelte](../frontend/src/routes/admin/+page.svelte). Use this doc as the design
+> reference; treat code as the source of truth for current behaviour.
+
+## Overview
+
+EventSnap is intended for mobile use at live events. This document describes the full
+mobile-first design covering navigation, the feed/gallery, account page, host dashboard,
+and admin dashboard.
+
+---
+
+## 1. Navigation: Bottom Tab Bar
+
+Replace all per-page top-right icon links with a single **persistent bottom tab bar** present
+on every page. The bar sits at the very bottom with proper `padding-bottom` for iPhone home
+indicator (safe-area-inset-bottom).
+
+### Tab Composition by Role
+
+| Role  | Tabs |
+|-------|------|
+| Guest | 🏠 Feed · [📷+] · 👤 Account |
+| Host  | 🏠 Feed · [📷+] · 👤 Account |
+| Admin | 🏠 Feed · [📷+] · 👤 Account |
+
+All roles see the same three tabs. Role-specific dashboard links (Host, Admin) live inside
+the Account page — not as separate tabs. This keeps the bar simple and avoids conditional
+tab rendering.
+
+### Visual Style
+
+- Frosted glass background: `bg-white/85 backdrop-blur-md`
+- Thin top border: `border-t border-gray-200`
+- Subtle shadow upward
+- Active tab: colored icon + small label below
+- Inactive tab: gray icon, small gray label
+
+### Upload FAB (Floating Action Button)
+
+The center tab is an elevated circular button, not a flat tab icon:
+
+- Circle ~56 px diameter, `bg-blue-600`
+- Icon: camera outline with a small `+` badge overlaid at bottom-right
+- Raised above the bar with a drop shadow
+- Press: slight scale-down (`scale-95`) + haptic feedback where available
+- Communicates "capture new or upload existing"
+
+---
+
+## 2. Feed / Gallery Page
+
+### Header
+
+```
+┌─────────────────────────────────────────┐
+│  Sommerfest 2025              ≡  ⊞      │
+└─────────────────────────────────────────┘
+```
+
+- Event name left-aligned
+- List/grid view toggle icons right-aligned (≡ list, ⊞ grid)
+- Header collapses on downward scroll (only toggle remains visible), expands on upward scroll
+
+---
+
+### View A — Chronological List (default)
+
+Full-width post cards, newest at top, infinite scroll.
+
+```
+┌─────────────────────────────────────────┐
+│  👤 MaxMustermann          · vor 2 Min  │
+│ ┌───────────────────────────────────┐   │
+│ │                                   │   │
+│ │          [photo / video]          │   │
+│ │                                   │   │
+│ └───────────────────────────────────┘   │
+│  Tolle Stimmung! #party #spaß           │
+│  ❤️ 12   💬 3                           │
+└─────────────────────────────────────────┘
+```
+
+- Media: full-width, native aspect ratio, capped at 80 vh
+- Avatar: colored initial circle, no photo
+- Timestamp: relative ("vor 2 Min", "vor 1 Std")
+- Tap media → fullscreen lightbox, swipe left/right navigates feed
+- No search bar in list view
+
+---
+
+### View B — Grid View
+
+Transition animation when toggling: list collapses, grid fades/scales in (~200 ms).
+
+#### Search Bar (grid view only)
+
+```
+┌─────────────────────────────────────────┐
+│  🔍  Nutzer oder #Tag suchen…      ×   │
+└─────────────────────────────────────────┘
+```
+
+- Appears below the header only in grid view
+- Slides in as part of the view transition
+- `×` clears current input
+- Auto-focuses when grid view is activated
+
+#### Autocomplete Dropdown
+
+Appears immediately on focus and updates on every keystroke. Data source: the already-loaded
+posts in memory — **no extra API calls**.
+
+Two suggestion lists are derived at load time:
+- `allTags`: unique hashtags from all post captions, sorted by frequency descending
+- `allUploaders`: unique display names, sorted alphabetically
+
+| User input | Suggestions shown |
+|------------|-------------------|
+| (focus, empty) | Top 3 tags by frequency + top 3 uploaders |
+| `#` | All tags, frequency-sorted |
+| `#par` | Tags with prefix "par": `#party`, `#parade` |
+| `Max` | Uploaders matching "max" (case-insensitive) |
+| `a` | Uploaders containing "a" + tags containing "a" |
+
+Dropdown layout:
+
+```
+┌─────────────────────────────────────────┐
+│  👤 Nutzer                              │
+│     MaxMustermann                       │
+│     AnnaSchulz                          │
+│  # Tags                                 │
+│     #party    #tanz    #spaß            │
+└─────────────────────────────────────────┘
+```
+
+Max ~5 total suggestions. Tapping a suggestion adds it as an active filter chip and clears
+the search bar for another entry.
+
+#### Active Filter Chips
+
+```
+┌─────────────────────────────────────────┐
+│  👤 MaxMustermann ×   # party ×         │
+│  Alle Filter löschen                    │  ← shown when 2+ chips active
+└─────────────────────────────────────────┘
+```
+
+Filter combination logic:
+
+| Combination | Logic |
+|-------------|-------|
+| Two tags: `#party` + `#tanz` | OR — posts with either tag |
+| Two uploaders: Max + Anna | OR — posts from either |
+| Uploader + tag: Max + `#party` | AND — posts by Max that also have `#party` |
+
+#### Grid Layout
+
+```
+┌───────┬───────┬───────┐
+│       │       │       │
+│       │       │       │  3-column, equal square cells
+├───────┼───────┼───────┤  small gap (2 px)
+│       │  ▶    │       │  ← video: small ▶ badge + duration
+│       │  0:42 │       │
+└───────┴───────┴───────┘
+```
+
+- Tap cell → fullscreen lightbox, swipe navigates filtered set only
+- Virtualized grid for performance on large events
+
+---
+
+## 3. Upload Flow
+
+### Step 1 — Source Selection (Bottom Sheet)
+
+Tapping the FAB slides up a bottom sheet (~300 ms spring animation).
+Frosted glass, rounded top corners, drag handle at top. Tap outside or swipe down to dismiss.
+
+```
+┌──────────────────────────────────┐
+│          ▬  (drag handle)        │
+│                                  │
+│  📸  Kamera                      │
+│      Jetzt aufnehmen             │
+│                                  │
+│  🖼  Galerie                     │
+│      Foto oder Video wählen      │
+│                                  │
+│  [  Abbrechen  ]                 │
+└──────────────────────────────────┘
+```
+
+### Step 2a — Camera
+
+Triggers `<input type="file" accept="image/*,video/*" capture="environment">`.
+Native camera opens. After capture → Step 3.
+
+### Step 2b — Gallery
+
+Triggers `<input type="file" accept="image/*,video/*" multiple>`.
+Native gallery picker with multi-select (up to ~10 items). After selection → Step 3.
+
+### Step 3 — Preview & Metadata Screen
+
+Full-screen, pushes in from right. Bottom nav hidden (immersive).
+
+```
+┌──────────────────────────────────┐
+│  ×  Abbrechen      Hochladen →  │
+├──────────────────────────────────┤
+│                                  │
+│  ┌────┐ ┌────┐ ┌────┐  →        │  ← horizontal scroll, tap to preview
+│  │img │ │img │ │ ×  │           │    × on each thumbnail to remove
+│  └────┘ └────┘ └────┘           │
+│                                  │
+│  Beschreibung (optional)         │
+│  ┌────────────────────────────┐  │
+│  │                            │  │  ← auto-focused
+│  └────────────────────────────┘  │
+│                                  │
+│  # Schnell-Tags                  │
+│  [#Feier] [#Spaß] [#Party] …   │  ← tap to append to caption
+│                                  │
+├──────────────────────────────────┤
+│  ┌────────────────────────────┐  │
+│  │      📤 Hochladen          │  │  ← sticky, disabled until ≥1 file
+│  └────────────────────────────┘  │
+└──────────────────────────────────┘
+```
+
+### Step 4 — Background Upload + Feedback
+
+- Tapping "Hochladen" immediately returns to the feed (optimistic UX)
+- Slim progress bar above the bottom tab bar while queue is active
+- FAB gets a small spinning ring badge while uploads are in progress
+- On completion: brief toast near the bottom ("✓ Hochgeladen")
+- Rate-limit countdown banner anchored above the bottom bar (existing behavior)
+
+---
+
+## 4. Account Page
+
+Single entry point for profile info and role-based dashboard navigation.
+
+```
+┌─────────────────────────────────────────┐
+│              Mein Account               │
+├─────────────────────────────────────────┤
+│                                         │
+│        ┌───────┐                        │
+│        │  M    │   MaxMustermann        │
+│        └───────┘   🏷 Gast              │
+│                    Sommerfest 2025      │
+│                    7 Uploads            │
+│                                         │
+├── Dashboards ───────────────────────────┤  (entire section absent for guests)
+│                                         │
+│  ⭐  Host-Dashboard                →    │  (host + admin only)
+│  🛡  Admin-Dashboard               →    │  (admin only)
+│                                         │
+├── Konto ────────────────────────────────┤
+│                                         │
+│  ✏️  Anzeigename ändern             →    │
+│  🔑  PIN ändern                     →    │
+│  🚪  Event verlassen                →    │  (red text, confirm sheet)
+│                                         │
+└─────────────────────────────────────────┘
+│  🏠 Feed  ·  [📷+]  ·  👤 Account       │
+└─────────────────────────────────────────┘
+```
+
+- "Dashboards" section is entirely absent in the DOM for plain guests — not just hidden
+- "Event verlassen" triggers a bottom-sheet confirmation before action
+- Avatar: colored circle with initial letter
+
+---
+
+## 5. Host Dashboard
+
+Accessed via Account → ⭐ Host-Dashboard. Full-screen page, bottom nav visible.
+
+```
+┌─────────────────────────────────────────┐
+│  ←   🎉 Host-Dashboard                  │
+├─────────────────────────────────────────┤
+│                                         │
+│  ── Statistiken ──────────────────────  │
+│  ┌──────────┐  ┌──────────┐            │
+│  │    24    │  │   156    │            │
+│  │  Nutzer  │  │ Uploads  │            │
+│  └──────────┘  └──────────┘            │
+│                                         │
+│  ── Event-Einstellungen ──────────────  │  ← collapsible section
+│                                         │
+│  Neue Uploads sperren                   │
+│  ○────────────●  Gesperrt              │  ← toggle
+│  Keine neuen Uploads möglich            │
+│                                         │
+│  ── Nutzerverwaltung ─────────────────  │  ← collapsible section
+│                                         │
+│  🔍  Nutzer suchen…                     │
+│  ┌───────────────────────────────────┐  │
+│  │  👤  MaxMustermann   Gast  [🚫]   │  │
+│  │  👤  AnnaSchulz      Gast  [🚫]   │  │
+│  │  👤  GesperrterNutzer      [↩]    │  │  ← banned: undo icon
+│  └───────────────────────────────────┘  │
+│                                         │
+└─────────────────────────────────────────┘
+│  🏠 Feed  ·  [📷+]  ·  👤 Account       │
+└─────────────────────────────────────────┘
+```
+
+- Sections have a chevron toggle to collapse/expand (helps on small phones)
+- Ban/unban: icon tap + bottom sheet confirmation ("Nutzer wirklich sperren?")
+- User list virtualized for large events
+
+---
+
+## 6. Admin Dashboard
+
+Most complex page. Uses an **inner tab bar** directly below the header to divide the four
+functional areas. The inner tabs are independent of the bottom nav.
+
+```
+┌─────────────────────────────────────────┐
+│  ←   🛡 Admin-Dashboard                 │
+├─────────────────────────────────────────┤
+│  [Stats] [Config] [Export] [Nutzer]     │  ← inner tab bar (scrollable if needed)
+├─────────────────────────────────────────┤
+│                                         │
+│  [Tab content]                          │
+│                                         │
+└─────────────────────────────────────────┘
+│  🏠 Feed  ·  [📷+]  ·  👤 Account       │
+└─────────────────────────────────────────┘
+```
+
+### Stats Tab
+
+```
+  ┌──────────┐  ┌──────────┐
+  │  156     │  │  24      │
+  │ Uploads  │  │ Nutzer   │
+  └──────────┘  └──────────┘
+  ┌──────────┐  ┌──────────┐
+  │  2.1 GB  │  │  3       │
+  │ Speicher │  │ Gesperrt │
+  └──────────┘  └──────────┘
+```
+
+2×2 metric card grid. Values large and prominent. Optionally expandable to show time-series
+charts on tap.
+
+### Config Tab
+
+```
+  Upload-Limit / Nutzer
+  ┌────────────────────────────────┐
+  │  10                            │
+  └────────────────────────────────┘
+
+  Zeitfenster (Sek.)
+  ┌────────────────────────────────┐
+  │  60                            │
+  └────────────────────────────────┘
+
+  Max. Dateigröße (MB)
+  ┌────────────────────────────────┐
+  │  50                            │
+  └────────────────────────────────┘
+
+  ┌────────────────────────────────┐
+  │        💾 Speichern            │  ← sticky at bottom of tab scroll area
+  └────────────────────────────────┘
+```
+
+Each setting: full-width label + input. Save button always reachable without scrolling.
+
+### Export Tab
+
+```
+  ── Galerie ──────────────────────────
+  [  🔓 Galerie freigeben  ]
+
+  ── Export-Jobs ──────────────────────
+  [  🔄 Aktualisieren  ]
+
+  ┌───────────────────────────────────┐
+  │  HTML-Viewer   ● Fertig  [↓ ZIP]  │
+  │  JSON-Export   ⏳ Läuft…          │
+  │  ZIP-Archiv    ✗ Fehler   [↺]     │
+  └───────────────────────────────────┘
+
+  [  + Neuer Export-Job  ]
+```
+
+- Status chips: green (Fertig), amber (Läuft), red (Fehler)
+- Download button inline per completed job
+- Only the jobs list refreshes on "Aktualisieren" — no full page re-render
+
+### Nutzer Tab
+
+Same structure as Host Nutzerverwaltung, with any additional admin-only actions
+(e.g. role assignment) added as extra controls per row.
+
+---
+
+## Touch gestures vs. desktop buttons (planned extension)
+
+Where a gesture is more ergonomic on mobile than a button, EventSnap prefers the gesture
+on touch and mirrors it as an explicit button on desktop. Inspired by Instagram, WhatsApp
+and Telegram — long-press for context, swipe to dismiss, double-tap to react.
+
+| Surface                                 | Touch gesture                       | Desktop equivalent                       |
+|-----------------------------------------|-------------------------------------|------------------------------------------|
+| Post card                               | Long-press → context bottom sheet   | ⋯ kebab in the card corner               |
+| Comment row                             | Long-press → bottom sheet           | ⋯ next to the comment timestamp          |
+| User row (Host / Admin dashboards)      | Long-press → bottom sheet           | Inline buttons (ban, promote, reset PIN) |
+| Lightbox                                | Swipe left / right                  | ←/→ arrow keys + on-screen chevrons      |
+| Lightbox                                | Swipe down to close                 | Esc + ✕ button                           |
+| Bottom sheet                            | Swipe down to dismiss               | Click backdrop or × in the sheet header  |
+| Feed                                    | Pull to refresh                     | Refresh icon next to the view toggle     |
+| Post (any)                              | Double-tap → like                   | Click the heart icon                     |
+
+**Discoverability rule:** every gesture must have a visible button equivalent on the same
+page. Gestures are never the *only* path to an action. Helps with stylus users,
+accessibility, and people who don't know the gesture vocabulary.
+
+**Context bottom-sheet pattern** (used by every long-press above):
+
+```
+┌──────────────────────────────────┐
+│         ▬  (drag handle)         │
+│                                  │
+│  🗑  Löschen                     │  ← destructive action red
+│  📥  Original anzeigen           │
+│  🔗  Teilen                      │
+│  🚩  Melden                      │  (only on others' content)
+│                                  │
+│  [        Abbrechen        ]     │
+└──────────────────────────────────┘
+```
+
+Each sheet is composed from a shared `<ContextSheet>` component (planned) with a single
+`actions: ContextAction[]` prop. Adding a new gesture context = define the actions array
+where needed. Drop-in, one file.
+
+## Design Principles Summary
+
+| Principle | Application |
+|-----------|-------------|
+| Thumb zone | All primary actions in bottom ~20% of screen |
+| One-hand operation | FAB centered, bottom sheets dismissable with swipe |
+| Minimal taps to upload | Source → picker → preview → upload: 4 taps |
+| Immediate feedback | Optimistic return to feed, background upload |
+| Progressive disclosure | Caption/tags optional; CTA always reachable |
+| No role clutter in nav | Role links only in Account, bar stays clean |
+| Collapsible sections | Long management pages stay usable on small phones |
+| Inner tabs for complex pages | Admin dashboard split across 4 focused tabs |
+| Gestures over chrome | Long-press for context menus, swipe to dismiss, double-tap to react — always with a button fallback for desktop and accessibility |

docs/DECISION-media-auth.md

@@ -0,0 +1,79 @@
+# Decision: media serving — unauthenticated UUID vs. signed gateway
+
+**Status:** OPEN — needs a call. Written 2026-06-30 from the 2026-06-27 audit
+(`fix/audit-2026-06-27-critical-medium`), which implemented the signed-gateway option that
+`main` did not adopt.
+
+**Scope:** how `main` serves uploaded photos/videos (originals + previews/thumbnails) to the
+`<img>`/`<video>` tags in the feed, lightbox, and diashow.
+
+---
+
+## The two models
+
+### A. Current `main` — unauthenticated, UUID-as-capability
+- `/api/v1/upload/{id}/original` — **no auth**; the unguessable upload UUID *is* the capability.
+  (Documented as intentional in `frontend/src/lib/data-mode-store.ts`.)
+- `/media/*` — static `axum` `ServeDir`, **no auth layer**, serves preview/thumbnail files by path.
+- No expiry, no signature, no per-request authorization on the bytes.
+- **Works with** plain `<img src>` (no Authorization header needed) and browser caching, at zero
+  per-request backend cost.
+
+### B. Audit branch — authenticated signed gateway
+- `/media/{kind}/{id}?sig=…` via `handlers::media::serve`.
+- HMAC signature (keyed off `jwt_secret`), **time-boxed** (~24 h, bucketed to a 1 h
+  URL-stability window so warm fetches stay cacheable).
+- Authorizes by **signature + uploader visibility**, not requester identity.
+- Still `<img>`-compatible (capability rides in the query string, not a header).
+- Cost: token mint/verify, ~2 DB queries per *cold* fetch, and the HMAC key currently reuses
+  `jwt_secret` (see "Media HMAC domain separation" in [SECURITY-BACKLOG.md](SECURITY-BACKLOG.md)).
+
+---
+
+## What actually differs (the tradeoff)
+
+| | A. UUID (main) | B. Signed gateway (audit) |
+|---|---|---|
+| Leaked URL/UUID (forwarded link, browser history, referer, logs) | **Permanent** full-res access | Access **expires** (~24 h); URL can't be re-minted |
+| Banned / departed guest | Retains **permanent** access to every original whose UUID they hold | Can't mint new URLs; can replay **held** URLs ≤ TTL only |
+| Revocation | None (UUID is forever) | Rotate the signing key → all outstanding URLs die |
+| Enumeration | Mitigated by UUIDv4 unguessability | Same, plus signature |
+| Per-request cost | Zero (static serve) | Token verify + ~2 DB queries (cold) |
+| Plumbing / failure surface | Minimal | Token mint/verify, key mgmt, cache-bucket logic |
+
+Neither model solves the fundamental `<img>`-can't-send-a-Bearer constraint: once a browser holds
+a media URL, it is replayable for that URL's lifetime. Model B simply **bounds that lifetime** and
+**adds revocability**; Model A's lifetime is *forever*.
+
+## Threat model (this deployment)
+
+Private single-box event app, ~100 guests, ~1 000 files, one event at a time. Media is **shared
+with all guests by design** — it is personal but not secret *within* the event. The real risk is
+**a URL escaping the event boundary** (a guest forwards a link; it lands in chat history, a public
+post, or server/proxy logs) granting an outsider — or a removed guest — access.
+
+## Recommendation
+
+This is a judgment call, not a clear-cut bug, so it's yours to make. My leaning:
+
+- **Adopt Model B (signed gateway)** if post-event link leakage or removed-guest access is a real
+  concern for you — it's the materially stronger posture (bounded exposure + key-rotation
+  revocation) and the implementation already exists on the audit branch. If adopted, also do the
+  cheap **HKDF domain-separation** for the HMAC key (backlog 🅱).
+- **Keep Model A (UUID)** as an *explicitly accepted risk* if you're comfortable that a leaked
+  link = permanent access at this scale. If so, two cheap hardening steps are still worth doing:
+  1. **Log hygiene** — ensure the upload UUID never lands in access logs with enough context to
+     correlate (the `TraceLayer` logs the request *path*, and `/api/v1/upload/{id}/original`
+     puts the UUID *in the path*). Confirm logs aren't shipped/retained where that matters.
+  2. **Document the decision** in `README`/`PROJECT.md` so "unauthenticated media" is a recorded
+     choice, not an oversight.
+
+**Sharpest single fact to decide on:** in Model A, a guest you *ban* keeps full-resolution access
+to every original they ever loaded, forever. In Model B, that access dies within ~24 h. If that
+asymmetry matters for your events, adopt the gateway.
+
+## If you adopt B
+The implementation lives on `origin/fix/audit-2026-06-27-critical-medium`:
+`backend/src/handlers/media.rs` (+ `media_token`), the `/media/{kind}/{id}` route in `main.rs`, and
+the feed/upload/host changes that emit signed URLs (`models/upload.rs`, `handlers/feed.rs`). It
+would need re-basing onto current `main` (which has since diverged across ~80 files).

docs/FEATURES.md

@@ -0,0 +1,313 @@
+# EventSnap — Feature Set & Capability Matrix
+
+This document is the authoritative, code-cross-checked summary of what EventSnap can do today
+and what is planned. For the design rationale of each area see [PROJECT.md](../PROJECT.md);
+for journeys / step-by-step flows see [USER_JOURNEYS.md](USER_JOURNEYS.md).
+
+Status legend: **✓ shipped** · **◐ partial** · **◯ planned** · **✗ out of scope**
+
+---
+
+## 1. Capability matrix by role
+
+| Capability                                              | Guest | Host  | Admin | Notes                                                                 |
+|---------------------------------------------------------|:-----:|:-----:|:-----:|-----------------------------------------------------------------------|
+| **Onboarding & sessions**                               |       |       |       |                                                                       |
+| Join via shared event link / QR code                    |   ✓   |   ✓   |   ✓   | Name-only registration; server issues JWT + 4-digit PIN               |
+| First-visit guided tour (4 steps)                       |   ✓   |   ✓   |   ✓   | Dismissed once, flag in `localStorage`                                |
+| Persistent 30-day session                               |   ✓   |   ✓   |   ✓   | JWT in `localStorage`; refreshed on activity                          |
+| Sign in on another device using name + PIN              |   ✓   |   ✓   |   ✓   | 3 wrong PINs → 15-min lockout                                         |
+| "Ich habe bereits einen Account" link on the join page  |   ✓   |   ✓   |   ✓   | Small inline link → `/recover` (name + PIN)                           |
+| View / copy own PIN any time ("My Account")             |   ✓   |   ✓   |   ✓   | Read from `localStorage`; never sent back from the server             |
+| Log out / "Leave event"                                 |   ✓   |   ✓   |   ✓   | Confirmation bottom-sheet; invalidates the session row                |
+| Rename own display name                                 |   ◯   |   ◯   |   ◯   | Not yet wired; PIN-protected change                                   |
+| Pick **data mode** (Saver / Original) in My Account     |   ✓   |   ✓   |   ✓   | Saver = compressed (default). Original = full files + data-usage warning. Applies to feed and diashow. Per-device, in `localStorage` |
+| Read the **Datenschutzhinweis** (privacy note)          |   ✓   |   ✓   |   ✓   | Free text set by Admin during setup; rendered preformatted in My Account; first-visit guide briefly points to it |
+| Admin password login (separate route)                   |       |       |   ✓   | 1-day token; lives in `sessionStorage`                                |
+| Reset another user's PIN (one-time display modal)       |       |  ✓*   |   ✓   | Host: guests only. Admin: hosts + guests. New PIN shown once to the requester; user signs in with it; PIN is stored on their device on next login. \* Host cannot reset another Host's PIN |
+|                                                         |       |       |       |                                                                       |
+| **Posting**                                             |       |       |       |                                                                       |
+| Pick photos/videos from device library (multi-select)   |   ✓   |   ✓   |   ✓   | Bottom-sheet source picker                                            |
+| In-app camera capture (`getUserMedia`)                  |   ✓   |   ✓   |   ✓   | Front/back toggle, photo, `MediaRecorder` video                       |
+| Caption + `#hashtag` extraction                         |   ✓   |   ✓   |   ✓   | Optional; hashtags parsed server-side                                 |
+| Edit own caption / hashtags after upload                |   ✓   |   ✓   |   ✓   | `PATCH /api/v1/upload/{id}`                                           |
+| Delete own upload                                       |   ✓   |   ✓   |   ✓   | Long-press on the card (or the kebab menu on desktop) → **Löschen** in the context sheet. Comment-style trash icon also available on each post elsewhere as it's added. |
+| Delete own comment                                      |   ✓   |   ✓   |   ✓   | Trash icon in lightbox                                                |
+| Background upload queue (survives reload)               |   ✓   |   ✓   |   ✓   | IndexedDB-persisted, sequential, retry                                |
+| Rate-limit auto-resume banner                           |   ✓   |   ✓   |   ✓   | Countdown above bottom nav; resumes when window opens                 |
+| Chunked / resumable upload for > 100 MB                 |   ◯   |   ◯   |   ◯   | Planned (v1.x)                                                        |
+|                                                         |       |       |       |                                                                       |
+| **Feed & social**                                       |       |       |       |                                                                       |
+| Chronological list feed (full-width cards)              |   ✓   |   ✓   |   ✓   | Default view, infinite scroll                                         |
+| 3-column grid feed with toggle                          |   ✓   |   ✓   |   ✓   | Video play badges, duration                                           |
+| Search & autocomplete (uploader + hashtag)              |   ✓   |   ✓   |   ✓   | Grid view; derived in-memory, no extra API calls                      |
+| Active filter chips (OR within type, AND across types)  |   ✓   |   ✓   |   ✓   | Multiple hashtags = OR; uploader + hashtag = AND                      |
+| Fullscreen lightbox with swipe                          |   ✓   |   ✓   |   ✓   | Swipe navigates the filtered set                                      |
+| Like / unlike any post                                  |   ✓   |   ✓   |   ✓   | Single toggle; SSE `like-update`                                      |
+| Read comments on any post                               |   ✓   |   ✓   |   ✓   |                                                                       |
+| Add a comment                                           |   ✓   |   ✓   |   ✓   | Hashtags in comments also parsed                                      |
+| Real-time feed via SSE                                  |   ✓   |   ✓   |   ✓   | `new-upload`, `new-comment`, `like-update`, `upload-processed`, `pin-reset`, `event-updated`, etc.  |
+| Pause SSE when app is backgrounded                      |   ✓   |   ✓   |   ✓   | Page Visibility API; reconnect on foreground                          |
+| Delta-fetch (`/feed/delta?since=`) on reconnect         |   ✓   |   ✓   |   ✓   | Runs on every visibility-restore; merges new + deleted uploads        |
+| Individual file download button per post                |   ✓   |   ✓   |   ✓   | "Original anzeigen" in the post context sheet — streams via `/api/v1/upload/{id}/original` |
+|                                                         |       |       |       |                                                                       |
+| **Live diashow** (see [CONCEPT_DIASHOW.md](CONCEPT_DIASHOW.md)) |  |  |  |                                                                       |
+| Start fullscreen auto-advancing slideshow               |   ✓   |   ✓   |   ✓   | Two queues: live (SSE) drains first, shuffle as fallback. Crossfade + Ken Burns transitions; pluggable. Respects data mode.   |
+|                                                         |       |       |       |                                                                       |
+| **Moderation (Host)**                                   |       |       |       |                                                                       |
+| List all event users                                    |       |   ✓   |   ✓   | Includes upload count, total bytes                                    |
+| Ban / unban a user                                      |       |   ✓   |   ✓   | Modal asks: hide their existing uploads, or keep visible?             |
+| Delete any upload                                       |       |   ✓   |   ✓   |                                                                       |
+| Delete any comment                                      |       |   ✓   |   ✓   |                                                                       |
+| Promote guest to Host                                   |       |   ✓   |   ✓   |                                                                       |
+| Demote Host to guest                                    |       |   ✓   |   ✓   | Hosts may demote other Hosts. Cannot demote self. Admins cannot be demoted by hosts. |
+| Reset a guest's PIN (Host) / any non-admin PIN (Admin)  |       |   ✓   |   ✓   | New PIN shown once in modal; Host shows/shares it with the guest      |
+| Lock new uploads ("Event schließen")                    |       |   ✓   |   ✓   | Likes + comments + browsing remain open                               |
+| Unlock new uploads                                      |       |   ✓   |   ✓   |                                                                       |
+| Release gallery → trigger export generation             |       |   ✓   |   ✓   | Enqueues both ZIP and HTML-viewer jobs                                |
+|                                                         |       |       |       |                                                                       |
+| **Instance configuration (Admin)**                      |       |       |       |                                                                       |
+| Live disk-usage / user / upload / banned stats          |       |       |   ✓   | Stats tab; queries `sysinfo`                                          |
+| Edit per-file limits (image MB / video MB)              |       |       |   ✓   | Config tab; hot-reloadable from DB                                    |
+| Edit per-endpoint rate limits                           |       |       |   ✓   | Upload/hour, feed/min, export/day                                     |
+| Toggle **all** rate limits on/off                       |       |       |   ✓   | Master switch — when off, every limiter passes through                |
+| Toggle **individual** rate limits on/off                |       |       |   ✓   | Per-endpoint switch (upload / feed / export / join)                   |
+| Toggle quota enforcement on/off (master + per-area)     |       |       |   ✓   | Master switch + per-area (storage / upload count). When off, nothing is enforced |
+| Edit quota tolerance                                    |       |       |   ✓   | Live `(free_disk × tolerance) / active_uploaders` formula enforced on upload |
+| Edit estimated guest count                              |       |       |   ✓   |                                                                       |
+| Edit compression-worker concurrency                     |       |       |   ✓   |                                                                       |
+| Edit **Datenschutzhinweis** (privacy note, free text)   |       |       |   ✓   | Plain text, whitespace + newlines preserved, no HTML. SSE `event-updated` broadcasts edits live. |
+| Inspect export job list & progress                      |       |       |   ✓   |                                                                       |
+| Low-disk alert (< 10 GB free)                           |       |       |   ◯   | Planned                                                               |
+| Event banner / cover image                              |       |       |   ◯   | DB column exists, no UI                                               |
+|                                                         |       |       |       |                                                                       |
+| **Quota visibility (Guest-facing)**                     |       |       |       |                                                                       |
+| Show current per-user quota estimate                    |   ✓   |   ✓   |   ✓   | "Du hast X MB von Y MB genutzt." in My Account and on the upload screen. Computed from the live formula. Hidden when quota enforcement is toggled off |
+|                                                         |       |       |       |                                                                       |
+| **Export**                                              |       |       |       |                                                                       |
+| Wait at locked export page until released               |   ✓   |   ✓   |   ✓   | Friendly "not yet available" copy                                     |
+| Download `Gallery.zip` (full-quality originals)         |   ✓   |   ✓   |   ✓   | Streamed via `async-zip`; `Photos/` + `Videos/` folders               |
+| Download `Memories.zip` (offline HTML viewer)           |   ✓   |   ✓   |   ✓   | Self-contained SvelteKit-static app + `data.json` + `media/`          |
+| HTML-export in-app guide modal before download          |   ✓   |   ✓   |   ✓   | Explains: unzip first, open `index.html`                              |
+| Per-IP export download rate limit (3 / day)             |   ✓   |   ✓   |   ✓   |                                                                       |
+|                                                         |       |       |       |                                                                       |
+| **Banned guest** (subset)                               |       |       |       |                                                                       |
+| Cannot upload, like, or comment                         |   ✗   |       |       | Returns HTTP 403                                                      |
+| Can browse the feed                                     |   ✓   |       |       |                                                                       |
+| Can still download the export once released             |   ✓   |       |       | Spec design choice                                                    |
+
+---
+
+## 2. Feature areas in detail
+
+### 2.0 Touch-first interactions (mobile) vs. buttons (desktop)
+
+EventSnap is mobile-first. Where it makes the UI cleaner, primary actions are reached via
+**gestures** on touch devices, with conventional **buttons** mirrored on tablet/desktop:
+
+- **Long-press on a post** → context bottom sheet ("Löschen", "Original anzeigen", report,
+  share). On desktop the same actions are a kebab/⋯ menu in the card's corner.
+- **Long-press on a comment** → context sheet with "Löschen" (own comments only) and
+  "Kopieren".
+- **Swipe left/right in the lightbox** → navigate the filtered set.
+- **Swipe down on a bottom sheet** → dismiss.
+- **Pull-to-refresh on the feed** → force a delta-fetch even when SSE is up.
+- **Double-tap on a post** → like (Instagram-style), with a heart-burst animation. Tap the
+  heart icon as the explicit alternative.
+
+Design rule: **gestures should always have a discoverable button equivalent** somewhere on
+the page, so the app stays usable on a stylus, mouse, or for users who don't know the
+gesture vocabulary. Take inspiration from Instagram, WhatsApp, and Telegram for the
+"feels right" baseline — long-press for context, swipe to dismiss, double-tap to react.
+
+### 2.1 Authentication and identity
+
+EventSnap's identity model is "**a name + a 4-digit PIN, scoped to one event**". There is no
+email, no password, no account portal.
+
+- **Joining.** On the join page the user types a display name. The server creates a `user`
+  row, generates a 4-digit PIN, stores `bcrypt(pin)`, signs a 30-day JWT, and returns the
+  PIN in clear text **once** in the response. The client persists the JWT and the PIN to
+  `localStorage`.
+- **PIN visibility.** The PIN is shown to the user *prominently* once at registration, and
+  remains visible in the My-Account page (read directly from `localStorage` — never sent
+  back from the server).
+- **Returning on the same device.** A valid JWT in `localStorage` → straight to the feed.
+- **Returning on a new device.** Type the name on the join page → server detects the
+  existing user → user is prompted for their PIN. `bcrypt.verify` → new JWT, fresh device
+  is now bound to the same account.
+- **Lockout.** 3 wrong PIN attempts → 15-minute lockout per user (`pin_locked_until` column,
+  migration 006).
+- **Name collisions.** Names are unique per event (case-insensitive, migration 007). If
+  someone tries to join with a name already taken, the join page automatically presents the
+  PIN-recovery form for that account ("Already taken — sign in instead, or pick another
+  name"). The join page also surfaces an explicit **"Ich habe bereits einen Account"**
+  link routing to `/recover` for users who already know they want to sign in.
+- **PIN reset by Host / Admin.** Planned. If a guest loses their PIN and `localStorage` is
+  gone everywhere, a Host (for guests) or Admin (for hosts and guests) can hit a
+  **PIN zurücksetzen** action in the user list. A fresh PIN is generated server-side, its
+  bcrypt stored, and the plaintext is shown **once** in a modal to the requesting
+  operator. The operator shows / sends the new PIN to the user, who then signs in via
+  `/recover` — the PIN is persisted to `localStorage` on that device on a successful
+  recovery, exactly like a brand-new join. Host cannot reset another Host's PIN; only
+  Admins can.
+- **Roles.** `guest` (default), `host`, `admin`. The Admin role is seeded from the
+  `ADMIN_PASSWORD_HASH` env var; admins log in at `/admin/login` with a password (separate
+  JWT, 1-day expiry, in `sessionStorage`). Hosts are guests promoted by an admin. **Hosts
+  may also demote other Hosts to guests** (planned) — but never themselves, to avoid
+  locking the event out of moderation. Admins can demote anyone except admins.
+
+### 2.2 Posting pipeline
+
+The upload pipeline is built for flaky mobile networks:
+
+1. **Source picker** (bottom sheet from the FAB): camera or gallery.
+2. **Preview screen** — staged files appear as thumbnails; user can remove individuals, add
+   a caption (with `#hashtags`), and tap quick-tag chips derived from the caption.
+3. **Submit** — the client immediately returns to the feed (optimistic UX). Files enter an
+   IndexedDB-persisted queue.
+4. **Queue worker** — runs sequentially (one upload at a time), per-file progress via XHR.
+   Survives reloads and app backgrounding. A red badge on the FAB indicates active uploads.
+5. **Server-side processing** — multipart received → MIME-sniffed via `infer` → size
+   validated → original stored → compression worker (bounded by a `tokio::sync::Semaphore`)
+   resizes to an 800-px preview (images via the `image` crate + `oxipng` for PNG) or
+   extracts a frame at the 1-second mark (videos via `ffmpeg`). Status is tracked in the new
+   `compression_status` column (migration 008).
+6. **Real-time fan-out** — `new-upload` SSE first (no preview yet), then `upload-processed`
+   when the preview/thumbnail is ready, so clients can swap a placeholder for the real image
+   without re-fetching the feed.
+7. **Rate-limit-aware client** — when the server returns HTTP 429 with `Retry-After`, the
+   queue parks remaining items and shows an inline countdown banner; uploads resume
+   automatically.
+
+### 2.3 Feed
+
+- **Two layouts** — chronological list (default) and 3-column grid. Toggle in the header.
+- **List view** has no search; it's the consumption-focused mode (like an Instagram feed).
+- **Grid view** has the search bar — autocomplete suggestions are computed in-memory from
+  the loaded uploads, so typing never hits the server.
+- **Filter chips** — multiple hashtags combine with OR; multiple uploaders combine with OR;
+  hashtag + uploader combine with AND. Matches the redesign concept exactly.
+- **Lightbox** — fullscreen view, swipe navigates the *filtered* set, with embedded
+  like/comment UI.
+- **Real-time** — SSE delivers `new-upload`, `upload-processed`, `like-update`,
+  `new-comment`, `upload-deleted`, `event-closed`/`event-opened`, `export-progress`,
+  `export-available`. Client pauses SSE on `visibilitychange: hidden` and reopens on visible.
+
+### 2.4 Host / Admin tooling
+
+- **Host dashboard** — three collapsible sections: Stats, Event-Einstellungen,
+  Nutzerverwaltung. Ban modal asks explicitly whether to hide the user's existing uploads
+  from the public feed. Promote/demote, lock/unlock, release-gallery are one-tap.
+- **Admin dashboard** — same dashboard plus three more inner tabs (Stats, Config, Export,
+  Nutzer). Config form covers per-file limits, rate limits, quota tolerance, estimated
+  guest count, and compression concurrency — all stored in the `config` table and read on
+  each request, so changes take effect without a restart. Disk widget pulls from the
+  `sysinfo` crate live.
+
+### 2.5 Data mode (planned)
+
+Each device picks a **data mode** in My Account; the setting lives in `localStorage` so a
+guest can be on Saver on their phone and Original on their laptop.
+
+| Mode       | Default? | Feed loads...               | Lightbox / diashow loads... | Warning shown? |
+|------------|:--------:|-----------------------------|------------------------------|:-------------:|
+| Datensparer (Saver)  |  ✓ | preview (compressed)        | preview                      | no            |
+| Original             |    | original                    | original                     | yes — "kann mobile Datennutzung erhöhen" once on enable |
+
+Applies uniformly to the live app's feed/lightbox **and** the diashow. The viewer (offline
+HTML export) is unaffected — it's already a snapshot of pre-bundled media variants.
+
+### 2.6 Rate limits and quotas — toggleable (planned)
+
+The Admin Config tab gains explicit on/off toggles in addition to the numeric inputs:
+
+- **Master switch — all rate limits.** When off, every limiter middleware short-circuits to
+  pass-through. Useful for testing or trusted internal events.
+- **Per-endpoint switches.** Upload / feed / export / join each have their own toggle. The
+  numeric input becomes informational while the toggle is off.
+- **Master switch — quotas.** When off, no quota check ever runs.
+- **Per-area quota switch.** Storage-bytes quota and upload-count quota can be disabled
+  independently.
+
+When a feature is toggled off, the relevant UI in the guest-facing app should adapt: e.g.
+the "Du hast X von Y MB genutzt" widget hides itself when storage quota is disabled. The
+quota estimate is computed from the same formula the server uses
+(`(free_disk × tolerance) / max(active_uploaders, 1)`) — surfaced in My Account *and* on
+the upload preview screen so guests know before they pick files.
+
+### 2.7 Privacy note (Datenschutzhinweis, planned)
+
+Admin sets a free-text **Datenschutzhinweis** during instance setup (Admin Dashboard →
+Config). It's stored as a single config key (plain text, whitespace and newlines
+preserved, no HTML). Guests see it in their **My Account** page, rendered inside a
+preformatted block — no parsing, no markdown, just exactly what the admin typed. The
+first-visit onboarding guide gains a one-line nudge: *"Datenschutzhinweis findest du in
+deinem Account."*
+
+Rationale: many real events (in Germany especially) need a per-event privacy statement
+without the operator wanting to ship a separate static page or rebuild the app.
+
+### 2.8 Export
+
+Two artifacts, both generated on demand after the host taps "Release gallery":
+
+- **Gallery.zip** — full-quality originals only, structured into `Photos/` and `Videos/`,
+  filenames `{date}_{time}_{username}_{id}.{ext}`, streamed via `async-zip` with no full
+  archive in memory.
+- **Memories.zip** — the offline HTML viewer. Pre-built SvelteKit-static app from
+  [frontend/export-viewer/](../frontend/export-viewer/), bundled with a generated
+  `data.json` snapshot and a `media/` folder of thumbnails + full-size variants. Open
+  `index.html` in any browser — no server required, no internet required. List/grid views,
+  lightbox, hashtag chips, like counts, comments — all visually matched to the live app.
+
+The export page shows live progress (SSE) while jobs run, then becomes a download button
+when complete.
+
+---
+
+### 2.9 Maintainability and extensibility
+
+EventSnap is small enough to be a single-developer project; it should stay easy to extend.
+A few principles to keep adding features cheap:
+
+- **Diashow transitions are drop-in components.** Each animation implements a small
+  interface and lives under `frontend/src/lib/diashow/transitions/`. Adding a transition is
+  one file + one entry in a registry.
+- **Feature toggles live in the `config` table.** Today's rate-limit and quota switches
+  follow the same pattern any new opt-in feature would use — no redeploy to flip
+  behaviour.
+- **One Svelte store per cross-cutting concern.** Auth, upload queue, SSE, data mode,
+  diashow state — composable rather than copy-pasted into each route.
+- **Migrations are append-only.** Never edit a shipped migration; always add a new pair.
+- **Background jobs share one pipeline.** Export and compression already publish progress
+  via the `export_job` row + SSE; future long-running work (analytics, archival) should
+  plug into the same shape.
+
+See [IDEAS.md](IDEAS.md) for a longer riff on these patterns.
+
+## 3. Out of scope (intentionally not built)
+
+These are explicit non-goals from [PROJECT.md §4](../PROJECT.md):
+
+- Native iOS / Android apps
+- Multiple simultaneous events (multi-tenancy)
+- Email-based auth / password reset
+- Push notifications
+- User-to-user direct messaging
+- Payment / monetisation
+- CI/CD pipeline
+- "Save to camera roll" automation on iOS/Android — guests download the ZIP and use their
+  platform file manager
+
+---
+
+## 4. See also
+
+- [USER_JOURNEYS.md](USER_JOURNEYS.md) — step-by-step flows for every supported scenario.
+- [CONCEPT_MOBILE_UI.md](CONCEPT_MOBILE_UI.md) — design reference for the mobile layout.
+- [CONCEPT_HTML_VIEWER.md](CONCEPT_HTML_VIEWER.md) — export-viewer design.
+- [CONCEPT_DIASHOW.md](CONCEPT_DIASHOW.md) — planned diashow design.
+- [IDEAS.md](IDEAS.md) — speculative extensions (global diashow, reactions, multi-tenancy, ...).
+- [PROJECT.md](../PROJECT.md) — full architectural blueprint and rationale.
+- [TEST_GUIDE.md](../TEST_GUIDE.md) — manual smoke-test script for the main flows.

docs/IDEAS.md

@@ -0,0 +1,199 @@
+# EventSnap — Ideas & Future Extensions
+
+A dumping ground for design ideas that are **not yet on the roadmap**. Everything here is a
+v2+ candidate, brainstormed once the core experience is stable. For shipped or actively
+planned scope see [FEATURES.md](FEATURES.md) and the `CONCEPT_*.md` design docs.
+
+The bar to land here is low: "would be cool one day" qualifies. The bar to graduate to a
+`CONCEPT_*.md` is much higher (design committed, ready to build).
+
+---
+
+## Diashow extensions
+
+### Global / synchronised diashow
+
+Multiple devices show **the same slide at the same time** (e.g. a projector in the main
+hall plus tablets behind the bar plus a screen by the photo booth).
+
+Sketch:
+- Server holds a single authoritative "current slide" cursor for the event.
+- New SSE event `diashow-tick` broadcasts `{ slide_id, started_at, next_at }`.
+- Each subscribed client renders locally — server only chooses ordering and pace.
+- Live-queue / shuffle-queue logic (see [CONCEPT_DIASHOW.md](CONCEPT_DIASHOW.md)) lives
+  server-side instead of client-side.
+- A "leader device" can claim the diashow, or the server runs it head­lessly. Host UI lets
+  Host start / stop the global diashow.
+- Plays well with venues that already have multiple displays — no need for HDMI splitters
+  or chromecast hacks.
+
+### Audio bed
+
+- Host uploads or selects a background track (per-event).
+- Videos in the diashow auto-mute so they don't fight the music.
+- Optional ducking when a video has speech.
+
+### Curated diashow mode
+
+- Diashow filtered by a hashtag (`#highlights`) or to a Host-pinned set ("Story" feature).
+- Useful for the end-of-evening recap reel.
+
+### Animation pack
+
+- More transitions out of the box: zoom, slide, mosaic, dip-to-black, push.
+- Per-event "theme" preset — wedding-elegant, party-energetic, minimal, gallery-classic.
+- Builds on the maintainability principle below: each transition is a drop-in Svelte
+  component, so growing the pack is trivial.
+
+### Lower-third metadata
+
+- Subtle chyron at the bottom of each slide: uploader name + timestamp + caption.
+- Off by default; toggle in the diashow settings popover.
+
+### Smart pacing
+
+- Detect video duration and let videos play their full length (with a cap), pause stills
+  for the remainder. Avoids choppy 6-second cuts on a clip with key content at 0:08.
+- "Action density" heuristic — slow down for portraits, speed up for landscapes.
+
+---
+
+## Social
+
+### Per-guest gallery
+
+A first-class "All posts by Anna" view, navigable from a guest's avatar — not just a
+filter chip. Doubles as a personal "what did I post?" page.
+
+### Story-style highlights
+
+Host curates a best-of timeline pinned at the top of the feed (already in PROJECT.md's
+"Should Have"). Tap-through, fullscreen, ~5 s per story, like Instagram. Could double as
+the source for the curated-diashow mode above.
+
+### Reactions beyond like
+
+Multiple emoji reactions (❤️ 😂 😍 🎉 🥲) instead of just like. The DB design already keys
+the `like` table on `(upload_id, user_id)` — generalising to `(upload_id, user_id, kind)`
+is a small migration.
+
+### Mentions and reply-threads in comments
+
+- `@anna` in a comment becomes a tap-through to her profile / posts.
+- Threaded replies under each top-level comment.
+- Combined with PWA push, drives engagement.
+
+### Collaborative captions
+
+Co-authored captions when multiple uploaders are in the same photo — second tagger
+contributes their `#hashtags` to the same post.
+
+---
+
+## Notifications
+
+- PWA push for new comments on a guest's own posts.
+- Per-user opt-in; granular per-event preference (mute event, mute uploader, etc.).
+- Email digest after the event (1 message per guest) — optional, controversial vs. the
+  "no email" identity model. Could be opt-in only.
+
+---
+
+## Capture & posting
+
+- **Live-photo mode** — capture a 1–2 s clip alongside each still (Apple-style). Diashow
+  could animate stills using the live clip as the Ken Burns source.
+- **Boomerang / GIF capture** — short looping clips.
+- **Client-side filters and stickers** — Instagram-style.
+- **Voice notes** attached to a photo — "first dance" voice memo + the photo.
+- **Bulk-upload presets** — pre-fill a caption for a batch ("Photos from the ceremony").
+
+---
+
+## Privacy & moderation
+
+- **Per-post visibility** — "only visible to people with this hashtag" or
+  "private to my friend group".
+- **Pre-moderation queue** — Host approves posts before they hit the public feed (default
+  off; for sensitive events).
+- **Auto-blur** of detected faces of non-guests, or NSFW detection.
+- **Per-uploader watermark** on full-quality downloads.
+
+---
+
+## Multi-tenancy
+
+- **Multiple events per instance** — picked by URL slug. Today the binary is single-event.
+- **Org accounts** — a wedding photographer running 4 weddings a month against the same
+  deployment.
+- **Per-event admin** vs. **instance admin** roles.
+
+---
+
+## Internationalisation
+
+- Localisation beyond German — English, French, Spanish, ...
+- Admin picks UI language during setup; per-user override.
+- Strings extracted into a small JSON catalogue — works well with `svelte-i18n` or similar.
+
+---
+
+## Export
+
+- **Year-in-pictures PDF** — host-curated layout, printable.
+- **ICS calendar attachment** of the event, included in the export ZIP.
+- **Direct upload to a guest's chosen cloud** (iCloud, Google Photos) — needs OAuth, adds
+  a third-party integration where today there are none.
+
+---
+
+## Resilience / infrastructure
+
+- **Distributed rate limiting** (Redis) for multi-instance / multi-event deploys.
+- **Object-storage backend** (S3 / MinIO) behind a feature flag — out of scope for the
+  single-VPS use case but easy to add if multi-tenancy is ever pursued.
+- **Read replicas** for very large events.
+
+---
+
+## Maintainability principles to keep adding features cheap
+
+The codebase is small today and should stay friendly to extension. A few patterns to lean
+into as the surface grows:
+
+- **Diashow transitions as drop-in components.** Each transition implements a tiny
+  interface (`enter`, `leave`, optional `duration`). Adding a new animation is one file in
+  `frontend/src/lib/diashow/transitions/` and one line in a registry. Same idea for
+  hashtag-filter operators.
+- **Per-feature toggle flags in the `config` table.** Today rate limits and quotas are
+  individually toggleable (see [FEATURES.md](FEATURES.md)). The same pattern fits for any
+  future opt-in feature — no need to redeploy to flip behaviour.
+- **Background-task trait on the server.** Export, compression, and (future) analytics
+  jobs would all share a `BackgroundJob` interface that wires into the existing
+  `export_job` progress + SSE pipeline. New long-running work plugs in by implementing the
+  trait — no bespoke worker code per feature.
+- **One Svelte store per cross-cutting concern.** Auth, upload queue, SSE, data mode,
+  diashow state — each lives in its own store under `frontend/src/lib/`. New UI features
+  consume the stores; cross-feature behaviour is composed, not copy-pasted.
+- **DTOs in one file** ([frontend/src/lib/types.ts](../frontend/src/lib/types.ts)),
+  mirrored to the Rust DTOs. Changing a contract is exactly two edits.
+- **Migration-first schema evolution** — never edit an old migration; always add a new
+  `0NN_*.up.sql` / `.down.sql` pair. Already the discipline; just keep it.
+
+---
+
+## Speculative / "would be cool"
+
+Lower bar of plausibility — keep these around as conversation seeds:
+
+- **AI-generated event summary** at release time (3-paragraph recap, key moments,
+  funniest comment).
+- **AI auto-tagging** — suggested hashtags based on image content, opt-in per upload.
+- **Guest-of-honour mode** — special UI for the couple / birthday person showing
+  *everything they're in*, prioritised by face detection.
+- **Live caption translation** for international weddings — auto-translate comments
+  inline.
+- **Sound-reactive diashow** — slides advance in sync with music BPM picked up via the
+  device mic.
+- **Photo-booth integration** — a fixed iPad at the venue posts to the feed with a single
+  tap, no PIN.

docs/MOBILE_TESTING_GUIDE.md

@@ -0,0 +1,392 @@
+# Mobile Testing Guide — EventSnap v0.15.0
+
+## Setup
+
+### Dev Servers
+
+| Service | URL | Notes |
+|---------|-----|-------|
+| Frontend | `http://localhost:5173` | Vite dev server, hot-reload |
+| Backend API | `http://localhost:3000` | Rust/Axum |
+| Database | `localhost:5432` | PostgreSQL (Docker) |
+
+The frontend dev server proxies `/api` and `/media` to the backend automatically.
+
+**Mobile device access:** Connect your phone to the same Wi-Fi network.
+Find your machine's local IP (`ip a | grep 192.168` or `hostname -I`), then open
+`http://<your-ip>:5173` on your phone.
+
+### Browser DevTools Mobile Emulation (quick testing without a phone)
+
+1. Open Chrome → DevTools (`F12`) → Toggle device toolbar (`Ctrl+Shift+M`)
+2. Select **iPhone 14 Pro** or **Pixel 7** from the device dropdown
+3. Reload the page — safe-area insets and viewport are emulated
+4. To test touch gestures: enable "Touch" in the three-dot menu inside the device toolbar
+
+---
+
+## Test Accounts
+
+Use the following to get all three roles:
+
+| Role | How to get it |
+|------|--------------|
+| Guest | Join at `/join` with any name |
+| Host | Promote a guest via Host Dashboard, or set role in DB |
+| Admin | POST to `/api/v1/admin/login` or navigate to `/admin/login` |
+
+Admin password: `admin123` (set in `.env`)
+
+---
+
+## Section 1 — Bottom Navigation Bar
+
+**Goal:** Verify the tab bar is present, thumb-accessible, and correct per role.
+
+### 1.1 Bar Presence & Safe Area
+- [ ] Open `/feed` — a bottom tab bar with **Galerie**, a blue circle FAB, and **Konto** appears
+- [ ] On a real iPhone/Safari: the bar does **not** overlap the home indicator (safe-area padding)
+- [ ] On Chrome DevTools with an iPhone device: the bar is above the viewport bottom
+- [ ] Scroll down on a long feed — the bar stays **fixed** at the bottom at all times
+- [ ] The bar has a frosted-glass blur effect (`bg-white/90 backdrop-blur-md`)
+
+### 1.2 Active Tab Indicator
+- [ ] On `/feed` — the Galerie icon is **blue**; Konto icon is gray
+- [ ] Tap **Konto** — navigates to `/account`; Konto icon turns blue, Galerie goes gray
+- [ ] Tap **Galerie** — navigates back to `/feed`
+
+### 1.3 Role Gating
+- [ ] Log in as a **guest** — bar shows Galerie · FAB · Konto (3 items)
+- [ ] Log in as a **host** — same 3 items (dashboard links are inside Account, not the bar)
+- [ ] Log in as **admin** — same 3 items
+
+### 1.4 Auth Pages Hide the Bar
+- [ ] Visit `/join` — **no** bottom bar
+- [ ] Visit `/recover` — **no** bottom bar
+- [ ] Visit `/admin/login` — **no** bottom bar
+
+---
+
+## Section 2 — Upload FAB & Bottom Sheet
+
+**Goal:** Verify the FAB opens the upload sheet and both source options work.
+
+### 2.1 FAB Appearance
+- [ ] The FAB is a blue circle elevated ~12 px above the tab bar
+- [ ] A camera icon with an implicit "+" meaning is shown
+- [ ] When uploads are in the queue (pending or uploading), a **red badge number** appears on the FAB
+
+### 2.2 Sheet Opens & Closes
+- [ ] Tap the FAB — a bottom sheet slides up smoothly (~300 ms) from below
+- [ ] Sheet shows: **Galerie** (blue icon), **Kamera** (purple icon), **Abbrechen**
+- [ ] Tap the gray backdrop — sheet slides **back down** and closes
+- [ ] Tap **Abbrechen** — sheet closes
+- [ ] Swipe the drag handle downward — sheet closes *(if touch gestures are enabled)*
+
+### 2.3 Gallery Source
+- [ ] Tap **Galerie** — the native file picker opens
+- [ ] Select 1–3 images or videos
+- [ ] Sheet closes; you are navigated to `/upload` (the composer page)
+- [ ] Thumbnail strip at the top shows your selected files
+- [ ] **Bottom nav is gone** on this page (immersive full-screen)
+
+### 2.4 Camera Source
+- [ ] Tap the FAB → **Kamera**
+- [ ] Browser asks for camera permission — grant it
+- [ ] Full-screen camera UI appears (existing CameraCapture component)
+- [ ] Take a photo
+- [ ] Camera closes; you are navigated to `/upload` with the captured image in the strip
+
+### 2.5 Upload Composer Page
+- [ ] Back `×` button top-left → returns to `/feed`, clears pending files
+- [ ] Thumbnail strip scrolls horizontally when >3 files
+- [ ] Each thumbnail has a small `×` to remove it — tapping removes that file only
+- [ ] Caption `<textarea>` is **auto-focused** (keyboard opens on mobile)
+- [ ] Type `#party #spaß` in the caption — **quick-tag chips** appear below the textarea in real time
+- [ ] Quick-tag chips are read-only (they reflect what's already in the caption)
+- [ ] The **"Hochladen"** sticky button at the bottom shows the file count: "2 Dateien hochladen"
+- [ ] Button is **disabled** when the strip is empty
+- [ ] Also a smaller "Hochladen" button in the header (convenient on desktop/landscape)
+- [ ] Tap **Hochladen** — files are queued, you are returned to `/feed`
+- [ ] A **slim blue progress bar** appears just above the bottom tab bar while uploading
+- [ ] FAB shows a **red badge** during upload; badge disappears when done
+- [ ] A brief "Fertig" / completed state appears in the UploadQueue (check queue store)
+
+---
+
+## Section 3 — Feed: List View
+
+**Goal:** Verify the default chronological list view.
+
+### 3.1 Default State
+- [ ] Open `/feed` — **list view** is active by default (≡ icon highlighted in the toggle)
+- [ ] Posts appear as full-width cards in reverse-chronological order (newest first)
+
+### 3.2 List Card Anatomy
+For each card, verify:
+- [ ] **Avatar circle** with the uploader's initial letter and a deterministic color
+- [ ] **Display name** + **relative timestamp** ("vor 2 Min.", "vor 1 Std.", etc.)
+- [ ] **Media**: full-width image, or video with a play button overlay
+- [ ] **Caption** below the media (truncated to 3 lines with `...` if long)
+- [ ] **Like count** (❤️) and **Comment count** (💬) action buttons
+- [ ] Tapping the ❤️ toggles the like optimistically (count changes immediately)
+- [ ] Tapping 💬 or the media opens the **Lightbox Modal** (existing behavior, unchanged)
+
+### 3.3 Hashtag Chips (List View Only)
+- [ ] Below the main header, hashtag filter chips are visible in **list view**
+- [ ] Tap a hashtag chip — feed re-fetches filtered by that tag
+- [ ] Tap **Alle** — returns to unfiltered feed
+- [ ] Chips are **not visible** when grid view is active
+
+### 3.4 Infinite Scroll
+- [ ] Scroll to the bottom — more posts load automatically
+- [ ] A spinner appears briefly while loading more
+- [ ] Scroll sentinel triggers ~200 px before the actual bottom
+
+### 3.5 Real-Time Updates (SSE)
+- [ ] Open the feed on two devices/tabs simultaneously
+- [ ] Upload a photo on one — it appears at the **top** of the other's list view in real time
+- [ ] Like a post on one — the count updates on the other
+
+---
+
+## Section 4 — Feed: Grid View & Search
+
+**Goal:** Verify the 3-column grid, search bar, autocomplete, and filter chips.
+
+### 4.1 Switching to Grid View
+- [ ] Tap the ⊞ grid icon in the header — view switches to a 3-column grid
+- [ ] The ≡/⊞ toggle shows ⊞ as active (white background, shadow)
+- [ ] Hashtag chips **disappear**; a **search bar** slides in below the header
+
+### 4.2 Grid Layout
+- [ ] Grid is **3 columns** with equal square cells (no 2-column fallback on mobile)
+- [ ] Videos show a ▶ play button overlay
+- [ ] Tapping a cell opens the Lightbox Modal
+- [ ] Grid background is seamless (0.5px gap between cells)
+
+### 4.3 Search Bar
+- [ ] Search bar shows: 🔍 icon, placeholder "Nutzer oder #Tag suchen…", × clear button
+- [ ] Tapping the bar focuses it and opens the keyboard
+- [ ] × button appears only when there is text in the input; tapping it clears the query
+
+### 4.4 Autocomplete — On Focus (Empty)
+- [ ] Focus the search bar with no text — a dropdown appears with:
+  - Up to 3 uploader names (person icon)
+  - Up to 3 popular tags (#)
+- [ ] The dropdown disappears when the input loses focus (150 ms delay)
+
+### 4.5 Autocomplete — Tag Suggestions
+- [ ] Type `#` — only **tag suggestions** appear (no users), sorted by frequency
+- [ ] Type `#par` — only tags starting with "par" remain (e.g. `#party`, `#parade`)
+- [ ] Tap a suggestion — it's added as a **blue filter chip** below the search bar; input clears
+
+### 4.6 Autocomplete — User Suggestions
+- [ ] Type a partial name (e.g. `max`) — users matching "max" appear first, then tags containing "max"
+- [ ] Tap a user suggestion — chip added: shows the name without `#` prefix
+
+### 4.7 Filter Chips
+- [ ] After selecting a tag filter — grid shows only posts with that tag in the caption
+- [ ] Select a second tag — grid shows posts with **either** tag (OR logic)
+- [ ] Select a user **and** a tag — grid shows posts by that user **that also** have that tag (AND across types)
+- [ ] Each chip has an **× remove button**; tapping it removes only that chip
+- [ ] When 2+ chips are active: **"Alle löschen"** link appears; tapping clears all filters
+- [ ] When no results match: "Keine Treffer für die gewählten Filter." + "Filter zurücksetzen" button
+
+### 4.8 Switching Back to List View
+- [ ] Tap ≡ — list view returns; search bar gone; hashtag chips reappear
+- [ ] Active grid filters are **reset** when switching back to list (no stale state)
+
+---
+
+## Section 5 — Account Page
+
+**Goal:** Verify the profile card, dashboard links, and leave-confirm flow.
+
+### 5.1 Profile Card
+- [ ] Open `/account` via the Konto tab
+- [ ] **Avatar circle** shows your initial letter in a deterministic color
+- [ ] **Display name** and **role badge** (Gast / Gastgeber / Admin) shown
+- [ ] Session expiry date shown in small text below
+
+### 5.2 Dashboard Links (Host/Admin Only)
+- [ ] Log in as a **guest** — no "Dashboards" section visible at all
+- [ ] Log in as a **host** — "Dashboards" section shows ⭐ **Host-Dashboard** → chevron
+  - [ ] Tapping it navigates to `/host`
+  - [ ] No Admin-Dashboard link visible
+- [ ] Log in as **admin** — both links appear:
+  - [ ] ⭐ Host-Dashboard → `/host`
+  - [ ] 🛡 Admin-Dashboard → `/admin`
+
+### 5.3 PIN Card
+- [ ] Amber card shows the 4-digit PIN in large monospace font
+- [ ] **Kopieren** button copies to clipboard; label changes to "Kopiert!" for 2 seconds
+- [ ] If no PIN is stored: fallback message shown
+
+### 5.4 Konto Section
+- [ ] **Gerät wechseln / PIN nutzen** → navigates to `/recover`
+- [ ] **Event verlassen** (red text) → tapping opens a **leave-confirm bottom sheet**
+  - [ ] Sheet shows: "Event verlassen?", "Du wirst abgemeldet…", red "Abmelden" + "Abbrechen"
+  - [ ] Tap backdrop — sheet closes, you remain logged in
+  - [ ] Tap **Abbrechen** — same
+  - [ ] Tap **Abmelden** — you are logged out and redirected to `/join`
+
+### 5.5 No Stale Nav Links
+- [ ] **No** "Zur Galerie" link in the header (navigation is via the bottom bar)
+
+---
+
+## Section 6 — Host Dashboard
+
+**Goal:** Verify the back arrow, collapsible sections, and all existing host actions still work.
+
+### 6.1 Navigation
+- [ ] Open Host Dashboard via Account → ⭐ Host-Dashboard
+- [ ] Page shows a **← back arrow** in the top-left header
+- [ ] Tapping it navigates to `/account`
+- [ ] **No** shield/gallery header icons (removed)
+- [ ] Bottom tab bar is still visible
+
+### 6.2 Statistiken Section (Collapsible)
+- [ ] Section is **expanded** by default with a downward chevron
+- [ ] Shows a 2×2 grid of stat cards: Gäste, Uploads, Uploads status (Offen/Gesperrt), Freigegeben (Ja/Nein)
+- [ ] Numbers are large and readable on mobile
+- [ ] Tap the **Statistiken** header button — section collapses (smooth max-height animation)
+- [ ] Chevron rotates 180° to point upward when collapsed
+- [ ] Tap again — section expands
+
+### 6.3 Event-Einstellungen Section (Collapsible)
+- [ ] Collapse/expand works same as above
+- [ ] Shows **"Uploads sperren"** (amber) / **"Uploads wieder öffnen"** (green) button
+- [ ] Shows **"Galerie freigeben"** (blue) / "Galerie bereits freigegeben" (disabled gray)
+- [ ] Tap "Uploads sperren" — toast confirms, button switches to "Uploads wieder öffnen"
+- [ ] Existing functionality unchanged
+
+### 6.4 Nutzerverwaltung Section (Collapsible)
+- [ ] **Search bar** at top of section filters the user list in real time (client-side)
+- [ ] Each user row shows name, role badge, banned badge (if applicable), upload count/bytes
+- [ ] **Sperren** button triggers the existing ban modal (confirm + hide-uploads checkbox)
+- [ ] **Entsperren** appears for banned users
+- [ ] **Host** button promotes a guest to host role
+- [ ] **Degradieren** appears for hosts (admin only)
+- [ ] Toast notifications appear above the bottom bar (not obscured by it)
+
+---
+
+## Section 7 — Admin Dashboard
+
+**Goal:** Verify the inner tab bar, all 4 tabs, and the new Nutzer tab.
+
+### 7.1 Navigation
+- [ ] Open Admin Dashboard via Account → 🛡 Admin-Dashboard
+- [ ] Page shows **← back arrow** → `/account`
+- [ ] **No** star/gallery header icons
+- [ ] Bottom tab bar visible
+
+### 7.2 Inner Tab Bar
+- [ ] A second tab bar appears **below the main header**, sticky on scroll
+- [ ] 4 tabs: **Stats · Config · Export · Nutzer**
+- [ ] Active tab has a blue bottom border and blue text
+- [ ] Inactive tabs are gray
+- [ ] Tabs are scrollable horizontally (try narrowing viewport)
+- [ ] Switching tabs is instant with no page reload
+
+### 7.3 Stats Tab
+- [ ] Shows a **2×2 grid** of metric cards: Gäste, Uploads, Kommentare, Speicher %
+- [ ] Values are large (`text-3xl`)
+- [ ] Below the grid: a full-width disk usage bar with color coding
+  - Blue ≤ 74%, Amber 75–89%, Red ≥ 90%
+- [ ] Exact used/total/free values shown
+
+### 7.4 Config Tab
+- [ ] Shows stacked label + full-width input for each of the 8 config keys
+- [ ] Inputs are `type="number"` with large touch targets
+- [ ] A **"Speichern"** button is **sticky at the bottom** of the tab (always visible, even on long scroll)
+- [ ] Edit a value → tap Speichern → toast "Konfiguration gespeichert."
+- [ ] Tap Speichern with no changes → toast "Keine Änderungen."
+
+### 7.5 Export Tab
+- [ ] **"Galerie freigeben"** button triggers gallery release
+- [ ] **"Aktualisieren"** button refreshes the jobs list only (no full page flash)
+- [ ] Export jobs listed with status chips: Ausstehend (gray) / Läuft (blue) / Fertig (green) / Fehlgeschlagen (red)
+- [ ] Running jobs show a progress bar
+- [ ] Failed jobs show the error message in red
+
+### 7.6 Nutzer Tab (New)
+- [ ] Users are loaded from `/host/users` (admin shares host permissions)
+- [ ] **Search bar** filters list in real time
+- [ ] Same ban/unban/promote/demote actions as Host dashboard
+- [ ] After an action (e.g. ban) only the users list refreshes, not the whole page
+
+---
+
+## Section 8 — Toast Position
+
+- [ ] On host/admin pages, toasts appear at `bottom-24` (above the bottom nav bar)
+- [ ] Toasts are **not** obscured by the nav bar
+
+---
+
+## Section 9 — Desktop Usability (Second Citizen)
+
+**Goal:** Confirm all pages are still usable on a wide viewport.
+
+### 9.1 Layout Centering
+- [ ] On a 1280px+ viewport, all pages center their content at `max-w-2xl` or `max-w-3xl`
+- [ ] Bottom tab bar spans full width but content columns remain centered
+- [ ] No content is clipped or overflows horizontally
+
+### 9.2 Feed Desktop
+- [ ] List view: cards are centered, readable at 672px max width
+- [ ] Grid view: 3 columns at max-width — cells are larger and look good
+- [ ] Search bar is full-width within the max-width container
+
+### 9.3 Upload Composer Desktop
+- [ ] Upload page is full-height, centered column
+- [ ] Both the header "Hochladen" button AND the sticky bottom button are present
+- [ ] Desktop users can click the header button (more convenient without reaching to bottom)
+
+### 9.4 Host / Admin Desktop
+- [ ] Host collapsible sections work with mouse clicks
+- [ ] Admin inner tabs work with mouse clicks; all 4 tabs visible without scrolling at 1280px
+- [ ] Config tab sticky save is visible on desktop scroll
+
+---
+
+## Section 10 — Edge Cases
+
+### 10.1 Upload with No Files Selected
+- [ ] Navigate directly to `/upload` in the browser
+- [ ] No files pending → "Keine Dateien ausgewählt" screen shown with "Zurück" button
+- [ ] "Hochladen" button is disabled
+
+### 10.2 Rate Limiting
+- [ ] Upload rapidly beyond the configured limit (default: 10/hour)
+- [ ] A `429` response is received
+- [ ] The countdown banner appears above the bottom nav: "Upload-Limit erreicht. Wird in X Sek. automatisch fortgesetzt."
+- [ ] After the countdown, the queue resumes automatically
+
+### 10.3 SSE Reconnect
+- [ ] Stop the backend briefly and restart
+- [ ] The feed reconnects (SSE) — new uploads appear once the backend is back
+
+### 10.4 Back Navigation from Upload
+- [ ] Pick files → navigate to `/upload`
+- [ ] Tap `×` → files are discarded (`clearPending()` runs, object URLs are revoked)
+- [ ] Navigate back to `/upload` directly — "Keine Dateien ausgewählt" shown (not stale files)
+
+### 10.5 Grid Filter Persistence
+- [ ] Set a filter chip in grid view
+- [ ] Switch to list view — filter is cleared (list always shows full unfiltered feed)
+- [ ] Switch back to grid — search bar is empty, no stale chips
+
+---
+
+## Known Limitations (Not Bugs)
+
+| Item | Status |
+|------|--------|
+| "Anzeigename ändern" in Account | Deferred — shown as disabled; requires `/me` PATCH endpoint |
+| Upload count in Account profile card | Deferred — requires `/me` GET endpoint |
+| CSS collapse animation on host sections | Uses `max-h` trick; may be slightly sluggish for very large user lists |
+| Autocomplete results | Derived from currently-loaded posts only; new posts via SSE update the pool automatically |

docs/SECURITY-BACKLOG.md

@@ -0,0 +1,107 @@
+# Security & Hardening Backlog
+
+Tracks the deliberately-deferred items from the 2026-06-27 security audit and its review passes.
+
+**Provenance & reconciliation.** This document was extracted from the
+`fix/audit-2026-06-27-critical-medium` branch and reconciled against `main` on 2026-06-30.
+That audit's Critical→Medium findings were **largely re-implemented into `main`** through the
+later batch branches (security-review-followups, security-review-batch-2, the UX batches) rather
+than by merging the audit branch — so `git` shows no merge, but the controls are present. Each
+item below is tagged with its **current status in `main`**:
+
+- ✅ **Done in main** — addressed (possibly via a different implementation).
+- ⬜ **Open** — still applies to `main`.
+- 🔀 **Contingent** — only relevant if `main` adopts the audit's signed-media gateway (see
+  [DECISION-media-auth.md](DECISION-media-auth.md)).
+
+> The audit branch additionally implemented an **authenticated, signed media gateway** that `main`
+> did **not** adopt — `main` serves media unauthenticated (static `ServeDir` + UUID-capability
+> `/api/v1/upload/{id}/original`). That architectural choice is written up separately in
+> [DECISION-media-auth.md](DECISION-media-auth.md); the "by-design notes" at the bottom of this
+> file describe the *audit branch's* model and apply to `main` only if that gateway is adopted.
+
+---
+
+## 🅱 Worth a tracked ticket (real, not one-liners)
+
+- ⬜ **Moderation UI gap** — the backend `DELETE /host/upload/{id}` and `DELETE /host/comment/{id}`
+  endpoints have **no frontend caller**, so a host cannot remove a *guest's* content from the UI
+  (the feed `ContextSheet` only offers delete for the viewer's *own* uploads —
+  `target.user_id === myUserId`). Still a functional hole in `main`. Needs a host-facing "remove"
+  action wired to those endpoints, gated on host/admin role.
+
+- ✅/⬜ **Feed reactivity** — *Mostly fixed in `main`.* The full-reload-on-every-SSE-event problem
+  (a single `like-update`/`new-comment`/`upload-processed` calling `loadFeed(true)` and collapsing
+  a scrolled feed) was fixed in the UX batch: `main` now patches the affected card in place
+  (`patchCount`) and debounces processing (`scheduleInPlaceRefresh`). ⬜ **Remaining sub-item:**
+  owner-deleted uploads are still not broadcast — `delete_upload` returns `204` with no
+  `upload-deleted` SSE, so other clients keep showing a deleted post until refresh (only host
+  deletes broadcast). Emit `upload-deleted` from the owner delete path too.
+
+- 🔀 **Media HMAC domain separation** — *Only applies if the signed-media gateway is adopted.* The
+  audit's signed-media tokens reuse `jwt_secret` as the HMAC key; a dedicated derived key
+  (`HKDF(jwt_secret, "media-url")`) would isolate the domains so a future change to one can't
+  weaken the other. N/A to `main` as it stands (no signed media).
+
+- ⬜ **Quota mount-detection + low-disk guard** — *Still applies to `main`.* `compute_storage_quota`
+  (`backend/src/handlers/upload.rs`) and `admin.rs` pick the disk via `starts_with`, so root (`/`)
+  is a wildcard prefix that can match the wrong device; and there's no hard min-free-space
+  precheck when the quota is disabled. Use longest-prefix match; add an unconditional 507/429 when
+  free space is critically low.
+
+## ✅ Fixed in main since the audit (for the record)
+
+These audit findings are present in `main` today (verified 2026-06-30): server-side MIME/ext
+allowlist on upload, recovery-PIN lockout backoff, DB port no longer publicly exposed, bcrypt
+offloaded via `spawn_blocking`, bounded compression concurrency (semaphore), bounded feed queries
+(`LIMIT ≤ 100`), and the viewport-fit / reduced-motion / aria a11y pass. An image-decode
+decompression-bomb cap (`image::Limits` 12000×12000 / 256 MiB) lives in
+`backend/src/services/compression.rs`.
+
+**Fixed in the 2026-07 review pass** (previously *claimed* fixed here but were not — corrected):
+- **Effective JWT production-secret guard** — `APP_ENV=production` is now set for the app service,
+  and `config.rs::validate_secrets` rejects any placeholder-ish `JWT_SECRET`/`ADMIN_PASSWORD_HASH`
+  (not just the exact dev sentinel) and enforces `len ≥ 32` unconditionally in prod.
+- **Unspoofable client IP in the rate limiter** — `client_ip` now takes the right-most
+  `X-Forwarded-For` entry (the hop Caddy appends), so a client-supplied left-most value is ignored.
+- **Live role/ban re-check** — the auth extractor re-reads the user row and trusts the DB `role`
+  and `is_banned` flag rather than the JWT claim, so a demote/ban takes effect immediately (no
+  30-day token window). Bans are enforced on write handlers + the host/admin extractors, preserving
+  the documented read-only access for banned guests (USER_JOURNEYS §10); demoted hosts lose host
+  powers at once.
+- **Container healthchecks** — `app` and `frontend` now have healthchecks and Caddy waits on
+  `service_healthy`.
+- **Event-scoped `ban_user`** — the ban UPDATE is now scoped by `event_id` like its siblings.
+
+## 🅲 Consciously won't-fix at ~100-guest single-box scale
+
+Diminishing returns vs. the deployment's actual threat model. Revisit only if the scale or
+tenancy model changes.
+
+- Rate-limiter `HashMap` key LRU/cap (attacker-chosen `recover:{ip}:{name}` keys accumulate up to
+  the 24h prune ceiling) — bounded and pruned; not worth an LRU.
+- "Last host" / host↔host role-churn guard — operational, low blast radius.
+- Performance micro-indexes (`idx_like_user_upload`, comment pagination index) — current queries
+  are sub-ms at this row count.
+- Optimistic-like in-flight guard, ownership-snapshot-at-mount, assorted copy tweaks — UX polish.
+- **Mid-session ban does not tear down an already-open SSE stream.** The live-ban check lives in the
+  `AuthUser` extractor, but the stream authenticates via ticket→session (`handlers/sse.rs::stream`),
+  so a user banned while connected keeps receiving broadcast events until their stream drops. New
+  tickets *are* blocked (`issue_ticket` uses `AuthUser`), so they cannot reconnect. Low blast radius
+  (read-only feed events, no re-subscribe); tearing live streams down would need a per-session
+  broadcast filter. Revisit only if bans must take effect within seconds.
+
+## By-design notes (audit branch's signed-media model — see DECISION-media-auth.md)
+
+> These describe the **audit branch's** signed-gateway model. `main` does **not** serve media this
+> way; for `main`'s actual media-access posture and the tradeoff, see
+> [DECISION-media-auth.md](DECISION-media-auth.md).
+
+- **Banned user retains ≤24h media access via already-held signed URLs.** The audit's media gateway
+  authorizes by *signature + uploader visibility*, not requester identity, and signed URLs are
+  time-boxed (24h, bucketed). A banned/revoked-session user cannot mint new URLs (every API call
+  401/403s) but can replay URLs they already hold until expiry — only for content they already saw.
+  Accepted there: tightening would require per-request identity on every `<img>` load, which the
+  `<img>`-can't-send-a-Bearer constraint precludes.
+- **Two DB queries per *cold* media fetch** (`upload` row + uploader row). Mitigated by browser
+  caching and the stable bucketed URL. Could be one JOIN if it ever shows up in profiling.

docs/USER_JOURNEYS.md

@@ -0,0 +1,327 @@
+# EventSnap — User Journeys
+
+This document walks through every supported user scenario step-by-step. For a quick "who
+can do what" overview, see [FEATURES.md](FEATURES.md). For manual QA, see
+[TEST_GUIDE.md](../TEST_GUIDE.md).
+
+---
+
+## 1. First-time guest (the happy path)
+
+1. Guest scans the QR code / opens the event link.
+2. Lands on the **join page** (`/join`), sees the event name. A small
+   *"Ich habe bereits einen Account"* link is visible below the form for returning users
+   — it routes to `/recover`.
+3. Types display name → taps **Beitreten**.
+4. Server creates the account, generates a 4-digit PIN, stores `bcrypt(PIN)`, signs a
+   30-day JWT.
+5. A **PIN modal** appears: large monospace digits, a **Kopieren** button, a warning that
+   this PIN is the only way to sign in on another device. PIN is also written to
+   `localStorage`.
+6. Guest taps **Weiter zur Galerie** → lands in the feed (`/feed`).
+7. The **first-visit onboarding overlay** appears: dismissible steps (welcome, upload,
+   hashtags, PIN, and a brief pointer to the **Datenschutzhinweis** in My Account).
+   `localStorage('eventsnap_guide_seen') = 'true'` after dismiss.
+8. Guest sees the bottom nav: **🏠 Feed · [📷+ FAB] · 👤 Account**.
+
+## 2. Returning guest, same device
+
+1. App finds a valid JWT in `localStorage`.
+2. Redirected straight to `/feed`, no input required.
+
+## 3. Returning guest, new device or cleared storage
+
+1. Guest opens the event link on the new device → join page.
+2. Types the **same name** they used before.
+3. Server detects the existing account → the join page transforms into a recovery prompt:
+   *"„Name" ist bereits vergeben"* with a **PIN input** and an **Anmelden** button, plus
+   an **Anderen Namen wählen** escape hatch.
+4. Guest types their PIN → `bcrypt.verify` succeeds → new JWT issued for the existing
+   `user_id`. PIN is written to `localStorage` on this device too.
+5. Wrong PIN: up to 3 attempts. After the third, the account is locked for 15 minutes
+   (`pin_locked_until` is set; further attempts return HTTP 429 with a localized message).
+
+## 4. PIN forgotten — Host or Admin resets it (planned)
+
+The PIN is visible in **My Account** as long as `localStorage` is intact on at least one
+of the user's devices. If lost everywhere, the user asks a Host (or Admin) for a reset.
+
+1. Guest approaches the Host: *"I can't sign in on my new phone."*
+2. Host opens the **Host Dashboard → Nutzerverwaltung** and finds the user.
+3. Host taps **PIN zurücksetzen** on that row.
+4. A confirmation prompt explains what happens; on confirm the server generates a fresh
+   4-digit PIN, replaces `recovery_pin_hash` with the new bcrypt, clears any active
+   `pin_locked_until`, and returns the new plaintext PIN in the response.
+5. A **modal shows the new PIN ONCE** — large, with a copy button. The Host shows the
+   screen to the guest or sends it via another channel (SMS, slip of paper, …). Closing
+   the modal forgets the plaintext on the operator's device too.
+6. Guest goes to `/recover` (or taps "Ich habe bereits einen Account" on `/join`), enters
+   their name + the new PIN, signs in, and the PIN is persisted to `localStorage` on
+   their device — exactly like a fresh join.
+
+**Permission rules:**
+- Host can reset PINs for **guests** only.
+- Admin can reset PINs for **hosts and guests** (not other admins; admins use the
+  password login).
+- Anyone whose PIN was reset retains all their uploads, comments, and likes — only the
+  PIN changes.
+
+**If no Host or Admin is reachable**, the guest can still re-join under a new name (a
+clean account; their previous uploads remain attributed to the abandoned account, which
+the Host can clean up later).
+
+## 5. Posting a photo / video
+
+1. Guest taps the central **📷+ FAB** in the bottom nav.
+2. A **bottom sheet** slides up offering **Kamera** (in-app capture) or **Galerie** (file
+   picker, multi-select).
+3a. **Camera path** — [CameraCapture](../frontend/src/lib/components/CameraCapture.svelte)
+    opens the back camera (`facingMode: 'environment'`), with toggle for front camera,
+    photo button, and a video-record button using `MediaRecorder`.
+3b. **Gallery path** — native picker, multiple selection.
+4. **Preview screen** (`/upload`) shows staged files as horizontal thumbnails. The user can:
+   - Remove individual files.
+   - Type a caption with `#hashtags`.
+   - Tap quick-tag chips (derived from the caption) to copy a hashtag into the caption.
+5. Taps **Hochladen** → returns immediately to the feed (optimistic UX). The slim progress
+   bar above the bottom nav and the red badge on the FAB indicate active uploads.
+6. The client uploads files **one at a time** (XHR with progress) from an IndexedDB queue.
+7. Each upload triggers a server-side compression job; once the preview is ready the feed
+   updates via `upload-processed` SSE — placeholders swap for actual previews.
+
+## 6. Posting under rate limits
+
+1. Hit the per-hour upload limit (default 10 / hour, configurable).
+2. Server returns **HTTP 429** with a `Retry-After` header on the next upload attempt.
+3. Client parks pending items in **Wartend** state and shows an amber banner:
+   *"Upload-Limit erreicht. Wird in Xs automatisch fortgesetzt."*
+4. Countdown ticks down. When it reaches 0, the queue resumes automatically.
+
+## 7. Liking and commenting
+
+1. Tap the heart icon on a card or in the lightbox → like is recorded; count increments
+   optimistically; server returns the canonical count via `like-update` SSE.
+2. Tap the comment icon → opens the lightbox with the comments list.
+3. Type a comment → `POST /api/v1/upload/{id}/comment`. Hashtags inside the comment are
+   parsed and attached.
+4. The user can delete their own comments (trash icon next to them).
+
+## 8. Filtering the gallery
+
+1. Toggle to **grid view** (icon top-right of the feed header).
+2. A search bar appears below the header (auto-focused).
+3. Type a name or `#hashtag` — autocomplete suggestions are derived **in memory** from the
+   loaded uploads.
+4. Tap a suggestion → it becomes an **active filter chip** and the search bar clears.
+5. Filter logic:
+   - Multiple hashtag chips: OR
+   - Multiple uploader chips: OR
+   - One uploader + one hashtag: AND
+6. Open a post → swipe in the lightbox navigates the **filtered set**, not the full feed.
+
+## 9. Hosting the event — moderation
+
+1. Host opens **My Account** → taps **⭐ Host-Dashboard**.
+2. **Stats section** — guest count, upload count, lock status, release status.
+3. **Event settings** — toggle to lock new uploads (likes / comments / browsing stay open;
+   broadcasts `event-closed` SSE so all clients show a "uploads are locked" banner).
+4. **Galerie freigeben** — releases the export. Enqueues two export jobs (ZIP + HTML
+   viewer). Progress is visible in the Admin dashboard's Export tab; SSE
+   `export-progress` keeps it live; `export-available` notifies all guests when ready.
+5. **Nutzerverwaltung** — search users; per-user controls:
+   - **Sperren** opens a confirmation modal. Banning **always hides** the user's existing
+     uploads (a banned user's content is "gone" everywhere) — there is no opt-out. Submitting
+     calls `POST /host/users/{id}/ban` (no body).
+   - **Entsperren** lifts the ban. Same authority boundary as ban (below): a plain Host may
+     only unban Guests; only an Admin may unban a Host.
+   - **Host** promotes a guest to host (Hosts and Admins may do this).
+   - **Degradieren** — demote a Host back to guest. **Only an Admin may change a Host's
+     role.** A plain Host may *not* demote a peer Host: doing so would let them then ban or
+     PIN-reset (→ `/recover` account-takeover) that ex-peer, since those guards key off the
+     target's *current* role. So the backend rejects it (403) and the button is hidden for
+     non-admin Hosts. Nobody may change their own role (self-lockout / self-escalation guard),
+     and Admins are un-demotable and un-bannable by anyone. (This tightens an earlier
+     "Hosts may demote other Hosts" design — see the F1 security fix.)
+   - **Sperren / PIN zurücksetzen** — a plain Host may act on Guests only; an Admin may act on
+     Guests + Hosts; nobody may act on an Admin. The buttons are hidden where they'd 403.
+     PIN reset generates a new PIN, shows it once in a modal, and revokes the target's
+     existing sessions (forcing re-auth with the new PIN). See journey §4.
+6. **Deleting content** — Host can delete any upload or comment via the moderation routes
+   (`DELETE /host/upload/{id}`, `DELETE /host/comment/{id}`). On mobile this is also
+   reachable by long-pressing the content (planned, see §15).
+
+## 10. Banned-guest experience
+
+1. The ban takes effect immediately on the banned user's existing session — the auth
+   layer re-reads the live `is_banned` flag on every request rather than trusting the
+   JWT, so there's no 30-day token-lifetime window.
+2. Their next authenticated *write* request returns HTTP 403 with a clear message
+   ("Du bist gesperrt."). Ban enforcement lives on the write handlers and the
+   host/admin extractors, not the base auth extractor.
+3. They can still browse the read-only feed (and download the export once it's released).
+   Their sessions are **not** revoked — this is a read-only ban, not a logout. Their own
+   live SSE stream is dropped by the `is_banned` revalidation (no live pushes), but plain
+   feed reads still work.
+4. They cannot upload, like, or comment; a banned host also loses all host/admin
+   actions (including unbanning themselves).
+5. Banning **always hides**: the user's existing uploads are filtered out of the feed for
+   everyone (`v_feed`, `find_visible_media`, and the export query all enforce
+   `is_banned = FALSE`), and a live `user-hidden` SSE event evicts their cards from every
+   open feed + the diashow without a reload. A client that was offline/disconnected during
+   the live event doesn't miss the eviction: `uploads_hidden_at` is stamped at ban time
+   (migration 013) and the reconnect delta (`GET /feed/delta`) returns the banned users in
+   `hidden_user_ids`, so the feed and diashow replay the eviction on the next reconnect —
+   the projector-missed-the-live-push case is exactly why this exists.
+
+## 11. Admin — instance configuration
+
+1. Admin opens `/admin/login`, types the admin password (compared against
+   `ADMIN_PASSWORD_HASH`). Receives a separate 1-day admin JWT (in `sessionStorage`).
+2. Admin dashboard has four inner tabs:
+   - **Stats**: live counts and disk-usage widget (via `sysinfo`).
+   - **Config**: per-file limits (image MB / video MB), rate limits (upload / feed /
+     export), quota tolerance, estimated guest count, compression-worker concurrency,
+     plus the **Datenschutzhinweis** free-text editor and **on/off toggles** for the rate
+     limiters and quotas (planned — see §16). Whitelist on the server side rejects
+     unknown keys. Values are read from the `config` table on each request — no restart
+     needed.
+   - **Export**: list of past export jobs with status badges (pending / running / done /
+     failed) and progress bars; refresh button re-polls.
+   - **Nutzer**: same user list as Host, with the additional Demote action and (planned)
+     PIN-reset on host rows.
+
+## 12. Releasing the export and downloading
+
+1. Host (or Admin) taps **Galerie freigeben** in the dashboard.
+2. Server sets `event.export_released_at`, locks uploads, **bumps `event.export_epoch`**, and
+   enqueues two background jobs — all in ONE transaction (workers spawn after it commits, so a
+   client disconnecting mid-request can never leave the event released with no export to build).
+
+   The **epoch** is the whole generation model (migration 014). It is bumped in the same UPDATE as
+   any change to `export_released_at` — release and reopen are its only writers — and each
+   `export_job` row carries a copy of the epoch it was enqueued for. An export is downloadable
+   **iff** `released AND job.epoch = event.export_epoch AND job.status = 'done'`. Readiness is
+   therefore *derived*, never stored: it cannot drift, and a worker whose epoch has been retired
+   (by a reopen, a re-release, or a takedown) is inert — anything it writes is simply invisible.
+   A reopen retires the current keepsake instantly, which is why a reopened event serves no export
+   until the host releases again.
+3. ZIP job: streams `Gallery.zip` (`Photos/` + `Videos/`, full-quality originals) directly
+   to disk via `async-zip`. Progress updates via `export-progress` SSE.
+4. HTML-viewer job: copies the pre-built viewer assets from
+   [backend/static/export-viewer/](../backend/static/export-viewer/) (embedded via
+   `include_dir!`), generates `data.json` from the database, processes `_thumb`/`_full`
+   variants for each upload, and assembles `Memories.zip`.
+5. Both jobs complete → server broadcasts `export-available` SSE.
+   **Takedowns:** if a host deletes an upload (or a comment) while the gallery is released, the
+   epoch is bumped and the keepsake is REGENERATED without it. Otherwise a photo removed on request
+   would live on forever in the already-generated archive — the one place it most needs to be gone.
+   The download 404s for the few seconds it takes to rebuild, which is the correct answer: serving
+   the old archive would serve the deleted photo.
+6. Any user opens `/export`:
+   - Before release: friendly "Export not yet available" banner.
+   - During generation: progress bars per artifact.
+   - After completion: two cards (**ZIP-Archiv** and **HTML-Viewer**) with download
+     buttons. Tapping the HTML download first shows an in-app guide modal explaining:
+     "Entpacke die ZIP, öffne `index.html`". Tapping **Herunterladen** triggers the
+     browser download.
+7. Downloads are rate-limited per IP (default 3 / day).
+
+## 13. Diashow (planned)
+
+See [CONCEPT_DIASHOW.md](CONCEPT_DIASHOW.md). Summary of the planned flow:
+
+1. User taps a **Diashow / Präsentation** action (feed header on tablet/desktop, Account
+   on mobile).
+2. Navigates to `/diashow` — fullscreen, bottom nav hidden, screen wake-lock acquired.
+3. Initial pool fetched from `GET /api/v1/feed`. Slides crossfade every ~6 s.
+4. New uploads (`upload-processed` SSE) push to a live queue; the next slide transition
+   pops from the live queue first, otherwise from a shuffled queue.
+5. `upload-deleted` removes that ID from both queues; if it's the current slide, advance
+   immediately.
+6. Tap or Escape reveals an overlay (pause, dwell selector, exit).
+
+## 14. Picking a data mode (planned)
+
+1. Guest opens **My Account** → scrolls to **Datennutzung**.
+2. Two options: **Datensparer (empfohlen)** and **Original**. Saver is the default.
+3. Selecting **Original** shows a one-time warning bottom-sheet:
+   *"Original-Dateien werden geladen — das kann deine mobile Datennutzung deutlich
+   erhöhen. Trotzdem aktivieren?"* with **Abbrechen** / **Aktivieren** buttons.
+4. Choice persists in `localStorage` (per-device). The feed, lightbox, and diashow all
+   read this flag and load originals instead of compressed previews when Original is on.
+5. The viewer (offline HTML export) is unaffected — it already ships with its own pre-
+   bundled `_thumb` / `_full` variants.
+
+## 15. Leaving an event
+
+1. User opens **My Account** → taps **🚪 Event verlassen**.
+2. Bottom-sheet confirmation: "Event verlassen?" with **Abmelden** and **Bleiben**.
+3. Confirming calls `DELETE /api/v1/session` (invalidates the session row), clears the JWT
+   and PIN from `localStorage`, and redirects to the join page.
+
+## 16. Reading the Datenschutzhinweis (planned)
+
+1. User opens **My Account** → scrolls to **Datenschutzhinweis**.
+2. The note is rendered inside a preformatted block (`<pre>`-style: monospace, whitespace
+   and newlines preserved exactly as the Admin typed them). No HTML, no markdown — the
+   admin's plain text is shown verbatim.
+3. The first-visit onboarding overlay carries a one-line reminder of where to find this:
+   *"Datenschutzhinweis findest du in deinem Account."*
+4. Admin sets / edits the note in **Admin Dashboard → Config → Datenschutzhinweis**: a
+   tall textarea with a save button. Saved to a single `config` key.
+
+## 17. Mobile-first gestures (planned)
+
+EventSnap's UI is mobile-first; gestures replace explicit buttons where they're more
+ergonomic. Buttons are always present as fallback for desktop and accessibility.
+
+| Gesture                                   | Action                                                |
+|-------------------------------------------|-------------------------------------------------------|
+| Long-press on a post (own)                | Bottom sheet → Löschen, Original anzeigen, Teilen     |
+| Long-press on a post (other)              | Bottom sheet → Original anzeigen, Teilen, Melden (planned) |
+| Long-press on a comment (own)             | Bottom sheet → Löschen                                |
+| Long-press on a comment (other)           | Bottom sheet → Kopieren                               |
+| Long-press on a user row (Host)           | Bottom sheet → Sperren, Promote/Demote, PIN zurücksetzen |
+| Swipe left/right in the lightbox          | Navigate the filtered set                             |
+| Swipe down on any bottom sheet            | Dismiss                                               |
+| Pull-to-refresh on the feed               | Force a delta-fetch                                   |
+| Double-tap on a post                      | Like (heart-burst animation)                          |
+
+On desktop the same actions surface as kebab/⋯ menus, click-able icons in card corners,
+and keyboard shortcuts in the lightbox (← → for navigate, Esc to close).
+
+Inspiration: Instagram (double-tap heart, swipe stories), WhatsApp (long-press for
+context), Telegram (swipe-to-reply on messages — could inform comment threads if those
+land).
+
+## 18. Admin toggles a rate limit or quota off (planned)
+
+1. Admin opens **Admin Dashboard → Config**.
+2. **Rate-Limits** section: a master switch and per-endpoint switches (upload / feed /
+   export / join).
+3. Admin flips, e.g., **Upload-Limit aktiv** off. The numeric input for "uploads per hour"
+   stays visible but greyed out (still editable for when the toggle goes back on).
+4. **Speichern** persists to the `config` table. The next upload request bypasses the
+   limiter entirely.
+5. **Quoten** section mirrors the pattern: master toggle plus per-area toggles (storage
+   bytes / upload count).
+6. When the storage-quota toggle is off, the **"X von Y MB genutzt"** widget in the
+   guest's My Account and upload screen hides itself (no quota → no number to show).
+
+Suggested defaults at deploy time: all toggles **on**, sensible numeric limits.
+Toggling off is the explicit escape hatch for testing or trusted internal events.
+
+---
+
+## Edge cases worth knowing
+
+| Case                                                  | Behaviour                                                                       |
+|-------------------------------------------------------|---------------------------------------------------------------------------------|
+| Browser tab backgrounded for > 5 min                  | SSE closes on `visibilitychange: hidden`; reopens on visible                    |
+| Upload finishes while user is on `/account`           | Feed updates anyway — the queue + SSE are global stores                         |
+| Event "closed" while files are still in the queue     | Server rejects with a friendly error; client surfaces it in the queue UI        |
+| Network drops mid-upload                              | Queue retries the file; retry button available on permanent failure             |
+| New device but the PIN was lost                       | Either re-join under a new name, or Host manually re-links (no self-service)    |
+| Two guests pick the same name                         | Second one is offered the PIN-recovery form (case-insensitive UNIQUE, mig. 007) |
+| Compression fails for a file                          | Server emits `upload-error` SSE; the upload is still listed but marked degraded |
+| User deletes their own post (once UI is shipped)      | Soft delete (`deleted_at`); SSE `upload-deleted`; vanishes from feed everywhere |

e2e/.gitignore

@@ -0,0 +1,6 @@
+node_modules/
+playwright-report/
+test-results/
+.cache/
+.env.test
+*.log

e2e/.prettierignore

@@ -0,0 +1,4 @@
+node_modules/
+playwright-report/
+test-results/
+package-lock.json

e2e/.prettierrc

@@ -0,0 +1,7 @@
+{
+  "useTabs": false,
+  "tabWidth": 2,
+  "singleQuote": true,
+  "trailingComma": "es5",
+  "printWidth": 100
+}

e2e/Caddyfile.test

@@ -0,0 +1,24 @@
+# Caddyfile used only by the E2E test stack. Listens on the in-container :3101
+# (mapped to host :3101) and proxies API + media to the backend, everything else
+# to the SvelteKit frontend container — same layout as production but stripped
+# of HTTPS/Let's Encrypt.
+
+:3101 {
+    # Mirror prod: exclude the SSE stream from compression so buffering doesn't
+    # delay real-time events (and so the test stack exercises the real behavior).
+    @compressible not path /api/v1/stream
+    encode @compressible zstd gzip
+
+    # Mirror prod's security headers (minus HSTS, which is HTTPS-only).
+    header {
+        X-Content-Type-Options "nosniff"
+        X-Frame-Options "DENY"
+        Referrer-Policy "strict-origin-when-cross-origin"
+    }
+
+    reverse_proxy /api/* app:3000
+    reverse_proxy /media/* app:3000
+    reverse_proxy /health app:3000
+
+    reverse_proxy frontend:3001
+}

e2e/README.md

@@ -0,0 +1,290 @@
+# EventSnap E2E Suite
+
+Playwright-driven end-to-end tests for the EventSnap stack. The suite spins
+up an isolated docker-compose stack on ports `:3101` (Caddy → frontend +
+backend) and `:55432` (Postgres), and exercises the SvelteKit frontend
+against a real Rust backend with rate limits and quotas disabled.
+
+**Phases 1, 2, and 3-mobile-gestures are landed**:
+
+- **Phase 1** — happy-path coverage of every documented user journey, plus a
+  smoke matrix across nine browser/UA profiles to catch engine-level
+  divergences.
+- **Phase 2** — adversarial inputs (XSS, SQL-injection, JWT forgery, MIME
+  spoofing, oversize, brute-force) and browser chaos (storage purge,
+  offline/slow-3G, multi-tab, clock skew, no-JS, quota exhaustion).
+- **Phase 3 (gestures only)** — touch-target audit, safe-area structural
+  check, long-press → ContextSheet, double-tap → like, viewport reflow,
+  plus `test.fixme` stubs for planned gestures (lightbox swipe, swipe-down
+  dismiss, pull-to-refresh).
+
+Phase 3 real-device compat (Android emulator + Samsung Internet via
+`connectOverCDP`, BrowserStack), visual regression, and a11y audits are
+sketched in the **Roadmap** at the bottom.
+
+## Quickstart
+
+```bash
+cd e2e
+npm install
+npm run install:browsers      # one-time: ~500 MB across chromium/firefox/webkit
+
+# 1. Boot the test stack (rebuilds backend + frontend Docker images)
+npm run stack:up
+
+# 2. Wait ~20s for migrations + warmup, then run tests
+npm run test:e2e              # full Phase 1 suite on chromium-desktop
+npm run test:e2e:smoke        # cross-UA smoke matrix (~9 projects × 1 test)
+npm run test:e2e:ui           # interactive Playwright UI mode
+
+# 3. After: tear the stack down (deletes volumes)
+npm run stack:down
+```
+
+The CI workflow at `.github/workflows/e2e.yml` runs both jobs on every PR.
+
+## What's tested
+
+Every spec covers a journey from [`docs/USER_JOURNEYS.md`](../docs/USER_JOURNEYS.md)
+or a security/chaos scenario. One folder per area:
+
+| Folder                    | Phase | Journeys / Topic                                                                                | Tests     | Notes                                                         |
+| ------------------------- | ----- | ----------------------------------------------------------------------------------------------- | --------- | ------------------------------------------------------------- |
+| `specs/01-auth/`          | 1     | §1, §2, §3, §11, §15                                                                            | 13        | Join, recover, PIN lockout, admin login, leave event.         |
+| `specs/02-upload/`        | 1     | §5, §6, §18                                                                                     | 5         | Gallery picker, multi-file, rate-limit, admin toggle.         |
+| `specs/03-feed/`          | 1     | §7, §8, §17                                                                                     | 5         | Like/comment SSE, filter chips, SSE reconnect.                |
+| `specs/04-host/`          | 1     | §9                                                                                              | 5         | Event lock, ban/unban, role change.                           |
+| `specs/05-admin/`         | 1     | §11, §16                                                                                        | 11        | Config validation, foundational auth guards, stats.           |
+| `specs/06-export/`        | 1     | §12                                                                                             | 3         | Status, release, download stub.                               |
+| `specs/__smoke/`          | 1     | (matrix)                                                                                        | 1 × 9 UAs | `@smoke`-tagged happy-path on every UA project.               |
+| `specs/07-adversarial/`   | **2** | Input attacks, file upload boundaries, JWT forgery, brute-force, deep authorization, small DDoS | ~40       | See breakdown below.                                          |
+| `specs/08-browser-chaos/` | **2** | Storage purge, IndexedDB, offline/slow-3G, multi-tab, no-JS, clock skew, quota                  | ~20       | See breakdown below.                                          |
+| `specs/09-mobile/`        | **3** | Touch-target audit, safe-area, long-press, double-tap, viewport reflow, fixme stubs             | 23        | Runs only on `chromium-mobile` (Pixel 7 viewport). See below. |
+
+### Phase 2 — adversarial (`specs/07-adversarial/`)
+
+- **`xss-injection.spec.ts`** — 13 tests. Six XSS payloads × display-name path
+  - four SQLi patterns + length/encoding edge cases (NUL byte, RTL override,
+    caption overflow). Asserts `window.__xssFired` never gets set and no
+    `dialog` event fires.
+- **`ui-rendering.spec.ts`** — 2 tests. Belt-and-braces: even when a script-
+  payload sits in localStorage as the user's display name, rendering through
+  `/account` keeps it as text.
+- **`file-upload-attacks.spec.ts`** — 9 tests. ELF body claimed as JPEG,
+  oversize image vs `max_image_size_mb`, zero-byte, missing file field,
+  path-traversal filename, NUL filename, `application/*` declared category
+  bypass, SVG-with-script.
+- **`auth-tampering.spec.ts`** — 8 tests. `alg:none` forging admin role,
+  signature tamper, payload tamper with original signature, logged-out
+  session reuse, header without `Bearer `, missing Authorization,
+  PIN brute-force lockout, admin password brute-force (documented finding —
+  no lockout today, bcrypt cost is the only defense).
+- **`authorization-deep.spec.ts`** — 6 tests. Cross-user comment delete,
+  banned user across like/comment/feed-read, host→admin escalation attempts.
+- **`ddos.spec.ts`** — 4 small-scale abuse tests. 20 parallel /join, 10 MB
+  comment body, 10 concurrent SSE streams, malformed JSON.
+
+### Phase 2 — browser chaos (`specs/08-browser-chaos/`)
+
+- **`storage-purge.spec.ts`** — 5 tests. `localStorage.clear()` mid-session,
+  cookies cleared (JWT in localStorage still works), sessionStorage cleared,
+  admin force-relogin, PIN intentionally survives clearAuth.
+- **`indexeddb.spec.ts`** — 2 tests. Drop all IDB databases mid-session;
+  stub IDB to undefined before navigation.
+- **`offline-network.spec.ts`** — 4 tests. `setOffline(true)` → reconnect,
+  slow-3G via `page.route` delay, intermittent 503s, 429 from server (no
+  infinite retry storm).
+- **`multi-tab.spec.ts`** — 3 tests. Same user two tabs, two users two
+  contexts (storage isolated), logout in tab A doesn't sync to tab B
+  (documented gap).
+- **`environment.spec.ts`** — 5 tests. JS disabled, localStorage quota
+  exhausted, hostile CSS hiding nav, clock skew ±1h / -2d.
+
+Pending tests covering features that need a Node-side multipart upload helper
+are marked `test.fixme` and will activate when that helper lands.
+
+## Browser & UA matrix
+
+| Project               | Engine   | UA / Device                         | Why                                           |
+| --------------------- | -------- | ----------------------------------- | --------------------------------------------- |
+| `chromium-desktop`    | Chromium | Desktop Chrome                      | Baseline. Full suite runs here.               |
+| `chromium-pixel7`     | Chromium | Pixel 7 device descriptor           | Chrome Android.                               |
+| `chromium-galaxy-s22` | Chromium | Galaxy viewport + Samsung phone UA  | Chrome on Samsung hardware.                   |
+| `samsung-internet`    | Chromium | Galaxy viewport + SamsungBrowser UA | **Tier-A Samsung Internet baseline.**         |
+| `edge-android`        | Chromium | Pixel viewport + EdgA UA            | Edge Mobile (Blink-based).                    |
+| `chrome-ios`          | Chromium | iPhone viewport + CriOS UA          | Chrome iOS (actually WebKit, but UA differs). |
+| `webkit-iphone`       | WebKit   | iPhone 14 Pro                       | Real iOS Safari engine.                       |
+| `firefox-android`     | Firefox  | Pixel viewport + Firefox Android UA | Gecko engine.                                 |
+| `firefox-desktop`     | Firefox  | Desktop Firefox                     | FF-specific quirks.                           |
+
+Only the `@smoke` happy-path runs across all projects (controlled by
+`grep` in `playwright.config.ts`). The full Phase 1 suite is
+`chromium-desktop`-only by default to keep CI under 15 min.
+
+### Samsung Internet — three escalation tiers
+
+Samsung Internet ships on every Galaxy phone (~5% of mobile traffic in DE).
+It's **Blink-based**, so Tier-A catches ~90% of regressions. Real Samsung
+divergences (Smart Switch save-data mode, dark-mode injection, custom
+autoplay, in-browser ad blocking) are only reproducible at Tier B+:
+
+- **Tier A** _(this repo, free, in CI)_: Playwright Chromium with the
+  Samsung Internet user-agent + Galaxy viewport. See the `samsung-internet`
+  project in `playwright.config.ts`.
+- **Tier B** _(free, manual, future)_: Android Studio emulator on Linux →
+  install Samsung Internet APK → enable `--remote-debugging-port=9222` →
+  `chromium.connectOverCDP('http://localhost:9222')`. Setup docs live in
+  `docs/samsung-emulator.md` (to be written).
+- **Tier C** _(paid, optional)_: BrowserStack or LambdaTest cloud devices.
+  Real Galaxy S22/S23 hardware via Playwright's cloud integration.
+
+## Test isolation
+
+Every test runs against a **freshly truncated database**:
+
+1. `global-setup.ts` waits for `/health`, logs in admin, and disables every
+   rate-limit and quota toggle via `PATCH /admin/config`.
+2. The auto-fixture `truncate` in `fixtures/test.ts` calls
+   `POST /api/v1/admin/__truncate` before every test.
+3. The truncate endpoint is only registered when the backend is started
+   with `EVENTSNAP_TEST_MODE=1` (see `backend/src/main.rs` and
+   `backend/src/handlers/test_admin.rs`). Production builds return 404.
+
+Single-worker by design (`workers: 1` in the config). Per-worker isolated
+DBs are a Phase-2+ change.
+
+## Architecture
+
+```
+e2e/
+├── docker-compose.test.yml   # Isolated test stack: db :55432, caddy :3101
+├── Caddyfile.test            # Proxies /api/* /media/* /health to backend
+├── playwright.config.ts      # UA matrix + smoke grep
+├── global-setup.ts           # admin login, rate-limit disable
+├── global-teardown.ts        # (no-op; use `npm run stack:down`)
+├── fixtures/
+│   ├── api-client.ts         # Typed wrapper over /api/v1/*
+│   ├── db.ts                 # Direct Postgres escape hatch (locked-PIN, etc.)
+│   ├── test.ts               # Central test.extend (guest, host, signIn fixtures)
+│   └── media/                # sample.jpg, sample.mp4, not-an-image.jpg
+├── helpers/
+│   ├── sse-listener.ts       # Async SSE iterator with waitForEvent()
+│   ├── storage-helpers.ts    # localStorage/sessionStorage helpers
+│   └── fake-media.ts         # Camera permissions (Chromium only)
+├── page-objects/
+│   ├── join-page.ts          # /join
+│   ├── recover-page.ts       # /recover
+│   ├── admin-login-page.ts   # /admin/login
+│   ├── feed-page.ts          # /feed + bottom nav
+│   ├── upload-sheet.ts       # UploadSheet.svelte + /upload
+│   ├── lightbox.ts           # LightboxModal.svelte
+│   ├── account-page.ts       # /account
+│   ├── host-dashboard.ts     # /host
+│   ├── admin-dashboard.ts    # /admin
+│   └── export-page.ts        # /export
+└── specs/
+    ├── __smoke/              # @smoke cross-UA matrix (1 spec)
+    ├── 01-auth/
+    ├── 02-upload/
+    ├── 03-feed/
+    ├── 04-host/
+    ├── 05-admin/
+    └── 06-export/
+```
+
+## Debugging a failure
+
+- `npm run test:e2e:ui` — interactive UI with time-travel and selector probe.
+- `npm run test:e2e:headed` — watch the browser run live.
+- `npm run test:e2e:debug` — Playwright inspector with breakpoints.
+- `npm run stack:logs` — tail backend + Postgres logs during a failure.
+- `playwright-report/index.html` — opens the HTML report (auto-generated on every run).
+- Trace files (`test-results/**/trace.zip`) drag-and-drop into `https://trace.playwright.dev`.
+
+## Conventions
+
+- **One assertion per `expect`**. Bundling multiple expects in one statement
+  loses the line-level failure context.
+- **Wait on data, not time**. Use `expect.poll` for DB checks; never `waitForTimeout` in production specs.
+- **`@smoke` tag** on each suite's happiest path so the matrix run stays under 2 min.
+- **`test.fixme`** for features that need infrastructure not yet built (Node-side multipart upload helper, real video fixtures, etc.). Fixme tests don't fail the suite but show up in the report.
+- **Page objects own selectors**. Specs never use raw locators.
+- **German text in assertions** is fine — it's not going to change frequently. When it does, the page object is the only file to update.
+
+## Roadmap
+
+### Phase 2 — Adversarial & browser chaos ✅ landed
+
+See the **What's tested** table above and the per-file breakdown.
+Known findings surfaced (documented in tests, not silent failures):
+
+1. `/admin/login` has no rate-limit or lockout — bcrypt cost is the only defense.
+2. `localStorage` 'storage' event is not listened for, so logout in tab A
+   doesn't synchronously sign out tab B (the next 401 from any API call
+   clears it).
+3. SVG uploads currently pass the magic-byte check (depends on `infer`'s
+   detection coverage) — consider adding `X-Content-Type-Options: nosniff`
+   - CSP on `/media/*` if SVGs are ever expected as user content.
+
+### Phase 3 — Mobile gestures (`specs/09-mobile/`) ✅ landed
+
+Runs only on the `chromium-mobile` project (Pixel 7 device descriptor with
+`hasTouch` and `isMobile`). The `chromium-desktop` project explicitly
+ignores this folder via `testIgnore` in [playwright.config.ts](playwright.config.ts).
+
+- **`touch-targets.spec.ts`** — 4 tests. Audits ≥ 44×44 px on bottom nav,
+  FAB, join submit, admin-login submit, PIN-modal buttons. Uses
+  `expect.soft` so a single failure surfaces the actual bounding-box
+  dimensions instead of stopping the suite.
+- **`safe-area.spec.ts`** — 4 tests. Asserts `env(safe-area-inset-bottom)`
+  is present in the inline style of every bottom-anchored UI element
+  (bottom nav, UploadSheet, ContextSheet), and that the nav stays flush
+  with the viewport bottom on a no-notch emulated device.
+- **`gestures-longpress.spec.ts`** — 3 tests. A 600 ms hold on a
+  FeedListCard opens the ContextSheet; a 200 ms tap does not; the
+  click-suppression logic prevents the lightbox from also opening at
+  pointer-up. Driven via `page.mouse.down/up` because the `longpress`
+  action listens for pointer events (mouse/touch/pen unified).
+- **`gestures-doubletap.spec.ts`** — 2 tests. Double-tap on a feed card
+  image button records a like; double-tap inside the lightbox triggers
+  the heart-burst animation and records a like. Assertions read the like
+  count back via `/api/v1/feed` so they don't couple to specific badge
+  markup.
+- **`viewport-reflow.spec.ts`** — 5 tests. Portrait, landscape, narrow
+  (320×568), phablet (480×1024) — each asserts the bottom nav is
+  visible, the FAB stays roughly centered, and there's no horizontal
+  overflow on `<html>`. Plus a rotation test that confirms auth survives
+  a viewport resize.
+- **`planned-gestures.spec.ts`** — 5 **`test.fixme`** stubs documenting
+  the contracts for gestures from journey §17 that aren't shipped yet
+  (lightbox swipe L/R, swipe-down to dismiss UploadSheet,
+  pull-to-refresh, long-press on a comment). Flip `test.fixme` to `test`
+  when wiring each gesture.
+
+#### Driving gestures: the `helpers/touch.ts` module
+
+- `longPress(page, locator, durationMs)` — holds the pointer down for
+  the duration. Default 600 ms beats the action's 500 ms threshold.
+- `doubleTap(page, locator)` — two `mouse.down/up` pairs within the
+  `doubletap` action's 300 ms window.
+- `swipe(page, from, to, steps)` — gradual mouse-driven move (used by
+  the fixme stubs once swipe gestures land).
+- `inlineStyle(locator)` / `computedStyle(locator, prop)` — read raw
+  `style` attributes (where `env(...)` strings live) and computed
+  values.
+
+### Phase 3 — Real-device compat & visual / a11y (not landed)
+
+- Long-press own/other post, swipe lightbox L/R, swipe-down dismiss, pull-to-refresh, double-tap like.
+- Safe-area inset visual diff on iPhone notch.
+- Touch-target ≥ 44 px audit.
+- Tier B Samsung Internet via `connectOverCDP` on Android Studio emulator.
+- Tier C BrowserStack integration (paid, optional).
+- `@axe-core/playwright` accessibility audits.
+- Visual regression with screenshot diffs.
+
+### Out of scope (handed to other tools)
+
+- Load testing → k6 / Vegeta.
+- API contract testing → backend `cargo test` integration tests.
+- Static asset auditing → Lighthouse CI.

e2e/docker-compose.test.yml

@@ -0,0 +1,77 @@
+# Isolated EventSnap test stack. Mirrors production layout (Caddy → frontend +
+# backend) but on its own ports, its own volumes, with rate-limits disabled and
+# `EVENTSNAP_TEST_MODE=1` so the `/admin/__truncate` reset endpoint is live.
+#
+# Bring it up once before running the suite:
+#     npm run stack:up
+# Tear it down (and wipe all volumes) after:
+#     npm run stack:down
+#
+# Port 3101 is the only externally exposed port: Caddy fronts everything.
+
+services:
+  db:
+    image: postgres:16-alpine
+    environment:
+      POSTGRES_USER: eventsnap_test
+      POSTGRES_PASSWORD: eventsnap_test
+      POSTGRES_DB: eventsnap_test
+    healthcheck:
+      test: ['CMD-SHELL', 'pg_isready -U eventsnap_test -d eventsnap_test']
+      interval: 3s
+      timeout: 3s
+      retries: 30
+    ports:
+      - '55432:5432' # exposed so the e2e harness can connect via pg for fixture setup
+
+  app:
+    build:
+      context: ../backend
+      dockerfile: Dockerfile
+    depends_on:
+      db:
+        condition: service_healthy
+    environment:
+      DATABASE_URL: postgres://eventsnap_test:eventsnap_test@db:5432/eventsnap_test
+      JWT_SECRET: 00112233445566778899aabbccddeeff00112233445566778899aabbccddeeff
+      # bcrypt hash for the literal string "admin-test-pw" (cost 4 — fast for tests).
+      # Generated once and committed. Verified with `bcrypt.compareSync`.
+      # The $ characters are doubled to escape compose interpolation.
+      ADMIN_PASSWORD_HASH: $$2b$$04$$XKJJkNX6BOi6y3S42DA5JOWwk4oxc8DHPL6.MrPfJI2vpnccZjP32
+      EVENT_SLUG: e2e-test-event
+      EVENT_NAME: E2E Test Event
+      APP_PORT: '3000'
+      MEDIA_PATH: /media
+      SESSION_EXPIRY_DAYS: '30'
+      EVENTSNAP_TEST_MODE: '1' # ENABLES /admin/__truncate — never set in prod
+      RUST_LOG: eventsnap_backend=info,tower_http=warn
+    volumes:
+      - media_data:/media
+    expose:
+      - '3000'
+
+  frontend:
+    build:
+      context: ../frontend
+      dockerfile: Dockerfile
+    depends_on:
+      - app
+    environment:
+      PORT: '3001'
+      HOST: '0.0.0.0'
+      ORIGIN: 'http://localhost:3101'
+    expose:
+      - '3001'
+
+  caddy:
+    image: caddy:2-alpine
+    depends_on:
+      - app
+      - frontend
+    volumes:
+      - ./Caddyfile.test:/etc/caddy/Caddyfile:ro
+    ports:
+      - '3101:3101'
+
+volumes:
+  media_data:

e2e/eslint.config.js

@@ -0,0 +1,47 @@
+import js from '@eslint/js';
+import ts from 'typescript-eslint';
+import prettier from 'eslint-config-prettier';
+import globals from 'globals';
+
+/**
+ * Flat-config ESLint for the Playwright TypeScript suite. Prettier owns formatting (its config is
+ * last so it disables every stylistic rule). The rules kept here catch real test bugs — unused
+ * setup, floating promises that make a test race, `any` that hides a wrong assertion shape.
+ */
+export default ts.config(
+  js.configs.recommended,
+  ...ts.configs.recommended,
+  prettier,
+  {
+    languageOptions: {
+      globals: { ...globals.node },
+    },
+  },
+  {
+    rules: {
+      '@typescript-eslint/no-unused-vars': [
+        'error',
+        { argsIgnorePattern: '^_', varsIgnorePattern: '^_', caughtErrorsIgnorePattern: '^_' },
+      ],
+      // A floating promise in a test is a real hazard: an un-awaited request or assertion can let
+      // the test end before it runs, passing vacuously. Keep it on.
+      '@typescript-eslint/no-floating-promises': 'error',
+      // Off for the suite: this is all test code, where `any` is the honest type for an untyped
+      // `res.json()` body or a `page.evaluate()` return. Threading DTO types through every
+      // assertion is churn that buys nothing — the assertion values are what's checked, not the
+      // static shape. (no-floating-promises and no-unused-vars, which catch real test bugs, stay on.)
+      '@typescript-eslint/no-explicit-any': 'off',
+      // Off: Playwright fixtures with no dependencies are declared `async ({}, use) => {}` — the
+      // empty destructure is required by the fixtures API, not an accident.
+      'no-empty-pattern': 'off',
+    },
+  },
+  {
+    languageOptions: {
+      parserOptions: { projectService: true },
+    },
+  },
+  {
+    ignores: ['node_modules/', 'playwright-report/', 'test-results/', '*.config.js'],
+  }
+);

e2e/fixtures/api-client.ts

@@ -0,0 +1,189 @@
+/**
+ * Tiny typed wrapper around the EventSnap REST API for use inside tests.
+ * Used to seed data far faster than driving the UI through every join /
+ * upload, and to set up adversarial states (banned users, locked PINs) that
+ * the UI cannot reach.
+ *
+ * Auth: pass `token` on individual calls; no global state.
+ */
+
+export const ADMIN_PASSWORD = 'admin-test-pw';
+
+export class ApiClient {
+  constructor(private baseUrl: string = process.env.E2E_FRONTEND_URL ?? 'http://localhost:3101') {}
+
+  private async request<T>(
+    method: string,
+    path: string,
+    opts: { token?: string; body?: unknown; expectedStatus?: number | number[] } = {}
+  ): Promise<{ status: number; body: T }> {
+    const headers: Record<string, string> = {};
+    if (opts.token) headers['Authorization'] = `Bearer ${opts.token}`;
+    if (opts.body !== undefined) headers['Content-Type'] = 'application/json';
+
+    const res = await fetch(`${this.baseUrl}/api/v1${path}`, {
+      method,
+      headers,
+      body: opts.body !== undefined ? JSON.stringify(opts.body) : undefined,
+    });
+
+    const expected = opts.expectedStatus ?? [200, 201, 204];
+    const allowed = Array.isArray(expected) ? expected : [expected];
+    let body: unknown = undefined;
+    if (res.status !== 204) {
+      const text = await res.text();
+      try {
+        body = text.length > 0 ? JSON.parse(text) : undefined;
+      } catch {
+        body = text;
+      }
+    }
+    if (!allowed.includes(res.status)) {
+      throw new Error(
+        `API ${method} ${path} → ${res.status} (expected ${allowed.join('/')}). Body: ${JSON.stringify(body)}`
+      );
+    }
+    return { status: res.status, body: body as T };
+  }
+
+  // ── Auth ───────────────────────────────────────────────────────────────
+  async join(
+    displayName: string
+  ): Promise<{ jwt: string; pin: string; user_id: string; is_new: boolean }> {
+    const { body } = await this.request<any>('POST', '/join', {
+      body: { display_name: displayName },
+      expectedStatus: [201],
+    });
+    return body;
+  }
+
+  async recover(
+    displayName: string,
+    pin: string,
+    opts: { expectedStatus?: number | number[] } = {}
+  ) {
+    return this.request<any>('POST', '/recover', {
+      body: { display_name: displayName, pin },
+      expectedStatus: opts.expectedStatus ?? [200],
+    });
+  }
+
+  async adminLogin(password: string = ADMIN_PASSWORD): Promise<string> {
+    const { body } = await this.request<{ jwt: string }>('POST', '/admin/login', {
+      body: { password },
+    });
+    return body.jwt;
+  }
+
+  async logout(token: string) {
+    return this.request<void>('DELETE', '/session', { token, expectedStatus: [204] });
+  }
+
+  // ── Test-mode helpers ──────────────────────────────────────────────────
+  async truncate(adminToken: string) {
+    return this.request<void>('POST', '/admin/__truncate', {
+      token: adminToken,
+      expectedStatus: [204],
+    });
+  }
+
+  // ── Config ─────────────────────────────────────────────────────────────
+  async patchConfig(adminToken: string, patch: Record<string, string>) {
+    return this.request<void>('PATCH', '/admin/config', {
+      token: adminToken,
+      body: patch,
+      expectedStatus: [204],
+    });
+  }
+
+  async getConfig(adminToken: string): Promise<Record<string, string>> {
+    const { body } = await this.request<Record<string, string>>('GET', '/admin/config', {
+      token: adminToken,
+    });
+    return body;
+  }
+
+  // ── Host moderation ────────────────────────────────────────────────────
+  async listUsers(token: string) {
+    const { body } = await this.request<any[]>('GET', '/host/users', { token });
+    return body;
+  }
+
+  async setRole(token: string, userId: string, role: 'guest' | 'host') {
+    return this.request<void>('PATCH', `/host/users/${userId}/role`, {
+      token,
+      body: { role },
+      expectedStatus: [200, 204],
+    });
+  }
+
+  // A ban ALWAYS hides the user's uploads — the backend takes no body and ignores any
+  // `hide_uploads` flag (the old opt-out was removed). No per-request options.
+  async banUser(token: string, userId: string) {
+    return this.request<void>('POST', `/host/users/${userId}/ban`, {
+      token,
+      expectedStatus: [200, 204],
+    });
+  }
+
+  async unbanUser(
+    token: string,
+    userId: string,
+    opts: { expectedStatus?: number | number[] } = {}
+  ) {
+    return this.request<void>('POST', `/host/users/${userId}/unban`, {
+      token,
+      expectedStatus: opts.expectedStatus ?? [200, 204],
+    });
+  }
+
+  /** Reset another user's PIN. Returns the plaintext PIN the host must relay once. */
+  async resetUserPin(
+    token: string,
+    userId: string,
+    opts: { expectedStatus?: number | number[] } = {}
+  ): Promise<{ status: number; body: { pin?: string } }> {
+    return this.request<{ pin?: string }>('POST', `/host/users/${userId}/pin-reset`, {
+      token,
+      expectedStatus: opts.expectedStatus ?? [200],
+    });
+  }
+
+  async listPinResetRequests(token: string): Promise<any[]> {
+    const { body } = await this.request<any[]>('GET', '/host/pin-reset-requests', { token });
+    return body;
+  }
+
+  async closeEvent(token: string) {
+    return this.request<void>('POST', '/host/event/close', { token, expectedStatus: [200, 204] });
+  }
+
+  async openEvent(token: string) {
+    return this.request<void>('POST', '/host/event/open', { token, expectedStatus: [200, 204] });
+  }
+
+  // ── Feed ───────────────────────────────────────────────────────────────
+  async getFeed(token: string) {
+    const { body } = await this.request<any>('GET', '/feed', { token });
+    return body;
+  }
+
+  async getStats(adminToken: string) {
+    const { body } = await this.request<any>('GET', '/admin/stats', { token: adminToken });
+    return body;
+  }
+
+  // ── Health ─────────────────────────────────────────────────────────────
+  async waitForHealth(retries = 60): Promise<void> {
+    for (let i = 0; i < retries; i++) {
+      try {
+        const res = await fetch(`${this.baseUrl}/health`);
+        if (res.ok) return;
+      } catch {
+        /* keep retrying */
+      }
+      await new Promise((r) => setTimeout(r, 1000));
+    }
+    throw new Error(`Backend never became healthy at ${this.baseUrl}/health`);
+  }
+}

e2e/fixtures/db.ts

@@ -0,0 +1,149 @@
+/**
+ * Direct PostgreSQL escape hatch for setting up states the public API doesn't
+ * expose — e.g. forcing a user into the locked-PIN state to assert the 429
+ * recovery path, or expiring sessions for chaos tests.
+ *
+ * Most tests should NOT use this: prefer `ApiClient` so the tests exercise
+ * the same code paths real users do. Reach for direct SQL only when the API
+ * can't get you where you need to go.
+ */
+import { Client } from 'pg';
+
+const CONN = {
+  host: process.env.E2E_DB_HOST ?? 'localhost',
+  port: Number(process.env.E2E_DB_PORT ?? '55432'),
+  user: process.env.E2E_DB_USER ?? 'eventsnap_test',
+  password: process.env.E2E_DB_PASSWORD ?? 'eventsnap_test',
+  database: process.env.E2E_DB_NAME ?? 'eventsnap_test',
+};
+
+async function withClient<T>(fn: (c: Client) => Promise<T>): Promise<T> {
+  const client = new Client(CONN);
+  await client.connect();
+  try {
+    return await fn(client);
+  } finally {
+    await client.end();
+  }
+}
+
+export const db = {
+  async lockUserPin(userId: string, minutesFromNow = 15) {
+    await withClient((c) =>
+      c.query(
+        `UPDATE "user" SET pin_locked_until = NOW() + ($2 || ' minutes')::interval, failed_pin_attempts = 3 WHERE id = $1`,
+        [userId, String(minutesFromNow)]
+      )
+    );
+  },
+
+  async expireSession(userId: string) {
+    await withClient((c) =>
+      c.query(`UPDATE session SET expires_at = NOW() - interval '1 hour' WHERE user_id = $1`, [
+        userId,
+      ])
+    );
+  },
+
+  async setUploadCompressionStatus(
+    uploadId: string,
+    status: 'pending' | 'processing' | 'done' | 'failed'
+  ) {
+    await withClient((c) =>
+      c.query(`UPDATE upload SET compression_status = $2 WHERE id = $1`, [uploadId, status])
+    );
+  },
+
+  async countUploadsForUser(userId: string): Promise<number> {
+    return withClient(async (c) => {
+      const r = await c.query<{ count: string }>(
+        `SELECT COUNT(*)::text AS count FROM upload WHERE user_id = $1 AND deleted_at IS NULL`,
+        [userId]
+      );
+      return Number(r.rows[0].count);
+    });
+  },
+
+  async countSessionsForUser(userId: string): Promise<number> {
+    return withClient(async (c) => {
+      const r = await c.query<{ count: string }>(
+        `SELECT COUNT(*)::text AS count FROM session WHERE user_id = $1`,
+        [userId]
+      );
+      return Number(r.rows[0].count);
+    });
+  },
+
+  async countPinResetRequestsForUser(userId: string): Promise<number> {
+    return withClient(async (c) => {
+      const r = await c.query<{ count: string }>(
+        `SELECT COUNT(*)::text AS count FROM pin_reset_request WHERE user_id = $1`,
+        [userId]
+      );
+      return Number(r.rows[0].count);
+    });
+  },
+
+  async setExportReleased(slug: string, released: boolean) {
+    await withClient((c) =>
+      c.query(`UPDATE event SET export_released_at = $2 WHERE slug = $1`, [
+        slug,
+        released ? new Date() : null,
+      ])
+    );
+  },
+
+  /**
+   * Make an export "ready" (or not) in the epoch model. There is no `export_zip_ready` column any
+   * more — readiness is DERIVED (`released AND job.epoch = event.export_epoch AND status='done'`),
+   * so a job is ready exactly when its row carries the event's live epoch. To make a `done` job NOT
+   * ready we retire it to a dead epoch (-1), which is what a reopen effectively does.
+   *
+   * `file_path` is deliberately left NULL, so a "ready" job with no file on disk still exercises
+   * the download's missing-file 404 branch.
+   */
+  async setExportZipReady(slug: string, ready: boolean) {
+    await withClient((c) =>
+      c.query(
+        `UPDATE export_job ej
+            SET epoch = CASE WHEN $2 THEN e.export_epoch ELSE -1 END
+           FROM event e
+          WHERE e.id = ej.event_id AND e.slug = $1 AND ej.type = 'zip'`,
+        [slug, ready]
+      )
+    );
+  },
+
+  /**
+   * Insert a pre-baked export job row to skip the (slow) real compression path. Stamped with the
+   * event's CURRENT epoch so it counts as the live generation.
+   */
+  async fakeExportJob(
+    eventSlug: string,
+    type: 'zip' | 'html',
+    status: 'pending' | 'running' | 'done'
+  ) {
+    await withClient(async (c) => {
+      const ev = await c.query<{ id: string; export_epoch: string }>(
+        `SELECT id, export_epoch FROM event WHERE slug = $1`,
+        [eventSlug]
+      );
+      if (ev.rows.length === 0) throw new Error(`No event with slug ${eventSlug}`);
+      await c.query(
+        `INSERT INTO export_job (event_id, type, status, progress_pct, completed_at, epoch)
+         VALUES ($1, $2::export_type, $3::export_status, $4, $5, $6)
+         ON CONFLICT (event_id, type) DO UPDATE
+           SET status = EXCLUDED.status, progress_pct = EXCLUDED.progress_pct,
+               epoch = EXCLUDED.epoch`,
+        [
+          ev.rows[0].id,
+          type,
+          status,
+          status === 'done' ? 100 : 0,
+          status === 'done' ? new Date() : null,
+          ev.rows[0].export_epoch,
+        ]
+      );
+    });
+  },
+};

e2e/fixtures/test.ts

@@ -0,0 +1,100 @@
+/**
+ * Central `test.extend` for EventSnap E2E tests. Specs import `test` from here
+ * (not `@playwright/test` directly) so they get the shared fixtures: API
+ * client, DB helper, fresh admin token, and convenience factories.
+ *
+ * Every test runs against a freshly truncated database (via the `truncate`
+ * auto-fixture). Because truncate wipes the `session` table, any admin JWT
+ * obtained before truncate becomes invalid afterwards — so the
+ * `truncate` fixture always does its own admin login, and the `adminToken`
+ * fixture re-logs-in per test instead of caching. bcrypt cost 4 → ~10 ms,
+ * negligible against the ~1 s per-test setup overhead.
+ */
+import { test as base, expect, type Page } from '@playwright/test';
+import { ApiClient } from './api-client';
+import { db } from './db';
+
+type GuestHandle = {
+  jwt: string;
+  pin: string;
+  userId: string;
+  displayName: string;
+};
+
+type Fixtures = {
+  api: ApiClient;
+  db: typeof db;
+  adminToken: string;
+  truncate: void;
+  guest: (displayName?: string) => Promise<GuestHandle>;
+  host: GuestHandle;
+  /** Apply an existing guest's JWT + PIN to the page's localStorage and reload. */
+  signIn: (page: Page, handle: GuestHandle) => Promise<void>;
+};
+
+export const test = base.extend<Fixtures>({
+  api: async ({}, use) => {
+    await use(new ApiClient());
+  },
+
+  db: async ({}, use) => {
+    await use(db);
+  },
+
+  // Auto-fixture: runs before every test, truncates the DB so each test starts
+  // clean. Acquires its OWN admin token because the previous test's truncate
+  // wiped the session row that backs any cached token.
+  truncate: [
+    async ({ api }, use) => {
+      const token = await api.adminLogin();
+      await api.truncate(token);
+      await use();
+    },
+    { auto: true, scope: 'test' },
+  ],
+
+  // Fresh admin login per test that asks for it. Comes AFTER the truncate
+  // auto-fixture has run (truncate doesn't depend on adminToken, so the
+  // dependency-free truncate runs first; this fixture then logs in on a
+  // freshly reset DB).
+  adminToken: async ({ api }, use) => {
+    const token = await api.adminLogin();
+    await use(token);
+  },
+
+  guest: async ({ api }, use) => {
+    let counter = 0;
+    const factory = async (displayName?: string): Promise<GuestHandle> => {
+      const name = displayName ?? `Gast${++counter}_${Math.random().toString(36).slice(2, 6)}`;
+      const res = await api.join(name);
+      return { jwt: res.jwt, pin: res.pin, userId: res.user_id, displayName: name };
+    };
+    await use(factory);
+  },
+
+  host: async ({ api, guest, adminToken }, use) => {
+    const h = await guest('TestHost');
+    await api.setRole(adminToken, h.userId, 'host');
+    // Role is encoded in the JWT — re-recover to get a fresh token with role=host.
+    const { body } = await api.recover(h.displayName, h.pin);
+    await use({ ...h, jwt: body.jwt });
+  },
+
+  signIn: async ({}, use) => {
+    const fn = async (page: Page, handle: GuestHandle) => {
+      // Visit any in-app URL first so localStorage is scoped to the right origin.
+      await page.goto('/');
+      await page.evaluate(({ jwt, pin, userId, displayName }) => {
+        localStorage.setItem('eventsnap_jwt', jwt);
+        localStorage.setItem('eventsnap_pin', pin);
+        localStorage.setItem('eventsnap_user_id', userId);
+        localStorage.setItem('eventsnap_display_name', displayName);
+        localStorage.setItem('eventsnap_guide_seen', 'true');
+      }, handle);
+      await page.goto('/feed');
+    };
+    await use(fn);
+  },
+});
+
+export { expect };

e2e/global-setup.ts

@@ -0,0 +1,47 @@
+/**
+ * Runs once before all tests. Waits for the test stack to be healthy, logs
+ * in as admin, and flips all rate-limit/quota toggles off so tests don't
+ * trip over them. Individual tests that *want* to assert rate-limit
+ * behaviour re-enable the relevant flags in `beforeAll` and restore them
+ * in `afterAll`.
+ *
+ * The admin token is written to `.cache/admin-token` so per-worker
+ * fixtures can read it instead of logging in repeatedly.
+ */
+import { ApiClient } from './fixtures/api-client';
+import { mkdir, writeFile } from 'node:fs/promises';
+import { join } from 'node:path';
+
+const CACHE_DIR = join(process.cwd(), '.cache');
+
+export default async function globalSetup() {
+  const api = new ApiClient();
+
+  console.log('[e2e] waiting for backend health…');
+  await api.waitForHealth(120);
+
+  console.log('[e2e] logging in admin…');
+  const adminToken = await api.adminLogin();
+
+  console.log('[e2e] resetting database…');
+  await api.truncate(adminToken);
+
+  // Re-login because truncate wiped the session row backing the previous token.
+  const freshAdminToken = await api.adminLogin();
+
+  console.log('[e2e] disabling rate limits & quotas for the test run…');
+  await api.patchConfig(freshAdminToken, {
+    rate_limits_enabled: 'false',
+    upload_rate_enabled: 'false',
+    feed_rate_enabled: 'false',
+    export_rate_enabled: 'false',
+    join_rate_enabled: 'false',
+    quota_enabled: 'false',
+    storage_quota_enabled: 'false',
+    upload_count_quota_enabled: 'false',
+  });
+
+  await mkdir(CACHE_DIR, { recursive: true });
+  await writeFile(join(CACHE_DIR, 'admin-token'), freshAdminToken, 'utf8');
+  console.log('[e2e] global setup complete');
+}

e2e/global-teardown.ts

@@ -0,0 +1,8 @@
+/**
+ * Runs once after all tests. Currently a no-op: we leave the test stack
+ * running so the developer can inspect the post-run state (open the report,
+ * tail logs). Use `npm run stack:down` to actually tear the volumes down.
+ */
+export default async function globalTeardown() {
+  // intentionally empty
+}

e2e/helpers/env.ts

@@ -0,0 +1,5 @@
+/**
+ * Shared test environment constants. The frontend base URL was redeclared verbatim in ~23 specs;
+ * one source of truth means a port/scheme change is a single edit, not a sweep.
+ */
+export const BASE = process.env.E2E_FRONTEND_URL ?? 'http://localhost:3101';

e2e/helpers/fake-media.ts

@@ -0,0 +1,15 @@
+/**
+ * Camera/microphone faking. Chromium has the `--use-fake-device-for-media-stream`
+ * launch arg (set in playwright.config.ts) which makes `getUserMedia` succeed
+ * with a generated green-rectangle stream. WebKit and Firefox do not — those
+ * UA projects skip camera-path tests entirely.
+ */
+import type { BrowserContext } from '@playwright/test';
+
+export async function grantCameraPermissions(context: BrowserContext, origin: string) {
+  await context.grantPermissions(['camera', 'microphone'], { origin });
+}
+
+export function isFakeMediaSupported(browserName: string): boolean {
+  return browserName === 'chromium';
+}

e2e/helpers/seed.ts

@@ -0,0 +1,67 @@
+/**
+ * Shared seed helpers so specs don't each hand-roll the upload/comment create
+ * flow. Centralising the API contract (routes, field names, expected statuses)
+ * means an API change is a one-file edit, not a 7-file hunt.
+ */
+import { readFileSync } from 'node:fs';
+import { join } from 'node:path';
+import { uploadRaw } from './upload-client';
+import { db } from '../fixtures/db';
+import { BASE } from './env';
+
+// A real, decodable JPEG. Magic-bytes-only fakes now get auto-cleaned by the
+// backend when compression fails (a failed transcode refunds quota + deletes the
+// row), which would delete the seeded upload out from under the test. Read lazily
+// at call time (cwd is the e2e dir at runtime, like the gallery-path spec) so test
+// collection can't trip on a module-load read.
+let sampleJpg: Buffer | null = null;
+function sampleImage(): Buffer {
+  if (!sampleJpg) {
+    sampleJpg = readFileSync(join(process.cwd(), 'fixtures', 'media', 'sample.jpg'));
+  }
+  return sampleJpg;
+}
+
+export type SeedUploadOptions = {
+  caption?: string;
+  /** Mark compression done so the card is fully rendered in the feed. Default true. */
+  visible?: boolean;
+};
+
+/** Seed a real, accepted upload owned by `jwt` and return its id. */
+export async function seedUpload(jwt: string, opts: SeedUploadOptions = {}): Promise<string> {
+  const res = await uploadRaw(jwt, sampleImage(), {
+    filename: 'a.jpg',
+    contentType: 'image/jpeg',
+    caption: opts.caption,
+  });
+  if (res.status !== 201) throw new Error(`seedUpload failed: ${res.status} ${await res.text()}`);
+  const { id } = await res.json();
+  if (opts.visible !== false) await db.setUploadCompressionStatus(id, 'done');
+  return id;
+}
+
+/** Seed a comment on `uploadId` authored by `jwt`; return its id. */
+export async function seedComment(jwt: string, uploadId: string, body: string): Promise<string> {
+  const res = await fetch(`${BASE}/api/v1/upload/${uploadId}/comments`, {
+    method: 'POST',
+    headers: { Authorization: `Bearer ${jwt}`, 'Content-Type': 'application/json' },
+    body: JSON.stringify({ body }),
+  });
+  if (res.status !== 201) throw new Error(`seedComment failed: ${res.status} ${await res.text()}`);
+  return (await res.json()).id;
+}
+
+/** Read the comments for an upload as `jwt`. */
+export async function listComments(jwt: string, uploadId: string): Promise<any[]> {
+  const res = await fetch(`${BASE}/api/v1/upload/${uploadId}/comments`, {
+    headers: { Authorization: `Bearer ${jwt}` },
+  });
+  return res.json();
+}
+
+/** Resolve a single upload row from a feed response whose envelope shape isn't pinned. */
+export function findFeedRow(feed: any, id: string): any {
+  const list: any[] = feed.uploads ?? feed.items ?? feed;
+  return Array.isArray(list) ? list.find((u: any) => u.id === id) : undefined;
+}

e2e/helpers/sse-listener.ts

@@ -0,0 +1,97 @@
+/**
+ * Subscribe to /api/v1/stream and collect events. Tests can then assert "a
+ * `like-update` arrived for upload X within 5 seconds" without driving a
+ * second browser tab.
+ *
+ * The backend authenticates the SSE endpoint via a single-use `?ticket=` minted
+ * at POST /api/v1/stream/ticket (the raw JWT is never put in the URL). This helper
+ * does that exchange internally, so callers still just pass a JWT to `start()`.
+ */
+
+import { mintSseTicket } from './sse';
+
+export type SseEvent = { type: string; data: any; receivedAt: number };
+
+export class SseListener {
+  private controller = new AbortController();
+  private events: SseEvent[] = [];
+  private closed = false;
+
+  constructor(private baseUrl: string = process.env.E2E_FRONTEND_URL ?? 'http://localhost:3101') {}
+
+  async start(token: string): Promise<void> {
+    // Exchange the JWT for a single-use SSE ticket (the stream endpoint no longer
+    // accepts ?token=).
+    const ticket = await mintSseTicket(token);
+    const url = `${this.baseUrl}/api/v1/stream?ticket=${encodeURIComponent(ticket)}`;
+    // Use fetch with streaming since Node has no EventSource by default.
+    const res = await fetch(url, { signal: this.controller.signal });
+    if (!res.body) throw new Error('SSE response has no body');
+    const reader = res.body.getReader();
+    const decoder = new TextDecoder();
+    let buffer = '';
+    // Fire-and-forget reader loop; the caller drives lifecycle via stop()/the AbortController.
+    void (async () => {
+      try {
+        while (!this.closed) {
+          const { done, value } = await reader.read();
+          if (done) break;
+          buffer += decoder.decode(value, { stream: true });
+          // SSE messages are separated by \n\n
+          const parts = buffer.split('\n\n');
+          buffer = parts.pop() ?? '';
+          for (const part of parts) {
+            const evt = parseSseMessage(part);
+            if (evt) this.events.push({ ...evt, receivedAt: Date.now() });
+          }
+        }
+      } catch {
+        /* aborted */
+      }
+    })();
+  }
+
+  async waitForEvent(
+    type: string,
+    predicate: (e: SseEvent) => boolean = () => true,
+    timeoutMs = 10_000
+  ): Promise<SseEvent> {
+    const deadline = Date.now() + timeoutMs;
+    while (Date.now() < deadline) {
+      const hit = this.events.find((e) => e.type === type && predicate(e));
+      if (hit) return hit;
+      await new Promise((r) => setTimeout(r, 50));
+    }
+    throw new Error(
+      `SSE event ${type} did not arrive within ${timeoutMs}ms. Got: ${this.events.map((e) => e.type).join(', ')}`
+    );
+  }
+
+  allEvents(): SseEvent[] {
+    return [...this.events];
+  }
+
+  stop(): void {
+    this.closed = true;
+    this.controller.abort();
+  }
+}
+
+function parseSseMessage(raw: string): { type: string; data: any } | null {
+  const lines = raw.split('\n');
+  let type = 'message';
+  const dataLines: string[] = [];
+  for (const line of lines) {
+    if (line.startsWith('event:')) type = line.slice(6).trim();
+    else if (line.startsWith('data:')) dataLines.push(line.slice(5).trim());
+  }
+  if (dataLines.length === 0) return null;
+  const dataStr = dataLines.join('\n');
+  let data: any = dataStr;
+  try {
+    data = JSON.parse(dataStr);
+  } catch {
+    /* leave as string */
+  }
+  return { type, data };
+}

e2e/helpers/sse.ts

@@ -0,0 +1,43 @@
+/**
+ * Shared SSE-flow helpers. The stream auth flow (mint a single-use ticket, open
+ * with `?ticket=`) is security-sensitive and recently changed from `?token=`, so
+ * it lives in one place instead of being re-inlined per spec.
+ */
+import type { Page } from '@playwright/test';
+import { BASE } from './env';
+
+/** Exchange a JWT for a single-use SSE ticket via POST /api/v1/stream/ticket. */
+export async function mintSseTicket(jwt: string): Promise<string> {
+  const res = await fetch(`${BASE}/api/v1/stream/ticket`, {
+    method: 'POST',
+    headers: { Authorization: `Bearer ${jwt}` },
+  });
+  if (res.status !== 200)
+    throw new Error(`mintSseTicket failed: ${res.status} ${await res.text()}`);
+  return (await res.json()).ticket;
+}
+
+/** Open the SSE stream with a ticket, return the HTTP status, and tear the stream down. */
+export async function openStream(ticket: string): Promise<number> {
+  const c = new AbortController();
+  try {
+    const res = await fetch(`${BASE}/api/v1/stream?ticket=${encodeURIComponent(ticket)}`, {
+      signal: c.signal,
+    });
+    return res.status;
+  } finally {
+    c.abort();
+  }
+}
+
+/**
+ * Count EventSource opens (GET /api/v1/stream?ticket=…) on a page — NOT the ticket
+ * POST. Returns a getter for the running count.
+ */
+export function trackStreamOpens(page: Page): () => number {
+  let n = 0;
+  page.on('request', (req) => {
+    if (req.method() === 'GET' && req.url().includes('/api/v1/stream?')) n++;
+  });
+  return () => n;
+}