Chaim
c348903e4b
31 שורות case_law עם source_kind='cited_only' (ציטוט-בלבד, ללא full_text/chunks) נושאות halacha_extraction_status='pending' רק כברירת-מחדל ומזהמות את מונה ה-pending ובמתזמר/בדף-התפעול — אין להן מה לחלץ. תיקון (G1 — תיקון-במקור, G2 — מסנן יחיד משותף): - db.EXTRACTION_ELIGIBLE_PREDICATE — מקור-אמת יחיד ל"שורה ברת-חילוץ" (source_kind <> 'cited_only' AND יש precedent_chunks). מוחל ב-list_pending_extraction_requests; #139 יעשה בו שימוש-חוזר ל-reconcile (אותו כלל, לא כפול). - מוני-snapshot מסננים cited_only: halacha_drain_supervisor.db_snapshot, web/app.py meta+hal_ext (GROUP BY status). - reconcile_metadata_status.py מורחב לכסות גם את תור-ההלכות: cited_only→'skipped' (אותו terminal-state כמו צד-המטא, תור-תאום, G2). בוצע על ה-DB החי: 31 הועברו ל-'skipped' (metadata כבר היה מיושב — אידמפוטנטי). התפלגות-אחרי: halacha pending=9 (עבודה אמיתית), skipped=31, completed=309. בדיקות: test_extraction_queue_eligibility (predicate + list_pending מחיל אותו, שני ה-kinds). כל 345 בדיקות mcp עוברות. guards נקיים. Invariants: G1 (terminal-state אמיתי במקור), G2 (predicate יחיד, ללא תור מקביל), INV-DM1 (stub לא-searchable אינו מועמד-חילוץ), G12 (leak-guard נקי). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
67 lines
2.1 KiB
Python
67 lines
2.1 KiB
Python
"""Regression test for #140 — cited_only stubs must never enter the extraction
|
|
work queue.
|
|
|
|
``list_pending_extraction_requests`` must apply ``EXTRACTION_ELIGIBLE_PREDICATE``
|
|
so a citation-only stub (no full_text, no precedent_chunks) is excluded even if
|
|
it carries a stamped ``*_extraction_requested_at`` and a default 'pending'
|
|
status. The predicate is the single shared eligibility rule (#139 reuses it).
|
|
|
|
Runs OFFLINE — a fake pool captures the SQL and asserts the predicate is wired
|
|
into the WHERE clause (same style as test_halacha_reextract_preserves_approved).
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import asyncio
|
|
|
|
import pytest
|
|
|
|
from legal_mcp.services import db
|
|
|
|
|
|
class _FakePool:
|
|
def __init__(self) -> None:
|
|
self.fetched: list[str] = []
|
|
|
|
async def fetch(self, sql: str, *args): # noqa: ANN002
|
|
self.fetched.append(sql)
|
|
return []
|
|
|
|
|
|
@pytest.fixture()
|
|
def fake_pool(monkeypatch: pytest.MonkeyPatch) -> _FakePool:
|
|
pool = _FakePool()
|
|
|
|
async def _get_pool() -> _FakePool:
|
|
return pool
|
|
|
|
monkeypatch.setattr(db, "get_pool", _get_pool)
|
|
return pool
|
|
|
|
|
|
def _norm(sql: str) -> str:
|
|
return " ".join(sql.split())
|
|
|
|
|
|
def test_predicate_excludes_cited_only_and_requires_chunks() -> None:
|
|
pred = _norm(db.EXTRACTION_ELIGIBLE_PREDICATE)
|
|
assert "source_kind <> 'cited_only'" in pred
|
|
assert "precedent_chunks" in pred and "EXISTS" in pred.upper()
|
|
|
|
|
|
@pytest.mark.parametrize("kind", ["metadata", "halacha"])
|
|
def test_list_pending_applies_eligibility_predicate(fake_pool: _FakePool, kind: str) -> None:
|
|
loop = asyncio.new_event_loop()
|
|
try:
|
|
loop.run_until_complete(db.list_pending_extraction_requests(kind=kind))
|
|
finally:
|
|
loop.close()
|
|
|
|
assert fake_pool.fetched, "expected a queue query"
|
|
sql = _norm(fake_pool.fetched[0])
|
|
# The eligibility predicate must be ANDed into the queue WHERE clause.
|
|
assert _norm(db.EXTRACTION_ELIGIBLE_PREDICATE) in sql, sql
|
|
# ...alongside the requested_at gate, for the correct kind.
|
|
col = "metadata_extraction_requested_at" if kind == "metadata" else "halacha_extraction_requested_at"
|
|
assert f"{col} IS NOT NULL" in sql, sql
|