extract_phil_wilson_posts.py

import os
import re
import xml.etree.ElementTree as ET
import html2text

# This script extracts posts authored by Phil Wilson from the University of Bath's Web Services Blog RSS feed
# and converts them to Markdown format with appropriate frontmatter for Zola.
# It assumes the XML file is well-formed and contains the necessary elements.
# The output will be saved in a specified directory, organized by year and month based on the publication date.


# Path to the XML file and output directory
XML_PATH = 'webservices.xml'
OUTPUT_DIR = 'markdown'

# Ensure output directory exists
os.makedirs(OUTPUT_DIR, exist_ok=True)

def sanitize_filename(title):
    """Create a safe filename from the post title."""
    filename = re.sub(r'[^a-zA-Z0-9\-_ ]', '', title)
    filename = filename.strip().replace(' ', '-').lower()
    return filename + '.md'

def extract_phil_wilson_posts(xml_path, output_dir):
    # Parse the XML file
    tree = ET.parse(xml_path)
    root = tree.getroot()

    ns = {
        'dc': 'http://purl.org/dc/elements/1.1/',
        'ns2': 'http://purl.org/rss/1.0/modules/content/'
    }

    from datetime import datetime
    import email.utils

    for item in root.iter('item'):
        creator = item.find('dc:creator', ns)
        if creator is not None and creator.text == 'Phil Wilson':
            title_elem = item.find('title')
            encoded_elem = item.find('ns2:encoded', ns)
            desc_elem = item.find('description')
            link_elem = item.find('link')
            pubdate_elem = item.find('pubDate')
            categories = [cat.text for cat in item.findall('category') if cat.text]

            if title_elem is not None:
                title = title_elem.text or 'untitled'
                if encoded_elem is not None and encoded_elem.text and encoded_elem.text.strip():
                    content = encoded_elem.text.strip()
                elif desc_elem is not None and desc_elem.text:
                    content = desc_elem.text.strip()
                else:
                    content = ''
                filename = sanitize_filename(title)
                # Determine output subdirectory based on date
                subdir = OUTPUT_DIR
                if pubdate_elem is not None and pubdate_elem.text:
                    try:
                        # Parse RFC822 date and format as YYYY-MM-DD HH:MM:SS
                        parsed = email.utils.parsedate_to_datetime(pubdate_elem.text)
                        date_str = parsed.strftime('%Y-%m-%d %H:%M:%S')
                        year, month = date_str.split()[0].split('-')[:2]
                        subdir = os.path.join(OUTPUT_DIR, year, month)
                        os.makedirs(subdir, exist_ok=True)
                    except Exception:
                        pass
                out_path = os.path.join(subdir, filename)
                # Extract slug from link
                slug = ''
                if link_elem is not None and link_elem.text:
                    # Remove trailing slash, then take last path segment
                    parts = link_elem.text.rstrip('/').split('/')
                    if parts:
                        slug = parts[-1]
                # Build frontmatter
                frontmatter = ['---']
                frontmatter.append(f'title: "{title}"')
                if slug:
                    frontmatter.append(f'slug: {slug}')
                if date_str:
                    frontmatter.append(f'date: {date_str}')
                # Output categories as tags under taxonomies, always lowercase
                if categories or True:
                    frontmatter.append('taxonomies:')
                    frontmatter.append('  tags:')
                    for cat in categories:
                        frontmatter.append(f'    - {cat.lower()}')
                    frontmatter.append('  collections:')
                    frontmatter.append('    - university of bath')
                # Add extra metadata
                frontmatter.append('extra:')
                frontmatter.append('  originalLocationName: The University of Bath\'s Web Services Blog')
                if link_elem is not None and link_elem.text:
                    frontmatter.append(f'  originalUrl: {link_elem.text}')
                frontmatter.append('---\n')
                # Convert HTML content to Markdown with no line length limit
                h2t = html2text.HTML2Text()
                h2t.body_width = 0  # No line wrapping
                content_md = h2t.handle(content)
                with open(out_path, 'w', encoding='utf-8') as f:
                    f.write('\n'.join(frontmatter))
                    f.write(content_md + '\n')
                print(f'Wrote: {out_path}')

if __name__ == '__main__':
    extract_phil_wilson_posts(XML_PATH, OUTPUT_DIR)