import urllib.parse
import xml.etree.ElementTree as ET
import oracledb
import json
import re

def extract_siteorigin_html(content_text):
    # Find all occurrences of siteorigin-panels block
    # It might look like <!-- wp:siteorigin-panels/layout-block {"panelsData":{...}} /-->
    html_parts = []
    
    # regex to match <!-- wp:siteorigin-panels/layout-block {json} /-->
    matches = re.finditer(r'<!--\s*wp:siteorigin-panels/layout-block\s+(.*?)\s*/?-->', content_text, re.DOTALL)
    
    has_siteorigin = False
    for match in matches:
        has_siteorigin = True
        json_str = match.group(1).strip()
        if json_str.endswith('/'):
            json_str = json_str[:-1].strip()
            
        try:
            data = json.loads(json_str)
            if 'panelsData' in data and 'widgets' in data['panelsData']:
                widgets = data['panelsData']['widgets']
                for widget in widgets:
                    if 'text' in widget:
                        # Extract the HTML text from the widget
                        text = widget['text']
                        # The text might have unicode escapes like \u003c
                        # json.loads already decoded it!
                        html_parts.append(text)
        except Exception as e:
            print(f"Failed to parse JSON: {e}")
            pass

    if has_siteorigin:
        return "\n".join(html_parts)
    return content_text

def wrap_in_editorjs(html_content):
    if not html_content.strip():
        return ""
    editor_json = {
        "time": 1620000000,
        "blocks": [
            {
                "type": "raw",
                "data": {
                    "html": html_content
                }
            }
        ],
        "version": "2.22.2"
    }
    return json.dumps(editor_json)

def main(xml_file_path):
    print(f"Reading {xml_file_path} ...")
    tree = ET.parse(xml_file_path)
    root = tree.getroot()

    ns = {
        'wp': 'http://wordpress.org/export/1.2/',
        'content': 'http://purl.org/rss/1.0/modules/content/'
    }

    dsn = "localhost:1521/sisvietnam"
    conn = oracledb.connect(user="sisvietnam", password="sisvietnam", dsn=dsn)
    cursor = conn.cursor()

    count = 0

    for item in root.findall('.//item'):
        post_name_tag = item.find('wp:post_name', ns)
        if post_name_tag is not None and post_name_tag.text:
            slug = post_name_tag.text
        else:
            continue
            
        slug = urllib.parse.unquote(slug)
        if slug == "khoa-than-kinh-dot-quy-2":
            slug = "khoa-than-kinh-dot-quy"

        content_tag = item.find('content:encoded', ns)
        content_text = content_tag.text if content_tag is not None and content_tag.text else ""
        
        # Check if it has siteorigin
        if 'wp:siteorigin-panels/layout-block' in content_text:
            print(f"Found SiteOrigin content in: {slug}")
            extracted_html = extract_siteorigin_html(content_text)
            
            # Wrap in wrapper
            extracted_html = f'<div class="wp-content-container prose max-w-none">\n{extracted_html}\n</div>'
            
            final_content = wrap_in_editorjs(extracted_html)
            
            cursor.execute("UPDATE sis_page SET content = :1 WHERE slug = :2", [final_content, slug])
            count += 1

    conn.commit()
    print(f"\n--- TỔNG KẾT ---")
    print(f"Successfully processed {count} SiteOrigin pages.")

if __name__ == "__main__":
    import sys
    if len(sys.argv) > 1:
        main(sys.argv[1])
    else:
        print("Usage: python3 fix_siteorigin.py <xml_file>")