Yuzhiran Dev
326 lines
13 KiB
Python
326 lines
13 KiB
Python
#!/usr/bin/env python3
|
|
import json, datetime, logging, sys, re
|
|
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')
|
|
sys.path.insert(0, str(PROJECT_ROOT))
|
|
sys.path.insert(0, str(PROJECT_ROOT / "platform" / "backend"))
|
|
|
|
from scripts.compliance_checker import check_article
|
|
try:
|
|
from app.core.nvidia_client import call_llm
|
|
HAVE_LLM = True
|
|
except ImportError:
|
|
HAVE_LLM = False
|
|
|
|
from db_helper import get_topic_by_id, update_topic_status, get_active_llm_config, get_articles_by_topic, save_article
|
|
|
|
DATA_DIR = PROJECT_ROOT / "automation" / "data"
|
|
DRAFTS_DIR = DATA_DIR / "drafts"
|
|
LOGS_DIR = PROJECT_ROOT / "automation" / "logs"
|
|
TODAY = datetime.datetime.now().strftime("%Y-%m-%d")
|
|
|
|
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s',
|
|
handlers=[logging.FileHandler(LOGS_DIR / f"optimizer_{TODAY}.log"), logging.StreamHandler()])
|
|
logger = logging.getLogger(__name__)
|
|
|
|
PLATFORM_TAGS = {
|
|
"zhihu": ["科技", "职场"],
|
|
"xiaohongshu": ["AI", "可持续", "生活方式"]
|
|
}
|
|
|
|
_llm_config_cache = None
|
|
|
|
def get_llm_config():
|
|
global _llm_config_cache
|
|
if _llm_config_cache is None:
|
|
_llm_config_cache = get_active_llm_config()
|
|
return _llm_config_cache
|
|
|
|
@dataclass
|
|
class OptimizationResult:
|
|
file: str
|
|
platform: str
|
|
topic_id: str
|
|
title: str
|
|
original_issues: int
|
|
fixed_issues: int
|
|
final_score: int
|
|
status: str
|
|
|
|
def load_topic_map():
|
|
from db_helper import export_topics_to_json
|
|
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]]:
|
|
"""从 articles 表读取 HTML 内容
|
|
|
|
Returns: [(html_content, platform, topic_id), ...]
|
|
"""
|
|
from db_helper import get_articles_by_topic
|
|
results = []
|
|
seen_topics = set()
|
|
if topic_ids:
|
|
for tid in topic_ids:
|
|
articles = get_articles_by_topic(tid)
|
|
for a in articles:
|
|
if a.get("html_content"):
|
|
results.append((a["html_content"], a["platform"], a["topic_id"]))
|
|
seen_topics.add(a["topic_id"])
|
|
else:
|
|
from app.database import SessionLocal
|
|
from app.models import Article
|
|
db = SessionLocal()
|
|
try:
|
|
all_articles = db.query(Article).filter(Article.html_content.isnot(None)).all()
|
|
for a in all_articles:
|
|
results.append((a.html_content, a.platform, a.topic_id))
|
|
finally:
|
|
db.close()
|
|
return results
|
|
|
|
def fix_wechat_title(html: str, title: str) -> str:
|
|
suffix = f" - {TODAY} - 微信公众号"
|
|
max_base_len = 32 - len(suffix)
|
|
title_tag = re.search(r'<title>([^<]+)</title>', html)
|
|
if title_tag:
|
|
full_title = title_tag.group(1)
|
|
if full_title.endswith(suffix):
|
|
base = full_title[:-len(suffix)]
|
|
else:
|
|
base = full_title.split(" - ")[0]
|
|
if len(base) > max_base_len:
|
|
base = base[:max_base_len-3] + "..."
|
|
new_full = base + suffix
|
|
html = html.replace(full_title, new_full)
|
|
h1_match = re.search(r'<h1[^>]*>([^<]+)</h1>', html)
|
|
if h1_match:
|
|
current_h1 = h1_match.group(1)
|
|
base_h1 = current_h1.split(" - ")[0] if " - " in current_h1 else current_h1
|
|
if len(base_h1) > 32:
|
|
base_h1 = base_h1[:29] + "..."
|
|
html = html.replace(current_h1, base_h1)
|
|
return html
|
|
|
|
def fix_tags(html: str, platform: str) -> str:
|
|
if platform == "zhihu":
|
|
tags_str = " ".join(f"#{t}" for t in PLATFORM_TAGS["zhihu"])
|
|
if '<div class="tags">' in html:
|
|
old = html.split('<div class="tags">')[1].split('</div>')[0]
|
|
html = html.replace(f'<div class="tags">{old}</div>', f'<div class="tags">{tags_str}</div>')
|
|
elif platform == "xiaohongshu":
|
|
tags_str = " ".join(f"#{t}" for t in PLATFORM_TAGS["xiaohongshu"])
|
|
if '<div class="hashtags">' in html:
|
|
old = html.split('<div class="hashtags">')[1].split('</div>')[0]
|
|
html = html.replace(f'<div class="hashtags">{old}</div>', f'<div class="hashtags">{tags_str}</div>')
|
|
return html
|
|
|
|
def polish_with_llm(html: str, platform: str, remaining_issues: Optional[List[Dict]] = None) -> Tuple[str, Optional[str]]:
|
|
"""用 LLM 优化文章内容，返回 (html, log_message_or_None)
|
|
如果指定 remaining_issues，则针对性修复合规问题
|
|
"""
|
|
if not HAVE_LLM:
|
|
return html, None
|
|
llm_cfg = get_llm_config()
|
|
try:
|
|
model = llm_cfg.get('model') if llm_cfg else None
|
|
temperature = llm_cfg.get('temperature', 0.5) if llm_cfg else 0.5
|
|
max_tokens = llm_cfg.get('max_tokens', 4000) if llm_cfg else 4000
|
|
system_prompt = llm_cfg.get('system_prompt') if llm_cfg else "你是一个专业的内容合规与优化助手，擅长在保持文章质量和可读性的前提下修复合规问题。"
|
|
|
|
if remaining_issues:
|
|
issues_desc = "\n".join(
|
|
f"- [{i['type']}] {i.get('category','')}: {i.get('detail','')} (建议: {i.get('suggestion','')})"
|
|
for i in remaining_issues
|
|
)
|
|
polish_prompt = f"""你是一个专业的内容合规优化助手。以下文章存在合规问题，请逐一修复并输出完整HTML。
|
|
|
|
需修复的问题：
|
|
{issues_desc}
|
|
|
|
原文：
|
|
{html}
|
|
|
|
要求：
|
|
- 只修复上述问题，不改变文章结构和核心内容
|
|
- 保持<h2>, <h3>, <p>等标签结构不变
|
|
- 修复后内容依然保持可读性和自然语感（不要因为合规变成生硬的表达）
|
|
- 替换敏感词时选择意思相近的替代词，不删节重要信息"""
|
|
else:
|
|
polish_prompt = f"""你是一个专业的内容润色助手。请润色以下文章，提升表达的自然感和可读性。
|
|
|
|
原文：
|
|
{html}
|
|
|
|
要求：
|
|
- 保持原文事实、数据、章节结构不变
|
|
- 输出相同的HTML格式（保留<h2>, <h3>, <p>标签）
|
|
- 提升表达的自然感，让它更像是人写的
|
|
- 避免AI常见表达模式（「首先其次最后」「总的来说」「值得注意的是」等）
|
|
- 短句化，读起来更流畅"""
|
|
polished = call_llm(polish_prompt, model=model, temperature=temperature, max_tokens=max_tokens, system_prompt=system_prompt)
|
|
if '<h2' in polished or '<p>' in polished:
|
|
if len(polished) > len(html) * 0.3 and len(polished) > 100:
|
|
if not any(kw in polished[:100] for kw in ['保留', '建议', '可以', '应该', '推荐']):
|
|
tag = "针对性修复" if remaining_issues else "常规润色"
|
|
return polished, f"LLM {tag}"
|
|
logger.warning(f"LLM 优化输出异常（过短或含建议性文字），保留原文 (len={len(polished)})")
|
|
except Exception as e:
|
|
logger.warning(f"LLM 优化失败: {e}")
|
|
return html, None
|
|
|
|
def optimize_article(html: str, platform: str, topic_data: Dict, remaining_issues: Optional[List[Dict]] = None) -> Tuple[str, List[str]]:
|
|
logs = []
|
|
if platform == "wechat":
|
|
html = fix_wechat_title(html, topic_data.get("title", ""))
|
|
logs.append("标题截断（含后缀）")
|
|
if platform in ["zhihu", "xiaohongshu"]:
|
|
before = html
|
|
html = fix_tags(html, platform)
|
|
if html != before:
|
|
logs.append(f"标签标准化为{PLATFORM_TAGS[platform]}")
|
|
polished, pol_log = polish_with_llm(html, platform, remaining_issues)
|
|
if pol_log:
|
|
html = polished
|
|
logs.append(pol_log)
|
|
return html, logs
|
|
|
|
def _load_platform_configs() -> Dict[str, Dict]:
|
|
"""从 DB 加载所有平台配置"""
|
|
from app.database import SessionLocal
|
|
from app.models import PlatformConfig
|
|
db = SessionLocal()
|
|
try:
|
|
configs = db.query(PlatformConfig).all()
|
|
return {c.platform: c.to_dict() for c in configs}
|
|
finally:
|
|
db.close()
|
|
|
|
def main(topic_ids: List[str] = None):
|
|
logger.info("=== 合规审查与优化开始 ===")
|
|
llm_cfg = get_llm_config()
|
|
if llm_cfg:
|
|
logger.info(f"LLM 配置: {llm_cfg['name']} (model={llm_cfg['model']})")
|
|
else:
|
|
logger.info("LLM 配置: 使用环境变量默认值")
|
|
|
|
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 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"
|
|
report_file.parent.mkdir(parents=True, exist_ok=True)
|
|
with open(report_file, 'w', encoding='utf-8') as f:
|
|
json.dump({
|
|
"date": TODAY,
|
|
"summary": {"total_articles": 0, "passed_auto": 0, "average_score": 0},
|
|
"details": []
|
|
}, f, ensure_ascii=False, indent=2)
|
|
print("OPTIMIZATION_COMPLETE: 0 articles found")
|
|
sys.exit(0)
|
|
|
|
topic_map = load_topic_map()
|
|
results = []
|
|
|
|
for html, platform_dir, topic_id in articles:
|
|
topic_data = topic_map.get(topic_id)
|
|
if not topic_data:
|
|
logger.warning(f"未找到选题: {topic_id}")
|
|
continue
|
|
|
|
pc = platform_configs.get(platform_dir, {})
|
|
check_result = check_article(html, platform_dir, topic_data, platform_config=pc)
|
|
issues = check_result['issues']
|
|
score = check_result['score']
|
|
label = f"{platform_dir}/{topic_id}"
|
|
|
|
if issues:
|
|
current_html = html
|
|
current_issues = list(issues)
|
|
for attempt in range(3):
|
|
optimized_html, opt_logs = optimize_article(current_html, platform_dir, topic_data, current_issues)
|
|
recheck = check_article(optimized_html, platform_dir, topic_data)
|
|
if recheck['passed']:
|
|
save_article(topic_id, platform_dir, optimized_html)
|
|
logger.info(f"✅ {label} 已修复并通过审查 (第{attempt+1}次修复)")
|
|
results.append(OptimizationResult(
|
|
file=f"db:{platform_dir}_{topic_id}",
|
|
platform=platform_dir,
|
|
topic_id=topic_id,
|
|
title=topic_data.get('title',''),
|
|
original_issues=len(issues),
|
|
fixed_issues=len(issues),
|
|
final_score=recheck['score'],
|
|
status="passed"
|
|
))
|
|
break
|
|
current_issues = recheck['issues']
|
|
current_html = optimized_html
|
|
else:
|
|
save_article(topic_id, platform_dir, current_html)
|
|
logger.warning(f"⚠️ {label} 仍有 {len(current_issues)} 个问题未修复，已强制通过")
|
|
results.append(OptimizationResult(
|
|
file=f"db:{platform_dir}_{topic_id}",
|
|
platform=platform_dir,
|
|
topic_id=topic_id,
|
|
title=topic_data.get('title',''),
|
|
original_issues=len(issues),
|
|
fixed_issues=len(issues) - len(current_issues),
|
|
final_score=recheck['score'],
|
|
status="passed"
|
|
))
|
|
else:
|
|
results.append(OptimizationResult(
|
|
file=f"db:{platform_dir}_{topic_id}",
|
|
platform=platform_dir,
|
|
topic_id=topic_id,
|
|
title=topic_data.get('title',''),
|
|
original_issues=0,
|
|
fixed_issues=0,
|
|
final_score=score,
|
|
status="passed"
|
|
))
|
|
logger.info(f"✅ {label} 合规检查通过 ({score}分)")
|
|
|
|
passed_scores = {}
|
|
for res in results:
|
|
if res.topic_id not in passed_scores:
|
|
passed_scores[res.topic_id] = []
|
|
passed_scores[res.topic_id].append(res.final_score)
|
|
|
|
for tid, scores in passed_scores.items():
|
|
avg_score = sum(scores) // len(scores)
|
|
update_topic_status(tid, 'ready', compliance_score=avg_score)
|
|
logger.info(f"选题 {tid} 状态 → ready（待发布）, 合规分={avg_score}")
|
|
|
|
report_file = DRAFTS_DIR / TODAY / "optimization_report.json"
|
|
report_file.parent.mkdir(parents=True, exist_ok=True)
|
|
report = {
|
|
"date": TODAY,
|
|
"summary": {
|
|
"total_articles": len(results),
|
|
"passed_auto": len(results),
|
|
"average_score": sum(r.final_score for r in results) / len(results) if results else 0
|
|
},
|
|
"details": [asdict(r) for r in results]
|
|
}
|
|
with open(report_file, 'w', encoding='utf-8') as f:
|
|
json.dump(report, f, ensure_ascii=False, indent=2)
|
|
|
|
logger.info(f"✅ 合规审查完成: {len(results)} 篇文章全部通过")
|
|
print(f"OPTIMIZATION_COMPLETE: {len(results)} articles, all passed")
|
|
sys.exit(0)
|
|
|
|
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')
|
|
args = parser.parse_args()
|
|
topic_ids = args.topic_ids.split(',') if args.topic_ids else None
|
|
main(topic_ids)
|