"""Govdoc 公文模块服务实现（阶段骨架）。 本文件为 Phase 1 骨架实现，所有方法暂返回占位结果。 后续步骤将逐步接入： - govdoc_bridge 执行桥接 - govdoc_engine 引擎内核 - 文档主档复用 - OSS / Celery 集成 """ from __future__ import annotations from typing import Any from fastapi import UploadFile from fastapi_common.fastapi_common_logger import logger from fastapi_modules.fastapi_leaudit.services import IGovdocService class GovdocServiceImpl(IGovdocService): """公文处理与格式审查服务实现。""" # ── 文档 ────────────────────────────────────────────── async def UploadDocument( self, file: UploadFile, typeId: int | None = None, region: str = "default", autoRun: bool = False, speed: str = "normal", ruleVersionId: int | None = None, createdBy: int | None = None, ) -> dict[str, Any]: logger.info("[Govdoc] UploadDocument placeholder — file=%s region=%s", file.filename, region) return { "documentId": 0, "fileId": 0, "fileName": file.filename, "region": region, "engineType": "govdoc", "autoRunTriggered": autoRun, } async def ListDocuments( self, page: int = 1, pageSize: int = 20, keyword: str | None = None, region: str | None = None, status: str | None = None, resultStatus: str | None = None, createdBy: int | None = None, dateFrom: str | None = None, dateTo: str | None = None, userId: int | None = None, ) -> dict[str, Any]: logger.info("[Govdoc] ListDocuments placeholder — page=%s pageSize=%s", page, pageSize) return {"items": [], "total": 0, "page": page, "pageSize": pageSize} async def GetDocumentDetail(self, documentId: int, userId: int | None = None) -> dict[str, Any]: logger.info("[Govdoc] GetDocumentDetail placeholder — id=%s", documentId) return {"documentId": documentId} async def UpdateDocument(self, documentId: int, body: dict[str, Any], userId: int | None = None) -> dict[str, Any]: logger.info("[Govdoc] UpdateDocument placeholder — id=%s", documentId) return {"documentId": documentId, **body} async def DeleteDocument(self, documentId: int, userId: int | None = None) -> dict[str, Any]: logger.info("[Govdoc] DeleteDocument placeholder — id=%s", documentId) return {"documentId": documentId, "deleted": True} # ── 审查运行 ────────────────────────────────────────── async def CreateRun( self, documentId: int, ruleVersionId: int | None = None, speed: str = "normal", force: bool = False, triggerUserId: int | None = None, ) -> dict[str, Any]: logger.info("[Govdoc] CreateRun placeholder — documentId=%s", documentId) return { "runId": 0, "documentId": documentId, "status": "queued", "phase": "dispatch", } async def GetRunStatus(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetRunStatus placeholder — runId=%s", runId) return {"runId": runId, "status": "pending"} # ── 结果与报告 ──────────────────────────────────────── async def GetRunResult(self, runId: int) -> dict[str, Any]: """从 govdoc_runs + govdoc_rule_results 读取审查结果，含 structure/outline。""" import json as _json from fastapi_common.fastapi_common_sqlalchemy.database import GetAsyncSession from sqlalchemy import text async with GetAsyncSession() as session: run_row = await session.execute( text( """SELECT id, document_id, status, phase, total_score, passed_count, failed_count, skipped_count, result_status, result_summary_json, rules_path, started_at, finished_at FROM govdoc_runs WHERE id = :rid""" ), {"rid": runId}, ) run_data = run_row.mappings().first() if not run_data: return {"runId": runId, "summary": {}, "checkedRules": [], "findings": [], "structure": [], "outline": [], "entities": {}} rules_rows = await session.execute( text( """SELECT rule_id, rule_name, severity, category, result, skip_reason, message, suggestion, actual, expected, evidence, paragraph_index, paragraph_text, location_path, score FROM govdoc_rule_results WHERE run_id = :rid""" ), {"rid": runId}, ) rule_results = [dict(r._mapping) for r in rules_rows.fetchall()] aux_raw = run_data.get("result_summary_json") aux = {} if aux_raw: try: aux = _json.loads(aux_raw) if isinstance(aux_raw, str) else aux_raw except (TypeError, _json.JSONDecodeError): pass findings = [] for rr in rule_results: loc = {} if rr.get("paragraph_index") is not None: loc["paragraph_index"] = rr["paragraph_index"] if rr.get("paragraph_text"): loc["context"] = rr["paragraph_text"] if rr.get("location_path"): loc["role"] = rr["location_path"] findings.append({ "finding_id": f"{rr['rule_id']}-{rr['paragraph_index'] or 0}", "rule_id": rr["rule_id"], "rule_name": rr["rule_name"], "severity": rr["severity"], "category": rr["category"], "location": loc if loc else None, "actual": rr.get("actual") or {}, "expected": rr.get("expected") or {}, "message": rr.get("message") or "", "suggestion": rr.get("suggestion") or "", "evidence": rr.get("evidence") or "", "confidence": 1.0, }) checked_rules = [] seen = set() for rr in rule_results: rid = rr["rule_id"] if rid in seen: continue seen.add(rid) status = rr.get("result", "pass") checked_rules.append({ "rule_id": rid, "name": rr["rule_name"], "severity": rr["severity"], "category": rr["category"], "status": status if status in ("pass", "fail", "skipped") else "pass", "skip_reason": rr.get("skip_reason"), }) return { "runId": runId, "summary": { "score": run_data.get("total_score", 100), "total_findings": len(findings), "by_severity": {}, "by_category": {}, "passed_count": run_data.get("passed_count", 0), "failed_count": run_data.get("failed_count", 0), "skipped_count": run_data.get("skipped_count", 0), }, "checkedRules": checked_rules, "findings": findings, "structure": aux.get("structure", []), "outline": aux.get("outline", []), "entities": {}, } async def GetRunFindings(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetRunFindings placeholder — runId=%s", runId) return {"runId": runId, "findings": []} async def GetRunEntities(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetRunEntities placeholder — runId=%s", runId) return {"runId": runId, "entities": []} async def GetRunParagraphs(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetRunParagraphs placeholder — runId=%s", runId) return {"runId": runId, "paragraphs": []} async def GetRunStructure(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetRunStructure placeholder — runId=%s", runId) return {"runId": runId, "structure": []} async def GetRunOutline(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetRunOutline placeholder — runId=%s", runId) return {"runId": runId, "outline": []} async def GetReportHtml(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetReportHtml placeholder — runId=%s", runId) return {"runId": runId, "htmlUrl": ""} async def GetReportDocx(self, runId: int) -> dict[str, Any]: logger.info("[Govdoc] GetReportDocx placeholder — runId=%s", runId) return {"runId": runId, "docxUrl": ""} async def DownloadOriginal(self, documentId: int) -> dict[str, Any]: logger.info("[Govdoc] DownloadOriginal placeholder — documentId=%s", documentId) return {"documentId": documentId, "downloadUrl": ""} # ── 规则 ────────────────────────────────────────────── async def ListRules(self, rulesPath: str | None = None) -> dict[str, Any]: """从 govdoc 规则 YAML 文件加载规则清单。""" rules = await self._load_rules_list(rulesPath) return {"rules": rules, "total_rules": len(rules)} async def GetRuleDetail(self, ruleId: str, rulesPath: str | None = None) -> dict[str, Any]: """获取单条规则完整详情（名称、严重度、stages、消息等）。""" ruleset = await self._load_ruleset(rulesPath) if ruleset is None: return {"rule_id": ruleId, "name": ruleId, "severity": "info", "category": "", "group": ""} for rule in ruleset.all_rules(): if rule.rule_id == ruleId: return { "rule_id": rule.rule_id, "name": rule.name, "severity": rule.severity, "category": rule.category, "group": "", "applies_to": rule.applies_to.model_dump() if rule.applies_to else None, "target": rule.target, "on_missing": rule.on_missing, "stages": [s.model_dump(exclude_none=True) for s in (rule.stages or [])], "messages": rule.messages.model_dump() if rule.messages else {}, } return {"rule_id": ruleId, "name": ruleId, "severity": "info", "category": "", "group": ""} # ── 规则加载助手 ──────────────────────────────────── async def _resolve_rules_path(self, rulesPath: str | None = None) -> str | None: """解析规则 YAML 文件路径。 优先级：传入参数 > govdoc_runs 表记录 > None """ if rulesPath: return rulesPath # 尝试从最近的 completed run 中获取 rules_path try: from fastapi_common.fastapi_common_sqlalchemy.database import GetAsyncSession from sqlalchemy import text async with GetAsyncSession() as session: row = await session.execute( text( """SELECT rules_path FROM govdoc_runs WHERE rules_path IS NOT NULL AND status = 'completed' ORDER BY id DESC LIMIT 1""" ) ) result = row.mappings().first() if result and result.get("rules_path"): return result["rules_path"] except Exception: pass return None async def _load_ruleset(self, rulesPath: str | None = None): """加载 rules.yaml 为 RuleSet 对象。""" resolved = await self._resolve_rules_path(rulesPath) if not resolved: logger.warning("[Govdoc] Cannot resolve rules path for GetRuleDetail/ListRules") return None from fastapi_modules.fastapi_leaudit.govdoc_engine.dsl.loader import load_rules return load_rules(resolved) async def _load_rules_list(self, rulesPath: str | None = None) -> list[dict[str, Any]]: """加载规则列表（简要信息）。""" ruleset = await self._load_ruleset(rulesPath) if ruleset is None: return [] result = [] for rule in ruleset.all_rules(): result.append({ "rule_id": rule.rule_id, "name": rule.name, "severity": rule.severity, "category": rule.category, "group": "", }) return result