Add an extractor for rottentomatoes.com and improve InternetVideoArchiveIE to get...

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 0f38bdd541e53828521eb861ad02d246b6e6d675..9dc9651ad31e331d6b84325dc28787e93764954a 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -94,6 +94,7 @@ from .rbmaradio import RBMARadioIE
 from .redtube import RedTubeIE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
+from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtlnow import RTLnowIE
 from .sina import SinaIE

diff --git a/youtube_dl/extractor/internetvideoarchive.py b/youtube_dl/extractor/internetvideoarchive.py
index 52e3f9eecb9dbc7192e3298f0a5f766d84d62394..5986459d6dfdf7358b7d7a2e4bc139a579a01265 100644 (file)
--- a/youtube_dl/extractor/internetvideoarchive.py
+++ b/youtube_dl/extractor/internetvideoarchive.py
@@ -4,6 +4,7 @@ import xml.etree.ElementTree
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
+    compat_urllib_parse,
     xpath_with_ns,
     determine_ext,
 )
@@ -26,6 +27,16 @@ class InternetVideoArchiveIE(InfoExtractor):
     def _build_url(query):
         return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
 
+    @staticmethod
+    def _clean_query(query):
+        NEEDED_ARGS = ['publishedid', 'customerid']
+        query_dic = compat_urlparse.parse_qs(query)
+        cleaned_dic = dict((k,v[0]) for (k,v) in query_dic.items() if k in NEEDED_ARGS)
+        # Other player ids return m3u8 urls
+        cleaned_dic['playerid'] = '247'
+        cleaned_dic['videokbrate'] = '100000'
+        return compat_urllib_parse.urlencode(cleaned_dic)
+
     def _real_extract(self, url):
         query = compat_urlparse.urlparse(url).query
         query_dic = compat_urlparse.parse_qs(query)
@@ -37,6 +48,11 @@ class InternetVideoArchiveIE(InfoExtractor):
         flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
         file_url = flashconfiguration.find('file').text
         file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
+        # Replace some of the parameters in the query to get the best quality
+        # and http links (no m3u8 manifests)
+        file_url = re.sub(r'(?<=\?)(.+)$',
+            lambda m: self._clean_query(m.group()),
+            file_url)
         info_xml = self._download_webpage(file_url, video_id,
             u'Downloading video info')
         info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))

diff --git a/youtube_dl/extractor/rottentomatoes.py b/youtube_dl/extractor/rottentomatoes.py
new file mode 100644 (file)
index 0000000..c79c394
--- /dev/null
+++ b/youtube_dl/extractor/rottentomatoes.py
@@ -0,0 +1,16 @@
+from .videodetective import VideoDetectiveIE
+
+
+# It just uses the same method as videodetective.com,
+# the internetvideoarchive.com is extracted from the og:video property
+class RottenTomatoesIE(VideoDetectiveIE):
+    _VALID_URL = r'https?://www\.rottentomatoes\.com/m/[^/]+/trailers/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.rottentomatoes.com/m/toy_story_3/trailers/11028566/',
+        u'file': '613340.mp4',
+        u'info_dict': {
+            u'title': u'TOY STORY 3',
+            u'description': u'From the creators of the beloved TOY STORY films, comes a story that will reunite the gang in a whole new way.',
+        },
+    }

diff --git a/youtube_dl/extractor/videodetective.py b/youtube_dl/extractor/videodetective.py
index 265dd5b91fd9e5c4fc5a0cac8a9f36dd36731cfe..d89f8409443675f4359b85b96ffc0c2d49ec32e5 100644 (file)
--- a/youtube_dl/extractor/videodetective.py
+++ b/youtube_dl/extractor/videodetective.py
@@ -16,7 +16,7 @@ class VideoDetectiveIE(InfoExtractor):
         u'info_dict': {
             u'title': u'KICK-ASS 2',
             u'description': u'md5:65ba37ad619165afac7d432eaded6013',
-            u'duration': 135,
+            u'duration': 138,
         },
     }