]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/gamespot.py
debian/control: Add recommends on aria2 | wget | curl to use external downloaders.
[youtubedl] / youtube_dl / extractor / gamespot.py
1 from __future__ import unicode_literals
2
3 import re
4 import json
5
6 from .common import InfoExtractor
7 from ..compat import (
8 compat_urllib_parse,
9 compat_urlparse,
10 )
11 from ..utils import (
12 unescapeHTML,
13 )
14
15
16 class GameSpotIE(InfoExtractor):
17 _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
18 _TEST = {
19 'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
20 'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
21 'info_dict': {
22 'id': 'gs-2300-6410818',
23 'ext': 'mp4',
24 'title': 'Arma 3 - Community Guide: SITREP I',
25 'description': 'Check out this video where some of the basics of Arma 3 is explained.',
26 }
27 }
28
29 def _real_extract(self, url):
30 page_id = self._match_id(url)
31 webpage = self._download_webpage(url, page_id)
32 data_video_json = self._search_regex(
33 r'data-video=["\'](.*?)["\']', webpage, 'data video')
34 data_video = json.loads(unescapeHTML(data_video_json))
35
36 # Transform the manifest url to a link to the mp4 files
37 # they are used in mobile devices.
38 f4m_url = data_video['videoStreams']['f4m_stream']
39 f4m_path = compat_urlparse.urlparse(f4m_url).path
40 QUALITIES_RE = r'((,\d+)+,?)'
41 qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
42 http_path = f4m_path[1:].split('/', 1)[1]
43 http_template = re.sub(QUALITIES_RE, r'%s', http_path)
44 http_template = http_template.replace('.csmil/manifest.f4m', '')
45 http_template = compat_urlparse.urljoin(
46 'http://video.gamespotcdn.com/', http_template)
47 formats = []
48 for q in qualities:
49 formats.append({
50 'url': http_template % q,
51 'ext': 'mp4',
52 'format_id': q,
53 })
54
55 return {
56 'id': data_video['guid'],
57 'display_id': page_id,
58 'title': compat_urllib_parse.unquote(data_video['title']),
59 'formats': formats,
60 'description': self._html_search_meta('description', webpage),
61 'thumbnail': self._og_search_thumbnail(webpage),
62 }