feat: phase 3 of AI pipeline hardening — extraction quality
All checks were successful
On Pull Request to Main / test (pull_request) Successful in 31s
On Pull Request to Main / publish (pull_request) Successful in 1m1s
On Pull Request to Main / deploy-dev (pull_request) Successful in 1m32s

Replace stateless one-shot extraction with a stateful, paced, cancellable
pipeline. Six subtasks:

- 3.1 Sentence-aware chunking with 800-char overlap (was paragraph-only
  at 4000 chars). Hard-split fallback for runaway sentences.
- 3.2 Stateful extraction: chunks 2+ receive an "already-extracted topic
  IDs" hint capped at 200 IDs, so the model reuses IDs instead of
  inventing variants like software-developer vs software-engineer.
- 3.3 Token-bucket limiter in llmRetry.js (extractionLimiter, 5 req/min).
  callLLM awaits the limiter before fetch; 429+Retry-After calls
  pauseUntil. Replaces hard setTimeout(12000) and setTimeout(15000).
- 3.4 relevance_locked column on topics — admin edits to relevance are
  sticky across re-extraction. Migration + merge respects the flag +
  unlock checkbox in KnowledgeGraph edit form.
- 3.5 Unify relation vocabulary — handbook prompt no longer mentions
  legacy "executes"; one-shot migration rewrites existing executes rows
  to executed_by with source/target swapped.
- 3.6 Cancellation — Cancel button on UploadZone wired to an
  AbortController threaded into callLLM; aborted runs persist status =
  "cancelled" rather than "failed".

Tests: 16 new unit tests for chunkText, buildKnownIdsHint, and
createLimiter. All 61 tests pass, 0 lint errors, build clean.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
RaymondVerhoef
2026-05-20 17:56:45 +02:00
parent 9771928926
commit aeb197d5f4
10 changed files with 509 additions and 48 deletions

View File

@@ -0,0 +1,23 @@
/// <reference path="../pb_data/types.d.ts" />
migrate((app) => {
const collection = app.findCollectionByNameOrId("pbc_2800040823")
// add field — relevance_locked is set to true whenever an admin edits
// learning_relevance via the UI; mergeKnowledgeGraph must never overwrite
// learning_relevance on a locked topic during re-extraction.
collection.fields.addAt(5, new Field({
"hidden": false,
"id": "bool_relevance_locked",
"name": "relevance_locked",
"presentable": false,
"required": false,
"system": false,
"type": "bool"
}))
return app.save(collection)
}, (app) => {
const collection = app.findCollectionByNameOrId("pbc_2800040823")
collection.fields.removeById("bool_relevance_locked")
return app.save(collection)
})

View File

@@ -0,0 +1,43 @@
/// <reference path="../pb_data/types.d.ts" />
// One-shot data migration: rewrite legacy "executes" relations to the
// canonical "executed_by" vocabulary by swapping source and target.
// Previously `role --executes--> process`; canonical is
// `process --executed_by--> role`.
migrate((app) => {
const records = app.findRecordsByFilter(
"pbc_1883724256", // relations collection
'type = "executes"',
'',
0,
0,
)
for (const rec of records) {
const source = rec.get("source")
const target = rec.get("target")
rec.set("type", "executed_by")
rec.set("source", target)
rec.set("target", source)
app.save(rec)
}
}, (app) => {
// Reverse: turn executed_by back into executes (best-effort — only those
// created before this migration would have been "executes"; rolling back
// will affect any newer executed_by rows too).
const records = app.findRecordsByFilter(
"pbc_1883724256",
'type = "executed_by"',
'',
0,
0,
)
for (const rec of records) {
const source = rec.get("source")
const target = rec.get("target")
rec.set("type", "executes")
rec.set("source", target)
rec.set("target", source)
app.save(rec)
}
})