feat(regulatory): 按实际处理数量更新节点进度

This commit is contained in:
2026-06-07 13:32:06 +08:00
parent 32d258bb75
commit 3e8720e521
5 changed files with 286 additions and 62 deletions

View File

@@ -2,6 +2,7 @@ from __future__ import annotations
import re
from collections import defaultdict
from collections.abc import Callable
from review_agent.regulatory_review.schemas import Finding
@@ -17,27 +18,40 @@ FIELDS = {
}
def run_consistency_check(document_texts: dict[str, str]) -> list[Finding]:
def run_consistency_check(
document_texts: dict[str, str],
progress_callback: Callable[[dict[str, object]], None] | None = None,
) -> list[Finding]:
findings: list[Finding] = []
for label, pattern in FIELDS.items():
fields = list(FIELDS.items())
total = len(fields)
for index, (label, pattern) in enumerate(fields, start=1):
values: dict[str, list[str]] = defaultdict(list)
for file_name, text in document_texts.items():
match = re.search(pattern, text)
if match:
values[_normalize(match.group(1))].append(file_name)
if len(values) <= 1:
continue
findings.append(
Finding(
rule_code=f"consistency:{label}",
category="consistency",
severity="high",
title=f"{label}在不同文件中不一致",
detail=f"发现 {len(values)} 个不同{label}取值",
suggestion=f"请统一各注册资料中的{label}",
evidence={"field": label, "values": dict(values)},
if len(values) > 1:
findings.append(
Finding(
rule_code=f"consistency:{label}",
category="consistency",
severity="high",
title=f"{label}在不同文件中不一致",
detail=f"发现 {len(values)} 个不同的{label}取值。",
suggestion=f"请统一各注册资料中{label}",
evidence={"field": label, "values": dict(values)},
)
)
if progress_callback:
progress_callback(
{
"processed": index,
"total": total,
"label": label,
"finding_count": len(findings),
}
)
)
return findings