]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/mtv.py
24a79ae130e7ca52983ae9846532f643eda60085
[youtubedl] / youtube_dl / extractor / mtv.py
1 import re
2 import xml.etree.ElementTree
3
4 from .common import InfoExtractor
5 from ..utils import (
6 compat_urllib_parse,
7 ExtractorError,
8 )
9
10 def _media_xml_tag(tag):
11 return '{http://search.yahoo.com/mrss/}%s' % tag
12
13 class MTVIE(InfoExtractor):
14 _VALID_URL = r'^https?://(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$'
15
16 _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
17
18 _TESTS = [
19 {
20 u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
21 u'file': u'853555.mp4',
22 u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8',
23 u'info_dict': {
24 u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"',
25 u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
26 },
27 },
28 {
29 u'add_ie': ['Vevo'],
30 u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
31 u'file': u'USCJY1331283.mp4',
32 u'md5': u'73b4e7fcadd88929292fe52c3ced8caf',
33 u'info_dict': {
34 u'title': u'Everything Has Changed',
35 u'upload_date': u'20130606',
36 u'uploader': u'Taylor Swift',
37 },
38 u'skip': u'VEVO is only available in some countries',
39 },
40 ]
41
42 @staticmethod
43 def _id_from_uri(uri):
44 return uri.split(':')[-1]
45
46 # This was originally implemented for ComedyCentral, but it also works here
47 @staticmethod
48 def _transform_rtmp_url(rtmp_video_url):
49 m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
50 if not m:
51 raise ExtractorError(u'Cannot transform RTMP url')
52 base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
53 return base + m.group('finalid')
54
55 def _get_thumbnail_url(self, uri, itemdoc):
56 return 'http://mtv.mtvnimages.com/uri/' + uri
57
58 def _extract_video_formats(self, metadataXml):
59 if '/error_country_block.swf' in metadataXml:
60 raise ExtractorError(u'This video is not available from your country.', expected=True)
61 mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
62 renditions = mdoc.findall('.//rendition')
63
64 formats = []
65 for rendition in mdoc.findall('.//rendition'):
66 try:
67 _, _, ext = rendition.attrib['type'].partition('/')
68 rtmp_video_url = rendition.find('./src').text
69 formats.append({'ext': ext,
70 'url': self._transform_rtmp_url(rtmp_video_url),
71 'format_id': rendition.get('bitrate'),
72 'width': int(rendition.get('width')),
73 'height': int(rendition.get('height')),
74 })
75 except (KeyError, TypeError):
76 raise ExtractorError('Invalid rendition field.')
77 return formats
78
79 def _get_video_info(self, itemdoc):
80 uri = itemdoc.find('guid').text
81 video_id = self._id_from_uri(uri)
82 self.report_extraction(video_id)
83 mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
84 # Remove the templates, like &device={device}
85 mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', u'', mediagen_url)
86 if 'acceptMethods' not in mediagen_url:
87 mediagen_url += '&acceptMethods=fms'
88 mediagen_page = self._download_webpage(mediagen_url, video_id,
89 u'Downloading video urls')
90
91 description_node = itemdoc.find('description')
92 if description_node is not None:
93 description = description_node.text.strip()
94 else:
95 description = None
96
97 info = {
98 'title': itemdoc.find('title').text,
99 'formats': self._extract_video_formats(mediagen_page),
100 'id': video_id,
101 'thumbnail': self._get_thumbnail_url(uri, itemdoc),
102 'description': description,
103 }
104
105 # TODO: Remove when #980 has been merged
106 info.update(info['formats'][-1])
107
108 return info
109
110 def _get_videos_info(self, uri):
111 video_id = self._id_from_uri(uri)
112 data = compat_urllib_parse.urlencode({'uri': uri})
113 infoXml = self._download_webpage(self._FEED_URL +'?' + data, video_id,
114 u'Downloading info')
115 idoc = xml.etree.ElementTree.fromstring(infoXml.encode('utf-8'))
116 return [self._get_video_info(item) for item in idoc.findall('.//item')]
117
118 def _real_extract(self, url):
119 mobj = re.match(self._VALID_URL, url)
120 video_id = mobj.group('videoid')
121
122 webpage = self._download_webpage(url, video_id)
123
124 # Some videos come from Vevo.com
125 m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
126 webpage, re.DOTALL)
127 if m_vevo:
128 vevo_id = m_vevo.group(1);
129 self.to_screen(u'Vevo video detected: %s' % vevo_id)
130 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
131
132 uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri')
133 return self._get_videos_info(uri)