Files
yu-zhi-ran/platform/backend/app/initial_data.py
T
Yuzhiran Dev 233e23016c feat: 内容数据迁移至数据库,合规审查全链路打通
- 文章 HTML 存储从文件系统迁移至 articles 表,删除 releases 目录
- 合规审查从 DB 读取 HTML,审查结果写回 DB,通过后自动推进至待发布
- 新增 todayCount 筛选按钮,与系统概览统计数据一致
- 全屏预览修复:提升 z-index 超过侧边栏,添加退出全屏/关闭按钮
- 统一 '优化' → '审查' 命名,消除前后端术语不一致
- 调度器创作完成后自动触发审查(生成 → 审查 → 待发布)
- 清理旧备份/调试文件、过期大纲和研究笔记
2026-05-13 17:33:56 +08:00

241 lines
11 KiB
Python

import json
import os
from datetime import datetime
from pathlib import Path
from .database import SessionLocal, init_db
from .models import (
Topic, TopicField, TopicConfigField, TopicStatusConfig,
User, Case, LLMConfig, SystemConfig, PlatformConfig
)
import bcrypt
PROJECT_ROOT = Path(__file__).resolve().parents[1]
if os.getenv('PROJECT_ROOT'):
PROJECT_ROOT = Path(os.getenv('PROJECT_ROOT'))
TOPICS_FILE = PROJECT_ROOT / "automation" / "data" / "sustainability_topics.json"
CASES_FILE = PROJECT_ROOT / "automation" / "data" / "initial_cases.json"
DEFAULT_ADMIN_USERNAME = os.getenv('DEFAULT_ADMIN_USERNAME', 'admin')
DEFAULT_ADMIN_PASSWORD = os.getenv('DEFAULT_ADMIN_PASSWORD', 'admin123')
def import_initial_data():
db = SessionLocal()
try:
if db.query(User).filter(User.username == DEFAULT_ADMIN_USERNAME).first() is None:
hashed = bcrypt.hashpw(DEFAULT_ADMIN_PASSWORD.encode('utf-8'), bcrypt.gensalt())
admin = User(
username=DEFAULT_ADMIN_USERNAME,
password_hash=hashed.decode('utf-8'),
role="admin"
)
db.add(admin)
db.commit()
print(f"✅ 创建默认管理员: {DEFAULT_ADMIN_USERNAME}")
if db.query(LLMConfig).count() == 0:
default_llm = LLMConfig(
name="default_expand",
system_prompt="你是一个专业的内容创作者。",
user_prompt_template="""你是一个专业的内容创作者,风格精炼、直接、切中要点。请将以下大纲扩展为完整的文章章节,要求如下:
### 选题信息
标题:{topic.get('title')}
领域:{topic.get('field_name')}
核心观点:{topic.get('core_concept', '')}
受众痛点:{topic.get('audience_pain', '')}
独特视角:{topic.get('unique_angle', '')}
### 当前章节
## {section_title}
{section_content}
### 输出要求
- 以 "## {section_title}" 开始
- 字数:200-300 字(精炼为主)
- 语言:直白、有冲击力,避免空洞套话
- 使用 Markdown 格式
- 每个论点配具体案例或数据支撑
- 确保与整体文章调性一致
直接输出完整 Markdown 章节(包括标题和正文)。""",
temperature=0.8,
max_tokens=1000,
model="stepfun-ai/step-3.5-flash",
is_active=True
)
db.add(default_llm)
db.commit()
print("✅ 插入默认 LLM 配置")
default_system_configs = [
{"key": "collector_enabled", "value": "false", "description": "是否启用采集器"},
{"key": "scheduler_interval", "value": "daily", "description": "调度间隔:daily/hourly/weekly"},
]
for cfg in default_system_configs:
if db.query(SystemConfig).filter(SystemConfig.key == cfg["key"]).first() is None:
db.add(SystemConfig(**cfg))
if db.query(SystemConfig).filter(SystemConfig.key == "review_llm_id").first() is None:
first_llm = db.query(LLMConfig).filter(LLMConfig.is_active == True).first()
if first_llm:
db.add(SystemConfig(key="review_llm_id", value=str(first_llm.id), description="审查使用的 LLM 配置 ID(留空则用环境变量默认值)"))
db.commit()
print("✅ 插入默认系统配置")
if db.query(PlatformConfig).count() == 0:
platforms = [
{
"platform": "zhihu",
"name": "知乎",
"icon": "🔍",
"default_format": "长文深度分析,1500-3000字,有数据支撑",
"compliance_rules": {
"max_length": 50000,
"requires_authentication": False,
"sensitive_words": ["敏感词示例1", "敏感词示例2"]
},
"is_active": True
},
{
"platform": "wechat",
"name": "微信公众号",
"icon": "💚",
"default_format": "公众号图文,800-1500字,亲切口语化",
"compliance_rules": {
"max_length": 20000,
"requires_authentication": True
},
"is_active": True
},
{
"platform": "xiaohongshu",
"name": "小红书",
"icon": "📕",
"default_format": "图文笔记,300-800字,emoji+标签",
"compliance_rules": {
"max_length": 1000,
"requires_tags": True,
"max_tags": 10
},
"is_active": True
}
]
for p in platforms:
db.add(PlatformConfig(**p))
db.commit()
print("✅ 插入平台配置")
if db.query(TopicField).count() == 0:
fields = [
{"name": "未来工作方式", "icon": "💼", "color": "#667eea", "description": "远程工作、零工经济、职业转型", "sort_order": 1},
{"name": "AI与效率", "icon": "🤖", "color": "#764ba2", "description": "AI工具、数字助手、效率方法", "sort_order": 2},
{"name": "可持续生活", "icon": "🌿", "color": "#67c23a", "description": "环保、低碳、自然生活方式", "sort_order": 3},
{"name": "数字游民", "icon": "🌍", "color": "#409eff", "description": "旅行、地理自由、海外生活", "sort_order": 4},
{"name": "个人成长", "icon": "📚", "color": "#e6a23c", "description": "学习、技能、认知升级", "sort_order": 5},
{"name": "科技人文", "icon": "🔬", "color": "#f56c6c", "description": "科技伦理、数字生活反思", "sort_order": 6},
]
for f in fields:
db.add(TopicField(**f))
db.commit()
print("✅ 插入默认领域配置")
if db.query(TopicStatusConfig).count() == 0:
statuses = [
{"status": "pending", "label": "待处理", "color": "#E6A23C", "icon": "", "sort_order": 1, "is_default": True},
{"status": "review", "label": "待审查", "color": "#F56C6C", "icon": "🔍", "sort_order": 2},
{"status": "draft", "label": "草稿", "color": "#909399", "icon": "📝", "sort_order": 3},
{"status": "ready", "label": "待发布", "color": "#67C23A", "icon": "", "sort_order": 4},
{"status": "published", "label": "已发布", "color": "#409EFF", "icon": "🚀", "sort_order": 5},
]
for s in statuses:
db.add(TopicStatusConfig(**s))
db.commit()
print("✅ 插入状态配置")
field_map = {}
for f in db.query(TopicField).all():
field_map[f.name] = f.id
if db.query(Topic).count() == 0:
if os.path.exists(TOPICS_FILE):
topics = json.loads(open(TOPICS_FILE, encoding='utf-8').read())
seen = {}
for t in topics:
seen[t['id']] = t
unique_topics = list(seen.values())
for t in unique_topics:
field_id = field_map.get(t.get('field'))
field_name = t.get('field')
topic = Topic(
id=t['id'],
field_id=field_id,
field_name=field_name,
title=t['title'],
format=t.get('format'),
core_concept=t.get('core_concept'),
audience_pain=t.get('audience_pain'),
unique_angle=t.get('unique_angle'),
priority=t.get('priority'),
priority_score=t.get('priority_score', 0),
total_score=t.get('total_score'),
status=t.get('status', 'pending'),
cases=t.get('cases', []),
tags=t.get('tags', []),
source_file=t.get('source_file'),
ready_at=datetime.strptime(t['ready_at'], '%Y-%m-%d').date() if t.get('ready_at') else None,
published_at=datetime.strptime(t['published_at'], '%Y-%m-%d').date() if t.get('published_at') else None,
compliance_score=t.get('compliance_score'),
platform_urls=t.get('platform_urls', {})
)
db.add(topic)
db.commit()
print(f"✅ 导入 {len(unique_topics)} 个选题")
else:
print(f"⚠️ 选题文件不存在: {TOPICS_FILE}")
if db.query(Case).count() == 0:
if os.path.exists(CASES_FILE):
cases_data = json.loads(open(CASES_FILE, encoding='utf-8').read())
for c in cases_data:
case = Case(
id=c['id'],
title=c['title'],
field=c['field'],
summary=c['summary'],
key_metrics=c.get('key_metrics'),
date=c.get('date'),
source=c['source'],
source_url=c.get('source_url'),
credibility_rating=c.get('credibility_rating'),
china_applicability=c.get('china_applicability')
)
db.add(case)
db.commit()
print(f"✅ 导入 {len(cases_data)} 条案例")
# 同步 PostgreSQL 自增序列
if os.getenv('USE_POSTGRES', 'true').lower() == 'true':
try:
from sqlalchemy import text
tables = ["cases", "users", "content_calendar", "media_assets", "content_metrics", "content_tasks", "audit_logs", "task_logs", "topic_config_fields"]
for table in tables:
db.execute(text(f"SELECT setval(pg_get_serial_sequence('{table}', 'id'), COALESCE((SELECT MAX(id) FROM {table}), 0) + 1, false)"))
db.commit()
print("✅ PostgreSQL 自增序列已同步")
except Exception as e:
print(f"⚠️ 序列同步警告: {e}")
print("✅ 初始化完成")
except Exception as e:
import traceback
traceback.print_exc()
print(f"初始化失败: {e}")
db.rollback()
finally:
db.close()
if __name__ == "__main__":
init_db()
import_initial_data()