]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/zapiks.py
[brightcove:new] extract subtitles and strip video title
[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': 're:^https?://.*\.jpg$',
28                 'duration': 528,
29                 'timestamp': 1359044972,
30                 'upload_date': '20130124',
31                 'view_count': int,
32                 'comment_count': int,
33             },
34         },
35         {
36             'url': 'http://www.zapiks.com/ep3s5-bon-appetit-baqueira-m-1.html',
37             'only_matching': True,
38         },
39         {
40             'url': 'http://www.zapiks.com/nl/ep3s5-bon-appetit-baqueira-m-1.html',
41             'only_matching': True,
42         },
43         {
44             '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',
45             'only_matching': True,
46         },
47     ]
48
49     def _real_extract(self, url):
50         mobj = re.match(self._VALID_URL, url)
51         video_id = mobj.group('id')
52         display_id = mobj.group('display_id') or video_id
53
54         webpage = self._download_webpage(url, display_id)
55
56         if not video_id:
57             video_id = self._search_regex(
58                 r'data-media-id="(\d+)"', webpage, 'video id')
59
60         playlist = self._download_xml(
61             'http://www.zapiks.fr/view/index.php?action=playlist&media_id=%s&lang=en' % video_id,
62             display_id)
63
64         NS_MAP = {
65             'jwplayer': 'http://rss.jwpcdn.com/'
66         }
67
68         def ns(path):
69             return xpath_with_ns(path, NS_MAP)
70
71         item = playlist.find('./channel/item')
72
73         title = xpath_text(item, 'title', 'title') or self._og_search_title(webpage)
74         description = self._og_search_description(webpage, default=None)
75         thumbnail = xpath_text(
76             item, ns('./jwplayer:image'), 'thumbnail') or self._og_search_thumbnail(webpage, default=None)
77         duration = parse_duration(self._html_search_meta(
78             'duration', webpage, 'duration', default=None))
79         timestamp = parse_iso8601(self._html_search_meta(
80             'uploadDate', webpage, 'upload date', default=None), ' ')
81
82         view_count = int_or_none(self._search_regex(
83             r'UserPlays:(\d+)', webpage, 'view count', default=None))
84         comment_count = int_or_none(self._search_regex(
85             r'UserComments:(\d+)', webpage, 'comment count', default=None))
86
87         formats = []
88         for source in item.findall(ns('./jwplayer:source')):
89             format_id = source.attrib['label']
90             f = {
91                 'url': source.attrib['file'],
92                 'format_id': format_id,
93             }
94             m = re.search(r'^(?P<height>\d+)[pP]', format_id)
95             if m:
96                 f['height'] = int(m.group('height'))
97             formats.append(f)
98         self._sort_formats(formats)
99
100         return {
101             'id': video_id,
102             'title': title,
103             'description': description,
104             'thumbnail': thumbnail,
105             'duration': duration,
106             'timestamp': timestamp,
107             'view_count': view_count,
108             'comment_count': comment_count,
109             'formats': formats,
110         }