debian/NEWS: Add link to my webpage to give more details.

[youtubedl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 7a14c98f9b6ef9d550606c72c330d0730ec1233e..5bcc78bf79734ddd47ee93b2215c31bdc28e59d3 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -1,9 +1,12 @@
 # encoding: utf-8
 
 # encoding: utf-8
 

+from __future__ import unicode_literals
+

 import os
 import re
 
 from .common import InfoExtractor
 import os
 import re
 
 from .common import InfoExtractor

+from .youtube import YoutubeIE

 from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
 from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,


@@ -22,78 +25,79 @@ from .ooyala import OoyalaIE
 
 
 class GenericIE(InfoExtractor):
 
 
 class GenericIE(InfoExtractor):

-    IE_DESC = u'Generic downloader that works on some sites'
+    IE_DESC = 'Generic downloader that works on some sites'

     _VALID_URL = r'.*'
     _VALID_URL = r'.*'

-    IE_NAME = u'generic'
+    IE_NAME = 'generic'

     _TESTS = [
         {
     _TESTS = [
         {

-            u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
-            u'file': u'13601338388002.mp4',
-            u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd',
-            u'info_dict': {
-                u"uploader": u"www.hodiho.fr",
-                u"title": u"R\u00e9gis plante sa Jeep"
-            }
-        },
-        # embedded vimeo video
-        {
-            u'add_ie': ['Vimeo'],
-            u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
-            u'file': u'22444065.mp4',
-            u'md5': u'2903896e23df39722c33f015af0666e2',
-            u'info_dict': {
-                u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
-                u"uploader_id": u"skillsmatter",
-                u"uploader": u"Skills Matter",
+            'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
+            'file': '13601338388002.mp4',
+            'md5': '6e15c93721d7ec9e9ca3fdbf07982cfd',
+            'info_dict': {
+                'uploader': 'www.hodiho.fr',
+                'title': 'R\u00e9gis plante sa Jeep',

             }
         },
         # bandcamp page with custom domain
         {
             }
         },
         # bandcamp page with custom domain
         {

-            u'add_ie': ['Bandcamp'],
-            u'url': u'http://bronyrock.com/track/the-pony-mash',
-            u'file': u'3235767654.mp3',
-            u'info_dict': {
-                u'title': u'The Pony Mash',
-                u'uploader': u'M_Pallante',
+            'add_ie': ['Bandcamp'],
+            'url': 'http://bronyrock.com/track/the-pony-mash',
+            'file': '3235767654.mp3',
+            'info_dict': {
+                'title': 'The Pony Mash',
+                'uploader': 'M_Pallante',

             },
             },

-            u'skip': u'There is a limit of 200 free downloads / month for the test song',
+            'skip': 'There is a limit of 200 free downloads / month for the test song',

         },
         # embedded brightcove video
         # it also tests brightcove videos that need to set the 'Referer' in the
         # http requests
         {
         },
         # embedded brightcove video
         # it also tests brightcove videos that need to set the 'Referer' in the
         # http requests
         {

-            u'add_ie': ['Brightcove'],
-            u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
-            u'info_dict': {
-                u'id': u'2765128793001',
-                u'ext': u'mp4',
-                u'title': u'Le cours de bourse : l’analyse technique',
-                u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9',
-                u'uploader': u'BFM BUSINESS',
+            'add_ie': ['Brightcove'],
+            'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
+            'info_dict': {
+                'id': '2765128793001',
+                'ext': 'mp4',
+                'title': 'Le cours de bourse : l’analyse technique',
+                'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
+                'uploader': 'BFM BUSINESS',

             },
             },

-            u'params': {
-                u'skip_download': True,
+            'params': {
+                'skip_download': True,
+            },
+        },
+        {
+            # https://github.com/rg3/youtube-dl/issues/2253
+            'url': 'http://bcove.me/i6nfkrc3',
+            'file': '3101154703001.mp4',
+            'md5': '0ba9446db037002366bab3b3eb30c88c',
+            'info_dict': {
+                'title': 'Still no power',
+                'uploader': 'thestar.com',
+                'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',

             },
             },

+            'add_ie': ['Brightcove'],

         },
         # Direct link to a video
         {
         },
         # Direct link to a video
         {

-            u'url': u'http://media.w3.org/2010/05/sintel/trailer.mp4',
-            u'file': u'trailer.mp4',
-            u'md5': u'67d406c2bcb6af27fa886f31aa934bbe',
-            u'info_dict': {
-                u'id': u'trailer',
-                u'title': u'trailer',
-                u'upload_date': u'20100513',
+            'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
+            'file': 'trailer.mp4',
+            'md5': '67d406c2bcb6af27fa886f31aa934bbe',
+            'info_dict': {
+                'id': 'trailer',
+                'title': 'trailer',
+                'upload_date': '20100513',

             }
         },
         # ooyala video
         {
             }
         },
         # ooyala video
         {

-            u'url': u'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
-            u'md5': u'5644c6ca5d5782c1d0d350dad9bd840c',
-            u'info_dict': {
-                u'id': u'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
-                u'ext': u'mp4',
-                u'title': u'2cc213299525360.mov', #that's what we get
+            'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
+            'file': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ.mp4',
+            'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
+            'info_dict': {
+                'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
+                'ext': 'mp4',
+                'title': '2cc213299525360.mov',  # that's what we get

             },
         },
     ]
             },
         },
     ]


@@ -101,12 +105,12 @@ class GenericIE(InfoExtractor):
     def report_download_webpage(self, video_id):
         """Report webpage download."""
         if not self._downloader.params.get('test', False):
     def report_download_webpage(self, video_id):
         """Report webpage download."""
         if not self._downloader.params.get('test', False):

-            self._downloader.report_warning(u'Falling back on generic information extractor.')
+            self._downloader.report_warning('Falling back on generic information extractor.')

         super(GenericIE, self).report_download_webpage(video_id)
 
     def report_following_redirect(self, new_url):
         """Report information extraction."""
         super(GenericIE, self).report_download_webpage(video_id)
 
     def report_following_redirect(self, new_url):
         """Report information extraction."""

-        self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url)
+        self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)

 
     def _send_head(self, url):
         """Check if it is a redirect, like url shorteners, in case return the new url."""
 
     def _send_head(self, url):
         """Check if it is a redirect, like url shorteners, in case return the new url."""


@@ -152,16 +156,29 @@ class GenericIE(InfoExtractor):
 
         response = opener.open(HEADRequest(url))
         if response is None:
 
         response = opener.open(HEADRequest(url))
         if response is None:

-            raise ExtractorError(u'Invalid URL protocol')
+            raise ExtractorError('Invalid URL protocol')

         return response
 
     def _real_extract(self, url):
         parsed_url = compat_urlparse.urlparse(url)
         if not parsed_url.scheme:
         return response
 
     def _real_extract(self, url):
         parsed_url = compat_urlparse.urlparse(url)
         if not parsed_url.scheme:

-            self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
-            return self.url_result('http://' + url)
+            default_search = self._downloader.params.get('default_search')
+            if default_search is None:
+                default_search = 'auto'
+
+            if default_search == 'auto':
+                if '/' in url:
+                    self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
+                    return self.url_result('http://' + url)
+                else:
+                    return self.url_result('ytsearch:' + url)
+            else:
+                assert ':' in default_search
+                return self.url_result(default_search + url)

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

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

         try:
             response = self._send_head(url)
 
         try:
             response = self._send_head(url)
 


@@ -184,7 +201,7 @@ class GenericIE(InfoExtractor):
                     'formats': [{
                         'format_id': m.group('format_id'),
                         'url': url,
                     'formats': [{
                         'format_id': m.group('format_id'),
                         'url': url,

-                        'vcodec': u'none' if m.group('type') == 'audio' else None
+                        'vcodec': 'none' if m.group('type') == 'audio' else None

                     }],
                     'upload_date': upload_date,
                 }
                     }],
                     'upload_date': upload_date,
                 }


@@ -198,7 +215,7 @@ class GenericIE(InfoExtractor):
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here

-            raise ExtractorError(u'Failed to download URL: %s' % url)
+            raise ExtractorError('Failed to download URL: %s' % url)

 
         self.report_extraction(video_id)
 
 
         self.report_extraction(video_id)
 


@@ -209,22 +226,33 @@ class GenericIE(InfoExtractor):
         #   Video Title - Tagline | Site Name
         # and so on and so forth; it's just not practical
         video_title = self._html_search_regex(
         #   Video Title - Tagline | Site Name
         # and so on and so forth; it's just not practical
         video_title = self._html_search_regex(

-            r'(?s)<title>(.*?)</title>', webpage, u'video title',
-            default=u'video')
+            r'(?s)<title>(.*?)</title>', webpage, 'video title',
+            default='video')

 
         # video uploader is domain name
         video_uploader = self._search_regex(
 
         # video uploader is domain name
         video_uploader = self._search_regex(

-            r'^(?:https?://)?([^/]*)/.*', url, u'video uploader')
+            r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')

 
         # Look for BrightCove:
 
         # Look for BrightCove:

-        bc_url = BrightcoveIE._extract_brightcove_url(webpage)
-        if bc_url is not None:
-            self.to_screen(u'Brightcove video detected.')
-            return self.url_result(bc_url, 'Brightcove')
+        bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
+        if bc_urls:
+            self.to_screen('Brightcove video detected.')
+            entries = [{
+                '_type': 'url',
+                'url': smuggle_url(bc_url, {'Referer': url}),
+                'ie_key': 'Brightcove'
+            } for bc_url in bc_urls]
+
+            return {
+                '_type': 'playlist',
+                'title': video_title,
+                'id': video_id,
+                'entries': entries,
+            }

 
         # Look for embedded (iframe) Vimeo player
         mobj = re.search(
 
         # Look for embedded (iframe) Vimeo player
         mobj = re.search(

-            r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage)
+            r'<iframe[^>]+?src="((?:https?:)?//player\.vimeo\.com/video/.+?)"', webpage)

         if mobj:
             player_url = unescapeHTML(mobj.group(1))
             surl = smuggle_url(player_url, {'Referer': url})
         if mobj:
             player_url = unescapeHTML(mobj.group(1))
             surl = smuggle_url(player_url, {'Referer': url})


@@ -232,7 +260,7 @@ class GenericIE(InfoExtractor):
 
         # Look for embedded (swf embed) Vimeo player
         mobj = re.search(
 
         # Look for embedded (swf embed) Vimeo player
         mobj = re.search(

-            r'<embed[^>]+?src="(https?://(?:www\.)?vimeo.com/moogaloop.swf.+?)"', webpage)
+            r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)

         if mobj:
             return self.url_result(mobj.group(1), 'Vimeo')
 
         if mobj:
             return self.url_result(mobj.group(1), 'Vimeo')
 


@@ -271,16 +299,12 @@ class GenericIE(InfoExtractor):
             }
 
         # Look for embedded blip.tv player
             }
 
         # Look for embedded blip.tv player

-        mobj = re.search(r'<meta\s[^>]*https?://api.blip.tv/\w+/redirect/\w+/(\d+)', webpage)
+        mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)

         if mobj:
         if mobj:

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

-            player_url = 'http://blip.tv/play/%s.x?p=1' % mobj.group(1)
-            player_page = self._download_webpage(player_url, mobj.group(1))
-            blip_video_id = self._search_regex(r'data-episode-id="(\d+)', player_page, u'blip_video_id', fatal=False)
-            if blip_video_id:
-                return self.url_result('http://blip.tv/seo/-'+blip_video_id, 'BlipTV')
+            return self.url_result(mobj.group(1), 'BlipTV')

 
         # Look for Bandcamp pages with custom domain
         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
 
         # Look for Bandcamp pages with custom domain
         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)


@@ -301,18 +325,44 @@ class GenericIE(InfoExtractor):
             return OoyalaIE._build_url_result(mobj.group(1))
 
         # Look for Aparat videos
             return OoyalaIE._build_url_result(mobj.group(1))
 
         # Look for Aparat videos

-        mobj = re.search(r'<iframe src="(http://www.aparat.com/video/[^"]+)"', webpage)
+        mobj = re.search(r'<iframe src="(http://www\.aparat\.com/video/[^"]+)"', webpage)

         if mobj is not None:
             return self.url_result(mobj.group(1), 'Aparat')
 
         if mobj is not None:
             return self.url_result(mobj.group(1), 'Aparat')
 

+        # Look for MPORA videos
+        mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group(1), 'Mpora')
+
+        # Look for embedded Novamov player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?novamov\.com/embed\.php.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Novamov')
+
+        # Look for embedded Facebook player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Facebook')
+
+        # Look for embedded Huffington Post player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'HuffPost')
+

         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)

+        if mobj is None:
+            # Look for gorilla-vid style embedding
+            mobj = re.search(r'(?s)(?:jw_plugins|JWPlayerOptions).*?file\s*:\s*["\'](.*?)["\']', webpage)

         if mobj is None:
             # Broaden the search a little bit
             mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
         if mobj is None:
             # Broaden the search a little bit: JWPlayer JS loader
         if mobj is None:
             # Broaden the search a little bit
             mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
         if mobj is None:
             # Broaden the search a little bit: JWPlayer JS loader

-            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"]*)', webpage)
+            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)

         if mobj is None:
             # Try to find twitter cards info
             mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
         if mobj is None:
             # Try to find twitter cards info
             mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)


@@ -327,23 +377,27 @@ class GenericIE(InfoExtractor):
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:

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

 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:
 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:

-            raise ExtractorError(u'Did not find a valid video URL at %s' % url)
+            raise ExtractorError('Did not find a valid video URL at %s' % url)

 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)
         video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 
 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)
         video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 

+        # Sometimes, jwplayer extraction will result in a YouTube URL
+        if YoutubeIE.suitable(video_url):
+            return self.url_result(video_url, 'Youtube')
+

         # here's a fun little line of code for you:
         video_id = os.path.splitext(video_id)[0]
 
         return {
         # here's a fun little line of code for you:
         video_id = os.path.splitext(video_id)[0]
 
         return {

-            'id':       video_id,
-            'url':      video_url,
+            'id': video_id,
+            'url': video_url,

             'uploader': video_uploader,
             'uploader': video_uploader,

-            'title':    video_title,
+            'title': video_title,

         }
         }