Merge pull request #2274 from z00nx/master

[youtube-dl] / youtube_dl / extractor / bliptv.py

diff --git a/youtube_dl/extractor/bliptv.py b/youtube_dl/extractor/bliptv.py
index c87f9160bd676ea13bd59e8381ef43b08abb4a76..ffe6f89b11506a3ae834a02021edd7e59ad12caf 100644 (file)
--- a/youtube_dl/extractor/bliptv.py
+++ b/youtube_dl/extractor/bliptv.py
@@ -8,10 +8,8 @@ import socket
 from .common import InfoExtractor
 from ..utils import (
     compat_http_client,
-    compat_parse_qs,
     compat_str,
     compat_urllib_error,
-    compat_urllib_parse_urlparse,
     compat_urllib_request,
 
     ExtractorError,
@@ -22,12 +20,12 @@ from ..utils import (
 class BlipTVIE(InfoExtractor):
     """Information extractor for blip.tv"""
 
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
-    IE_NAME = 'blip.tv'
+    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
+
     _TEST = {
         'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
-        'file': '5779306.m4v',
-        'md5': '80baf1ec5c3d2019037c1c707d676b9f',
+        'file': '5779306.mov',
+        'md5': 'c6934ad0b6acf2bd920720ec888eb812',
         'info_dict': {
             'upload_date': '20111205',
             'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
@@ -46,17 +44,12 @@ class BlipTVIE(InfoExtractor):
             raise ExtractorError('Invalid URL: %s' % url)
 
         # See https://github.com/rg3/youtube-dl/issues/857
-        api_mobj = re.match(r'http://a\.blip\.tv/api\.swf#(?P<video_id>[\d\w]+)', url)
-        if api_mobj is not None:
-            url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id')
-        urlp = compat_urllib_parse_urlparse(url)
-        if urlp.path.startswith('/play/'):
-            response = self._request_webpage(url, None, False)
-            redirecturl = response.geturl()
-            rurlp = compat_urllib_parse_urlparse(redirecturl)
-            file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2]
-            url = 'http://blip.tv/a/a-' + file_id
-            return self._real_extract(url)
+        embed_mobj = re.search(r'^(?:https?://)?(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
+        if embed_mobj:
+            info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
+            info_page = self._download_webpage(info_url, embed_mobj.group(1))
+            video_id = self._search_regex(r'data-episode-id="(\d+)', info_page,  'video_id')
+            return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
 
         if '?' in url:
             cchar = '&'
@@ -85,8 +78,8 @@ class BlipTVIE(InfoExtractor):
             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
             formats = []
             if 'additionalMedia' in data:
-                for f in sorted(data['additionalMedia'], key=lambda f: int(f['media_height'])):
-                    if not int(f['media_width']): # filter m3u8
+                for f in sorted(data['additionalMedia'], key=lambda f: int(0 if f['media_height'] is None else f['media_height'])):
+                    if not int(0 if f['media_height'] is None else f['media_height']): # filter out m3u8 and srt files
                         continue
                     formats.append({
                         'url': f['url'],
@@ -101,6 +94,8 @@ class BlipTVIE(InfoExtractor):
                     'height': int(data['media']['height']),
                 })
 
+            self._sort_formats(formats)
+
             return {
                 'id': compat_str(data['item_id']),
                 'uploader': data['display_name'],