Claude Opus 4.8
91dfccf226
All checks were successful
CD / Build and push images (push) Successful in 3m5s
CI / Lint, typecheck, test (push) Successful in 2m19s
CI / Auth e2e pack (push) Successful in 2m51s
CI / Build container images (push) Has been skipped
CD / Deploy to Test (push) Successful in 9s
CD / Smoke tests against Test (push) Successful in 1m13s
CD / Promote to Int (push) Successful in 12s
Full-text search behind a swappable interface (ADR 0010).
- prisma: `page_content_cache.search_vector tsvector` (Unsupported column);
migration adds it plus a GIN index (raw SQL — the index is a production
perf optimization; correctness holds without it, so schema-pushed test DBs
work unchanged).
- shared: `normalizeForSearch` (NFKD + strip diacritics + lowercase) folds
both the indexed text and the query, so 'Baume' finds 'Bäume' without the
Postgres `unaccent` extension; search query schema + result view + highlight
sentinels.
- api search module:
- abstract `SearchProvider` (DI token: indexPage / removePage / search /
reindexAll) so an external engine can replace the binding — a fake proves
the seam in a test.
- `PostgresSearchProvider`: weighted vector (title A, labels B, body C),
`websearch_to_tsquery`, `ts_headline` snippets, results filtered to the
ponds the user may read; `GET /search?q=&pondId=&labels=`.
- `search:reindex` CLI (rebuilds from the content cache, idempotent).
- reindex hooks: page create/rename (title) and label assign/unassign/
rename/delete (labels are weight-B).
- collab: the persistence hook maintains `search_vector` in the same
transaction as the content cache (same weighting, normalized).
- tests: shared normalize/schema; api db (title ranks above body, highlight,
diacritic-insensitive match, permission filter, idempotent reindex) and the
fake-provider DI test; collab persistence already covers the write path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PGdhRiwU1WRL4XxJfZYipY
53 lines
2.1 KiB
TypeScript
53 lines
2.1 KiB
TypeScript
import { z } from 'zod';
|
||
|
||
/**
|
||
* Search schemas and views (issue #49/#50, ADR 0010). Full-text search runs on
|
||
* PostgreSQL behind the `SearchProvider` interface. Diacritic-insensitive
|
||
* matching ('Baume' finds 'Bäume') is achieved by normalizing both the indexed
|
||
* text and the query in application code — so no Postgres `unaccent` extension
|
||
* is required (which keeps the schema-pushed test databases working).
|
||
*/
|
||
|
||
// Combining diacritical marks (U+0300–U+036F), removed after NFKD decomposition.
|
||
const COMBINING_MARKS = /[̀-ͯ]/g;
|
||
|
||
/**
|
||
* Fold a string for search: strip diacritics (NFKD + drop combining marks) and
|
||
* lowercase, leaving tokenization to Postgres `to_tsvector('simple', …)`. Used
|
||
* for BOTH the stored search vector and the query, so they always agree.
|
||
*/
|
||
export function normalizeForSearch(text: string): string {
|
||
return text.normalize('NFKD').replace(COMBINING_MARKS, '').toLowerCase();
|
||
}
|
||
|
||
/** Sentinels wrapping a matched span in a result snippet (`ts_headline`), split
|
||
* by the UI to render highlights without trusting HTML from the content.
|
||
* Private-use codepoints that never occur in page text. */
|
||
export const SEARCH_HIGHLIGHT_START = String.fromCodePoint(0xe000);
|
||
export const SEARCH_HIGHLIGHT_END = String.fromCodePoint(0xe001);
|
||
|
||
/** How many results one search returns (v1, ADR 0010 target scale). */
|
||
export const SEARCH_RESULT_LIMIT = 30;
|
||
|
||
export const searchQuerySchema = z.object({
|
||
q: z.string().trim().min(1, 'validation.required').max(200, 'validation.tooLong'),
|
||
/** Restrict to one pond; omitted = all ponds the user may read. */
|
||
pondId: z.string().min(1).optional(),
|
||
/** Restrict to pages carrying any of these label ids. */
|
||
labels: z.array(z.string().min(1)).optional(),
|
||
});
|
||
export type SearchQuery = z.infer<typeof searchQuerySchema>;
|
||
|
||
/** One hit: the page, its pond, its labels, and a highlighted snippet. */
|
||
export interface SearchResultView {
|
||
pageId: string;
|
||
title: string;
|
||
slug: string;
|
||
pondId: string;
|
||
pondSlug: string;
|
||
pondName: string;
|
||
labelIds: string[];
|
||
/** Snippet with matches wrapped in the highlight sentinels above. */
|
||
snippet: string;
|
||
}
|