Schulcloud-MCP/test/store.test.ts

import assert from 'node:assert/strict';
import { after, before, describe, it } from 'node:test';
import { Store } from '../src/store/store.ts';
import type { Snapshot } from '../src/core/crawl.ts';

/**
 * Exercises the store against a real Postgres — the generation/diff semantics
 * are entirely SQL, so a mock would test nothing. Skipped when TEST_DATABASE_URL
 * is unset so `npm test` stays offline by default.
 */
const DB_URL = process.env.TEST_DATABASE_URL;

/**
 * These tests TRUNCATE. Pointing them at a real database destroys it — which
 * happened once during development, when TEST_DATABASE_URL was aimed at the dev
 * instance and the fixtures ended up in live data. Requiring "test" in the
 * database name makes that mistake impossible to repeat by accident.
 */
function assertDisposable(url: string): void {
	const name = new globalThis.URL(url).pathname.replace(/^\//, '');
	if (!/test/i.test(name)) {
		throw new Error(
			`Refusing to run: TEST_DATABASE_URL points at database "${name}", which is not obviously ` +
				`disposable. These tests TRUNCATE. Use a database with "test" in its name.`,
		);
	}
}

function snapshot(
	courses: { id: string; title: string; boardText?: string; files?: { id: string; name: string; size: number }[] }[],
	rooms: { id: string; name: string; boardText?: string }[] = [],
): Snapshot {
	return {
		crawledAt: new Date(),
		schoolId: 'school1',
		failures: [],
		rooms: rooms.map((r) => ({
			id: r.id,
			name: r.name,
			boards: r.boardText
				? [{ id: `${r.id}-b`, title: 'Raum-Board', courseId: r.id, text: r.boardText,
				     board: { id: `${r.id}-b`, title: 'Raum-Board', columns: [], fileCount: 0 } }]
				: [],
		})),
		courses: courses.map((c) => ({
			course: { id: c.id, title: c.title, shortTitle: c.title.slice(0, 2), displayColor: '#000' },
			title: c.title,
			boards: c.boardText
				? [{ id: `${c.id}-b`, title: 'Board', courseId: c.id, text: c.boardText,
				     board: { id: `${c.id}-b`, title: 'Board', columns: [], fileCount: 0 } }]
				: [],
			lessons: [],
			tasks: [],
		})),
		files: courses.flatMap((c) =>
			(c.files ?? []).map((f) => ({
				record: { id: f.id, name: f.name, parentId: 'p', parentType: 'boardnodes' as const,
				          url: '', size: f.size, mimeType: 'application/pdf',
				          securityCheckStatus: 'verified', previewStatus: 'x' },
				parentType: 'boardnodes' as const,
				parentId: 'p',
				at: { courseId: c.id, courseTitle: c.title, containerTitle: 'Board' },
			})),
		),
	};
}

describe('Store', { skip: DB_URL ? false : 'set TEST_DATABASE_URL to run' }, () => {
	let store: Store;

	before(async () => {
		assertDisposable(DB_URL!);
		const opened = await Store.open(DB_URL);
		assert.ok(opened, 'store should open');
		store = opened;
		// Start from a clean slate so generation ids are predictable.
		await (store as never as { db: { query: (q: string) => Promise<unknown> } }).db.query(
			'TRUNCATE crawls, file_texts RESTART IDENTITY CASCADE',
		);
	});

	after(async () => {
		await store?.close();
	});

	it('returns undefined rather than throwing when the database is unreachable', async () => {
		const dead = await Store.open('postgresql://nobody@127.0.0.1:1/none');
		assert.equal(dead, undefined);
	});

	it('saves a generation and reports stats', async () => {
		const id = await store.saveSnapshot(snapshot([{ id: 'c1', title: 'Mathe', boardText: 'Bruchrechnung', files: [{ id: 'f1', name: 'a.pdf', size: 10 }] }]), 'full');
		assert.ok(id > 0);
		const stats = await store.stats();
		assert.equal(stats.crawlId, id);
		assert.equal(stats.files, 1);
	});

	it('diffs by identity, reporting additions and deletions', async () => {
		const first = await store.latestCrawlId();
		const second = await store.saveSnapshot(
			snapshot([{ id: 'c1', title: 'Mathe', boardText: 'Bruchrechnung', files: [{ id: 'f2', name: 'b.pdf', size: 20 }] }]),
			'full',
		);
		const diff = await store.diff(first!, second);
		assert.ok(diff.added.some((n) => n.nodeId === 'f2'), 'f2 added');
		assert.ok(diff.removed.some((n) => n.nodeId === 'f1'), 'f1 removed — timestamps could never show this');
	});

	it('reports a content change even when the id is unchanged', async () => {
		const before = await store.latestCrawlId();
		const after = await store.saveSnapshot(
			snapshot([{ id: 'c1', title: 'Mathe', boardText: 'Prozentrechnung', files: [{ id: 'f2', name: 'b.pdf', size: 20 }] }]),
			'full',
		);
		const diff = await store.diff(before!, after);
		assert.ok(diff.changed.some((n) => n.nodeId === 'c1-b'), 'board body change detected via digest');
	});

	it('reports a renamed file, which keeps its id and size', async () => {
		// `PATCH /file/rename/{id}` renames a record in place. The digest once
		// covered only id and size on the assumption that file records never
		// change, so a rename went unreported — the file just quietly appeared
		// under a new name.
		const before = await store.latestCrawlId();
		const after = await store.saveSnapshot(
			snapshot([{ id: 'c1', title: 'Mathe', boardText: 'Prozentrechnung', files: [{ id: 'f2', name: 'b-v2.pdf', size: 20 }] }]),
			'full',
		);
		const diff = await store.diff(before!, after);
		assert.ok(diff.changed.some((n) => n.nodeId === 'f2'), 'rename detected');
		assert.ok(!diff.added.some((n) => n.nodeId === 'f2'), 'a rename is not a new file');
		assert.ok(!diff.removed.some((n) => n.nodeId === 'f2'), 'and not a deleted one');
	});

	it('stores rooms alongside courses, and notices when one changes', async () => {
		// Rooms are a separate space, not a kind of course: they must land in the
		// index under their own kind, or room content becomes unsearchable the way
		// submitted files once were.
		const before = await store.saveSnapshot(
			snapshot([{ id: 'c1', title: 'Mathe', boardText: 'Prozentrechnung' }], [{ id: 'r1', name: 'Projektraum', boardText: 'Projektsteuerung' }]),
			'full',
		);
		const hits = await store.search('Projektsteuerung', { limit: 5 });
		assert.ok(hits.some((h) => h.nodeId === 'r1-b'), 'a room board is searchable');

		const after = await store.saveSnapshot(
			snapshot([{ id: 'c1', title: 'Mathe', boardText: 'Prozentrechnung' }], [{ id: 'r1', name: 'Projektraum Informatik', boardText: 'Projektsteuerung' }]),
			'full',
		);
		const diff = await store.diff(before, after);
		assert.ok(diff.changed.some((n) => n.nodeId === 'r1' && n.kind === 'room'), 'a renamed room is reported as changed');
	});

	it('carries other courses forward on a per-course crawl', async () => {
		await store.saveSnapshot(
			snapshot([
				{ id: 'c1', title: 'Mathe', boardText: 'Prozentrechnung' },
				{ id: 'c2', title: 'Physik', boardText: 'Optik' },
			]),
			'full',
		);
		const before = await store.latestCrawlId();
		// Re-crawl only c2; c1 must survive rather than looking deleted.
		const after = await store.saveSnapshot(snapshot([{ id: 'c2', title: 'Physik', boardText: 'Mechanik' }]), 'c2');
		const diff = await store.diff(before!, after);
		assert.equal(diff.removed.length, 0, 'a partial crawl must not look like a mass deletion');
		assert.ok(diff.changed.some((n) => n.nodeId === 'c2-b'));
	});

	it('finds German content with stemming', async () => {
		const hits = await store.search('Mechanik');
		assert.ok(hits.length > 0, 'expected a hit for Mechanik');
	});

	it('makes extracted file text searchable', async () => {
		await store.saveSnapshot(snapshot([{ id: 'c3', title: 'Info', files: [{ id: 'f9', name: 'skript.pdf', size: 99 }] }]), 'full');
		await store.recordFileText({
			fileId: 'f9', name: 'skript.pdf', mimeType: 'application/pdf', size: 99,
			content: 'Die Cäsar-Verschlüsselung verschiebt Buchstaben im Alphabet.',
			note: 'ok', mirrorPath: 'Info/Board/skript.pdf', mirrorSize: 99,
		});
		const hits = await store.search('Verschlüsselung');
		assert.ok(hits.some((h) => h.nodeId === 'f9'), 'PDF contents should be searchable, not just the filename');
	});

	it('stores extracted text containing NUL, which Postgres text refuses', async () => {
		await store.recordFileText({
			fileId: 'f9', name: 'skript.pdf', mimeType: 'application/pdf', size: 99,
			content: 'Webserver\u0000 und Proxy', note: 'ok\u0000', mirrorPath: 'Info/Board/skript.pdf', mirrorSize: 99,
		});
		const hits = await store.search('Proxy');
		assert.ok(hits.some((h) => h.nodeId === 'f9'), 'text with NUL bytes should still be stored and searchable');
	});

	it('keeps a file whose download failed queued for the next crawl, but not one that failed to extract', async () => {
		await store.recordFileText({
			fileId: 'f9', name: 'skript.pdf', mimeType: 'application/pdf', size: 99,
			content: null, note: 'download failed, retried on the next crawl: timeout',
			mirrorPath: null, mirrorSize: null, retry: true,
		});
		assert.ok((await store.filesNeedingText()).some((f) => f.fileId === 'f9'), 'a transient failure must be retried');

		await store.recordFileText({
			fileId: 'f9', name: 'skript.pdf', mimeType: 'application/pdf', size: 99,
			content: null, note: 'extraction failed: bad xref', mirrorPath: null, mirrorSize: null,
		});
		assert.ok(!(await store.filesNeedingText()).some((f) => f.fileId === 'f9'), 'a parser failure is final');
	});

	it('resolves a timestamp cursor to a generation', async () => {
		const id = await store.resolveCursor(new Date().toISOString());
		assert.ok(id && id > 0);
		assert.equal(await store.resolveCursor('not-a-date'), undefined);
	});

	it('builds a manifest with per-file change status', async () => {
		const { entries } = await store.manifest();
		assert.ok(entries.some((e) => e.fileId === 'f9' && e.path.startsWith('Info/')));
	});
})