]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/xstream.py
Update changelog.
[youtubedl] / youtube_dl / extractor / xstream.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 int_or_none,
9 parse_iso8601,
10 xpath_with_ns,
11 xpath_text,
12 find_xpath_attr,
13 )
14
15
16 class XstreamIE(InfoExtractor):
17 _VALID_URL = r'''(?x)
18 (?:
19 xstream:|
20 https?://frontend\.xstream\.(?:dk|net)/
21 )
22 (?P<partner_id>[^/]+)
23 (?:
24 :|
25 /feed/video/\?.*?\bid=
26 )
27 (?P<id>\d+)
28 '''
29 _TESTS = [{
30 'url': 'http://frontend.xstream.dk/btno/feed/video/?platform=web&id=86588',
31 'md5': 'd7d17e3337dc80de6d3a540aefbe441b',
32 'info_dict': {
33 'id': '86588',
34 'ext': 'mov',
35 'title': 'Otto Wollertsen',
36 'description': 'Vestlendingen Otto Fredrik Wollertsen',
37 'timestamp': 1430473209,
38 'upload_date': '20150501',
39 },
40 }, {
41 'url': 'http://frontend.xstream.dk/ap/feed/video/?platform=web&id=21039',
42 'only_matching': True,
43 }]
44
45 def _real_extract(self, url):
46 mobj = re.match(self._VALID_URL, url)
47 partner_id = mobj.group('partner_id')
48 video_id = mobj.group('id')
49
50 data = self._download_xml(
51 'http://frontend.xstream.dk/%s/feed/video/?platform=web&id=%s'
52 % (partner_id, video_id),
53 video_id)
54
55 NS_MAP = {
56 'atom': 'http://www.w3.org/2005/Atom',
57 'xt': 'http://xstream.dk/',
58 'media': 'http://search.yahoo.com/mrss/',
59 }
60
61 entry = data.find(xpath_with_ns('./atom:entry', NS_MAP))
62
63 title = xpath_text(
64 entry, xpath_with_ns('./atom:title', NS_MAP), 'title')
65 description = xpath_text(
66 entry, xpath_with_ns('./atom:summary', NS_MAP), 'description')
67 timestamp = parse_iso8601(xpath_text(
68 entry, xpath_with_ns('./atom:published', NS_MAP), 'upload date'))
69
70 formats = []
71 media_group = entry.find(xpath_with_ns('./media:group', NS_MAP))
72 for media_content in media_group.findall(xpath_with_ns('./media:content', NS_MAP)):
73 media_url = media_content.get('url')
74 if not media_url:
75 continue
76 tbr = int_or_none(media_content.get('bitrate'))
77 mobj = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', media_url)
78 if mobj:
79 formats.append({
80 'url': mobj.group('url'),
81 'play_path': 'mp4:%s' % mobj.group('playpath'),
82 'app': mobj.group('app'),
83 'ext': 'flv',
84 'tbr': tbr,
85 'format_id': 'rtmp-%d' % tbr,
86 })
87 else:
88 formats.append({
89 'url': media_url,
90 'tbr': tbr,
91 })
92 self._sort_formats(formats)
93
94 link = find_xpath_attr(
95 entry, xpath_with_ns('./atom:link', NS_MAP), 'rel', 'original')
96 if link is not None:
97 formats.append({
98 'url': link.get('href'),
99 'format_id': link.get('rel'),
100 })
101
102 thumbnails = [{
103 'url': splash.get('url'),
104 'width': int_or_none(splash.get('width')),
105 'height': int_or_none(splash.get('height')),
106 } for splash in media_group.findall(xpath_with_ns('./xt:splash', NS_MAP))]
107
108 return {
109 'id': video_id,
110 'title': title,
111 'description': description,
112 'timestamp': timestamp,
113 'formats': formats,
114 'thumbnails': thumbnails,
115 }