fix: 合规审查卡死修复 + 小红书复制格式 + today-only过滤

scripts/compliance_optimizer.py

@@ -8,7 +8,7 @@ from pathlib import Path
 from typing import Dict, List, Optional, Tuple
 from dataclasses import dataclass, asdict
 
-PROJECT_ROOT = Path('/root/openclaw-workspace/projects/yu-zhi-ran')
+PROJECT_ROOT = Path(__file__).resolve().parent.parent
 sys.path.insert(0, str(PROJECT_ROOT))
 sys.path.insert(0, str(PROJECT_ROOT / "platform" / "backend"))
 
@@ -92,9 +92,9 @@ def load_topic_map():
     topics = export_topics_to_json()
     return {t['id']: t for t in topics}
 
-def get_articles_from_db(topic_ids: Optional[List[str]] = None) -> List[Tuple[str, str, str]]:
+def get_articles_from_db(topic_ids: Optional[List[str]] = None, today_only: bool = False) -> List[Tuple[str, str, str]]:
     """从 articles 表读取 HTML 内容
-    
+
     Returns: [(html_content, platform, topic_id), ...]
     """
     from db_helper import get_articles_by_topic
@@ -110,9 +110,14 @@ def get_articles_from_db(topic_ids: Optional[List[str]] = None) -> List[Tuple[st
     else:
         from app.database import SessionLocal
         from app.models import Article
+        from sqlalchemy import func
         db = SessionLocal()
         try:
-            all_articles = db.query(Article).filter(Article.html_content.isnot(None)).all()
+            query = db.query(Article).filter(Article.html_content.isnot(None))
+            if today_only:
+                cutoff = datetime.datetime.now() - datetime.timedelta(hours=24)
+                query = query.filter(Article.created_at >= cutoff)
+            all_articles = query.all()
             for a in all_articles:
                 results.append((a.html_content, a.platform, a.topic_id))
         finally:
@@ -222,14 +227,17 @@ def _load_platform_configs() -> Dict[str, Dict]:
     finally:
         db.close()
 
-def main(topic_ids: List[str] = None):
+def main(topic_ids: List[str] = None, today_only: bool = False):
     logger.info("=== 合规审查与优化开始 ===")
     logger.info("LLM 配置: opencode-go (model=deepseek-v4-flash) — 固定用于合规审查")
 
     platform_configs = _load_platform_configs()
     logger.info(f"已加载 {len(platform_configs)} 个平台配置")
 
-    articles = get_articles_from_db(topic_ids)
+    if today_only:
+        logger.info("仅处理当天创建的选题文章")
+
+    articles = get_articles_from_db(topic_ids, today_only)
     if not articles:
         logger.warning("未找到任何文章（可能尚未创作或同步到 DB）")
         report_file = DRAFTS_DIR / TODAY / "optimization_report.json"
@@ -349,6 +357,7 @@ if __name__ == "__main__":
     import argparse
     parser = argparse.ArgumentParser(description='合规审查与优化任务')
     parser.add_argument('--topic-ids', help='逗号分隔的选题ID列表，例如: A01,B02')
+    parser.add_argument('--today-only', action='store_true', help='仅处理当天创建的选题文章')
     args = parser.parse_args()
     topic_ids = args.topic_ids.split(',') if args.topic_ids else None
-    main(topic_ids)
+    main(topic_ids, today_only=args.today_only)