feat(check): 文件要在、扫描件兜底(C10);响应方式存进清单并可改(C12)

- check/presence.py:「提供《授权委托书》」「须提交投标函」这类要求按标题里有没有这个文件名判断,不调模型。
- 扫描件兜底:要附证明材料的要求没找到、投标文件里有扫描件页时,改为提示级「可能在扫描件里,请人工核对」。
- 响应方式解析时识别、带依据句子存进 checklist.response_modes(迁移 0008);清单页显示、可改(写审计,
  重新解析不覆盖人工改过的);查标用存下来的。新接口 PUT /projects/{id}/checklist/response-modes/{category}。
- check_item_result.layer 加 presence(迁移 0007);契约同步默认接受的两种结论。

开发集 17 份:规矩投标文件的致命问题 120 → 51,致命级召回不变(22 / 32)(docs/architecture/check-eval.md §5)。

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
YANG JIANKUAN
2026-11-04 15:53:45 +08:00
parent cb0c2fcafe
commit 47c6a1258e
21 changed files with 529 additions and 24 deletions

View File

@@ -129,7 +129,15 @@ async def main() -> None:
docs = []
for n, pdf in enumerate(sorted(d.glob("v*.pdf")), start=1):
pages = text_layer.extract(f"{args.bid}-{pdf.stem}", pdf.read_bytes())
docs.append(match.BidDoc(uuid4(), n, pdf.stem, text_layer.body_sentences(pages)))
docs.append(
match.BidDoc(
uuid4(),
n,
pdf.stem,
text_layer.body_sentences(pages),
[p.page_no for p in pages if p.text_source == "empty"],
)
)
ai_service.run_json = baseline_run_json if args.baseline else cached_run_json # type: ignore[assignment]
if args.v1:
match.PROMPT = ("check.match", 1)
@@ -148,6 +156,7 @@ async def main() -> None:
results, stats = await match.run(ctx, uuid4(), items, docs, progress)
if decisions:
match.apply_modes(results, items, decisions)
match.scanned_fallback(results, items, decisions, docs)
rows = []
for item_id, item in all_items.items():
if item_id not in items: