import xml.etree.ElementTree as ET
import csv
import sys
import re

def clean_html(raw_html):
    if not raw_html:
        return ""
    cleanr = re.compile('<.*?>')
    cleantext = re.sub(cleanr, '', raw_html)
    return cleantext.strip().replace('\n', ' ')

try:
    tree = ET.parse('/home/x79/sisvietnamvn_01/sisvietnamvn_Trang chính thức hiện tại/bnhvinakhoaquctsiscnth.WordPress.2026-07-28 (5).xml')
    root = tree.getroot()
except Exception as e:
    print(f"Error parsing XML: {e}")
    sys.exit(1)

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

# First pass to build a dictionary of attachment URLs
attachments = {}
for channel in root.findall('channel'):
    for item in channel.findall('item'):
        post_type_el = item.find('wp:post_type', ns)
        if post_type_el is not None and post_type_el.text == 'attachment':
            post_id_el = item.find('wp:post_id', ns)
            attachment_url_el = item.find('wp:attachment_url', ns)
            if post_id_el is not None and attachment_url_el is not None:
                attachments[post_id_el.text] = attachment_url_el.text

csv_path = '/home/x79/sisvietnamvn_01/sisvietnamvn_main/doctors_from_xml.csv'
with open(csv_path, 'w', newline='', encoding='utf-8') as csvfile:
    fieldnames = ['Name', 'Position', 'Gender', 'Birthday', 'Phone', 'Email', 'Address', 'Order', 'ImageID', 'ImageURL', 'Content']
    writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
    writer.writeheader()
    
    for channel in root.findall('channel'):
        for item in channel.findall('item'):
            post_type_el = item.find('wp:post_type', ns)
            if post_type_el is not None and post_type_el.text == 'employee':
                title = item.find('title')
                name = title.text if title is not None else ""
                
                content_el = item.find('content:encoded', ns)
                content = clean_html(content_el.text) if content_el is not None else ""
                
                doctor_data = {
                    'Name': name,
                    'Position': '',
                    'Gender': '',
                    'Birthday': '',
                    'Phone': '',
                    'Email': '',
                    'Address': '',
                    'Order': '',
                    'ImageID': '',
                    'ImageURL': '',
                    'Content': content[:200] # truncate for CSV view
                }
                
                for meta in item.findall('wp:postmeta', ns):
                    key_el = meta.find('wp:meta_key', ns)
                    val_el = meta.find('wp:meta_value', ns)
                    
                    if key_el is not None and key_el.text and val_el is not None:
                        k = key_el.text
                        v = val_el.text or ""
                        if k == 'position': doctor_data['Position'] = v
                        elif k == 'gender': doctor_data['Gender'] = v
                        elif k == 'birthday': doctor_data['Birthday'] = v
                        elif k == 'phone_number': doctor_data['Phone'] = v
                        elif k == 'email': doctor_data['Email'] = v
                        elif k == 'address': doctor_data['Address'] = v
                        elif k == 'order': doctor_data['Order'] = v
                        elif k == 'image':
                            doctor_data['ImageID'] = v
                            doctor_data['ImageURL'] = attachments.get(v, '')
                            
                writer.writerow(doctor_data)

print(f"Successfully wrote CSV to {csv_path}")