Initiate a new release.

[youtubedl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index af769ab619eac9cb9b3d20869ba4a9e009977f6c..fbbc79a574ca03f1e483738a726f2fde0bf6b21d 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -17,12 +17,14 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     HEADRequest,
     ExtractorError,
     float_or_none,
     HEADRequest,

+    is_html,

     orderedSet,
     parse_xml,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
     unsmuggle_url,
     orderedSet,
     parse_xml,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
     unsmuggle_url,

+    UnsupportedError,

     url_basename,
 )
 from .brightcove import BrightcoveIE
     url_basename,
 )
 from .brightcove import BrightcoveIE


@@ -130,12 +132,26 @@ class GenericIE(InfoExtractor):
         # ooyala video
         {
             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
         # ooyala video
         {
             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',

-            'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
+            'md5': '166dd577b433b4d4ebfee10b0824d8ff',

             'info_dict': {
                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
                 'ext': 'mp4',
                 'title': '2cc213299525360.mov',  # that's what we get
             },
             'info_dict': {
                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
                 'ext': 'mp4',
                 'title': '2cc213299525360.mov',  # that's what we get
             },

+            'add_ie': ['Ooyala'],
+        },
+        # multiple ooyala embeds on SBN network websites
+        {
+            'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
+            'info_dict': {
+                'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
+                'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
+            },
+            'playlist_mincount': 3,
+            'params': {
+                'skip_download': True,
+            },
+            'add_ie': ['Ooyala'],

         },
         # google redirect
         {
         },
         # google redirect
         {


@@ -145,7 +161,7 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'upload_date': '20130224',
                 'uploader_id': 'TheVerge',
                 'ext': 'mp4',
                 'upload_date': '20130224',
                 'uploader_id': 'TheVerge',

-                'description': 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.',
+                'description': 're:^Chris Ziegler takes a look at the\.*',

                 'uploader': 'The Verge',
                 'title': 'First Firefox OS phones side-by-side',
             },
                 'uploader': 'The Verge',
                 'title': 'First Firefox OS phones side-by-side',
             },


@@ -180,6 +196,14 @@ class GenericIE(InfoExtractor):
                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
             },
         },
                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
             },
         },

+        # BBC iPlayer embeds
+        {
+            'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
+            'info_dict': {
+                'title': 'BBC - Blogs -  Adam Curtis - BUGGER',
+            },
+            'playlist_mincount': 18,
+        },

         # RUTV embed
         {
             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
         # RUTV embed
         {
             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',


@@ -351,7 +375,7 @@ class GenericIE(InfoExtractor):
             'info_dict': {
                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
                 'title': 'Zero Punctuation',
             'info_dict': {
                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
                 'title': 'Zero Punctuation',

-                'description': 're:'
+                'description': 're:.*groundbreaking video review series.*'

             },
             'playlist_mincount': 11,
         },
             },
             'playlist_mincount': 11,
         },


@@ -445,6 +469,62 @@ class GenericIE(InfoExtractor):
                 'title': 'Rosetta #CometLanding webcast HL 10',
             }
         },
                 'title': 'Rosetta #CometLanding webcast HL 10',
             }
         },

+        # LazyYT
+        {
+            'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
+            'info_dict': {
+                'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
+            },
+            'playlist_mincount': 2,
+        },
+        # Direct link with incorrect MIME type
+        {
+            'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
+            'md5': '4ccbebe5f36706d85221f204d7eb5913',
+            'info_dict': {
+                'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
+                'id': '5_Lennart_Poettering_-_Systemd',
+                'ext': 'webm',
+                'title': '5_Lennart_Poettering_-_Systemd',
+                'upload_date': '20141120',
+            },
+            'expected_warnings': [
+                'URL could be a direct video link, returning it as such.'
+            ]
+        },
+        # Cinchcast embed
+        {
+            'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
+            'info_dict': {
+                'id': '7141703',
+                'ext': 'mp3',
+                'upload_date': '20141126',
+                'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
+            }
+        },
+        # Cinerama player
+        {
+            'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
+            'info_dict': {
+                'id': '730m_DandD_1901_512k',
+                'ext': 'mp4',
+                'uploader': 'www.abc.net.au',
+                'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
+            }
+        },
+        # embedded viddler video
+        {
+            'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
+            'info_dict': {
+                'id': '4d03aad9',
+                'ext': 'mp4',
+                'uploader': 'deadspin',
+                'title': 'WALL-TO-GORTAT',
+                'timestamp': 1422285291,
+                'upload_date': '20150126',
+            },
+            'add_ie': ['Viddler'],
+        }

     ]
 
     def report_following_redirect(self, new_url):
     ]
 
     def report_following_redirect(self, new_url):


@@ -537,9 +617,9 @@ class GenericIE(InfoExtractor):
 
             if default_search in ('error', 'fixup_error'):
                 raise ExtractorError(
 
             if default_search in ('error', 'fixup_error'):
                 raise ExtractorError(

-                    ('%r is not a valid URL. '
-                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
-                    ) % (url, url), expected=True)
+                    '%r is not a valid URL. '
+                    'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
+                    % (url, url), expected=True)

             else:
                 if ':' not in default_search:
                     default_search += ':'
             else:
                 if ':' not in default_search:
                     default_search += ':'


@@ -598,10 +678,28 @@ class GenericIE(InfoExtractor):
         if not self._downloader.params.get('test', False) and not is_intentional:
             self._downloader.report_warning('Falling back on generic information extractor.')
 
         if not self._downloader.params.get('test', False) and not is_intentional:
             self._downloader.report_warning('Falling back on generic information extractor.')
 

-        if full_response:
-            webpage = self._webpage_read_content(full_response, url, video_id)
-        else:
-            webpage = self._download_webpage(url, video_id)
+        if not full_response:
+            full_response = self._request_webpage(url, video_id)
+
+        # Maybe it's a direct link to a video?
+        # Be careful not to download the whole thing!
+        first_bytes = full_response.read(512)
+        if not is_html(first_bytes):
+            self._downloader.report_warning(
+                'URL could be a direct video link, returning it as such.')
+            upload_date = unified_strdate(
+                head_response.headers.get('Last-Modified'))
+            return {
+                'id': video_id,
+                'title': os.path.splitext(url_basename(url))[0],
+                'direct': True,
+                'url': url,
+                'upload_date': upload_date,
+            }
+
+        webpage = self._webpage_read_content(
+            full_response, url, video_id, prefix=first_bytes)
+

         self.report_extraction(video_id)
 
         # Is it an RSS feed?
         self.report_extraction(video_id)
 
         # Is it an RSS feed?


@@ -647,9 +745,9 @@ class GenericIE(InfoExtractor):
             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 
         # Helper method
             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 
         # Helper method

-        def _playlist_from_matches(matches, getter, ie=None):
+        def _playlist_from_matches(matches, getter=None, ie=None):

             urlrs = orderedSet(
             urlrs = orderedSet(

-                self.url_result(self._proto_relative_url(getter(m)), ie)
+                self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)

                 for m in matches)
             return self.playlist_result(
                 urlrs, playlist_id=video_id, playlist_title=video_title)
                 for m in matches)
             return self.playlist_result(
                 urlrs, playlist_id=video_id, playlist_title=video_title)


@@ -702,6 +800,12 @@ class GenericIE(InfoExtractor):
             return _playlist_from_matches(
                 matches, lambda m: unescapeHTML(m[1]))
 
             return _playlist_from_matches(
                 matches, lambda m: unescapeHTML(m[1]))
 

+        # Look for lazyYT YouTube embed
+        matches = re.findall(
+            r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
+        if matches:
+            return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
+

         # Look for embedded Dailymotion player
         matches = re.findall(
             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
         # Look for embedded Dailymotion player
         matches = re.findall(
             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)


@@ -733,7 +837,7 @@ class GenericIE(InfoExtractor):
                 'title': video_title,
                 'id': video_id,
             }
                 'title': video_title,
                 'id': video_id,
             }

-            
+

         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
         if match:
             return {
         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
         if match:
             return {


@@ -748,7 +852,7 @@ class GenericIE(InfoExtractor):
         # Look for embedded blip.tv player
         mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
         if mobj:
         # Look for embedded blip.tv player
         mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
         if mobj:

-            return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
+            return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')

         mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
         if mobj:
             return self.url_result(mobj.group(1), 'BlipTV')
         mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
         if mobj:
             return self.url_result(mobj.group(1), 'BlipTV')


@@ -782,12 +886,28 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'))
 
         if mobj is not None:
             return self.url_result(mobj.group('url'))
 

+        # Look for embedded Viddler player
+        mobj = re.search(
+            r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'))
+

         # Look for Ooyala videos
         # Look for Ooyala videos

-        mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
-             re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
+        mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
+                re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
+                re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))

         if mobj is not None:
             return OoyalaIE._build_url_result(mobj.group('ec'))
 
         if mobj is not None:
             return OoyalaIE._build_url_result(mobj.group('ec'))
 

+        # Look for multiple Ooyala embeds on SBN network websites
+        mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
+        if mobj is not None:
+            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')
+

         # Look for Aparat videos
         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
         if mobj is not None:
         # Look for Aparat videos
         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
         if mobj is not None:


@@ -847,6 +967,11 @@ class GenericIE(InfoExtractor):
             return _playlist_from_matches(
                 matches, getter=unescapeHTML, ie='FunnyOrDie')
 
             return _playlist_from_matches(
                 matches, getter=unescapeHTML, ie='FunnyOrDie')
 

+        # Look for BBC iPlayer embed
+        matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
+        if matches:
+            return _playlist_from_matches(matches, ie='BBCCoUk')
+

         # Look for embedded RUTV player
         rutv_url = RUTVIE._extract_url(webpage)
         if rutv_url:
         # Look for embedded RUTV player
         rutv_url = RUTVIE._extract_url(webpage)
         if rutv_url:


@@ -854,7 +979,7 @@ class GenericIE(InfoExtractor):
 
         # Look for embedded TED player
         mobj = re.search(
 
         # Look for embedded TED player
         mobj = re.search(

-            r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)

         if mobj is not None:
             return self.url_result(mobj.group('url'), 'TED')
 
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'TED')
 


@@ -914,6 +1039,13 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'SBS')
 
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'SBS')
 

+        # Look for embedded Cinchcast player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Cinchcast')
+

         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
             webpage)
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
             webpage)


@@ -963,9 +1095,13 @@ class GenericIE(InfoExtractor):
             found = filter_video(re.findall(r'''(?xs)
                 flowplayer\("[^"]+",\s*
                     \{[^}]+?\}\s*,
             found = filter_video(re.findall(r'''(?xs)
                 flowplayer\("[^"]+",\s*
                     \{[^}]+?\}\s*,

-                    \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
+                    \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*

                         ["']?url["']?\s*:\s*["']([^"']+)["']
             ''', webpage))
                         ["']?url["']?\s*:\s*["']([^"']+)["']
             ''', webpage))

+        if not found:
+            # Cinerama player
+            found = re.findall(
+                r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)

         if not found:
             # Try to find twitter cards info
             found = filter_video(re.findall(
         if not found:
             # Try to find twitter cards info
             found = filter_video(re.findall(


@@ -979,7 +1115,7 @@ class GenericIE(InfoExtractor):
                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
         if not found:
             # HTML5 video
                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
         if not found:
             # HTML5 video

-            found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src="([^"]+)"', webpage)
+            found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)

         if not found:
             found = re.search(
                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
         if not found:
             found = re.search(
                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'


@@ -993,7 +1129,7 @@ class GenericIE(InfoExtractor):
                     'url': new_url,
                 }
         if not found:
                     'url': new_url,
                 }
         if not found:

-            raise ExtractorError('Unsupported URL: %s' % url)
+            raise UnsupportedError(url)

 
         entries = []
         for video_url in found:
 
         entries = []
         for video_url in found:


@@ -1025,4 +1161,3 @@ class GenericIE(InfoExtractor):
                 '_type': 'playlist',
                 'entries': entries,
             }
                 '_type': 'playlist',
                 'entries': entries,
             }

-