mirror of
https://github.com/garrytan/gbrain.git
synced 2026-08-17 02:12:40 +00:00
Compare commits
3
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
a42352081c | ||
|
|
070f4c5678 | ||
|
|
4b2df89935 |
@@ -5,8 +5,8 @@
|
||||
* checks if back-links exist, and optionally creates them.
|
||||
*
|
||||
* Usage:
|
||||
* gbrain check-backlinks check [dir] [--dir <brain-dir>] # report missing back-links
|
||||
* gbrain check-backlinks fix [dir] [--dir <brain-dir>] # create missing back-links
|
||||
* gbrain check-backlinks check [--dir <brain-dir>] # report missing back-links
|
||||
* gbrain check-backlinks fix [--dir <brain-dir>] # create missing back-links
|
||||
* gbrain check-backlinks fix --dry-run # preview fixes
|
||||
*/
|
||||
|
||||
@@ -201,40 +201,6 @@ export interface BacklinksResult {
|
||||
dryRun: boolean;
|
||||
}
|
||||
|
||||
export interface ParsedBacklinksArgs {
|
||||
subcommand: string | undefined;
|
||||
brainDir: string;
|
||||
dryRun: boolean;
|
||||
}
|
||||
|
||||
export function parseBacklinksArgs(args: string[]): ParsedBacklinksArgs {
|
||||
const subcommand = args[0];
|
||||
const dryRun = args.includes('--dry-run');
|
||||
const dirIdx = args.indexOf('--dir');
|
||||
const flagDir = dirIdx >= 0 && args[dirIdx + 1] && !args[dirIdx + 1].startsWith('--')
|
||||
? args[dirIdx + 1]
|
||||
: undefined;
|
||||
|
||||
let positionalDir: string | undefined;
|
||||
for (let i = 1; i < args.length; i++) {
|
||||
const arg = args[i];
|
||||
if (arg === '--dir') {
|
||||
i++;
|
||||
continue;
|
||||
}
|
||||
if (arg === '--dry-run') continue;
|
||||
if (arg.startsWith('--')) continue;
|
||||
positionalDir = arg;
|
||||
break;
|
||||
}
|
||||
|
||||
return {
|
||||
subcommand,
|
||||
brainDir: flagDir ?? positionalDir ?? '.',
|
||||
dryRun,
|
||||
};
|
||||
}
|
||||
|
||||
/**
|
||||
* Library-level backlinks check/fix. Throws on validation errors; returns a
|
||||
* structured result so Minions handlers + autopilot-cycle can surface counts.
|
||||
@@ -270,14 +236,16 @@ export async function runBacklinksCore(opts: BacklinksOpts): Promise<BacklinksRe
|
||||
}
|
||||
|
||||
export async function runBacklinks(args: string[]) {
|
||||
const { subcommand, brainDir, dryRun } = parseBacklinksArgs(args);
|
||||
const subcommand = args[0];
|
||||
const dirIdx = args.indexOf('--dir');
|
||||
const brainDir = dirIdx >= 0 ? args[dirIdx + 1] : '.';
|
||||
const dryRun = args.includes('--dry-run');
|
||||
|
||||
if (!subcommand || !['check', 'fix'].includes(subcommand)) {
|
||||
console.error('Usage: gbrain check-backlinks <check|fix> [dir] [--dir <brain-dir>] [--dry-run]');
|
||||
console.error('Usage: gbrain check-backlinks <check|fix> [--dir <brain-dir>] [--dry-run]');
|
||||
console.error(' check Report missing back-links');
|
||||
console.error(' fix Create missing back-links (appends to Timeline)');
|
||||
console.error(' dir Brain directory (default: current directory)');
|
||||
console.error(' --dir Brain directory override');
|
||||
console.error(' --dir Brain directory (default: current directory)');
|
||||
console.error(' --dry-run Preview fixes without writing');
|
||||
process.exit(1);
|
||||
}
|
||||
|
||||
@@ -1743,7 +1743,15 @@ async function extractStaleFromDB(
|
||||
// `page.updated_at.toISOString()` — the JS Date is ms-truncated, so the
|
||||
// µs-precision DB updated_at stayed strictly greater and the page never
|
||||
// cleared on Postgres. Stamping the exact value makes them equal.
|
||||
processedRefs.push({ slug: page.slug, source_id: page.source_id, extractedAt: page.updated_at_iso });
|
||||
//
|
||||
// Version-arm floor: a page last edited BEFORE LINK_EXTRACTOR_VERSION_TS
|
||||
// would otherwise be stamped below the version watermark and stay
|
||||
// permanently stale (`links_extracted_at < versionTs` re-fires every run).
|
||||
// Stamp max(updated_at, versionTs) — versionTs is always a past release
|
||||
// date, so a concurrent edit's now() still exceeds the stamp and D4 holds.
|
||||
// Tie at ms precision picks updated_at_iso (its µs ≥ versionTs's .000000).
|
||||
const stampTs = new Date(page.updated_at_iso) >= new Date(versionTs) ? page.updated_at_iso : versionTs;
|
||||
processedRefs.push({ slug: page.slug, source_id: page.source_id, extractedAt: stampTs });
|
||||
}
|
||||
|
||||
// Flush NON-swallowing (CDX-4): a throw here propagates out of the sweep so
|
||||
|
||||
@@ -28,7 +28,7 @@ import { ensureWellFormed } from './text-safe.ts';
|
||||
* OR updated_at > links_extracted_at`. It is an ISO-8601 string (NOT a number) —
|
||||
* the column is TIMESTAMPTZ and the predicate binds it as `::timestamptz`.
|
||||
*/
|
||||
export const LINK_EXTRACTOR_VERSION_TS = '2026-05-31T00:00:00Z';
|
||||
export const LINK_EXTRACTOR_VERSION_TS = '2026-07-21T00:00:00Z';
|
||||
|
||||
// ─── Entity references ──────────────────────────────────────────
|
||||
|
||||
@@ -80,10 +80,10 @@ export type LinkResolutionType = 'qualified' | 'unqualified';
|
||||
* Directory prefix whitelist. These are the top-level slug dirs the extractor
|
||||
* recognizes as entity references. Upstream canonical + our extensions:
|
||||
* - Gbrain canonical: people, companies, meetings, concepts, deal, civic, project, source, media, yc, projects
|
||||
* - Our domain extensions: tech, finance, personal, openclaw (domain-organized wikis)
|
||||
* - Our domain extensions: tech, finance, personal, openclaw, ops (domain-organized wikis)
|
||||
* - Our entity prefix: entities (we kept some legacy entities/projects/ pages)
|
||||
*/
|
||||
const DIR_PATTERN = '(?:people|companies|meetings|concepts|deal|civic|project|projects|source|media|yc|tech|finance|personal|openclaw|entities)';
|
||||
const DIR_PATTERN = '(?:people|companies|meetings|concepts|deal|civic|project|projects|source|media|yc|tech|finance|personal|openclaw|entities|ops)';
|
||||
|
||||
/**
|
||||
* Match `[Name](path)` markdown links pointing to entity directories.
|
||||
@@ -865,7 +865,16 @@ export function queryBasenameIndex(idx: Map<string, string[]>, name: string): st
|
||||
if (!name || typeof name !== 'string') return [];
|
||||
const trimmed = name.trim();
|
||||
if (!trimmed) return [];
|
||||
const hit = idx.get(trimmed) ?? idx.get(trimmed.toLowerCase()) ?? idx.get(normalizeBasename(trimmed));
|
||||
let hit = idx.get(trimmed) ?? idx.get(trimmed.toLowerCase()) ?? idx.get(normalizeBasename(trimmed));
|
||||
// Issue #2576 bug 2: path-style refs (`runbooks/2026-05-01-x`) from dirs
|
||||
// outside DIR_PATTERN reach here, but normalizeBasename strips slashes
|
||||
// into a garbage key (`runbooks2026-05-01-x`) that can never hit the
|
||||
// tail-keyed index. Fall back to the path tail so qualified refs resolve
|
||||
// by basename like everything else.
|
||||
if (!hit && trimmed.includes('/')) {
|
||||
const tail = trimmed.slice(trimmed.lastIndexOf('/') + 1).trim();
|
||||
if (tail) hit = idx.get(tail) ?? idx.get(tail.toLowerCase()) ?? idx.get(normalizeBasename(tail));
|
||||
}
|
||||
return hit ? [...hit].sort(basenameSort) : [];
|
||||
}
|
||||
|
||||
|
||||
@@ -4,7 +4,6 @@ import {
|
||||
extractPageTitle,
|
||||
hasBacklink,
|
||||
buildBacklinkEntry,
|
||||
parseBacklinksArgs,
|
||||
} from '../src/commands/backlinks.ts';
|
||||
|
||||
describe('extractEntityRefs', () => {
|
||||
@@ -105,26 +104,3 @@ describe('findBacklinkGaps dedupe (v0.36.x #967 regression)', () => {
|
||||
}
|
||||
});
|
||||
});
|
||||
|
||||
describe('parseBacklinksArgs', () => {
|
||||
test('uses positional dir for check and fix subcommands', () => {
|
||||
expect(parseBacklinksArgs(['check', '/tmp/brain']).brainDir).toBe('/tmp/brain');
|
||||
expect(parseBacklinksArgs(['fix', '/tmp/brain']).brainDir).toBe('/tmp/brain');
|
||||
});
|
||||
|
||||
test('defaults to cwd when no dir given', () => {
|
||||
expect(parseBacklinksArgs(['check']).brainDir).toBe('.');
|
||||
});
|
||||
|
||||
test('--dir overrides positional dir and preserves dry-run', () => {
|
||||
const parsed = parseBacklinksArgs(['fix', '/tmp/ignored', '--dir', '/tmp/brain', '--dry-run']);
|
||||
expect(parsed.subcommand).toBe('fix');
|
||||
expect(parsed.brainDir).toBe('/tmp/brain');
|
||||
expect(parsed.dryRun).toBe(true);
|
||||
});
|
||||
|
||||
test('--dir missing its value falls back to positional dir', () => {
|
||||
expect(parseBacklinksArgs(['check', '/tmp/brain', '--dir']).brainDir).toBe('/tmp/brain');
|
||||
expect(parseBacklinksArgs(['check', '--dir', '--dry-run']).brainDir).toBe('.');
|
||||
});
|
||||
});
|
||||
|
||||
@@ -209,6 +209,24 @@ describe('gbrain extract --stale', () => {
|
||||
expect(usRows[0]?.eq).toBe(true);
|
||||
});
|
||||
|
||||
test('version-arm floor: page edited BEFORE LINK_EXTRACTOR_VERSION_TS clears after --stale (issue #2576 bug 3)', async () => {
|
||||
// A page whose updated_at predates the version watermark used to be
|
||||
// stamped at its updated_at (< versionTs), so the version arm re-fired
|
||||
// every run — permanently stale. The sweep now floors the stamp at
|
||||
// versionTs. (The #1768 test above also covers this since the v0.42.x
|
||||
// VERSION_TS bump moved its date below the watermark, but this pins the
|
||||
// behavior explicitly so a date "repair" there can't drop coverage.)
|
||||
await engine.putPage('people/alice', personPage('Alice'));
|
||||
await engine.executeRaw(`UPDATE pages SET updated_at = '2000-01-01T00:00:00Z' WHERE slug = 'people/alice'`);
|
||||
expect(await engine.countStalePagesForExtraction({ versionTs: LINK_EXTRACTOR_VERSION_TS })).toBe(1);
|
||||
|
||||
await runExtract(engine, ['--stale']);
|
||||
// Pre-floor this stayed 1 forever (stamp < versionTs → version arm re-fires).
|
||||
expect(await engine.countStalePagesForExtraction({ versionTs: LINK_EXTRACTOR_VERSION_TS })).toBe(0);
|
||||
await runExtract(engine, ['--stale']);
|
||||
expect(await engine.countStalePagesForExtraction({ versionTs: LINK_EXTRACTOR_VERSION_TS })).toBe(0);
|
||||
});
|
||||
|
||||
test('CDX-4 (D2): a link-flush throw aborts the sweep and leaves pages UNSTAMPED', async () => {
|
||||
await engine.putPage('people/alice', personPage('Alice'));
|
||||
await engine.putPage('companies/acme', companyPage('Acme', '[Alice](people/alice) founded [Acme](companies/acme).'));
|
||||
|
||||
@@ -140,6 +140,15 @@ describe('extractEntityRefs', () => {
|
||||
expect(wikiRefs[0].needsResolution).toBe(true);
|
||||
});
|
||||
|
||||
test('recognizes ops/ qualified wikilinks (issue #2576 bug 2)', () => {
|
||||
// `ops` was missing from DIR_PATTERN, so [[ops/...]] fell through to
|
||||
// the generic 2c pass (needsResolution) instead of being a real ref.
|
||||
const refs = extractEntityRefs('Deployed via [[ops/services/pointer-agent]].');
|
||||
expect(refs.length).toBe(1);
|
||||
expect(refs[0].slug).toBe('ops/services/pointer-agent');
|
||||
expect(refs[0].needsResolution).toBeUndefined();
|
||||
});
|
||||
|
||||
test('skips qualified-syntax tokens (those belong to 2a)', () => {
|
||||
// [[wiki:topics/ai]] looks like 2a's qualified shape — even though
|
||||
// it wouldn't satisfy DIR_PATTERN, 2c must not claim it either
|
||||
@@ -1069,6 +1078,15 @@ describe('makeResolver — fallback chain', () => {
|
||||
]);
|
||||
});
|
||||
|
||||
test('resolveBasenameMatches: path-style ref falls back to the tail (issue #2576 bug 2)', async () => {
|
||||
// normalizeBasename strips slashes, so `runbooks/2026-05-01-pointer-agent`
|
||||
// used to normalize to a garbage key that never hit the tail-keyed index.
|
||||
const engine = makeFakeEngineWithSlugs(['ops/changes/2026-05-01-pointer-agent']);
|
||||
const r = makeResolver(engine);
|
||||
expect(await r.resolveBasenameMatches!('runbooks/2026-05-01-pointer-agent'))
|
||||
.toEqual(['ops/changes/2026-05-01-pointer-agent']);
|
||||
});
|
||||
|
||||
test('resolveBasenameMatches: case-insensitive fallback', async () => {
|
||||
const engine = makeFakeEngineWithSlugs(['companies/fast-weigh']);
|
||||
const r = makeResolver(engine);
|
||||
|
||||
Reference in New Issue
Block a user