Add support for https for all extractors as preventive and future-proof measure
[youtube-dl] / youtube_dl / extractor / ubu.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     int_or_none,
8     qualities,
9 )
10
11
12 class UbuIE(InfoExtractor):
13     _VALID_URL = r'https?://(?:www\.)?ubu\.com/film/(?P<id>[\da-z_-]+)\.html'
14     _TEST = {
15         'url': 'http://ubu.com/film/her_noise.html',
16         'md5': '138d5652618bf0f03878978db9bef1ee',
17         'info_dict': {
18             'id': 'her_noise',
19             'ext': 'm4v',
20             'title': 'Her Noise - The Making Of (2007)',
21             'duration': 3600,
22         },
23     }
24
25     def _real_extract(self, url):
26         video_id = self._match_id(url)
27         webpage = self._download_webpage(url, video_id)
28
29         title = self._html_search_regex(
30             r'<title>.+?Film &amp; Video: ([^<]+)</title>', webpage, 'title')
31
32         duration = int_or_none(self._html_search_regex(
33             r'Duration: (\d+) minutes', webpage, 'duration', fatal=False),
34             invscale=60)
35
36         formats = []
37         FORMAT_REGEXES = [
38             ('sq', r"'flashvars'\s*,\s*'file=([^']+)'"),
39             ('hq', r'href="(http://ubumexico\.centro\.org\.mx/video/[^"]+)"'),
40         ]
41         preference = qualities([fid for fid, _ in FORMAT_REGEXES])
42         for format_id, format_regex in FORMAT_REGEXES:
43             m = re.search(format_regex, webpage)
44             if m:
45                 formats.append({
46                     'url': m.group(1),
47                     'format_id': format_id,
48                     'preference': preference(format_id),
49                 })
50         self._sort_formats(formats)
51
52         return {
53             'id': video_id,
54             'title': title,
55             'duration': duration,
56             'formats': formats,
57         }