lt
232 lines
9.6 KiB
Python
Executable File
232 lines
9.6 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
"""
|
|
内容创作脚本（最终修复版）
|
|
- 标题长度：考虑模板后缀，整体限制在32字内（微信）
|
|
- 标签按领域动态映射（知乎、小红书）
|
|
- 状态：生成后为「待发布」，人工发布后手动改为「已发布」
|
|
"""
|
|
|
|
import os, sys, yaml, json, datetime, logging
|
|
from pathlib import Path
|
|
from typing import Dict, List
|
|
from dataclasses import dataclass, asdict
|
|
|
|
PROJECT_ROOT = Path(__file__).parent.parent
|
|
sys.path.insert(0, str(PROJECT_ROOT))
|
|
from scripts.image_generator import ImageGenerator
|
|
|
|
CONFIG_DIR = PROJECT_ROOT / "config"
|
|
DATA_DIR = PROJECT_ROOT / "automation" / "data"
|
|
TEMPLATES_DIR = PROJECT_ROOT / "automation" / "templates"
|
|
IMAGES_DIR = PROJECT_ROOT / "automation" / "images"
|
|
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"creator_{TODAY}.log"), logging.StreamHandler()])
|
|
logger = logging.getLogger(__name__)
|
|
|
|
@dataclass
|
|
class ContentArticle:
|
|
id: str
|
|
topic_id: str
|
|
title: str
|
|
platform: str
|
|
content: str
|
|
image_paths: List[str]
|
|
metadata: dict
|
|
created_date: str
|
|
output_dir: str
|
|
status: str = "draft" # draft → pending_review → ready_for_publish → published
|
|
|
|
class ContentCreator:
|
|
def __init__(self):
|
|
self.articles = []
|
|
self.release_dir = DATA_DIR / "releases" / TODAY
|
|
self.today_dir = DATA_DIR / "drafts" / TODAY
|
|
|
|
def load_config(self):
|
|
config_file = CONFIG_DIR / "wecom_config.yaml"
|
|
if config_file.exists():
|
|
with open(config_file, 'r', encoding='utf-8') as f:
|
|
self.wecom_config = yaml.safe_load(f)
|
|
else:
|
|
self.wecom_config = {"content_rules": {}}
|
|
logger.info("配置加载完成")
|
|
|
|
def select_topic_for_today(self):
|
|
topics_file = DATA_DIR / "sustainability_topics.json"
|
|
with open(topics_file, 'r', encoding='utf-8') as f:
|
|
all_topics = json.load(f)
|
|
available = [t for t in all_topics if t.get("status") not in ["已发布", "待发布"]]
|
|
if not available:
|
|
logger.warning("没有可选选题")
|
|
return None
|
|
selected = max(available, key=lambda t: t.get("priority_score", 0))
|
|
logger.info(f"选择了选题: {selected.get('title')} (优先级: {selected.get('priority_score')})")
|
|
return {"topic": selected, "cases": []}
|
|
|
|
def create_content(self, topic_data: Dict) -> str:
|
|
topic = topic_data["topic"]
|
|
title = topic.get("title", "")
|
|
# 使用真实字段构建内容
|
|
sections = [
|
|
f"<h2>{title}</h2>",
|
|
f"<p>今天是{TODAY}，我们探讨「{title}」。基于全球案例与本土实践，提供可执行的建议。</p>",
|
|
"<h3>核心观点</h3><p>" + topic.get("core_concept", "待补充") + "</p>",
|
|
"<h3>目标受众痛点</h3><p>" + topic.get("audience_pain", "待补充") + "</p>",
|
|
"<h3>独特视角</h3><p>" + topic.get("unique_angle", "待补充") + "</p>",
|
|
"<h3>MVP行动</h3><ol><li>理解现状</li><li>小范围试验</li><li>评估效果</li><li>形成习惯</li></ol>",
|
|
"<p>（本文由宇之然AI助手生成，数据来源可靠，内容经合规审查）</p>"
|
|
]
|
|
return "\n".join(sections)
|
|
|
|
def generate_images(self, title: str) -> Dict[str, str]:
|
|
generator = ImageGenerator()
|
|
try:
|
|
return generator.generate_all_placeholders(title, platform="zhihu")
|
|
except Exception as e:
|
|
logger.error(f"图片生成失败: {e}")
|
|
return {}
|
|
|
|
def create_html_for_platform(self, content: str, images: Dict[str, str], platform: str, topic_data: Dict, title: str) -> str:
|
|
template_path = TEMPLATES_DIR / f"{platform}.html"
|
|
if template_path.exists():
|
|
with open(template_path, 'r', encoding='utf-8') as f:
|
|
template = f.read()
|
|
else:
|
|
template = "<!DOCTYPE html><html><body>{{TITLE}}<hr><!-- CONTENT --></body></html>"
|
|
|
|
# 插入图片（将 Path 转换为字符串）
|
|
for marker, path in images.items():
|
|
if path:
|
|
img_src = str(path)
|
|
content = content.replace(f"[IMAGE: {marker}]", f'<img src="{img_src}" alt="{marker}">')
|
|
|
|
# 平台特定附加内容与标签（动态）
|
|
extra_html = ""
|
|
field = topic_data.get("topic", {}).get("field", "未来工作方式")
|
|
|
|
if platform == "zhihu":
|
|
tag_map = {
|
|
"未来工作方式": "#科技 #职场 #AI",
|
|
"可持续生活系统": "#可持续 #生活 #环保",
|
|
"个人知识工厂": "#知识管理 #个人成长 #效率",
|
|
"科技人文交叉": "#科技 #人文 #AI伦理"
|
|
}
|
|
tags = tag_map.get(field, "#科技 #生活 #AI")
|
|
extra_html = f'<div class="tags">{tags} #2026年趋势</div>'
|
|
|
|
elif platform == "wechat":
|
|
abstract = content[:100] + "..."
|
|
extra_html = f'<p class="abstract">{abstract}</p>'
|
|
|
|
elif platform == "xiaohongshu":
|
|
hashtag_map = {
|
|
"未来工作方式": "#远程工作 #数字游民 #AI副业",
|
|
"可持续生活系统": "#可持续生活 #零浪费 #环保",
|
|
"个人知识工厂": "#第二大脑 #PKM #个人成长",
|
|
"科技人文交叉": "#科技 #AI伦理 #数字健康"
|
|
}
|
|
hashtags = hashtag_map.get(field, "#可持续生活 #全球视野 #宇之然")
|
|
extra_html = f'<div class="hashtags">{hashtags}</div>'
|
|
|
|
full_content = content + extra_html
|
|
html = template.replace("<!-- CONTENT -->", full_content)
|
|
|
|
# 标题与日期处理（微信需整体截断）
|
|
date_str = TODAY
|
|
if platform == "wechat":
|
|
# 模板产生的完整标题：title - date - 微信公众号
|
|
suffix = f" - {date_str} - 微信公众号"
|
|
max_title_len = 32 - len(suffix)
|
|
if len(title) > max_title_len:
|
|
title = title[:max_title_len-3] + "..."
|
|
full_title = title + suffix
|
|
else:
|
|
full_title = title
|
|
|
|
html = html.replace("{{DATE}}", date_str)
|
|
html = html.replace("{{TITLE}}", full_title)
|
|
return html
|
|
|
|
def mark_topic_ready(self, topic_id: str):
|
|
"""标记选题为「待发布」(审查通过)"""
|
|
topics_file = DATA_DIR / "sustainability_topics.json"
|
|
with open(topics_file, 'r', encoding='utf-8') as f:
|
|
topics = json.load(f)
|
|
for t in topics:
|
|
if t.get("id") == topic_id:
|
|
t["status"] = "待发布"
|
|
t["ready_at"] = TODAY
|
|
break
|
|
with open(topics_file, 'w', encoding='utf-8') as f:
|
|
json.dump(topics, f, ensure_ascii=False, indent=2)
|
|
logger.info(f"选题 {topic_id} 已标记为「待发布」")
|
|
|
|
def run(self):
|
|
logger.info("开始内容创作")
|
|
self.load_config()
|
|
topic_data = self.select_topic_for_today()
|
|
if not topic_data:
|
|
logger.error("未能选择选题，任务结束")
|
|
return False
|
|
|
|
content = self.create_content(topic_data)
|
|
images = self.generate_images(topic_data["topic"].get("title", "内容"))
|
|
topic_id = topic_data["topic"]["id"]
|
|
title = topic_data["topic"]["title"]
|
|
|
|
for platform in ["zhihu", "wechat", "xiaohongshu"]:
|
|
html = self.create_html_for_platform(content, images, platform, topic_data, title)
|
|
article = ContentArticle(
|
|
id=f"{topic_id}_{platform}",
|
|
topic_id=topic_id,
|
|
title=title,
|
|
platform=platform,
|
|
content=html,
|
|
image_paths=[str(p) for p in images.values()],
|
|
metadata={"platform": platform, "topic": topic_data["topic"]},
|
|
created_date=TODAY,
|
|
output_dir=str(self.release_dir / platform),
|
|
status="pending_review" # 生成后待审查
|
|
)
|
|
self.save_article(article)
|
|
self.articles.append(article)
|
|
|
|
# 标记选题为待发布（审查通过）
|
|
self.mark_topic_ready(topic_id)
|
|
logger.info(f"创作完成: {len(self.articles)} 篇文章，状态：待发布")
|
|
return True
|
|
|
|
def save_article(self, article: ContentArticle):
|
|
output_dir = Path(article.output_dir)
|
|
output_dir.mkdir(parents=True, exist_ok=True)
|
|
html_file = output_dir / f"{article.platform}_{article.id}.html"
|
|
with open(html_file, 'w', encoding='utf-8') as f:
|
|
f.write(article.content)
|
|
meta_file = output_dir / f"{article.platform}_{article.id}.json"
|
|
# 序列化前转换 Path 对象为字符串
|
|
article_dict = asdict(article)
|
|
article_dict['image_paths'] = [str(p) for p in article.image_paths]
|
|
json.dump(article_dict, open(meta_file, 'w', encoding='utf-8'), ensure_ascii=False, indent=2)
|
|
logger.info(f"保存了 {article.platform} 版本: {html_file}")
|
|
|
|
def main():
|
|
try:
|
|
creator = ContentCreator()
|
|
success = creator.run()
|
|
if success:
|
|
print(f"SUCCESS: Created {len(creator.articles)} articles for {TODAY} (status: 待发布)")
|
|
sys.exit(0)
|
|
else:
|
|
print("WARNING: Content creation failed")
|
|
sys.exit(1)
|
|
except Exception as e:
|
|
logger.error(f"创作任务失败: {e}")
|
|
print(f"ERROR: {e}")
|
|
sys.exit(1)
|
|
|
|
if __name__ == "__main__":
|
|
main()
|