X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Ftheplatform.py;h=9a57b49df66422a7a7828f2a12ff1385016951ef;hb=ac45505528a71e729a47153ce60053f861e644a9;hp=25edc310008ef0da7b407c97db86ce5e49c78a50;hpb=c687ac745b3c94b2fd246214e78c92a31bd9fc0f;p=youtube-dl

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index 25edc3100..9a57b49df 100644
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -16,11 +16,13 @@ from ..compat import (
 from ..utils import (
     determine_ext,
     ExtractorError,
-    xpath_with_ns,
-    unsmuggle_url,
-    int_or_none,
-    url_basename,
     float_or_none,
+    int_or_none,
+    sanitized_Request,
+    unsmuggle_url,
+    xpath_with_ns,
+    mimetype2ext,
+    find_xpath_attr,
 )
 
 default_ns = 'http://www.w3.org/2005/SMIL21/Language'
@@ -30,15 +32,11 @@ _x = lambda p: xpath_with_ns(p, {'smil': default_ns})
 class ThePlatformBaseIE(InfoExtractor):
     def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'):
         meta = self._download_xml(smil_url, video_id, note=note)
-        try:
-            error_msg = next(
-                n.attrib['abstract']
-                for n in meta.findall(_x('.//smil:ref'))
-                if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired')
-        except StopIteration:
-            pass
-        else:
-            raise ExtractorError(error_msg, expected=True)
+        error_element = find_xpath_attr(
+            meta, _x('.//smil:ref'), 'src',
+            'http://link.theplatform.com/s/errorFiles/Unavailable.mp4')
+        if error_element is not None:
+            raise ExtractorError(error_element.attrib['abstract'], expected=True)
 
         formats = self._parse_smil_formats(
             meta, smil_url, video_id, namespace=default_ns,
@@ -68,7 +66,7 @@ class ThePlatformBaseIE(InfoExtractor):
             for caption in captions:
                 lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type')
                 subtitles[lang] = [{
-                    'ext': 'srt' if mime == 'text/srt' else 'ttml',
+                    'ext': mimetype2ext(mime),
                     'url': src,
                 }]
 
@@ -84,7 +82,7 @@ class ThePlatformBaseIE(InfoExtractor):
 class ThePlatformIE(ThePlatformBaseIE):
     _VALID_URL = r'''(?x)
         (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
-           (?:(?P<media>(?:[^/]+/)+select/media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
+           (?:(?P<media>(?:(?:[^/]+/)+select/)?media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
          |theplatform:)(?P<id>[^/\?&]+)'''
 
     _TESTS = [{
@@ -139,6 +137,11 @@ class ThePlatformIE(ThePlatformBaseIE):
             'upload_date': '20150701',
             'categories': ['Today/Shows/Orange Room', 'Today/Sections/Money', 'Today/Topics/Tech', "Today/Topics/Editor's picks"],
         },
+    }, {
+        # From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1
+        # geo-restricted (US), HLS encrypted with AES-128
+        'url': 'http://player.theplatform.com/p/NnzsPC/onsite_universal/select/media/guid/2410887629/2928790?fwsitesection=nbc_the_blacklist_video_library&autoPlay=true&carouselID=137781',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -182,8 +185,12 @@ class ThePlatformIE(ThePlatformBaseIE):
             # Seems there's no pattern for the interested script filename, so
             # I try one by one
             for script in reversed(scripts):
-                feed_script = self._download_webpage(script, video_id, 'Downloading feed script')
-                feed_id = self._search_regex(r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, 'default feed id', default=None)
+                feed_script = self._download_webpage(
+                    self._proto_relative_url(script, 'http:'),
+                    video_id, 'Downloading feed script')
+                feed_id = self._search_regex(
+                    r'defaultFeedId\s*:\s*"([^"]+)"', feed_script,
+                    'default feed id', default=None)
                 if feed_id is not None:
                     break
             if feed_id is None:
@@ -193,6 +200,20 @@ class ThePlatformIE(ThePlatformBaseIE):
 
         if smuggled_data.get('force_smil_url', False):
             smil_url = url
+        # Explicitly specified SMIL (see https://github.com/rg3/youtube-dl/issues/7385)
+        elif '/guid/' in url:
+            headers = {}
+            source_url = smuggled_data.get('source_url')
+            if source_url:
+                headers['Referer'] = source_url
+            request = sanitized_Request(url, headers=headers)
+            webpage = self._download_webpage(request, video_id)
+            smil_url = self._search_regex(
+                r'<link[^>]+href=(["\'])(?P<url>.+?)\1[^>]+type=["\']application/smil\+xml',
+                webpage, 'smil url', group='url')
+            path = self._search_regex(
+                r'link\.theplatform\.com/s/((?:[^/?#&]+/)+[^/?#&]+)', smil_url, 'path')
+            smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4&format=SMIL'
         elif mobj.group('config'):
             config_url = url + '&form=json'
             config_url = config_url.replace('swf/', 'config/')
@@ -259,8 +280,8 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
         first_video_id = None
         duration = None
         for item in entry['media$content']:
-            smil_url = item['plfile$url'] + '&format=SMIL&Tracking=true&Embedded=true&formats=MPEG4,F4M'
-            cur_video_id = url_basename(smil_url)
+            smil_url = item['plfile$url'] + '&format=SMIL&mbr=true'
+            cur_video_id = ThePlatformIE._match_id(smil_url)
             if first_video_id is None:
                 first_video_id = cur_video_id
                 duration = float_or_none(item.get('plfile$duration'))