Merge pull request #6081 from yan12125/skip_problematic_sites

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

diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py
index 47d58330b6cfd5b1df32374e9debb11a89347554..96f0ed9ad19756b3d380d8023af6c61bee1c18a9 100644 (file)
--- a/youtube_dl/extractor/dailymotion.py
+++ b/youtube_dl/extractor/dailymotion.py
@@ -52,6 +52,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                 'ext': 'mp4',
                 'uploader': 'IGN',
                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
+                'upload_date': '20150306',
             }
         },
         # Vevo video
@@ -85,7 +86,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        url = 'http://www.dailymotion.com/video/%s' % video_id
+        url = 'https://www.dailymotion.com/video/%s' % video_id
 
         # Retrieve video webpage to extract further information
         request = self._build_request(url)
@@ -106,11 +107,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         age_limit = self._rta_search(webpage)
 
         video_upload_date = None
-        mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
+        mobj = re.search(r'<meta property="video:release_date" content="([0-9]{4})-([0-9]{2})-([0-9]{2}).+?"/>', webpage)
         if mobj is not None:
-            video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
+            video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 
-        embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
+        embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id
         embed_request = self._build_request(embed_url)
         embed_page = self._download_webpage(
             embed_request, video_id, 'Downloading embed page')
@@ -224,7 +225,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 
 class DailymotionUserIE(DailymotionPlaylistIE):
     IE_NAME = 'dailymotion:user'
-    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$'
     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
     _TESTS = [{
         'url': 'https://www.dailymotion.com/user/nqtv',
@@ -238,7 +239,8 @@ class DailymotionUserIE(DailymotionPlaylistIE):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         user = mobj.group('user')
-        webpage = self._download_webpage(url, user)
+        webpage = self._download_webpage(
+            'https://www.dailymotion.com/user/%s' % user, user)
         full_user = unescapeHTML(self._html_search_regex(
             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
             webpage, 'user'))
@@ -249,3 +251,45 @@ class DailymotionUserIE(DailymotionPlaylistIE):
             'title': full_user,
             'entries': self._extract_entries(user),
         }
+
+
+class DailymotionCloudIE(DailymotionBaseInfoExtractor):
+    _VALID_URL = r'http://api\.dmcloud\.net/embed/[^/]+/(?P<id>[^/?]+)'
+
+    _TEST = {
+        # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
+        # Tested at FranceTvInfo_2
+        'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
+        'only_matching': True,
+    }
+
+    @classmethod
+    def _extract_dmcloud_url(self, webpage):
+        mobj = re.search(r'<iframe[^>]+src=[\'"](http://api\.dmcloud\.net/embed/[^/]+/[^\'"]+)[\'"]', webpage)
+        if mobj:
+            return mobj.group(1)
+
+        mobj = re.search(r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](http://api\.dmcloud\.net/embed/[^/]+/[^\'"]+)[\'"]', webpage)
+        if mobj:
+            return mobj.group(1)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        request = self._build_request(url)
+        webpage = self._download_webpage(request, video_id)
+
+        title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
+
+        video_info = self._parse_json(self._search_regex(
+            r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
+
+        # TODO: parse ios_url, which is in fact a manifest
+        video_url = video_info['mp4_url']
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'thumbnail': video_info.get('thumbnail_url'),
+        }