#!/usr/bin/env python3 import dateutil.parser import io import feedparser import re from urllib.parse import urlparse def main(): non_slug_chars = re.compile(r'[^a-zA-Z0-9-_]') # Get feed for d in feed['entries']: # Generate file name url = urlparse(d['link']).path slug = url while non_slug_chars.match(slug[0]): slug = slug[1:] slug = non_slug_chars.sub('_', slug) file_name = 'scholarship/%s.html' % slug # Parse date date = dateutil.parser.parse(d['date']) # Open output file with io.open(file_name, 'w', encoding='utf-8') as fp: fp.write(u"""\

{text}

""".format(title=d['title'].strip(), slug=slug, date=date.strftime('%Y-%m-%d %H:%M:%S'), link=d['link'].strip(), text=d['summary'].strip())) if __name__ == '__main__': main()