#!/usr/bin/env python3
"""
将 content/ideas/ 目录下的 Markdown 选题文件转换为 JSON 格式
供 content creator 脚本使用
"""

import os
import sys
import json
import re
from pathlib import Path
from datetime import datetime

PROJECT_ROOT = Path(__file__).parent.parent
IDEAS_DIR = PROJECT_ROOT / "content" / "ideas"
DATA_DIR = PROJECT_ROOT / "automation" / "data"
OUTPUT_FILE = DATA_DIR / "sustainability_topics.json"

def extract_field(content, field_name):
    """从 Markdown 中提取字段值"""
    # 支持 **字段名**：值 或 字段名：值 格式
    patterns = [
        rf"\*\*{re.escape(field_name)}\*\*\s*[:：]\s*(.+?)(?:\n|$)",
        rf"{re.escape(field_name)}\s*[:：]\s*(.+?)(?:\n|$)",
    ]
    for pattern in patterns:
        match = re.search(pattern, content, re.MULTILINE)
        if match:
            return match.group(1).strip()
    return None

def extract_list(content, start_keyword):
    """提取列表数据（如数据/案例）"""
    lines = content.split('\n')
    result = []
    capturing = False
    for line in lines:
        if start_keyword in line:
            capturing = True
            continue
        if capturing:
            if line.strip().startswith(('**', '#', '-', '*', '1.', '2.')):
                if re.match(r'^(#|\*\*|-|\*|\d+\.)\s', line):
                    result.append(line.strip())
            elif line.strip() == '' or line.startswith('##'):
                break
    return result

def parse_evaluation_matrix(content):
    """解析选题评估矩阵表格"""
    scores = {}
    lines = content.split('\n')
    in_table = False
    for line in lines:
        if '|' in line and '---' not in line and '维度' not in line:
            parts = [p.strip() for p in line.split('|')]
            if len(parts) >= 3:
                dimension = parts[1]
                score_str = parts[2]
                try:
                    score = int(score_str)
                    scores[dimension] = score
                except:
                    pass
        if '**总分**' in line:
            total_match = re.search(r'\*\*总分\*\*\s*\|\s*\*\*(\d+)\*\*', line)
            if total_match:
                scores['总分'] = int(total_match.group(1))
    return scores

def md_to_topic(md_path):
    """将单个 Markdown 文件转换为 topic 字典"""
    with open(md_path, 'r', encoding='utf-8') as f:
        content = f.read()

    # 提取标题 (第一行 # 开头)
    title_match = re.search(r'^#\s+(.+)$', content, re.MULTILINE)
    title = title_match.group(1).strip() if title_match else md_path.stem

    # 提取基础字段
    field = extract_field(content, '领域')
    format_type = extract_field(content, '形式')
    word_count = extract_field(content, '预估字数')
    core_concept = extract_field(content, '核心观点')
    audience_pain = extract_field(content, '受众痛点')
    unique_angle = extract_field(content, '独特角度')
    data_cases = extract_list(content, '数据/案例')
    estimated_days = extract_field(content, '预估完成时间')
    priority_str = extract_field(content, '优先级')
    publish_date = extract_field(content, '预计发布时间')
    status = extract_field(content, '状态') or '待处理'

    # 解析优先级为分数
    priority_map = {'高': 10, '中': 7, '低': 4}
    priority_score = priority_map.get(priority_str, 5)

    # 解析评估矩阵
    evaluation = parse_evaluation_matrix(content)
    total_score = evaluation.get('总分', 0)

    # 生成 topic ID
    topic_id = md_path.stem.split('-')[0]  # 如 "001-上海阳台种菜一年.md" -> "001"

    # 构建 topic 对象
    topic = {
        "id": topic_id,
        "title": title,
        "field": field or "未知",
        "format": format_type or "未指定",
        "word_count": word_count,
        "core_concept": core_concept,
        "audience_pain": audience_pain,
        "unique_angle": unique_angle,
        "data_cases": data_cases,
        "estimated_days": estimated_days,
        "priority": priority_str,
        "priority_score": priority_score if priority_score > 0 else (total_score if total_score > 0 else 5),
        "publish_date": publish_date,
        "status": status,
        "evaluation": evaluation,
        "total_score": total_score,
        "cases": [],  # 关联的案例ID列表，待填充
        "source_file": md_path.name,
        "created_at": datetime.now().isoformat()
    }

    return topic

def main():
    """主函数：导入所有 Markdown 选题文件"""
    if not IDEAS_DIR.exists():
        print(f"错误：选题目录不存在 {IDEAS_DIR}")
        return

    # 只导入主选题文件（格式：NNN-标题.md），排除 research/compliance 等辅助文件
    md_files = []
    for f in IDEAS_DIR.glob("*.md"):
        if f.name == "README.md":
            continue
        # 排除 research 和 compliance 文件
        if f.name.endswith('-research.md') or f.name.endswith('-compliance.md'):
            continue
        # 匹配 001-xxx.md 格式
        if re.match(r'^\d{3}-.+\.md$', f.name):
            md_files.append(f)

    if not md_files:
        print("未找到选题文件")
        return

    print(f"找到 {len(md_files)} 个选题文件，开始导入...")

    topics = []
    for md_file in sorted(md_files):
        print(f"  处理: {md_file.name}")
        topic = md_to_topic(md_file)
        topics.append(topic)
        print(f"    标题: {topic['title']}")
        print(f"    总分: {topic['total_score']}")
        print(f"    状态: {topic['status']}")

    # 确保输出目录存在
    DATA_DIR.mkdir(parents=True, exist_ok=True)

    # 写入 JSON
    with open(OUTPUT_FILE, 'w', encoding='utf-8') as f:
        json.dump(topics, f, ensure_ascii=False, indent=2)

    print(f"\n✅ 已导入 {len(topics)} 个选题到 {OUTPUT_FILE}")

    # 统计
    ready_topics = [t for t in topics if t['status'] != '已发布']
    print(f"📊 可用选题数: {len(ready_topics)}")
    avg_score = sum(t['total_score'] for t in ready_topics) / len(ready_topics) if ready_topics else 0
    print(f"🎯 平均评分: {avg_score:.1f}")

if __name__ == "__main__":
    main()