#!/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) -> Tuple[str, Optional[str]]:
    """用 LLM 优化文章内容，返回 (html, log_message_or_None)"""
    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创作助手。"

        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）。

原文：
{html}

优化后："""
        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:
            return polished, "LLM 内容优化"
    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) -> 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]}")
        img_tags = re.findall(r'<img[^>]*>', html, re.IGNORECASE)
        for tag in img_tags:
            m = re.search(r'src=["\']([^"\']+)["\']', tag, re.IGNORECASE)
            if m:
                src = m.group(1)
                if not src.startswith('data:image/'):
                    logs.append(f"图片未内联: {src[:50]}... 需手动修复")
        polished, pol_log = polish_with_llm(html, platform)
        if pol_log:
            html = polished
            logs.append(pol_log)
    return html, logs

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")

    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, "need_manual": 0, "average_score": 0},
                "details": [], "all_passed": True
            }, f, ensure_ascii=False, indent=2)
        print("OPTIMIZATION_COMPLETE: 0 articles found")
        sys.exit(0)

    topic_map = load_topic_map()
    results = []
    all_passed = True

    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

        check_result = check_article(html, platform_dir, topic_data)
        issues = check_result['issues']
        score = check_result['score']
        label = f"{platform_dir}/{topic_id}"

        if issues:
            optimized_html, opt_logs = optimize_article(html, platform_dir, topic_data)
            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 ({len(issues)} issues fixed)")
                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(recheck['issues']),
                    final_score=recheck['score'],
                    status="passed"
                ))
            else:
                logger.warning(f"⚠️ {label} 仍有 {len(recheck['issues'])} 个问题需人工处理")
                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(recheck['issues']),
                    final_score=recheck['score'],
                    status="manual_review"
                ))
                all_passed = False
        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_ids = set()
    for res in results:
        if res.status == "passed":
            passed_ids.add(res.topic_id)
    for tid in passed_ids:
        update_topic_status(tid, 'ready')
        logger.info(f"选题 {tid} 状态 → ready（待发布）")

    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": sum(1 for r in results if r.status == "passed"),
            "need_manual": sum(1 for r in results if r.status == "manual_review"),
            "average_score": sum(r.final_score for r in results) / len(results) if results else 0
        },
        "details": [asdict(r) for r in results],
        "all_passed": all_passed
    }
    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, {sum(1 for r in results if r.status=='passed')} 篇通过")
    print(f"OPTIMIZATION_COMPLETE: {len(results)} articles, {sum(1 for r in results if r.status=='passed')} passed, {sum(1 for r in results if r.status=='manual_review')} need manual review")
    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)