]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/steam.py
[pornhub:uservideos] Add support for multipage videos (Closes #9006)
[youtube-dl] / youtube_dl / extractor / steam.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     ExtractorError,
8     unescapeHTML,
9 )
10
11
12 class SteamIE(InfoExtractor):
13     _VALID_URL = r"""(?x)
14         https?://store\.steampowered\.com/
15             (agecheck/)?
16             (?P<urltype>video|app)/ #If the page is only for videos or for a game
17             (?P<gameID>\d+)/?
18             (?P<videoID>\d*)(?P<extra>\??) # For urltype == video we sometimes get the videoID
19         |
20         https?://(?:www\.)?steamcommunity\.com/sharedfiles/filedetails/\?id=(?P<fileID>[0-9]+)
21     """
22     _VIDEO_PAGE_TEMPLATE = 'http://store.steampowered.com/video/%s/'
23     _AGECHECK_TEMPLATE = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970'
24     _TESTS = [{
25         'url': 'http://store.steampowered.com/video/105600/',
26         'playlist': [
27             {
28                 'md5': 'f870007cee7065d7c76b88f0a45ecc07',
29                 'info_dict': {
30                     'id': '81300',
31                     'ext': 'flv',
32                     'title': 'Terraria 1.1 Trailer',
33                     'playlist_index': 1,
34                 }
35             },
36             {
37                 'md5': '61aaf31a5c5c3041afb58fb83cbb5751',
38                 'info_dict': {
39                     'id': '80859',
40                     'ext': 'flv',
41                     'title': 'Terraria Trailer',
42                     'playlist_index': 2,
43                 }
44             }
45         ],
46         'params': {
47             'playlistend': 2,
48         }
49     }, {
50         'url': 'http://steamcommunity.com/sharedfiles/filedetails/?id=242472205',
51         'info_dict': {
52             'id': 'WB5DvDOOvAY',
53             'ext': 'mp4',
54             'upload_date': '20140329',
55             'title': 'FRONTIERS - Final Greenlight Trailer',
56             'description': 'md5:dc96a773669d0ca1b36c13c1f30250d9',
57             'uploader': 'AAD Productions',
58             'uploader_id': 'AtomicAgeDogGames',
59         }
60     }]
61
62     def _real_extract(self, url):
63         m = re.match(self._VALID_URL, url)
64         fileID = m.group('fileID')
65         if fileID:
66             videourl = url
67             playlist_id = fileID
68         else:
69             gameID = m.group('gameID')
70             playlist_id = gameID
71             videourl = self._VIDEO_PAGE_TEMPLATE % playlist_id
72         webpage = self._download_webpage(videourl, playlist_id)
73
74         if re.search('<h2>Please enter your birth date to continue:</h2>', webpage) is not None:
75             videourl = self._AGECHECK_TEMPLATE % playlist_id
76             self.report_age_confirmation()
77             webpage = self._download_webpage(videourl, playlist_id)
78
79         if fileID:
80             playlist_title = self._html_search_regex(
81                 r'<div class="workshopItemTitle">(.+)</div>', webpage, 'title')
82             mweb = re.finditer(r'''(?x)
83                 'movie_(?P<videoID>[0-9]+)':\s*\{\s*
84                 YOUTUBE_VIDEO_ID:\s*"(?P<youtube_id>[^"]+)",
85                 ''', webpage)
86             videos = [{
87                 '_type': 'url',
88                 'url': vid.group('youtube_id'),
89                 'ie_key': 'Youtube',
90             } for vid in mweb]
91         else:
92             playlist_title = self._html_search_regex(
93                 r'<h2 class="pageheader">(.*?)</h2>', webpage, 'game title')
94
95             mweb = re.finditer(r'''(?x)
96                 'movie_(?P<videoID>[0-9]+)':\s*\{\s*
97                 FILENAME:\s*"(?P<videoURL>[\w:/\.\?=]+)"
98                 (,\s*MOVIE_NAME:\s*\"(?P<videoName>[\w:/\.\?=\+-]+)\")?\s*\},
99                 ''', webpage)
100             titles = re.finditer(
101                 r'<span class="title">(?P<videoName>.+?)</span>', webpage)
102             thumbs = re.finditer(
103                 r'<img class="movie_thumb" src="(?P<thumbnail>.+?)">', webpage)
104             videos = []
105
106             for vid, vtitle, thumb in zip(mweb, titles, thumbs):
107                 video_id = vid.group('videoID')
108                 title = vtitle.group('videoName')
109                 video_url = vid.group('videoURL')
110                 video_thumb = thumb.group('thumbnail')
111                 if not video_url:
112                     raise ExtractorError('Cannot find video url for %s' % video_id)
113                 videos.append({
114                     'id': video_id,
115                     'url': video_url,
116                     'ext': 'flv',
117                     'title': unescapeHTML(title),
118                     'thumbnail': video_thumb
119                 })
120         if not videos:
121             raise ExtractorError('Could not find any videos')
122
123         return self.playlist_result(videos, playlist_id, playlist_title)