Merge tag 'upstream/2013.08.08'

[youtubedl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 7a877b3bcb4e51dfb68a3cbc2720ce52d17c4be9..b633e896c6eb3d5b2158585d01ffba9615fe8f31 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
 


@@ -9,12 +11,34 @@ from ..utils import (
 
     ExtractorError,
 )
 
     ExtractorError,
 )

+from .brightcove import BrightcoveIE

 
 class GenericIE(InfoExtractor):
 
 class GenericIE(InfoExtractor):

-    """Generic last-resort information extractor."""
-
+    IE_DESC = u'Generic downloader that works on some sites'

     _VALID_URL = r'.*'
     IE_NAME = u'generic'
     _VALID_URL = r'.*'
     IE_NAME = u'generic'

+    _TESTS = [
+        {
+            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"
+            }
+        },
+        {
+            u'url': u'http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/',
+            u'file': u'2371591881001.mp4',
+            u'md5': u'9e80619e0a94663f0bdc849b4566af19',
+            u'note': u'Test Brightcove downloads and detection in GenericIE',
+            u'info_dict': {
+                u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
+                u'uploader': u'8TV',
+                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
+            }
+        },
+    ]

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


@@ -95,6 +119,13 @@ class GenericIE(InfoExtractor):
             raise ExtractorError(u'Invalid URL: %s' % url)
 
         self.report_extraction(video_id)
             raise ExtractorError(u'Invalid URL: %s' % url)
 
         self.report_extraction(video_id)

+        # Look for BrigthCove:
+        m_brightcove = re.search(r'<object.+?class=([\'"]).*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL)
+        if m_brightcove is not None:
+            self.to_screen(u'Brightcove video detected.')
+            bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group())
+            return self.url_result(bc_url, 'Brightcove')
+

         # 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:


@@ -102,7 +133,7 @@ class GenericIE(InfoExtractor):
             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|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[^\'"&]*)', 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)


@@ -135,7 +166,7 @@ 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(r'<title>(.*)</title>',
         #   Video Title - Tagline | Site Name
         # and so on and so forth; it's just not practical
         video_title = self._html_search_regex(r'<title>(.*)</title>',

-            webpage, u'video title')
+            webpage, u'video title', default=u'video', flags=re.DOTALL)

 
         # video uploader is domain name
         video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*',
 
         # video uploader is domain name
         video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*',