import urllib.parse
import xml.etree.ElementTree as ET
import oracledb
from datetime import datetime
import sys
import json

def wrap_in_editorjs(html_content):
    editor_json = {
        "time": int(datetime.now().timestamp() * 1000),
        "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/',
        'dc': 'http://purl.org/dc/elements/1.1/'
    }

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

    dt = datetime.now()
    layout = "SPECIALTY_DETAIL"
    page_type = "CUSTOM"
    status = "PUBLISHED"

    count = 0
    warnings = []

    for channel in root.findall('channel'):
        for item in channel.findall('item'):
            title_tag = item.find('title')
            title = title_tag.text if title_tag is not None else "Untitled"
            
            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:
                link_tag = item.find('link')
                slug = link_tag.text.strip('/').split('/')[-1] if link_tag is not None else "unknown"
                
            slug = urllib.parse.unquote(slug)
            
            if slug == "khoa-than-kinh-dot-quy-2":
                slug = "khoa-than-kinh-dot-quy"
            
            # No longer skipping khoa-cap-cuu

            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 missing container
            if '<div class="wp-content-container' not in content_text and content_text.strip():
                warnings.append(f"CẢNH BÁO: Bài viết '{title}' ({slug}) thiếu thẻ container. Hệ thống đã tự động bổ sung.")
                content_text = f'<div class="wp-content-container prose max-w-none">\n{content_text}\n</div>'

            # Wrap into Editor.js JSON
            final_content = wrap_in_editorjs(content_text)

            cursor.execute("SELECT id FROM sis_page WHERE slug = :1", [slug])
            row = cursor.fetchone()
            
            if row:
                sql = "UPDATE sis_page SET title = :1, content = :2, layout = :3, page_type = :4 WHERE slug = :5"
                cursor.execute(sql, [title, final_content, layout, page_type, slug])
            else:
                sql = """
                    INSERT INTO sis_page (
                        id, title, slug, content, status, page_type, display_order, layout,
                        created_date, last_modified_date, created_by, last_modified_by
                    ) VALUES (
                        sequence_generator.nextval, :title, :slug, :content, :status, :page_type, 0, :layout,
                        :created_date, :last_modified_date, 'system', 'system'
                    )
                """
                cursor.execute(sql, {
                    'title': title,
                    'slug': slug,
                    'content': final_content,
                    'status': status,
                    'page_type': page_type,
                    'layout': layout,
                    'created_date': dt,
                    'last_modified_date': dt
                })
            count += 1

    conn.commit()
    print(f"\n--- TỔNG KẾT ---")
    print(f"Successfully processed {count} specialty pages.")
    for w in warnings:
        print(w)

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