feat: 数据源统一与前端预览修复
=== 后端核心 === - db_helper: 统一数据库访问抽象层 - system.py API: * 参数绑定修复: 使用 Body(embed=True) 接收 JSON * 添加请求日志记录 - sync.py: 仅导出 DB→JSON（备份） === 合规与流水线 === - compliance_checker: 标签检测优化（仅检查容器，避免正文误判） - 所有脚本（creator/collector/writer/outline/research等）统一使用数据库 === 前端改版 === - topics.html: * 创作/优化 API 路径修正 * 预览弹窗重设计：多平台并行加载、富文本显示、单复制按钮 * 状态中文映射（getStatusLabel） * 认证检查 - 所有 HTML 静态资源路径修复（移除 /static 前缀） === 数据一致性 === - 数据库状态统一为英文（pending/review/ready/published） - 前端显示中文化映射 已测试 A03 流水线完整通过。
This commit is contained in:
lt
@@ -1,11 +1,11 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
合规审查与优化任务
|
||||
合规审查与优化任务（数据库版）
|
||||
每天 05:45 运行，处理当天所有 draft 文章：
|
||||
1. 执行合规检查（compliance_checker）
|
||||
2. 自动修复已知问题（标题、标签）
|
||||
3. 重写合规版本
|
||||
4. 更新选题状态为「审查通过待发布」
|
||||
4. 更新选题状态为「待发布」
|
||||
5. 生成优化报告通知
|
||||
"""
|
||||
|
||||
@@ -26,10 +26,12 @@ try:
|
||||
except ImportError:
|
||||
HAVE_LLM = False
|
||||
|
||||
# 导入数据库辅助模块
|
||||
from db_helper import get_topic_by_id, update_topic_status
|
||||
|
||||
DATA_DIR = PROJECT_ROOT / "automation" / "data"
|
||||
RELEASES_DIR = DATA_DIR / "releases"
|
||||
DRAFTS_DIR = DATA_DIR / "drafts"
|
||||
TOPICS_FILE = DATA_DIR / "sustainability_topics.json"
|
||||
LOGS_DIR = PROJECT_ROOT / "automation" / "logs"
|
||||
TODAY = datetime.datetime.now().strftime("%Y-%m-%d")
|
||||
|
||||
@@ -55,42 +57,15 @@ class OptimizationResult:
|
||||
status: str
|
||||
|
||||
def load_topic_map():
|
||||
with open(TOPICS_FILE, 'r', encoding='utf-8') as f:
|
||||
topics = json.load(f)
|
||||
"""从数据库加载所有选题数据"""
|
||||
from db_helper import export_topics_to_json
|
||||
topics = export_topics_to_json()
|
||||
return {t['id']: t for t in topics}
|
||||
|
||||
|
||||
def update_topic_status(topic_id: str, status: str):
|
||||
"""更新选题状态（JSON + 数据库）"""
|
||||
# 更新 JSON
|
||||
with open(TOPICS_FILE, 'r', encoding='utf-8') as f:
|
||||
topics = json.load(f)
|
||||
updated = False
|
||||
for t in topics:
|
||||
if t.get('id') == topic_id:
|
||||
t['status'] = status
|
||||
updated = True
|
||||
break
|
||||
if updated:
|
||||
with open(TOPICS_FILE, 'w', encoding='utf-8') as f:
|
||||
json.dump(topics, f, ensure_ascii=False, indent=2)
|
||||
# 更新数据库
|
||||
try:
|
||||
import sys
|
||||
from pathlib import Path
|
||||
backend_path = Path(__file__).resolve().parents[2] / 'platform' / 'backend'
|
||||
if str(backend_path) not in sys.path:
|
||||
sys.path.insert(0, str(backend_path))
|
||||
from app.database import SessionLocal
|
||||
from app.models import Topic
|
||||
db = SessionLocal()
|
||||
topic_db = db.query(Topic).filter(Topic.id == topic_id).first()
|
||||
if topic_db:
|
||||
topic_db.status = status
|
||||
db.commit()
|
||||
db.close()
|
||||
except Exception as e:
|
||||
logger.error(f"更新数据库失败: {e}")
|
||||
def update_topic_status_db_only(topic_id: str, status: str):
|
||||
"""仅更新数据库状态（不更新JSON）"""
|
||||
from db_helper import update_topic_status
|
||||
update_topic_status(topic_id, status)
|
||||
|
||||
def fix_wechat_title(html: str, title: str) -> str:
|
||||
"""微信标题优化：<title>和<h1>都控制长度（考虑后缀）"""
|
||||
@@ -101,7 +76,6 @@ def fix_wechat_title(html: str, title: str) -> str:
|
||||
title_tag = re.search(r'<title>([^<]+)</title>', html)
|
||||
if title_tag:
|
||||
full_title = title_tag.group(1)
|
||||
# 提取 base（去掉后缀）
|
||||
if full_title.endswith(suffix):
|
||||
base = full_title[:-len(suffix)]
|
||||
else:
|
||||
@@ -111,11 +85,10 @@ def fix_wechat_title(html: str, title: str) -> str:
|
||||
new_full = base + suffix
|
||||
html = html.replace(full_title, new_full)
|
||||
|
||||
# 处理 <h1>...</h1>（不含后缀，但要截断）
|
||||
# 处理 <h1>...</h1>
|
||||
h1_match = re.search(r'<h1[^>]*>([^<]+)</h1>', html)
|
||||
if h1_match:
|
||||
current_h1 = h1_match.group(1)
|
||||
# 如果 h1 包含后缀（不应该），去掉
|
||||
base_h1 = current_h1.split(" - ")[0] if " - " in current_h1 else current_h1
|
||||
if len(base_h1) > 32:
|
||||
base_h1 = base_h1[:29] + "..."
|
||||
@@ -127,7 +100,6 @@ def fix_tags(html: str, platform: str) -> str:
|
||||
"""强制替换标签为平台白名单"""
|
||||
if platform == "zhihu":
|
||||
tags_str = " ".join(f"#{t}" for t in PLATFORM_TAGS["zhihu"])
|
||||
# 替换 <div class="tags">...</div>
|
||||
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>')
|
||||
@@ -140,18 +112,14 @@ def fix_tags(html: str, platform: str) -> str:
|
||||
|
||||
def optimize_article(html: str, platform: str, topic_data: Dict) -> (str, List[str]):
|
||||
logs = []
|
||||
# 1. 标题优化（微信）
|
||||
if platform == "wechat":
|
||||
html = fix_wechat_title(html, topic_data.get("title", ""))
|
||||
logs.append("标题截断（含后缀）")
|
||||
# 2. 标签优化
|
||||
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]}")
|
||||
# 3. 图片内联检查
|
||||
# 提取所有 img 标签
|
||||
img_tags = re.findall(r'<img[^>]*>', html, re.IGNORECASE)
|
||||
for tag in img_tags:
|
||||
m = re.search(r'src=["\']([^"\']+)["\']', tag, re.IGNORECASE)
|
||||
@@ -160,7 +128,6 @@ def optimize_article(html: str, platform: str, topic_data: Dict) -> (str, List[s
|
||||
if not src.startswith('data:image/'):
|
||||
logs.append(f"图片未内联: {src[:50]}... 需手动修复")
|
||||
|
||||
# 4. LLM 内容优化（使用 NVIDIA step-3.5-flash）
|
||||
if HAVE_LLM:
|
||||
try:
|
||||
polish_prompt = f"""你是一个专业的内容润色助手。请优化以下文章内容，提升表达的专业性和可读性，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）。
|
||||
@@ -198,7 +165,6 @@ def main(topic_ids: List[str] = None):
|
||||
if len(parts) < 2:
|
||||
continue
|
||||
topic_id = parts[1]
|
||||
# 如果指定了 topic_ids，则只处理匹配的
|
||||
if topic_ids is not None and topic_id not in topic_ids:
|
||||
continue
|
||||
topic_data = topic_map.get(topic_id)
|
||||
@@ -227,7 +193,7 @@ def main(topic_ids: List[str] = None):
|
||||
final_score=recheck['score'],
|
||||
status="passed"
|
||||
))
|
||||
update_topic_status(topic_id, '待发布')
|
||||
update_topic_status_db_only(topic_id, 'pending')
|
||||
else:
|
||||
logger.warning(f"⚠️ {html_file.name} 优化后仍有问题，需人工审核")
|
||||
results.append(OptimizationResult(
|
||||
@@ -255,20 +221,28 @@ def main(topic_ids: List[str] = None):
|
||||
if not check_result['passed']:
|
||||
all_passed = False
|
||||
|
||||
# 更新选题状态
|
||||
# 更新选题状态 (JSON + 数据库)
|
||||
for res in results:
|
||||
if res.status == "passed":
|
||||
tid = res.topic_id
|
||||
with open(TOPICS_FILE, 'r', encoding='utf-8') as f:
|
||||
topics = json.load(f)
|
||||
for t in topics:
|
||||
if t.get('id') == tid:
|
||||
t['status'] = '待发布'
|
||||
t['ready_at'] = TODAY
|
||||
t['compliance_score'] = res.final_score
|
||||
break
|
||||
with open(TOPICS_FILE, 'w', encoding='utf-8') as f:
|
||||
json.dump(topics, f, ensure_ascii=False, indent=2)
|
||||
# 更新数据库状态为 'pending'（待发布）
|
||||
update_topic_status(tid, 'ready')
|
||||
# 可选：同时更新 JSON 以保持兼容
|
||||
# (已废弃，但保留更新，避免其他组件出错)
|
||||
try:
|
||||
json_path = DATA_DIR / "sustainability_topics.json"
|
||||
with open(json_path, 'r', encoding='utf-8') as f:
|
||||
topics = json.load(f)
|
||||
for t in topics:
|
||||
if t.get('id') == tid:
|
||||
t['status'] = 'ready'
|
||||
t['ready_at'] = TODAY
|
||||
t['compliance_score'] = res.final_score
|
||||
break
|
||||
with open(json_path, 'w', encoding='utf-8') as f:
|
||||
json.dump(topics, f, ensure_ascii=False, indent=2)
|
||||
except Exception as e:
|
||||
logger.warning(f"更新 JSON 失败: {e}")
|
||||
|
||||
# 生成报告
|
||||
report_file = DRAFTS_DIR / TODAY / "optimization_report.json"
|
||||
|
||||
Reference in New Issue
Block a user