Update changelog.

[youtubedl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 20bc533300aa38d5d8b2d6a13eefee44fe439f72..c7552fddb587a60454bec6faa174c36bd4aa9a4a 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -1,3 +1,5 @@
+# encoding: utf-8
+

 import os
 import re
 
 import os
 import re
 


@@ -6,23 +8,70 @@ from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,

+    compat_urlparse,

 
     ExtractorError,
 
     ExtractorError,

+    smuggle_url,
+    unescapeHTML,

 )
 )

+from .brightcove import BrightcoveIE
+

 
 class GenericIE(InfoExtractor):
     IE_DESC = u'Generic downloader that works on some sites'
     _VALID_URL = r'.*'
     IE_NAME = u'generic'
 
 class GenericIE(InfoExtractor):
     IE_DESC = u'Generic downloader that works on some sites'
     _VALID_URL = r'.*'
     IE_NAME = u'generic'

-    _TEST = {
-        u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
-        u'file': u'13601338388002.mp4',
-        u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89',
-        u'info_dict': {
-            u"uploader": u"www.hodiho.fr", 
-            u"title": u"R\u00e9gis plante sa Jeep"
-        }
-    }
+    _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",
+            }
+        },
+        # 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',
+            },
+            u'skip': u'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
+        {
+            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',
+            },
+            u'params': {
+                u'skip_download': True,
+            },
+        },
+    ]

 
     def report_download_webpage(self, video_id):
         """Report webpage download."""
 
     def report_download_webpage(self, video_id):
         """Report webpage download."""


@@ -91,8 +140,18 @@ class GenericIE(InfoExtractor):
         return new_url
 
     def _real_extract(self, url):
         return new_url
 
     def _real_extract(self, url):

-        new_url = self._test_redirect(url)
-        if new_url: return [self.url_result(new_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)
+
+        try:
+            new_url = self._test_redirect(url)
+            if new_url:
+                return [self.url_result(new_url)]
+        except compat_urllib_error.HTTPError:
+            # This may be a stupid server that doesn't like HEAD, our UA, or so
+            pass

 
         video_id = url.split('/')[-1]
         try:
 
         video_id = url.split('/')[-1]
         try:


@@ -100,9 +159,36 @@ 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'Invalid URL: %s' % url)
+            raise ExtractorError(u'Failed to download URL: %s' % url)

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

+        # 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')
+
+        # Look for embedded Vimeo player
+        mobj = re.search(
+            r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage)
+        if mobj:
+            player_url = unescapeHTML(mobj.group(1))
+            surl = smuggle_url(player_url, {'Referer': url})
+            return self.url_result(surl, 'Vimeo')
+
+        # Look for embedded YouTube player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?youtube.com/embed/.+?)\1', webpage)
+        if mobj:
+            surl = unescapeHTML(mobj.group(u'url'))
+            return self.url_result(surl, 'Youtube')
+
+        # Look for Bandcamp pages with custom domain
+        mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
+        if mobj is not None:
+            burl = unescapeHTML(mobj.group(1))
+            return self.url_result(burl, 'Bandcamp')
+

         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:


@@ -122,15 +208,19 @@ class GenericIE(InfoExtractor):
             if m_video_type is not None:
                 mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
         if mobj is None:
             if m_video_type is not None:
                 mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
         if mobj is None:

-            raise ExtractorError(u'Invalid URL: %s' % url)
+            # HTML5 video
+            mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
+        if mobj is None:
+            raise ExtractorError(u'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'Invalid URL: %s' % url)
+            raise ExtractorError(u'Did not find a valid video URL at %s' % url)

 
 

-        video_url = compat_urllib_parse.unquote(mobj.group(1))
-        video_id = 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))

 
         # here's a fun little line of code for you:
         video_extension = os.path.splitext(video_id)[1][1:]
 
         # here's a fun little line of code for you:
         video_extension = os.path.splitext(video_id)[1][1:]