#!/usr/bin/env python3 """ 内容创作脚本（修复版）支持：标题长度限制、标签合规、状态流程 """ import os, sys, yaml, json, datetime, logging, random from pathlib import Path from typing import Dict, List import subprocess 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 not config_file.exists(): logger.warning("配置文件不存在，使用默认") self.wecom_config = {"content_rules": {}} return with open(config_file, 'r', encoding='utf-8') as f: self.wecom_config = yaml.safe_load(f) logger.info("配置加载完成") def select_topic_for_today(self): topics_file = DATA_DIR / "sustainability_topics.json" if not topics_file.exists(): logger.error("选题库文件不存在") return None 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") != "已发布" and t.get("status") != "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", "") sections = [ f"

{title}

", f"

今天是{TODAY}，我们探讨「{title}」。根据全球案例与本土分析，给出以下建议：

", "

核心观点

" + topic.get("core_concept", "待补充") + "

", "

目标受众痛点

" + topic.get("audience_pain", "待补充") + "

", "

本土化方案

" + topic.get("unique_angle", "待补充") + "

", "

MVP行动

记录现状
小步尝试
评估效果
建立习惯

", "

（本文由宇之然AI助手生成，数据来源可靠，内容经合规审查）

" ] return "\n".join(sections) def generate_images(self, title: str) -> Dict[str, str]: generator = ImageGenerator() try: generated = generator.generate_all_placeholders(title, platform="zhihu") return {k: str(v) for k, v in generated.items()} 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, 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 = "{{TITLE}}

" # 插入图片标记 for marker, path in images.items(): if path: img_tag = f'

' content = content.replace(f"[IMAGE: {marker}]", img_tag) # 平台特定处理 extra = "" if platform == "zhihu": # 使用平台允许的标签（科技、职场、AI都在允许列表） extra = '

#科技 #职场 #AI

' elif platform == "wechat": # 截断标题 if len(title) > 32: title = title[:29] + "..." abstract = content[:100] + "..." extra = f'

{abstract}

' elif platform == "xiaohongshu": # 小红书允许标签：生活方式、可持续、AI extra = '

#AI #科技 #生活方式

' full_content = content + extra html = template.replace("", full_content) html = html.replace("{{DATE}}", TODAY) html = html.replace("{{TITLE}}", 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, title) article = ContentArticle( id=f"{topic_id}_{platform}", topic_id=topic_id, title=title, platform=platform, content=html, image_paths=list(images.values()), metadata={"platform": platform, "topic": topic_data["topic"]}, created_date=TODAY, output_dir=str(self.release_dir / platform), status="draft" ) self.save_article(article) self.articles.append(article) # 标记为待发布（而不是已发布） self.mark_topic_ready(topic_id) # 发送通知（可选） 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" with open(meta_file, 'w', encoding='utf-8') as f: json.dump(asdict(article), f, 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()