Switch codebase to use sanitized_Request instead of

[youtube-dl] / youtube_dl / extractor / dailymotion.py

diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py
index 8fcae7402b95012479ed16c01044615184a5d496..ab7f3aec42ff7807f03095ba269bc5f989541870 100644 (file)
--- a/youtube_dl/extractor/dailymotion.py
+++ b/youtube_dl/extractor/dailymotion.py
@@ -7,16 +7,13 @@ import itertools
 
 from .common import InfoExtractor
 
-from ..compat import (
-    compat_str,
-    compat_urllib_request,
-)
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     determine_ext,
     int_or_none,
-    orderedSet,
     parse_iso8601,
+    sanitized_Request,
     str_to_int,
     unescapeHTML,
 )
@@ -26,10 +23,14 @@ class DailymotionBaseInfoExtractor(InfoExtractor):
     @staticmethod
     def _build_request(url):
         """Build a request with the family filter disabled"""
-        request = compat_urllib_request.Request(url)
+        request = sanitized_Request(url)
         request.add_header('Cookie', 'family_filter=off; ff=off')
         return request
 
+    def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
+        request = self._build_request(url)
+        return self._download_webpage_handle(request, *args, **kwargs)
+
     def _download_webpage_no_ff(self, url, *args, **kwargs):
         request = self._build_request(url)
         return self._download_webpage(request, *args, **kwargs)
@@ -93,6 +94,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                 'uploader': 'HotWaves1012',
                 'age_limit': 18,
             }
+        },
+        # geo-restricted, player v5
+        {
+            'url': 'http://www.dailymotion.com/video/xhza0o',
+            'only_matching': True,
         }
     ]
 
@@ -116,11 +122,14 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             webpage, 'comment count', fatal=False))
 
         player_v5 = self._search_regex(
-            r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
+            [r'buildPlayer\(({.+?})\);', r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);'],
             webpage, 'player v5', default=None)
         if player_v5:
             player = self._parse_json(player_v5, video_id)
             metadata = player['metadata']
+
+            self._check_error(metadata)
+
             formats = []
             for quality, media_list in metadata['qualities'].items():
                 for media in media_list:
@@ -130,9 +139,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                     type_ = media.get('type')
                     if type_ == 'application/vnd.lumberjack.manifest':
                         continue
-                    if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8':
-                        formats.extend(self._extract_m3u8_formats(
-                            media_url, video_id, 'mp4', m3u8_id='hls'))
+                    ext = determine_ext(media_url)
+                    if type_ == 'application/x-mpegURL' or ext == 'm3u8':
+                        m3u8_formats = self._extract_m3u8_formats(
+                            media_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
+                        if m3u8_formats:
+                            formats.extend(m3u8_formats)
+                    elif type_ == 'application/f4m' or ext == 'f4m':
+                        f4m_formats = self._extract_f4m_formats(
+                            media_url, video_id, preference=-1, f4m_id='hds', fatal=False)
+                        if f4m_formats:
+                            formats.extend(f4m_formats)
                     else:
                         f = {
                             'url': media_url,
@@ -198,9 +215,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                 'video info', flags=re.MULTILINE),
             video_id)
 
-        if info.get('error') is not None:
-            msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
-            raise ExtractorError(msg, expected=True)
+        self._check_error(info)
 
         formats = []
         for (key, format_id) in self._FORMATS:
@@ -243,6 +258,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             'duration': info['duration']
         }
 
+    def _check_error(self, info):
+        if info.get('error') is not None:
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
+
     def _get_subtitles(self, video_id, webpage):
         try:
             sub_list = self._download_webpage(
@@ -274,18 +294,26 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
     }]
 
     def _extract_entries(self, id):
-        video_ids = []
+        video_ids = set()
+        processed_urls = set()
         for pagenum in itertools.count(1):
-            webpage = self._download_webpage_no_ff(
-                self._PAGE_TEMPLATE % (id, pagenum),
-                id, 'Downloading page %s' % pagenum)
+            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())
 
-            video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
+            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, 'Dailymotion')
+                    video_ids.add(video_id)
 
             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
                 break
-        return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
-                for video_id in orderedSet(video_ids)]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -302,7 +330,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 
 class DailymotionUserIE(DailymotionPlaylistIE):
     IE_NAME = 'dailymotion:user'
-    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$'
+    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
     _TESTS = [{
         'url': 'https://www.dailymotion.com/user/nqtv',
@@ -311,6 +339,17 @@ class DailymotionUserIE(DailymotionPlaylistIE):
             'title': 'Rémi Gaillard',
         },
         'playlist_mincount': 100,
+    }, {
+        'url': 'http://www.dailymotion.com/user/UnderProject',
+        'info_dict': {
+            'id': 'UnderProject',
+            'title': 'UnderProject',
+        },
+        'playlist_mincount': 1800,
+        'expected_warnings': [
+            'Stopped at duplicated page',
+        ],
+        'skip': 'Takes too long time',
     }]
 
     def _real_extract(self, url):