[theplatform] Rework on <switch> inside <par>

[youtube-dl.git] / youtube_dl / extractor / theplatform.py

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index 1579822f2c8611d6b9e8e464624321009d897251..bcddd23e4b85d3e0f33bb5b8267e4c616c8a0296 100644 (file)
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -8,7 +8,7 @@ import binascii
 import hashlib
 
 
-from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor
 from ..compat import (
     compat_str,
 )
@@ -22,13 +22,13 @@ from ..utils import (
 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
 
 
-class ThePlatformIE(SubtitlesInfoExtractor):
+class ThePlatformIE(InfoExtractor):
     _VALID_URL = r'''(?x)
         (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
          |theplatform:)(?P<id>[^/\?&]+)'''
 
-    _TEST = {
+    _TESTS = [{
         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
         'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
         'info_dict': {
@@ -42,7 +42,20 @@ class ThePlatformIE(SubtitlesInfoExtractor):
             # rtmp download
             'skip_download': True,
         },
-    }
+        # from http://www.cnet.com/videos/tesla-model-s-a-second-step-towards-a-cleaner-motoring-future/
+    }, {
+        'url': 'http://link.theplatform.com/s/kYEXFC/22d_qsQ6MIRT',
+        'info_dict': {
+            'id': '22d_qsQ6MIRT',
+            'ext': 'flv',
+            'description': 'md5:ac330c9258c04f9d7512cf26b9595409',
+            'title': 'Tesla Model S: A second step towards a cleaner motoring future',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }]
 
     @staticmethod
     def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
@@ -71,7 +84,9 @@ class ThePlatformIE(SubtitlesInfoExtractor):
         if not provider_id:
             provider_id = 'dJ5BDC'
 
-        if mobj.group('config'):
+        if smuggled_data.get('force_smil_url', False):
+            smil_url = url
+        elif mobj.group('config'):
             config_url = url + '&form=json'
             config_url = config_url.replace('swf/', 'config/')
             config_url = config_url.replace('onsite/', 'onsite/config/')
@@ -90,7 +105,7 @@ class ThePlatformIE(SubtitlesInfoExtractor):
             error_msg = next(
                 n.attrib['abstract']
                 for n in meta.findall(_x('.//smil:ref'))
-                if n.attrib.get('title') == 'Geographic Restriction')
+                if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired')
         except StopIteration:
             pass
         else:
@@ -104,15 +119,11 @@ class ThePlatformIE(SubtitlesInfoExtractor):
         captions = info.get('captions')
         if isinstance(captions, list):
             for caption in captions:
-                lang, src = caption.get('lang'), caption.get('src')
-                if lang and src:
-                    subtitles[lang] = src
-
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, subtitles)
-            return
-
-        subtitles = self.extract_subtitles(video_id, subtitles)
+                lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type')
+                subtitles[lang] = [{
+                    'ext': 'srt' if mime == 'text/srt' else 'ttml',
+                    'url': src,
+                }]
 
         head = meta.find(_x('smil:head'))
         body = meta.find(_x('smil:body'))
@@ -129,6 +140,8 @@ class ThePlatformIE(SubtitlesInfoExtractor):
         else:
             formats = []
             switch = body.find(_x('smil:switch'))
+            if switch is None:
+                switch = body.find(_x('smil:par//smil:switch'))
             if switch is not None:
                 base_url = head.find(_x('smil:meta')).attrib['base']
                 for f in switch.findall(_x('smil:video')):