]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bliptv.py
acfc4ad736d9deecf1ed9cdadd4063e2fc8e7243
[youtubedl] / youtube_dl / extractor / bliptv.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from .subtitles import SubtitlesInfoExtractor
7 from ..utils import (
8 compat_urllib_request,
9 unescapeHTML,
10 parse_iso8601,
11 compat_urlparse,
12 clean_html,
13 compat_str,
14 )
15
16
17 class BlipTVIE(SubtitlesInfoExtractor):
18 _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+]+)))'
19
20 _TESTS = [
21 {
22 'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
23 'md5': 'c6934ad0b6acf2bd920720ec888eb812',
24 'info_dict': {
25 'id': '5779306',
26 'ext': 'mov',
27 'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
28 'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
29 'timestamp': 1323138843,
30 'upload_date': '20111206',
31 'uploader': 'cbr',
32 'uploader_id': '679425',
33 'duration': 81,
34 }
35 },
36 {
37 # https://github.com/rg3/youtube-dl/pull/2274
38 'note': 'Video with subtitles',
39 'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
40 'md5': '309f9d25b820b086ca163ffac8031806',
41 'info_dict': {
42 'id': '6586561',
43 'ext': 'mp4',
44 'title': 'Red vs. Blue Season 11 Episode 1',
45 'description': 'One-Zero-One',
46 'timestamp': 1371261608,
47 'upload_date': '20130615',
48 'uploader': 'redvsblue',
49 'uploader_id': '792887',
50 'duration': 279,
51 }
52 }
53 ]
54
55 def _real_extract(self, url):
56 mobj = re.match(self._VALID_URL, url)
57 lookup_id = mobj.group('lookup_id')
58
59 # See https://github.com/rg3/youtube-dl/issues/857
60 if lookup_id:
61 info_page = self._download_webpage(
62 'http://blip.tv/play/%s.x?p=1' % lookup_id, lookup_id, 'Resolving lookup id')
63 video_id = self._search_regex(r'data-episode-id="([0-9]+)', info_page, 'video_id')
64 else:
65 video_id = mobj.group('id')
66
67 rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
68
69 def blip(s):
70 return '{http://blip.tv/dtd/blip/1.0}%s' % s
71
72 def media(s):
73 return '{http://search.yahoo.com/mrss/}%s' % s
74
75 def itunes(s):
76 return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s
77
78 item = rss.find('channel/item')
79
80 video_id = item.find(blip('item_id')).text
81 title = item.find('./title').text
82 description = clean_html(compat_str(item.find(blip('puredescription')).text))
83 timestamp = parse_iso8601(item.find(blip('datestamp')).text)
84 uploader = item.find(blip('user')).text
85 uploader_id = item.find(blip('userid')).text
86 duration = int(item.find(blip('runtime')).text)
87 media_thumbnail = item.find(media('thumbnail'))
88 thumbnail = media_thumbnail.get('url') if media_thumbnail is not None else item.find(itunes('image')).text
89 categories = [category.text for category in item.findall('category')]
90
91 formats = []
92 subtitles = {}
93
94 media_group = item.find(media('group'))
95 for media_content in media_group.findall(media('content')):
96 url = media_content.get('url')
97 role = media_content.get(blip('role'))
98 msg = self._download_webpage(
99 url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
100 video_id, 'Resolving URL for %s' % role)
101 real_url = compat_urlparse.parse_qs(msg)['message'][0]
102
103 media_type = media_content.get('type')
104 if media_type == 'text/srt' or url.endswith('.srt'):
105 LANGS = {
106 'english': 'en',
107 }
108 lang = role.rpartition('-')[-1].strip().lower()
109 langcode = LANGS.get(lang, lang)
110 subtitles[langcode] = url
111 elif media_type.startswith('video/'):
112 formats.append({
113 'url': real_url,
114 'format_id': role,
115 'format_note': media_type,
116 'vcodec': media_content.get(blip('vcodec')),
117 'acodec': media_content.get(blip('acodec')),
118 'filesize': media_content.get('filesize'),
119 'width': int(media_content.get('width')),
120 'height': int(media_content.get('height')),
121 })
122 self._sort_formats(formats)
123
124 # subtitles
125 video_subtitles = self.extract_subtitles(video_id, subtitles)
126 if self._downloader.params.get('listsubtitles', False):
127 self._list_available_subtitles(video_id, subtitles)
128 return
129
130 return {
131 'id': video_id,
132 'title': title,
133 'description': description,
134 'timestamp': timestamp,
135 'uploader': uploader,
136 'uploader_id': uploader_id,
137 'duration': duration,
138 'thumbnail': thumbnail,
139 'categories': categories,
140 'formats': formats,
141 'subtitles': video_subtitles,
142 }
143
144 def _download_subtitle_url(self, sub_lang, url):
145 # For some weird reason, blip.tv serves a video instead of subtitles
146 # when we request with a common UA
147 req = compat_urllib_request.Request(url)
148 req.add_header('Youtubedl-user-agent', 'youtube-dl')
149 return self._download_webpage(req, None, note=False)
150
151
152 class BlipTVUserIE(InfoExtractor):
153 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
154 _PAGE_SIZE = 12
155 IE_NAME = 'blip.tv:user'
156
157 def _real_extract(self, url):
158 mobj = re.match(self._VALID_URL, url)
159 username = mobj.group(1)
160
161 page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
162
163 page = self._download_webpage(url, username, 'Downloading user page')
164 mobj = re.search(r'data-users-id="([^"]+)"', page)
165 page_base = page_base % mobj.group(1)
166
167 # Download video ids using BlipTV Ajax calls. Result size per
168 # query is limited (currently to 12 videos) so we need to query
169 # page by page until there are no video ids - it means we got
170 # all of them.
171
172 video_ids = []
173 pagenum = 1
174
175 while True:
176 url = page_base + "&page=" + str(pagenum)
177 page = self._download_webpage(
178 url, username, 'Downloading video ids from page %d' % pagenum)
179
180 # Extract video identifiers
181 ids_in_page = []
182
183 for mobj in re.finditer(r'href="/([^"]+)"', page):
184 if mobj.group(1) not in ids_in_page:
185 ids_in_page.append(unescapeHTML(mobj.group(1)))
186
187 video_ids.extend(ids_in_page)
188
189 # A little optimization - if current page is not
190 # "full", ie. does not contain PAGE_SIZE video ids then
191 # we can assume that this page is the last one - there
192 # are no more ids on further pages - no need to query
193 # again.
194
195 if len(ids_in_page) < self._PAGE_SIZE:
196 break
197
198 pagenum += 1
199
200 urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
201 url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
202 return [self.playlist_result(url_entries, playlist_title=username)]