Merge branch 'googledrive' of github.com:remitamine/youtube-dl into remitamine-google...

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

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 6cffde20d907fb96b9ecf53e11e610b9ca0acd2c..7cf13fddfe37fa1bf2d2e6329c060f9d4c6286f7 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -11,7 +11,6 @@ from .youtube import YoutubeIE
 from ..compat import (
     compat_etree_fromstring,
     compat_urllib_parse_unquote,
-    compat_urllib_request,
     compat_urlparse,
     compat_xml_parse_error,
 )
@@ -22,6 +21,7 @@ from ..utils import (
     HEADRequest,
     is_html,
     orderedSet,
+    sanitized_Request,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
@@ -54,6 +54,8 @@ from .onionstudios import OnionStudiosIE
 from .snagfilms import SnagFilmsEmbedIE
 from .screenwavemedia import ScreenwaveMediaIE
 from .mtv import MTVServicesEmbeddedIE
+from .pladform import PladformIE
+from .googledrive import GoogleDriveIE
 
 
 class GenericIE(InfoExtractor):
@@ -339,6 +341,7 @@ class GenericIE(InfoExtractor):
                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
                 'ext': 'mp4',
                 'title': '2cc213299525360.mov',  # that's what we get
+                'duration': 238.231,
             },
             'add_ie': ['Ooyala'],
         },
@@ -350,6 +353,7 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': '"Steve Jobs: Man in the Machine" trailer',
                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
+                'duration': 135.427,
             },
             'params': {
                 'skip_download': True,
@@ -960,8 +964,9 @@ class GenericIE(InfoExtractor):
             'info_dict': {
                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
                 'ext': 'mp4',
-                'description': 'VIDEO: Index/Match versus VLOOKUP.',
+                'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
                 'title': 'This is what separates the Excel masters from the wannabes',
+                'duration': 191.933,
             },
             'params': {
                 # m3u8 downloads
@@ -1058,6 +1063,20 @@ class GenericIE(InfoExtractor):
                 'description': 'Tabletop: Dread, Last Thoughts',
                 'duration': 51690,
             },
+        },
+        # JWPlayer with M3U8
+        {
+            'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video',
+            'info_dict': {
+                'id': 'playlist',
+                'ext': 'mp4',
+                'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ',
+                'uploader': 'ren.tv',
+            },
+            'params': {
+                # m3u8 downloads
+                'skip_download': True,
+            }
         }
     ]
 
@@ -1201,7 +1220,7 @@ class GenericIE(InfoExtractor):
 
         full_response = None
         if head_response is False:
-            request = compat_urllib_request.Request(url)
+            request = sanitized_Request(url)
             request.add_header('Accept-Encoding', '*')
             full_response = self._request_webpage(request, video_id)
             head_response = full_response
@@ -1230,7 +1249,7 @@ class GenericIE(InfoExtractor):
                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
 
         if not full_response:
-            request = compat_urllib_request.Request(url)
+            request = sanitized_Request(url)
             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
             # making it impossible to download only chunk of the file (yet we need only 512kB to
             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
@@ -1487,7 +1506,7 @@ class GenericIE(InfoExtractor):
                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
         if mobj is not None:
-            return OoyalaIE._build_url_result(mobj.group('ec'))
+            return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
 
         # Look for multiple Ooyala embeds on SBN network websites
         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
@@ -1495,7 +1514,7 @@ class GenericIE(InfoExtractor):
             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
             if embeds:
                 return _playlist_from_matches(
-                    embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
+                    embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
 
         # Look for Aparat videos
         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
@@ -1724,10 +1743,9 @@ class GenericIE(InfoExtractor):
             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
 
         # Look for Pladform embeds
-        mobj = re.search(
-            r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
-        if mobj is not None:
-            return self.url_result(mobj.group('url'), 'Pladform')
+        pladform_url = PladformIE._extract_url(webpage)
+        if pladform_url:
+            return self.url_result(pladform_url)
 
         # Look for Playwire embeds
         mobj = re.search(
@@ -1752,6 +1770,11 @@ class GenericIE(InfoExtractor):
         if nbc_sports_url:
             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
 
+        # Look for Google Drive embeds
+        google_drive_url = GoogleDriveIE._extract_url(webpage)
+        if google_drive_url:
+            return self.url_result(google_drive_url, 'GoogleDrive')
+
         # Look for UDN embeds
         mobj = re.search(
             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
@@ -1886,25 +1909,24 @@ class GenericIE(InfoExtractor):
             # here's a fun little line of code for you:
             video_id = os.path.splitext(video_id)[0]
 
+            entry_info_dict = {
+                'id': video_id,
+                'uploader': video_uploader,
+                'title': video_title,
+                'age_limit': age_limit,
+            }
+
             ext = determine_ext(video_url)
             if ext == 'smil':
-                entries.append({
-                    'id': video_id,
-                    'formats': self._extract_smil_formats(video_url, video_id),
-                    'uploader': video_uploader,
-                    'title': video_title,
-                    'age_limit': age_limit,
-                })
+                entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
             elif ext == 'xspf':
                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
+            elif ext == 'm3u8':
+                entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
             else:
-                entries.append({
-                    'id': video_id,
-                    'url': video_url,
-                    'uploader': video_uploader,
-                    'title': video_title,
-                    'age_limit': age_limit,
-                })
+                entry_info_dict['url'] = video_url
+
+            entries.append(entry_info_dict)
 
         if len(entries) == 1:
             return entries[0]