Files
Mangalord/backend/src/repo/author.rs
MechaCat02 c320eda7cd chore: dedupe is_unique_violation, lift SQL into repo, centralise URL parsing
Three layering cleanups from REVIEW.md §5 / §3:

- Drop the three private `is_unique_violation` helpers in
  repo::{user,chapter,bookmark} in favour of sqlx 0.8's
  `DatabaseError::is_unique_violation()` method (already used by
  repo::collection).
- Remove the unreachable 23505 branch in repo::chapter::create — the
  (manga_id, number) UNIQUE was dropped in 0013, so the defensive arm
  could no longer fire. A doc note records what to do if uniqueness
  is re-added.
- Move three inline SQL queries out of handlers/daemon into repo
  functions: bookmarks' chapter-belongs-to-manga guard
  (`repo::chapter::belongs_to_manga`), the daemon's dispatch lookup
  (`repo::chapter::dispatch_target`), and the daemon's page_count
  safety net (`repo::chapter::page_count`). Restores the
  handlers→repo layering invariant in CLAUDE.md.
- New `crawler::url_utils` module consolidates host_of / origin_of /
  registrable_domain — they used to live in three crawler submodules
  with diverging edge-case behaviour. Tests moved with them.
- Doc cross-references on repo::author::set_for_manga and
  repo::genre::set_for_manga pointing to the crawler's name-keyed
  variants, so the intentional duplication is discoverable.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-28 20:24:05 +02:00

211 lines
6.2 KiB
Rust

//! Author persistence. Same plain-function style as the other repos.
//!
//! Names are stored with the casing of the first submission; lookups go
//! through the case-insensitive unique index on `lower(name)` so
//! "Kentaro Miura" and "kentaro miura" share one row.
use std::collections::BTreeMap;
use sqlx::{PgConnection, PgExecutor, PgPool};
use uuid::Uuid;
use crate::domain::author::{Author, AuthorRef, AuthorWithCount};
use crate::domain::manga::Manga;
use crate::error::{AppError, AppResult};
/// Insert-or-find by name. Returns the canonical row whether or not
/// this call created it. The `ON CONFLICT … DO UPDATE` no-op is what
/// makes `RETURNING` come back even on the existing-row path.
pub async fn upsert_by_name<'e, E: PgExecutor<'e>>(
executor: E,
name: &str,
) -> AppResult<Author> {
let trimmed = name.trim();
if trimmed.is_empty() {
return Err(AppError::ValidationFailed {
message: "author name cannot be empty".into(),
details: serde_json::json!({ "authors": "non-empty names required" }),
});
}
let row = sqlx::query_as::<_, Author>(
r#"
INSERT INTO authors (name)
VALUES ($1)
ON CONFLICT (lower(name)) DO UPDATE SET name = authors.name
RETURNING id, name, created_at
"#,
)
.bind(trimmed)
.fetch_one(executor)
.await?;
Ok(row)
}
pub async fn find_by_id(pool: &PgPool, id: Uuid) -> AppResult<Author> {
sqlx::query_as::<_, Author>(
"SELECT id, name, created_at FROM authors WHERE id = $1",
)
.bind(id)
.fetch_optional(pool)
.await?
.ok_or(AppError::NotFound)
}
pub async fn find_with_count(pool: &PgPool, id: Uuid) -> AppResult<AuthorWithCount> {
sqlx::query_as::<_, AuthorWithCount>(
r#"
SELECT a.id,
a.name,
a.created_at,
(SELECT count(*) FROM manga_authors ma WHERE ma.author_id = a.id) AS manga_count
FROM authors a
WHERE a.id = $1
"#,
)
.bind(id)
.fetch_optional(pool)
.await?
.ok_or(AppError::NotFound)
}
/// Autocomplete-friendly: substring + trigram match on `name`, ordered
/// most-similar first when a search term is present.
pub async fn list(
pool: &PgPool,
search: Option<&str>,
limit: i64,
offset: i64,
) -> AppResult<Vec<Author>> {
let rows = sqlx::query_as::<_, Author>(
r#"
SELECT id, name, created_at
FROM authors
WHERE $1::text IS NULL
OR name ILIKE '%' || $1 || '%'
OR name % $1
ORDER BY CASE WHEN $1::text IS NULL THEN 0 ELSE similarity(name, $1) END DESC,
lower(name) ASC
LIMIT $2 OFFSET $3
"#,
)
.bind(search)
.bind(limit)
.bind(offset)
.fetch_all(pool)
.await?;
Ok(rows)
}
/// Atomically replace the set of authors on a manga. Caller passes a
/// `&mut PgConnection` (`&mut *tx` works) so the delete+upserts run in
/// one transaction with whatever called us.
///
/// Note: `crawler::repo::sync_authors` does a similar replace with the
/// same semantics on names. The duplication is intentional — handler
/// callers want the `Vec<AuthorRef>` for the API response; the
/// crawler doesn't need it and stays inside its own transaction.
pub async fn set_for_manga(
conn: &mut PgConnection,
manga_id: Uuid,
names: &[String],
) -> AppResult<Vec<AuthorRef>> {
sqlx::query("DELETE FROM manga_authors WHERE manga_id = $1")
.bind(manga_id)
.execute(&mut *conn)
.await?;
let mut refs = Vec::with_capacity(names.len());
for (position, name) in names.iter().enumerate() {
let author = upsert_by_name(&mut *conn, name).await?;
sqlx::query(
"INSERT INTO manga_authors (manga_id, author_id, position) VALUES ($1, $2, $3)
ON CONFLICT (manga_id, author_id) DO NOTHING",
)
.bind(manga_id)
.bind(author.id)
.bind(position as i32)
.execute(&mut *conn)
.await?;
refs.push(AuthorRef { id: author.id, name: author.name });
}
Ok(refs)
}
pub async fn list_for_manga(pool: &PgPool, manga_id: Uuid) -> AppResult<Vec<AuthorRef>> {
let rows = sqlx::query_as::<_, AuthorRef>(
r#"
SELECT a.id, a.name
FROM manga_authors ma
JOIN authors a ON a.id = ma.author_id
WHERE ma.manga_id = $1
ORDER BY ma.position, lower(a.name)
"#,
)
.bind(manga_id)
.fetch_all(pool)
.await?;
Ok(rows)
}
/// Batch-load authors for a list of manga ids and group by manga. Used
/// by the list endpoint to avoid N+1 round-trips.
pub async fn load_for_mangas(
pool: &PgPool,
manga_ids: &[Uuid],
) -> AppResult<BTreeMap<Uuid, Vec<AuthorRef>>> {
if manga_ids.is_empty() {
return Ok(BTreeMap::new());
}
let rows: Vec<(Uuid, Uuid, String)> = sqlx::query_as(
r#"
SELECT ma.manga_id, a.id, a.name
FROM manga_authors ma
JOIN authors a ON a.id = ma.author_id
WHERE ma.manga_id = ANY($1)
ORDER BY ma.manga_id, ma.position, lower(a.name)
"#,
)
.bind(manga_ids)
.fetch_all(pool)
.await?;
let mut grouped: BTreeMap<Uuid, Vec<AuthorRef>> = BTreeMap::new();
for (manga_id, author_id, name) in rows {
grouped
.entry(manga_id)
.or_default()
.push(AuthorRef { id: author_id, name });
}
Ok(grouped)
}
pub async fn list_mangas_for_author(
pool: &PgPool,
author_id: Uuid,
limit: i64,
offset: i64,
) -> AppResult<(Vec<Manga>, i64)> {
let rows = sqlx::query_as::<_, Manga>(
r#"
SELECT m.id, m.title, m.status, m.alt_titles, m.description,
m.cover_image_path, m.created_at, m.updated_at
FROM manga_authors ma
JOIN mangas m ON m.id = ma.manga_id
WHERE ma.author_id = $1
ORDER BY m.created_at DESC, m.id
LIMIT $2 OFFSET $3
"#,
)
.bind(author_id)
.bind(limit)
.bind(offset)
.fetch_all(pool)
.await?;
let (total,): (i64,) = sqlx::query_as(
"SELECT count(*) FROM manga_authors WHERE author_id = $1",
)
.bind(author_id)
.fetch_one(pool)
.await?;
Ok((rows, total))
}