[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / zapiks.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8     parse_duration,
9     parse_iso8601,
10     xpath_with_ns,
11     xpath_text,
12     int_or_none,
13 )
14
15
16 class ZapiksIE(InfoExtractor):
17     _VALID_URL = r'https?://(?:www\.)?zapiks\.(?:fr|com)/(?:(?:[a-z]{2}/)?(?P<display_id>.+?)\.html|index\.php\?.*\bmedia_id=(?P<id>\d+))'
18     _TESTS = [
19         {
20             'url': 'http://www.zapiks.fr/ep2s3-bon-appetit-eh-be-viva.html',
21             'md5': 'aeb3c473b2d564b2d46d664d28d5f050',
22             'info_dict': {
23                 'id': '80798',
24                 'ext': 'mp4',
25                 'title': 'EP2S3 - Bon Appétit - Eh bé viva les pyrénées con!',
26                 'description': 'md5:7054d6f6f620c6519be1fe710d4da847',
27                 'thumbnail': r're:^https?://.*\.jpg$',
28                 'duration': 528,
29                 'timestamp': 1359044972,
30                 'upload_date': '20130124',
31                 'view_count': int,
32             },
33         },
34         {
35             'url': 'http://www.zapiks.com/ep3s5-bon-appetit-baqueira-m-1.html',
36             'only_matching': True,
37         },
38         {
39             'url': 'http://www.zapiks.com/nl/ep3s5-bon-appetit-baqueira-m-1.html',
40             'only_matching': True,
41         },
42         {
43             'url': 'http://www.zapiks.fr/index.php?action=playerIframe&amp;media_id=118046&amp;width=640&amp;height=360&amp;autoStart=false&amp;language=fr',
44             'only_matching': True,
45         },
46     ]
47
48     def _real_extract(self, url):
49         mobj = re.match(self._VALID_URL, url)
50         video_id = mobj.group('id')
51         display_id = mobj.group('display_id') or video_id
52
53         webpage = self._download_webpage(url, display_id)
54
55         if not video_id:
56             video_id = self._search_regex(
57                 r'data-media-id="(\d+)"', webpage, 'video id')
58
59         playlist = self._download_xml(
60             'http://www.zapiks.fr/view/index.php?action=playlist&media_id=%s&lang=en' % video_id,
61             display_id)
62
63         NS_MAP = {
64             'jwplayer': 'http://rss.jwpcdn.com/'
65         }
66
67         def ns(path):
68             return xpath_with_ns(path, NS_MAP)
69
70         item = playlist.find('./channel/item')
71
72         title = xpath_text(item, 'title', 'title') or self._og_search_title(webpage)
73         description = self._og_search_description(webpage, default=None)
74         thumbnail = xpath_text(
75             item, ns('./jwplayer:image'), 'thumbnail') or self._og_search_thumbnail(webpage, default=None)
76         duration = parse_duration(self._html_search_meta(
77             'duration', webpage, 'duration', default=None))
78         timestamp = parse_iso8601(self._html_search_meta(
79             'uploadDate', webpage, 'upload date', default=None), ' ')
80
81         view_count = int_or_none(self._search_regex(
82             r'UserPlays:(\d+)', webpage, 'view count', default=None))
83         comment_count = int_or_none(self._search_regex(
84             r'UserComments:(\d+)', webpage, 'comment count', default=None))
85
86         formats = []
87         for source in item.findall(ns('./jwplayer:source')):
88             format_id = source.attrib['label']
89             f = {
90                 'url': source.attrib['file'],
91                 'format_id': format_id,
92             }
93             m = re.search(r'^(?P<height>\d+)[pP]', format_id)
94             if m:
95                 f['height'] = int(m.group('height'))
96             formats.append(f)
97         self._sort_formats(formats)
98
99         return {
100             'id': video_id,
101             'title': title,
102             'description': description,
103             'thumbnail': thumbnail,
104             'duration': duration,
105             'timestamp': timestamp,
106             'view_count': view_count,
107             'comment_count': comment_count,
108             'formats': formats,
109         }