debian/changelog: Annotate bugs to close.

[youtubedl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 3d756e8481e0aba09d2290cfd5a4a8b21369aa7d..96ca398de7ce46599e532a5d1adb4a05ac2fb939 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -9,6 +9,8 @@ from .common import InfoExtractor
 from .youtube import YoutubeIE
 from ..compat import (
     compat_urllib_parse,
 from .youtube import YoutubeIE
 from ..compat import (
     compat_urllib_parse,

+    compat_urllib_parse_unquote,
+    compat_urllib_request,

     compat_urlparse,
     compat_xml_parse_error,
 )
     compat_urlparse,
     compat_xml_parse_error,
 )


@@ -32,6 +34,7 @@ from .brightcove import BrightcoveIE
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE

+from .sportbox import SportBoxEmbedIE

 from .smotri import SmotriIE
 from .condenast import CondeNastIE
 from .udn import UDNEmbedIE
 from .smotri import SmotriIE
 from .condenast import CondeNastIE
 from .udn import UDNEmbedIE


@@ -45,6 +48,97 @@ class GenericIE(InfoExtractor):
     _VALID_URL = r'.*'
     IE_NAME = 'generic'
     _TESTS = [
     _VALID_URL = r'.*'
     IE_NAME = 'generic'
     _TESTS = [

+        # Direct link to a video
+        {
+            'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
+            'md5': '67d406c2bcb6af27fa886f31aa934bbe',
+            'info_dict': {
+                'id': 'trailer',
+                'ext': 'mp4',
+                'title': 'trailer',
+                'upload_date': '20100513',
+            }
+        },
+        # Direct link to media delivered compressed (until Accept-Encoding is *)
+        {
+            'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
+            'md5': '128c42e68b13950268b648275386fc74',
+            'info_dict': {
+                'id': 'FictionJunction-Parallel_Hearts',
+                'ext': 'flac',
+                'title': 'FictionJunction-Parallel_Hearts',
+                'upload_date': '20140522',
+            },
+            'expected_warnings': [
+                'URL could be a direct video link, returning it as such.'
+            ]
+        },
+        # Direct download with broken HEAD
+        {
+            'url': 'http://ai-radio.org:8000/radio.opus',
+            'info_dict': {
+                'id': 'radio',
+                'ext': 'opus',
+                'title': 'radio',
+            },
+            'params': {
+                'skip_download': True,  # infinite live stream
+            },
+            'expected_warnings': [
+                r'501.*Not Implemented'
+            ],
+        },
+        # 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.'
+            ]
+        },
+        # RSS feed
+        {
+            'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
+            'info_dict': {
+                'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
+                'title': 'Zero Punctuation',
+                'description': 're:.*groundbreaking video review series.*'
+            },
+            'playlist_mincount': 11,
+        },
+        # RSS feed with enclosure
+        {
+            'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
+            'info_dict': {
+                'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
+                'ext': 'm4v',
+                'upload_date': '20150228',
+                'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
+            }
+        },
+        # google redirect
+        {
+            'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
+            'info_dict': {
+                'id': 'cmQHVoWB5FY',
+                'ext': 'mp4',
+                'upload_date': '20130224',
+                'uploader_id': 'TheVerge',
+                'description': 're:^Chris Ziegler takes a look at the\.*',
+                'uploader': 'The Verge',
+                'title': 'First Firefox OS phones side-by-side',
+            },
+            'params': {
+                'skip_download': False,
+            }
+        },

         {
             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
         {
             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',


@@ -124,17 +218,6 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,  # m3u8 download
             },
         },
                 'skip_download': True,  # m3u8 download
             },
         },

-        # Direct link to a video
-        {
-            'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
-            'md5': '67d406c2bcb6af27fa886f31aa934bbe',
-            'info_dict': {
-                'id': 'trailer',
-                'ext': 'mp4',
-                'title': 'trailer',
-                'upload_date': '20100513',
-            }
-        },

         # 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',


@@ -159,22 +242,6 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Ooyala'],
         },
             },
             'add_ie': ['Ooyala'],
         },

-        # google redirect
-        {
-            'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
-            'info_dict': {
-                'id': 'cmQHVoWB5FY',
-                'ext': 'mp4',
-                'upload_date': '20130224',
-                'uploader_id': 'TheVerge',
-                'description': 're:^Chris Ziegler takes a look at the\.*',
-                'uploader': 'The Verge',
-                'title': 'First Firefox OS phones side-by-side',
-            },
-            'params': {
-                'skip_download': False,
-            }
-        },

         # embed.ly video
         {
             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
         # embed.ly video
         {
             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',


@@ -224,6 +291,37 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
                 'skip_download': True,
             },
         },

+        # SportBox embed
+        {
+            'url': 'http://www.vestifinance.ru/articles/25753',
+            'info_dict': {
+                'id': '25753',
+                'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
+            },
+            'playlist': [{
+                'info_dict': {
+                    'id': '370908',
+                    'title': 'Госзаказ. День 3',
+                    'ext': 'mp4',
+                }
+            }, {
+                'info_dict': {
+                    'id': '370905',
+                    'title': 'Госзаказ. День 2',
+                    'ext': 'mp4',
+                }
+            }, {
+                'info_dict': {
+                    'id': '370902',
+                    'title': 'Госзаказ. День 1',
+                    'ext': 'mp4',
+                }
+            }],
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },

         # Embedded TED video
         {
             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
         # Embedded TED video
         {
             'url': 'http://en.support.wordpress.com/videos/ted-talks/',


@@ -375,16 +473,6 @@ class GenericIE(InfoExtractor):
                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
             }
         },
                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
             }
         },

-        # RSS feed
-        {
-            'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
-            'info_dict': {
-                'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
-                'title': 'Zero Punctuation',
-                'description': 're:.*groundbreaking video review series.*'
-            },
-            'playlist_mincount': 11,
-        },

         # Multiple brightcove videos
         # https://github.com/rg3/youtube-dl/issues/2283
         {
         # Multiple brightcove videos
         # https://github.com/rg3/youtube-dl/issues/2283
         {


@@ -438,21 +526,6 @@ class GenericIE(InfoExtractor):
                 'uploader': 'thoughtworks.wistia.com',
             },
         },
                 'uploader': 'thoughtworks.wistia.com',
             },
         },

-        # Direct download with broken HEAD
-        {
-            'url': 'http://ai-radio.org:8000/radio.opus',
-            'info_dict': {
-                'id': 'radio',
-                'ext': 'opus',
-                'title': 'radio',
-            },
-            'params': {
-                'skip_download': True,  # infinite live stream
-            },
-            'expected_warnings': [
-                r'501.*Not Implemented'
-            ],
-        },

         # Soundcloud embed
         {
             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
         # Soundcloud embed
         {
             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',


@@ -484,21 +557,6 @@ class GenericIE(InfoExtractor):
             },
             'playlist_mincount': 2,
         },
             },
             '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/',
         # Cinchcast embed
         {
             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',


@@ -657,16 +715,6 @@ class GenericIE(InfoExtractor):
                 'age_limit': 0,
             },
         },
                 'age_limit': 0,
             },
         },

-        # RSS feed with enclosure
-        {
-            'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
-            'info_dict': {
-                'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
-                'ext': 'm4v',
-                'upload_date': '20150228',
-                'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
-            }
-        },

         # Crooks and Liars embed
         {
             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
         # Crooks and Liars embed
         {
             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',


@@ -862,7 +910,7 @@ class GenericIE(InfoExtractor):
             force_videoid = smuggled_data['force_videoid']
             video_id = force_videoid
         else:
             force_videoid = smuggled_data['force_videoid']
             video_id = force_videoid
         else:

-            video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
+            video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])

 
         self.to_screen('%s: Requesting header' % video_id)
 
 
         self.to_screen('%s: Requesting header' % video_id)
 


@@ -884,7 +932,9 @@ class GenericIE(InfoExtractor):
 
         full_response = None
         if head_response is False:
 
         full_response = None
         if head_response is False:

-            full_response = self._request_webpage(url, video_id)
+            request = compat_urllib_request.Request(url)
+            request.add_header('Accept-Encoding', '*')
+            full_response = self._request_webpage(request, video_id)

             head_response = full_response
 
         # Check for direct link to a video
             head_response = full_response
 
         # Check for direct link to a video


@@ -895,7 +945,7 @@ class GenericIE(InfoExtractor):
                 head_response.headers.get('Last-Modified'))
             return {
                 'id': video_id,
                 head_response.headers.get('Last-Modified'))
             return {
                 'id': video_id,

-                'title': os.path.splitext(url_basename(url))[0],
+                'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),

                 'direct': True,
                 'formats': [{
                     'format_id': m.group('format_id'),
                 'direct': True,
                 'formats': [{
                     'format_id': m.group('format_id'),


@@ -909,7 +959,17 @@ class GenericIE(InfoExtractor):
             self._downloader.report_warning('Falling back on generic information extractor.')
 
         if not full_response:
             self._downloader.report_warning('Falling back on generic information extractor.')
 
         if not full_response:

-            full_response = self._request_webpage(url, video_id)
+            request = compat_urllib_request.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
+            # that will always result in downloading the whole file that is not desirable.
+            # Therefore for extraction pass we have to override Accept-Encoding to any in order
+            # to accept raw bytes and being able to download only a chunk.
+            # It may probably better to solve this by checking Content-Type for application/octet-stream
+            # after HEAD request finishes, but not sure if we can rely on this.
+            request.add_header('Accept-Encoding', '*')
+            full_response = self._request_webpage(request, video_id)

 
         # Maybe it's a direct link to a video?
         # Be careful not to download the whole thing!
 
         # Maybe it's a direct link to a video?
         # Be careful not to download the whole thing!


@@ -921,7 +981,7 @@ class GenericIE(InfoExtractor):
                 head_response.headers.get('Last-Modified'))
             return {
                 'id': video_id,
                 head_response.headers.get('Last-Modified'))
             return {
                 'id': video_id,

-                'title': os.path.splitext(url_basename(url))[0],
+                'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),

                 'direct': True,
                 'url': url,
                 'upload_date': upload_date,
                 'direct': True,
                 'url': url,
                 'upload_date': upload_date,


@@ -1229,6 +1289,11 @@ class GenericIE(InfoExtractor):
         if rutv_url:
             return self.url_result(rutv_url, 'RUTV')
 
         if rutv_url:
             return self.url_result(rutv_url, 'RUTV')
 

+        # Look for embedded SportBox player
+        sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
+        if sportbox_urls:
+            return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
+

         # Look for embedded TED player
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
         # Look for embedded TED player
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)


@@ -1388,7 +1453,7 @@ class GenericIE(InfoExtractor):
         # Look for Senate ISVP iframe
         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
         if senate_isvp_url:
         # Look for Senate ISVP iframe
         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
         if senate_isvp_url:

-            return self.url_result(surl, 'SenateISVP')
+            return self.url_result(senate_isvp_url, 'SenateISVP')

 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):