import requests
import xml.etree.ElementTree as ET

# 取得したい法令名
LAW_NAME = "著作権法" 

def fetch_law_data(law_name):
    # 1. 法令リストからLawIdを検索
    print(f"[{law_name}] を検索中...")
    list_url = "https://elaws.e-gov.go.jp/api/1/lawlists/1"
    response = requests.get(list_url)
    root = ET.fromstring(response.content)
    
    law_id = None
    for law in root.findall(".//LawNameListInfo"):
        if law.find("LawName").text == law_name:
            law_id = law.find("LawId").text
            break
            
    if not law_id:
        print(f"Error: {law_name} が見つかりませんでした。")
        return None

    # 2. 条文データを取得
    print(f"条文データをダウンロード中... (ID: {law_id})")
    detail_url = f"https://elaws.e-gov.go.jp/api/1/lawdata/{law_id}"
    law_response = requests.get(detail_url)
    return law_response.content

def parse_to_markdown(xml_content):
    print("Markdownに変換中...")
    root = ET.fromstring(xml_content)
    markdown_text = f"# {LAW_NAME}\n\n"
    
    # 条文（Article）の取得
    articles = root.findall(".//Article")
    
    for article in articles:
        # 条数
        article_caption = article.find("ArticleCaption")
        caption_text = article_caption.text if article_caption is not None else ""
        
        # 見出し
        article_title = article.find("ArticleTitle")
        title_text = article_title.text if article_title is not None else ""
        
        markdown_text += f"## {title_text} {caption_text}\n"
        
        # 項（Paragraph）の取得
        paragraphs = article.findall(".//Paragraph")
        for p in paragraphs:
            p_num = p.find("ParagraphNum")
            p_sent = p.find(".//Sentence") # 項の本文
            
            num_text = p_num.text if p_num is not None else ""
            sent_text = p_sent.text if p_sent is not None and p_sent.text else ""
            
            markdown_text += f"- **{num_text}**: {sent_text}\n"
            
            # --- 【修正点】号（Item）の取得 ---
            items = p.findall("Item")
            for item in items:
                i_title = item.find("ItemTitle") # 号名（例：一）
                i_sent = item.find(".//Sentence") # 号の本文
                
                it_text = i_title.text if i_title is not None else ""
                is_text = i_sent.text if i_sent is not None else ""
                
                # インデントを下げてリスト表示
                markdown_text += f"    - **{it_text}**: {is_text}\n"

                # さらにその下（イロハ等：Subitem1）がある場合
                subitems1 = item.findall("Subitem1")
                for sub1 in subitems1:
                    s1_title = sub1.find("Subitem1Title")
                    s1_sent = sub1.find(".//Sentence")
                    
                    s1_t_text = s1_title.text if s1_title is not None else ""
                    s1_s_text = s1_sent.text if s1_sent is not None else ""
                    
                    markdown_text += f"        - **{s1_t_text}**: {s1_s_text}\n"

        markdown_text += "\n"
        
    return markdown_text

# 実行
xml_data = fetch_law_data(LAW_NAME)
if xml_data:
    md_output = parse_to_markdown(xml_data)
    
    filename = f"{LAW_NAME}_full.md"
    with open(filename, "w", encoding="utf-8") as f:
        f.write(md_output)
    print(f"完了！ファイルを保存しました: {filename}")

# JSONとして保存するイメージ
law_dict = {
    "law_name": "著作権法",
    "articles": [
        {"id": "1", "title": "目的", "text": "この法律は..."},
        {"id": "2", "title": "定義", "text": "この法律において..."},
    ]
}