]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/theplatform.py
cec65261bfffd2a25702634047a99526fa3a7d10
[youtubedl] / youtube_dl / extractor / theplatform.py
1 import re
2 import json
3
4 from .common import InfoExtractor
5 from ..utils import (
6 ExtractorError,
7 xpath_with_ns,
8 )
9
10 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
11
12
13 class ThePlatformIE(InfoExtractor):
14 _VALID_URL = r'(?:https?://link\.theplatform\.com/s/[^/]+/|theplatform:)(?P<id>[^/\?]+)'
15
16 _TEST = {
17 # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
18 u'url': u'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
19 u'info_dict': {
20 u'id': u'e9I_cZgTgIPd',
21 u'ext': u'flv',
22 u'title': u'Blackberry\'s big, bold Z30',
23 u'description': u'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
24 u'duration': 247,
25 },
26 u'params': {
27 # rtmp download
28 u'skip_download': True,
29 },
30 }
31
32 def _get_info(self, video_id):
33 smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
34 'format=smil&mbr=true'.format(video_id))
35 meta = self._download_xml(smil_url, video_id)
36
37 try:
38 error_msg = next(
39 n.attrib['abstract']
40 for n in meta.findall(_x('.//smil:ref'))
41 if n.attrib.get('title') == u'Geographic Restriction')
42 except StopIteration:
43 pass
44 else:
45 raise ExtractorError(error_msg, expected=True)
46
47 info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
48 info_json = self._download_webpage(info_url, video_id)
49 info = json.loads(info_json)
50
51 head = meta.find(_x('smil:head'))
52 body = meta.find(_x('smil:body'))
53 base_url = head.find(_x('smil:meta')).attrib['base']
54 switch = body.find(_x('smil:switch'))
55 formats = []
56 for f in switch.findall(_x('smil:video')):
57 attr = f.attrib
58 formats.append({
59 'url': base_url,
60 'play_path': 'mp4:' + attr['src'],
61 'ext': 'flv',
62 'width': int(attr['width']),
63 'height': int(attr['height']),
64 'vbr': int(attr['system-bitrate']),
65 })
66 formats.sort(key=lambda f: (f['height'], f['width'], f['vbr']))
67
68 return {
69 'id': video_id,
70 'title': info['title'],
71 'formats': formats,
72 'description': info['description'],
73 'thumbnail': info['defaultThumbnailUrl'],
74 'duration': info['duration']//1000,
75 }
76
77 def _real_extract(self, url):
78 mobj = re.match(self._VALID_URL, url)
79 video_id = mobj.group('id')
80 return self._get_info(video_id)