feat(principles): decision-level panel extraction regime — cap-5 + dedup-frees-slot (Phase B, #152)
extract() routes to _extract_via_panel when HALACHA_PANEL_REGIME_ENABLED: the 3-model panel proposes → votes/score → approval rule → dedup vs corpus (known links as citation, frees a cap slot) → cap HALACHA_PANEL_MAX_NEW genuinely-new principles/decision (by score), rest dropped. Replaces single-model auto-approve; legacy path kept as <2-judge fallback. db.store_panel_principles persists the pre-decided verdict + source-aware canonical create/link (G9 reviewer=panel:...). Dry-run validated on 29468-08-23: ~18 → 4 principles. 6 new tests; full suite 422 green. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -31,7 +31,7 @@ import asyncpg
|
||||
from legal_mcp import config
|
||||
from legal_mcp.config import parse_llm_json
|
||||
from legal_mcp.services import (
|
||||
claude_session, db, embeddings, halacha_quality, proofreader,
|
||||
claude_session, db, embeddings, halacha_quality, panel_extraction, proofreader,
|
||||
)
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
@@ -603,6 +603,15 @@ async def extract(case_law_id: UUID | str, force: bool = False,
|
||||
stop_keepalive = asyncio.Event()
|
||||
keepalive_task = asyncio.create_task(_lock_keepalive(lock_conn, stop_keepalive))
|
||||
try:
|
||||
if config.HALACHA_PANEL_REGIME_ENABLED:
|
||||
# #152 Phase B — decision-level 3-model panel (votes+cap+source label).
|
||||
res = await _extract_via_panel(case_law_id, force=force)
|
||||
if res is not None:
|
||||
return res
|
||||
# panel unavailable (all judges down) → degrade to legacy path so
|
||||
# extraction still makes progress instead of stalling the queue.
|
||||
logger.warning("panel regime returned no result for %s — "
|
||||
"falling back to legacy per-chunk extraction", case_law_id)
|
||||
return await _extract_impl(case_law_id, force=force, effort=effort)
|
||||
finally:
|
||||
# Stop the keepalive and await it BEFORE reusing lock_conn for unlock —
|
||||
@@ -894,3 +903,122 @@ async def _extract_impl(case_law_id: UUID, force: bool = False,
|
||||
"preserved_approved": preserved_approved,
|
||||
"total_chunks": len(chunks),
|
||||
}
|
||||
|
||||
|
||||
# Cap the text sent to each judge — extractable chunks already exclude
|
||||
# facts/intro/arguments, but a very long decision could still blow a context.
|
||||
_PANEL_MAX_CHARS = 80_000
|
||||
|
||||
# Canonical states a new extraction may dedup-link against (frees a cap slot).
|
||||
_PANEL_DEDUP_STATES = ("pending_synthesis", "pending_review", "approved", "published")
|
||||
|
||||
|
||||
async def _extract_via_panel(
|
||||
case_law_id: UUID, force: bool = False, dry_run: bool = False,
|
||||
) -> dict | None:
|
||||
"""Decision-level tri-model panel extraction (#152, Phase B).
|
||||
|
||||
Replaces the per-chunk single-model auto-approve with: 3 models propose →
|
||||
cross-model votes + mean-score → chair's approval rule → dedup vs corpus
|
||||
(link known → frees a slot) → cap of HALACHA_PANEL_MAX_NEW genuinely-new
|
||||
principles per decision (by score). A principle from a binding higher court
|
||||
is a הלכה; from the appeals committee a כלל פרשני (labeling via source).
|
||||
|
||||
Returns the result dict, or **None** when fewer than 2 judges are reachable
|
||||
(caller falls back to the legacy path so the queue still drains). ``dry_run``
|
||||
computes the full plan WITHOUT writing — used for validation/chair preview.
|
||||
"""
|
||||
record = await db.get_case_law(case_law_id)
|
||||
if not record:
|
||||
return {"status": "not_found", "extracted": 0, "stored": 0}
|
||||
|
||||
# Idempotency: panel extraction is decision-level (no per-chunk checkpoints).
|
||||
# Without force, skip if this decision already has halachot (avoid dup re-run).
|
||||
if not force and not dry_run:
|
||||
existing = await db.list_halachot(case_law_id=case_law_id, limit=1)
|
||||
if existing:
|
||||
total = len(await db.list_halachot(case_law_id=case_law_id, limit=10_000))
|
||||
await db.set_case_law_halacha_status(case_law_id, "completed")
|
||||
return {"status": "completed", "extracted": total, "stored": total,
|
||||
"resumed": True, "panel": True}
|
||||
|
||||
source_kind = record.get("source_kind") or "external_upload"
|
||||
is_binding = bool(record.get("is_binding"))
|
||||
full_text = record.get("full_text") or ""
|
||||
|
||||
chunks, used_fallback = await _select_extractable_chunks(case_law_id)
|
||||
if not chunks:
|
||||
if not dry_run:
|
||||
await db.set_case_law_halacha_status(case_law_id, "completed")
|
||||
return {"status": "no_chunks", "extracted": 0, "stored": 0, "panel": True}
|
||||
|
||||
preserved = 0
|
||||
if force and not dry_run:
|
||||
reset = await db.reset_halacha_extraction(case_law_id)
|
||||
preserved = reset.get("preserved", 0)
|
||||
|
||||
if not dry_run:
|
||||
await db.set_case_law_halacha_status(case_law_id, "processing")
|
||||
|
||||
text = "\n\n".join(c["content"] for c in chunks)[:_PANEL_MAX_CHARS]
|
||||
clusters = await panel_extraction.panel_extract(
|
||||
text, source_kind=source_kind, is_binding=is_binding,
|
||||
)
|
||||
if not clusters:
|
||||
# distinguish "judges down" (→ fallback) from "genuinely nothing found".
|
||||
if sum(panel_extraction.panel_judges.available().values()) < 2:
|
||||
return None
|
||||
if not dry_run:
|
||||
await db.mark_all_chunks_extracted(case_law_id)
|
||||
await db.set_case_law_halacha_status(case_law_id, "completed")
|
||||
return {"status": "completed", "extracted": 0, "stored": 0,
|
||||
"new": 0, "linked": 0, "panel": True, "preserved_approved": preserved}
|
||||
|
||||
kept = [c for c in clusters if c["verdict"] in ("approved", "pending_review")]
|
||||
max_new = config.HALACHA_PANEL_MAX_NEW
|
||||
new_count = linked = dropped_cap = 0
|
||||
to_store: list[dict] = []
|
||||
for c in kept: # strongest first (panel_extract sorts by votes,score)
|
||||
emb = c.get("embedding")
|
||||
canonical_id, instance_type = None, "original"
|
||||
if emb is not None and config.HALACHA_CANONICAL_LOOKUP_ENABLED:
|
||||
match = await db.nearest_canonical_halacha(
|
||||
emb, threshold=config.HALACHA_CANONICAL_THRESHOLD,
|
||||
status_filter=_PANEL_DEDUP_STATES,
|
||||
)
|
||||
if match:
|
||||
canonical_id, instance_type = match[0], "citation"
|
||||
if instance_type == "original":
|
||||
if new_count >= max_new: # cap: linked don't count, only new
|
||||
dropped_cap += 1
|
||||
continue
|
||||
new_count += 1
|
||||
else:
|
||||
linked += 1
|
||||
to_store.append({
|
||||
"rule_statement": c["rule_statement"], "supporting_quote": c["supporting_quote"],
|
||||
"reasoning_summary": c["reasoning_summary"], "rule_type": c["rule_type"],
|
||||
"confidence": c["score"], "score": c["score"], "votes": c["votes"],
|
||||
"voters": c["voters"], "review_status": c["verdict"], "embedding": emb,
|
||||
"instance_type": instance_type, "canonical_id": canonical_id,
|
||||
"quote_verified": _verify_quote(c["supporting_quote"], full_text),
|
||||
})
|
||||
|
||||
if dry_run:
|
||||
return {"status": "dry_run", "panel": True, "source_kind": source_kind,
|
||||
"candidates": clusters, "to_store": to_store,
|
||||
"new": new_count, "linked": linked, "dropped_over_cap": dropped_cap}
|
||||
|
||||
res = await db.store_panel_principles(case_law_id, to_store)
|
||||
await db.mark_all_chunks_extracted(case_law_id)
|
||||
await db.set_case_law_halacha_status(case_law_id, "completed")
|
||||
total = len(await db.list_halachot(case_law_id=case_law_id, limit=10_000))
|
||||
logger.info(
|
||||
"halacha panel: case_law=%s (%s) — %d new + %d linked stored, "
|
||||
"%d dropped over cap-%d", case_law_id, source_kind,
|
||||
res["created_new"], res["linked"], dropped_cap, max_new,
|
||||
)
|
||||
return {"status": "completed", "extracted": total,
|
||||
"stored": res["created_new"] + res["linked"], "new": res["created_new"],
|
||||
"linked": res["linked"], "dropped_over_cap": dropped_cap,
|
||||
"panel": True, "preserved_approved": preserved}
|
||||
|
||||
Reference in New Issue
Block a user