]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/svt.py
[utils] Add extract_attributes for extracting html tag attributes
[youtube-dl] / youtube_dl / extractor / svt.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     determine_ext,
9 )
10
11
12 class SVTBaseIE(InfoExtractor):
13     def _extract_video(self, url, video_id):
14         info = self._download_json(url, video_id)
15
16         title = info['context']['title']
17         thumbnail = info['context'].get('thumbnailImage')
18
19         video_info = info['video']
20         formats = []
21         for vr in video_info['videoReferences']:
22             vurl = vr['url']
23             ext = determine_ext(vurl)
24             if ext == 'm3u8':
25                 formats.extend(self._extract_m3u8_formats(
26                     vurl, video_id,
27                     ext='mp4', entry_protocol='m3u8_native',
28                     m3u8_id=vr.get('playerType')))
29             elif ext == 'f4m':
30                 formats.extend(self._extract_f4m_formats(
31                     vurl + '?hdcore=3.3.0', video_id,
32                     f4m_id=vr.get('playerType')))
33             else:
34                 formats.append({
35                     'format_id': vr.get('playerType'),
36                     'url': vurl,
37                 })
38         self._sort_formats(formats)
39
40         subtitles = {}
41         subtitle_references = video_info.get('subtitleReferences')
42         if isinstance(subtitle_references, list):
43             for sr in subtitle_references:
44                 subtitle_url = sr.get('url')
45                 if subtitle_url:
46                     subtitles.setdefault('sv', []).append({'url': subtitle_url})
47
48         duration = video_info.get('materialLength')
49         age_limit = 18 if video_info.get('inappropriateForChildren') else 0
50
51         return {
52             'id': video_id,
53             'title': title,
54             'formats': formats,
55             'subtitles': subtitles,
56             'thumbnail': thumbnail,
57             'duration': duration,
58             'age_limit': age_limit,
59         }
60
61
62 class SVTIE(SVTBaseIE):
63     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
64     _TEST = {
65         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
66         'md5': '9648197555fc1b49e3dc22db4af51d46',
67         'info_dict': {
68             'id': '2900353',
69             'ext': 'flv',
70             'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
71             'duration': 27,
72             'age_limit': 0,
73         },
74     }
75
76     @staticmethod
77     def _extract_url(webpage):
78         mobj = re.search(
79             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
80         if mobj:
81             return mobj.group('url')
82
83     def _real_extract(self, url):
84         mobj = re.match(self._VALID_URL, url)
85         widget_id = mobj.group('widget_id')
86         article_id = mobj.group('id')
87         return self._extract_video(
88             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
89             article_id)
90
91
92 class SVTPlayIE(SVTBaseIE):
93     IE_DESC = 'SVT Play and Öppet arkiv'
94     _VALID_URL = r'https?://(?:www\.)?(?P<host>svtplay|oppetarkiv)\.se/video/(?P<id>[0-9]+)'
95     _TEST = {
96         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
97         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
98         'info_dict': {
99             'id': '5996901',
100             'ext': 'mp4',
101             'title': 'Flygplan till Haile Selassie',
102             'duration': 3527,
103             'thumbnail': 're:^https?://.*[\.-]jpg$',
104             'age_limit': 0,
105             'subtitles': {
106                 'sv': [{
107                     'ext': 'wsrt',
108                 }]
109             },
110         },
111     }
112
113     def _real_extract(self, url):
114         mobj = re.match(self._VALID_URL, url)
115         video_id = mobj.group('id')
116         host = mobj.group('host')
117         return self._extract_video(
118             'http://www.%s.se/video/%s?output=json' % (host, video_id),
119             video_id)