dorfteich/apps/api/src/search/search.service.db.test.ts

import { randomUUID } from 'node:crypto';

import { SEARCH_HIGHLIGHT_START } from '@dorfteich/shared';
import { INestApplication } from '@nestjs/common';
import { PrismaClient, User } from '@prisma/client';
import { afterAll, beforeAll, describe, expect, it } from 'vitest';
import * as Y from 'yjs';

import { createTestApp } from '../testing/test-app';
import { createTestPrisma, grantOwnerAdmin, hasTestDb, uniqueSuffix } from '../testing/test-db';
import { SearchProvider } from './search.provider';

describe.skipIf(!hasTestDb)('PostgresSearchProvider (db, issue #49)', () => {
  let app: INestApplication;
  let prisma: PrismaClient;
  let search: SearchProvider;
  const suffix = uniqueSuffix();
  const term = `zeb${suffix}`; // a term unique to this run
  let owner: User;
  let outsider: User;
  let pondId: string;
  const pageIds: string[] = [];

  /** Creates a page with an indexed content cache, then indexes it. */
  async function makePage(title: string, plainText: string): Promise<string> {
    const id = randomUUID();
    await prisma.page.create({
      data: {
        id,
        pondId,
        title,
        slug: `p-${id.slice(0, 8)}`,
        ydocState: new Uint8Array(Y.encodeStateAsUpdate(new Y.Doc())),
        sortKey: `a${pageIds.length}`,
        createdBy: owner.id,
        contentCache: { create: { plainText, markdown: plainText, html: plainText, outline: [] } },
      },
    });
    pageIds.push(id);
    await search.indexPage(id);
    return id;
  }

  beforeAll(async () => {
    prisma = createTestPrisma();
    app = await createTestApp();
    search = app.get(SearchProvider);

    owner = await prisma.user.create({
      data: {
        username: `srch-owner-${suffix}`,
        email: `srch-owner-${suffix}@example.test`,
        displayName: 'Search Owner',
      },
    });
    outsider = await prisma.user.create({
      data: {
        username: `srch-out-${suffix}`,
        email: `srch-out-${suffix}@example.test`,
        displayName: 'Search Outsider',
      },
    });
    const pond = await prisma.pond.create({
      data: {
        slug: `srch-pond-${suffix}`,
        name: 'Search Pond',
        type: 'PERSONAL',
        ownerId: owner.id,
      },
    });
    pondId = pond.id;
    await grantOwnerAdmin(prisma, pondId, owner.id);
  });

  afterAll(async () => {
    await prisma.page.deleteMany({ where: { pondId } });
    await prisma.pond.deleteMany({ where: { id: pondId } });
    await prisma.user.deleteMany({ where: { id: { in: [owner.id, outsider.id] } } });
    await prisma.$disconnect();
    await app.close();
  });

  it('ranks a title match above a body match', async () => {
    const titleHit = await makePage(`${term} in the title`, 'unrelated body text');
    await makePage('unrelated title', `a long story that mentions ${term} in the body only`);

    const results = await search.search({ q: term }, owner);
    expect(results.length).toBeGreaterThanOrEqual(2);
    // Weight A (title) outranks weight C (body).
    expect(results[0]!.pageId).toBe(titleHit);
  });

  it('carries the classification with every hit — a classified snippet is never unmarked (issue #211)', async () => {
    const classifiedId = await makePage(`classified ${term} note`, `secret ${term} content`);
    await prisma.page.update({
      where: { id: classifiedId },
      data: { classification: 'VS_NFD' },
    });
    const results = await search.search({ q: term }, owner);
    const classified = results.find((r) => r.pageId === classifiedId);
    expect(classified?.classification).toBe('vs_nfd');
    // Every other hit carries the field too, as `unclassified`.
    const other = results.find((r) => r.pageId !== classifiedId);
    expect(other?.classification).toBe('unclassified');
  });

  it('highlights the match in the snippet', async () => {
    const results = await search.search({ q: term }, owner);
    const bodyHit = results.find((r) => r.snippet.includes(SEARCH_HIGHLIGHT_START));
    expect(bodyHit).toBeDefined();
  });

  it('matches diacritics-insensitively (Baume finds Bäume)', async () => {
    const page = await makePage('Wald', `viele Bäume-${suffix} im Wald`);
    const results = await search.search({ q: `Baume-${suffix}` }, owner);
    expect(results.map((r) => r.pageId)).toContain(page);
  });

  it('matches partial words in title and body (M10 follow-up)', async () => {
    const titlePage = await makePage(`Quakfrosch${suffix} Titel`, 'nichts weiter');
    const bodyPage = await makePage('anderer Titel', `hier lebt ein Teichmolch${suffix}`);

    // A mid-word fragment matches nothing via the tsquery — the LIKE branch
    // has to find both pages (title and body).
    const byTitle = await search.search({ q: `akfrosch${suffix}` }, owner);
    expect(byTitle.map((r) => r.pageId)).toContain(titlePage);
    const byBody = await search.search({ q: `eichmolch${suffix}` }, owner);
    expect(byBody.map((r) => r.pageId)).toContain(bodyPage);

    // The outsider still sees nothing through the substring branch.
    expect(await search.search({ q: `eichmolch${suffix}` }, outsider)).toEqual([]);
  });

  it('never returns pages the requester may not read', async () => {
    const results = await search.search({ q: term }, outsider);
    expect(results).toEqual([]);
  });

  it('scopes results to a single pond when pondId is given', async () => {
    // A second pond of the same owner with a page matching the same term.
    const other = await prisma.pond.create({
      data: {
        slug: `srch-pond2-${suffix}`,
        name: 'Second Pond',
        type: 'SHARED',
        ownerId: owner.id,
      },
    });
    await grantOwnerAdmin(prisma, other.id, owner.id);
    const otherPage = await prisma.page.create({
      data: {
        id: randomUUID(),
        pondId: other.id,
        title: `${term} elsewhere`,
        slug: `q-${suffix}`,
        ydocState: new Uint8Array(Y.encodeStateAsUpdate(new Y.Doc())),
        sortKey: 'a0',
        createdBy: owner.id,
        contentCache: { create: { plainText: '', markdown: '', html: '', outline: [] } },
      },
    });
    await search.indexPage(otherPage.id);

    // All ponds: the second pond's page is included.
    const all = await search.search({ q: term }, owner);
    expect(all.map((r) => r.pageId)).toContain(otherPage.id);
    // Scoped to the first pond: it is excluded.
    const scoped = await search.search({ q: term, pondId }, owner);
    expect(scoped.map((r) => r.pondId).every((id) => id === pondId)).toBe(true);
    expect(scoped.map((r) => r.pageId)).not.toContain(otherPage.id);

    await prisma.page.deleteMany({ where: { pondId: other.id } });
    await prisma.pond.deleteMany({ where: { id: other.id } });
  });

  it('reindexAll rebuilds from the cache and is idempotent', async () => {
    const before = await search.search({ q: term }, owner);
    await search.reindexAll();
    await search.reindexAll();
    const after = await search.search({ q: term }, owner);
    expect(after.map((r) => r.pageId).sort()).toEqual(before.map((r) => r.pageId).sort());
  });
});