]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/wistia.py
fa142b974ae873a7883ede4894421f4f8a010549
[youtubedl] / youtube_dl / extractor / wistia.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7 ExtractorError,
8 int_or_none,
9 float_or_none,
10 unescapeHTML,
11 )
12
13
14 class WistiaIE(InfoExtractor):
15 _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/)(?P<id>[a-z0-9]+)'
16 _API_URL = 'http://fast.wistia.com/embed/medias/%s.json'
17 _IFRAME_URL = 'http://fast.wistia.net/embed/iframe/%s'
18
19 _TESTS = [{
20 'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
21 'md5': 'cafeb56ec0c53c18c97405eecb3133df',
22 'info_dict': {
23 'id': 'sh7fpupwlt',
24 'ext': 'mov',
25 'title': 'Being Resourceful',
26 'description': 'a Clients From Hell Video Series video from worldwidewebhosting',
27 'upload_date': '20131204',
28 'timestamp': 1386185018,
29 'duration': 117,
30 },
31 }, {
32 'url': 'wistia:sh7fpupwlt',
33 'only_matching': True,
34 }, {
35 # with hls video
36 'url': 'wistia:807fafadvk',
37 'only_matching': True,
38 }, {
39 'url': 'http://fast.wistia.com/embed/iframe/sh7fpupwlt',
40 'only_matching': True,
41 }, {
42 'url': 'http://fast.wistia.net/embed/medias/sh7fpupwlt.json',
43 'only_matching': True,
44 }]
45
46 @staticmethod
47 def _extract_url(webpage):
48 match = re.search(
49 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/iframe/.+?)\1', webpage)
50 if match:
51 return unescapeHTML(match.group('url'))
52
53 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
54 if match:
55 return 'wistia:%s' % match.group('id')
56
57 match = re.search(
58 r'''(?sx)
59 <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
60 <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2
61 ''', webpage)
62 if match:
63 return 'wistia:%s' % match.group('id')
64
65 def _real_extract(self, url):
66 video_id = self._match_id(url)
67
68 data_json = self._download_json(
69 self._API_URL % video_id, video_id,
70 # Some videos require this.
71 headers={
72 'Referer': url if url.startswith('http') else self._IFRAME_URL % video_id,
73 })
74
75 if data_json.get('error'):
76 raise ExtractorError(
77 'Error while getting the playlist', expected=True)
78
79 data = data_json['media']
80 title = data['name']
81
82 formats = []
83 thumbnails = []
84 for a in data['assets']:
85 aurl = a.get('url')
86 if not aurl:
87 continue
88 astatus = a.get('status')
89 atype = a.get('type')
90 if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
91 continue
92 elif atype in ('still', 'still_image'):
93 thumbnails.append({
94 'url': aurl,
95 'width': int_or_none(a.get('width')),
96 'height': int_or_none(a.get('height')),
97 })
98 else:
99 aext = a.get('ext')
100 is_m3u8 = a.get('container') == 'm3u8' or aext == 'm3u8'
101 formats.append({
102 'format_id': atype,
103 'url': aurl,
104 'tbr': int_or_none(a.get('bitrate')),
105 'vbr': int_or_none(a.get('opt_vbitrate')),
106 'width': int_or_none(a.get('width')),
107 'height': int_or_none(a.get('height')),
108 'filesize': int_or_none(a.get('size')),
109 'vcodec': a.get('codec'),
110 'container': a.get('container'),
111 'ext': 'mp4' if is_m3u8 else aext,
112 'protocol': 'm3u8' if is_m3u8 else None,
113 'preference': 1 if atype == 'original' else None,
114 })
115
116 self._sort_formats(formats)
117
118 return {
119 'id': video_id,
120 'title': title,
121 'description': data.get('seoDescription'),
122 'formats': formats,
123 'thumbnails': thumbnails,
124 'duration': float_or_none(data.get('duration')),
125 'timestamp': int_or_none(data.get('createdAt')),
126 }