d/rules: Run offline tests.

[youtubedl] / youtube_dl / extractor / dailymotion.py

diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py
index 9a74906cb625f590bd5d1bc8862fc5c652fac52c..745971900b23f1cbf44ce286b9c4a4a7e527a3da 100644 (file)
--- a/youtube_dl/extractor/dailymotion.py
+++ b/youtube_dl/extractor/dailymotion.py
@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 import base64
 from __future__ import unicode_literals
 
 import base64

+import functools

 import hashlib
 import itertools
 import json
 import hashlib
 import itertools
 import json


@@ -16,11 +17,16 @@ from ..utils import (
     error_to_compat_str,
     ExtractorError,
     int_or_none,
     error_to_compat_str,
     ExtractorError,
     int_or_none,

+    mimetype2ext,
+    OnDemandPagedList,

     parse_iso8601,
     sanitized_Request,
     str_to_int,
     parse_iso8601,
     sanitized_Request,
     str_to_int,

+    try_get,

     unescapeHTML,
     unescapeHTML,

-    mimetype2ext,
+    update_url_query,
+    url_or_none,
+    urlencode_postdata,

 )
 
 
 )
 
 


@@ -42,7 +48,14 @@ class DailymotionBaseInfoExtractor(InfoExtractor):
 
 
 class DailymotionIE(DailymotionBaseInfoExtractor):
 
 
 class DailymotionIE(DailymotionBaseInfoExtractor):

-    _VALID_URL = r'(?i)https?://(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|#)/)?video|swf)/(?P<id>[^/?_]+)'
+    _VALID_URL = r'''(?ix)
+                    https?://
+                        (?:
+                            (?:(?:www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|\#)/)?video|swf)|
+                            (?:www\.)?lequipe\.fr/video
+                        )
+                        /(?P<id>[^/?_]+)
+                    '''

     IE_NAME = 'dailymotion'
 
     _FORMATS = [
     IE_NAME = 'dailymotion'
 
     _FORMATS = [


@@ -127,14 +140,26 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
     }, {
         'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
         'only_matching': True,
     }, {
         'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
         'only_matching': True,

+    }, {
+        'url': 'https://www.lequipe.fr/video/x791mem',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.lequipe.fr/video/k7MtHciueyTcrFtFKA2',
+        'only_matching': True,

     }]
 
     @staticmethod
     def _extract_urls(webpage):
     }]
 
     @staticmethod
     def _extract_urls(webpage):

+        urls = []

         # Look for embedded Dailymotion player
         # Look for embedded Dailymotion player

-        matches = re.findall(
-            r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
-        return list(map(lambda m: unescapeHTML(m[1]), matches))
+        # https://developer.dailymotion.com/player#player-parameters
+        for mobj in re.finditer(
+                r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage):
+            urls.append(unescapeHTML(mobj.group('url')))
+        for mobj in re.finditer(
+                r'(?s)DM\.player\([^,]+,\s*{.*?video[\'"]?\s*:\s*["\']?(?P<id>[0-9a-zA-Z]+).+?}\s*\);', webpage):
+            urls.append('https://www.dailymotion.com/embed/video/' + mobj.group('id'))
+        return urls

 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)


@@ -144,7 +169,8 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
 
         age_limit = self._rta_search(webpage)
 
 
         age_limit = self._rta_search(webpage)
 

-        description = self._og_search_description(webpage) or self._html_search_meta(
+        description = self._og_search_description(
+            webpage, default=None) or self._html_search_meta(

             'description', webpage, 'description')
 
         view_count_str = self._search_regex(
             'description', webpage, 'description')
 
         view_count_str = self._search_regex(


@@ -159,18 +185,33 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             webpage, 'comment count', default=None))
 
         player_v5 = self._search_regex(
             webpage, 'comment count', default=None))
 
         player_v5 = self._search_regex(

-            [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
+            [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/ytdl-org/youtube-dl/issues/7826

              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
              r'buildPlayer\(({.+?})\);',
              r'var\s+config\s*=\s*({.+?});',
              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
              r'buildPlayer\(({.+?})\);',
              r'var\s+config\s*=\s*({.+?});',

-             # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580)
+             # New layout regex (see https://github.com/ytdl-org/youtube-dl/issues/13580)

              r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
             webpage, 'player v5', default=None)
         if player_v5:
              r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
             webpage, 'player v5', default=None)
         if player_v5:

-            player = self._parse_json(player_v5, video_id)
-            metadata = player['metadata']
-
-            if metadata.get('error', {}).get('type') == 'password_protected':
+            player = self._parse_json(player_v5, video_id, fatal=False) or {}
+            metadata = try_get(player, lambda x: x['metadata'], dict)
+            if not metadata:
+                metadata_url = url_or_none(try_get(
+                    player, lambda x: x['context']['metadata_template_url1']))
+                if metadata_url:
+                    metadata_url = metadata_url.replace(':videoId', video_id)
+                else:
+                    metadata_url = update_url_query(
+                        'https://www.dailymotion.com/player/metadata/video/%s'
+                        % video_id, {
+                            'embedder': url,
+                            'integration': 'inline',
+                            'GK_PV5_NEON': '1',
+                        })
+                metadata = self._download_json(
+                    metadata_url, video_id, 'Downloading metadata JSON')
+
+            if try_get(metadata, lambda x: x['error']['type']) == 'password_protected':

                 password = self._downloader.params.get('videopassword')
                 if password:
                     r = int(metadata['id'][1:], 36)
                 password = self._downloader.params.get('videopassword')
                 if password:
                     r = int(metadata['id'][1:], 36)


@@ -342,58 +383,73 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
 
 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
     IE_NAME = 'dailymotion:playlist'
 
 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
     IE_NAME = 'dailymotion:playlist'

-    _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>[^/?#&]+)'
-    _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
-    _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>x[0-9a-z]+)'

     _TESTS = [{
         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
         'info_dict': {
             'title': 'SPORT',
     _TESTS = [{
         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
         'info_dict': {
             'title': 'SPORT',

-            'id': 'xv4bw_nqtv_sport',
+            'id': 'xv4bw',

         },
         'playlist_mincount': 20,
     }]
         },
         'playlist_mincount': 20,
     }]

-
-    def _extract_entries(self, id):
-        video_ids = set()
-        processed_urls = set()
-        for pagenum in itertools.count(1):
-            page_url = self._PAGE_TEMPLATE % (id, pagenum)
-            webpage, urlh = self._download_webpage_handle_no_ff(
-                page_url, id, 'Downloading page %s' % pagenum)
-            if urlh.geturl() in processed_urls:
-                self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
-                    page_url, urlh.geturl()), id)
-                break
-
-            processed_urls.add(urlh.geturl())
-
-            for video_id in re.findall(r'data-xid="(.+?)"', webpage):
-                if video_id not in video_ids:
-                    yield self.url_result(
-                        'http://www.dailymotion.com/video/%s' % video_id,
-                        DailymotionIE.ie_key(), video_id)
-                    video_ids.add(video_id)
-
-            if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
-                break
+    _PAGE_SIZE = 100
+
+    def _fetch_page(self, playlist_id, authorizaion, page):
+        page += 1
+        videos = self._download_json(
+            'https://graphql.api.dailymotion.com',
+            playlist_id, 'Downloading page %d' % page,
+            data=json.dumps({
+                'query': '''{
+  collection(xid: "%s") {
+    videos(first: %d, page: %d) {
+      pageInfo {
+        hasNextPage
+        nextPage
+      }
+      edges {
+        node {
+          xid
+          url
+        }
+      }
+    }
+  }
+}''' % (playlist_id, self._PAGE_SIZE, page)
+            }).encode(), headers={
+                'Authorization': authorizaion,
+                'Origin': 'https://www.dailymotion.com',
+            })['data']['collection']['videos']
+        for edge in videos['edges']:
+            node = edge['node']
+            yield self.url_result(
+                node['url'], DailymotionIE.ie_key(), node['xid'])

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        playlist_id = mobj.group('id')
+        playlist_id = self._match_id(url)

         webpage = self._download_webpage(url, playlist_id)
         webpage = self._download_webpage(url, playlist_id)

-
-        return {
-            '_type': 'playlist',
-            'id': playlist_id,
-            'title': self._og_search_title(webpage),
-            'entries': self._extract_entries(playlist_id),
-        }
-
-
-class DailymotionUserIE(DailymotionPlaylistIE):
+        api = self._parse_json(self._search_regex(
+            r'__PLAYER_CONFIG__\s*=\s*({.+?});',
+            webpage, 'player config'), playlist_id)['context']['api']
+        auth = self._download_json(
+            api.get('auth_url', 'https://graphql.api.dailymotion.com/oauth/token'),
+            playlist_id, data=urlencode_postdata({
+                'client_id': api.get('client_id', 'f1a362d288c1b98099c7'),
+                'client_secret': api.get('client_secret', 'eea605b96e01c796ff369935357eca920c5da4c5'),
+                'grant_type': 'client_credentials',
+            }))
+        authorizaion = '%s %s' % (auth.get('token_type', 'Bearer'), auth['access_token'])
+        entries = OnDemandPagedList(functools.partial(
+            self._fetch_page, playlist_id, authorizaion), self._PAGE_SIZE)
+        return self.playlist_result(
+            entries, playlist_id,
+            self._og_search_title(webpage))
+
+
+class DailymotionUserIE(DailymotionBaseInfoExtractor):

     IE_NAME = 'dailymotion:user'
     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
     IE_NAME = 'dailymotion:user'
     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'

+    _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'

     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
     _TESTS = [{
         'url': 'https://www.dailymotion.com/user/nqtv',
     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
     _TESTS = [{
         'url': 'https://www.dailymotion.com/user/nqtv',


@@ -415,6 +471,30 @@ class DailymotionUserIE(DailymotionPlaylistIE):
         'skip': 'Takes too long time',
     }]
 
         'skip': 'Takes too long time',
     }]
 

+    def _extract_entries(self, id):
+        video_ids = set()
+        processed_urls = set()
+        for pagenum in itertools.count(1):
+            page_url = self._PAGE_TEMPLATE % (id, pagenum)
+            webpage, urlh = self._download_webpage_handle_no_ff(
+                page_url, id, 'Downloading page %s' % pagenum)
+            if urlh.geturl() in processed_urls:
+                self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
+                    page_url, urlh.geturl()), id)
+                break
+
+            processed_urls.add(urlh.geturl())
+
+            for video_id in re.findall(r'data-xid="(.+?)"', webpage):
+                if video_id not in video_ids:
+                    yield self.url_result(
+                        'http://www.dailymotion.com/video/%s' % video_id,
+                        DailymotionIE.ie_key(), video_id)
+                    video_ids.add(video_id)
+
+            if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
+                break
+

     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         user = mobj.group('user')
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         user = mobj.group('user')