]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/dailymotion.py
Implement a different adult sites checking algorithm
[youtube-dl] / youtube_dl / extractor / dailymotion.py
1 import re
2 import json
3 import itertools
4
5 from .common import InfoExtractor
6 from .subtitles import SubtitlesInfoExtractor
7
8 from ..utils import (
9     compat_urllib_request,
10     compat_str,
11     get_element_by_attribute,
12     get_element_by_id,
13     orderedSet,
14     str_to_int,
15
16     ExtractorError,
17 )
18
19 class DailymotionBaseInfoExtractor(InfoExtractor):
20     @staticmethod
21     def _build_request(url):
22         """Build a request with the family filter disabled"""
23         request = compat_urllib_request.Request(url)
24         request.add_header('Cookie', 'family_filter=off')
25         request.add_header('Cookie', 'ff=off')
26         return request
27
28 class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
29     """Information Extractor for Dailymotion"""
30
31     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
32     IE_NAME = u'dailymotion'
33
34     _FORMATS = [
35         (u'stream_h264_ld_url', u'ld'),
36         (u'stream_h264_url', u'standard'),
37         (u'stream_h264_hq_url', u'hq'),
38         (u'stream_h264_hd_url', u'hd'),
39         (u'stream_h264_hd1080_url', u'hd180'),
40     ]
41
42     _TESTS = [
43         {
44             u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
45             u'file': u'x33vw9.mp4',
46             u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
47             u'info_dict': {
48                 u"uploader": u"Amphora Alex and Van .", 
49                 u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
50             }
51         },
52         # Vevo video
53         {
54             u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
55             u'file': u'USUV71301934.mp4',
56             u'info_dict': {
57                 u'title': u'Roar (Official)',
58                 u'uploader': u'Katy Perry',
59                 u'upload_date': u'20130905',
60             },
61             u'params': {
62                 u'skip_download': True,
63             },
64             u'skip': u'VEVO is only available in some countries',
65         },
66         # age-restricted video
67         {
68             u'url': u'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
69             u'file': u'xyh2zz.mp4',
70             u'md5': u'0d667a7b9cebecc3c89ee93099c4159d',
71             u'info_dict': {
72                 u'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
73                 u'uploader': 'HotWaves1012',
74                 u'age_limit': 18,
75             }
76
77         }
78     ]
79
80     def _real_extract(self, url):
81         # Extract id and simplified title from URL
82         mobj = re.match(self._VALID_URL, url)
83
84         video_id = mobj.group('id')
85
86         url = 'http://www.dailymotion.com/video/%s' % video_id
87
88         # Retrieve video webpage to extract further information
89         request = self._build_request(url)
90         webpage = self._download_webpage(request, video_id)
91
92         # Extract URL, uploader and title from webpage
93         self.report_extraction(video_id)
94
95         # It may just embed a vevo video:
96         m_vevo = re.search(
97             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
98             webpage)
99         if m_vevo is not None:
100             vevo_id = m_vevo.group('id')
101             self.to_screen(u'Vevo video detected: %s' % vevo_id)
102             return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
103
104         age_limit = self._rta_search(webpage)
105
106         video_upload_date = None
107         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
108         if mobj is not None:
109             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
110
111         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
112         embed_page = self._download_webpage(embed_url, video_id,
113                                             u'Downloading embed page')
114         info = self._search_regex(r'var info = ({.*?}),$', embed_page,
115             'video info', flags=re.MULTILINE)
116         info = json.loads(info)
117         if info.get('error') is not None:
118             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
119             raise ExtractorError(msg, expected=True)
120
121         formats = []
122         for (key, format_id) in self._FORMATS:
123             video_url = info.get(key)
124             if video_url is not None:
125                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
126                 if m_size is not None:
127                     width, height = m_size.group(1), m_size.group(2)
128                 else:
129                     width, height = None, None
130                 formats.append({
131                     'url': video_url,
132                     'ext': 'mp4',
133                     'format_id': format_id,
134                     'width': width,
135                     'height': height,
136                 })
137         if not formats:
138             raise ExtractorError(u'Unable to extract video URL')
139
140         # subtitles
141         video_subtitles = self.extract_subtitles(video_id, webpage)
142         if self._downloader.params.get('listsubtitles', False):
143             self._list_available_subtitles(video_id, webpage)
144             return
145
146         view_count = self._search_regex(
147             r'video_views_count[^>]+>\s+([\d\.,]+)', webpage, u'view count', fatal=False)
148         if view_count is not None:
149             view_count = str_to_int(view_count)
150
151         return {
152             'id':       video_id,
153             'formats': formats,
154             'uploader': info['owner_screenname'],
155             'upload_date':  video_upload_date,
156             'title':    self._og_search_title(webpage),
157             'subtitles':    video_subtitles,
158             'thumbnail': info['thumbnail_url'],
159             'age_limit': age_limit,
160             'view_count': view_count,
161         }
162
163     def _get_available_subtitles(self, video_id, webpage):
164         try:
165             sub_list = self._download_webpage(
166                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
167                 video_id, note=False)
168         except ExtractorError as err:
169             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
170             return {}
171         info = json.loads(sub_list)
172         if (info['total'] > 0):
173             sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
174             return sub_lang_list
175         self._downloader.report_warning(u'video doesn\'t have subtitles')
176         return {}
177
178
179 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
180     IE_NAME = u'dailymotion:playlist'
181     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
182     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
183     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
184
185     def _extract_entries(self, id):
186         video_ids = []
187         for pagenum in itertools.count(1):
188             request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
189             webpage = self._download_webpage(request,
190                                              id, u'Downloading page %s' % pagenum)
191
192             playlist_el = get_element_by_attribute(u'class', u'row video_list', webpage)
193             video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
194
195             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
196                 break
197         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
198                    for video_id in orderedSet(video_ids)]
199
200     def _real_extract(self, url):
201         mobj = re.match(self._VALID_URL, url)
202         playlist_id = mobj.group('id')
203         webpage = self._download_webpage(url, playlist_id)
204
205         return {'_type': 'playlist',
206                 'id': playlist_id,
207                 'title': get_element_by_id(u'playlist_name', webpage),
208                 'entries': self._extract_entries(playlist_id),
209                 }
210
211
212 class DailymotionUserIE(DailymotionPlaylistIE):
213     IE_NAME = u'dailymotion:user'
214     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
215     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/user/.+?".*?>.*?</a>.*?</div>'
216     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
217
218     def _real_extract(self, url):
219         mobj = re.match(self._VALID_URL, url)
220         user = mobj.group('user')
221         webpage = self._download_webpage(url, user)
222         full_user = self._html_search_regex(
223             r'<a class="label" href="/%s".*?>(.*?)</' % re.escape(user),
224             webpage, u'user', flags=re.DOTALL)
225
226         return {
227             '_type': 'playlist',
228             'id': user,
229             'title': full_user,
230             'entries': self._extract_entries(user),
231         }