Implement C-SPAN Live information extractor

This commit introduces the `CSpanLiveIE` class to add support for C-SPAN
live streams. These streams are based on the `BrightcoveNewIE` class
which requires Adobe Pass MSO authentication.

In order to support this new information extractor, `BrightcoveNewIE`
had to be updated to support an optional Akamai token ('hdnts' query
parameter) in the final m3u8 URL used by Brightcove.

youtube_dl/extractor/brightcove.py

@@ -12,6 +12,7 @@ from ..compat import (
     compat_etree_fromstring,
     compat_parse_qs,
     compat_str,
+    compat_urllib_parse_urlencode,
     compat_urllib_parse_urlparse,
     compat_urlparse,
     compat_xml_parse_error,
@@ -611,14 +612,14 @@ class BrightcoveNewIE(AdobePassIE):
 
         return entries
 
-    def _parse_brightcove_metadata(self, json_data, video_id, headers={}):
+    def _parse_brightcove_metadata(self, json_data, video_id, headers={}, options={}):
         title = json_data['name'].strip()
 
         formats = []
         for source in json_data.get('sources', []):
             container = source.get('container')
             ext = mimetype2ext(source.get('type'))
-            src = source.get('src')
+            src = self._preprocess_metadata_url(source.get('src'), options)
             # https://support.brightcove.com/playback-api-video-fields-reference#key_systems_object
             if ext == 'ism' or container == 'WVM' or source.get('key_systems'):
                 continue
@@ -723,6 +724,17 @@ class BrightcoveNewIE(AdobePassIE):
             'is_live': is_live,
         }
 
+    def _preprocess_metadata_url(self, url, options={}):
+        url = compat_urllib_parse_urlparse(url)._asdict()
+        query = dict(compat_parse_qs(url['query']))
+
+        if options.get('akamai_token') is not None:
+            query['hdnts'] = options.get('akamai_token')
+
+        url['query'] = compat_urllib_parse_urlencode(query)
+
+        return compat_urlparse.urlunparse(tuple(url.values()))
+
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
         self._initialize_geo_bypass({
@@ -786,12 +798,16 @@ class BrightcoveNewIE(AdobePassIE):
                     'tveToken': tve_token,
                 })
 
+        options = {
+            'akamai_token': smuggled_data.get('akamai_token')
+        }
+
         if content_type == 'playlist':
             return self.playlist_result(
-                [self._parse_brightcove_metadata(vid, vid.get('id'), headers)
+                [self._parse_brightcove_metadata(vid, vid.get('id'), headers=headers, options=options)
                  for vid in json_data.get('videos', []) if vid.get('id')],
                 json_data.get('id'), json_data.get('name'),
                 json_data.get('description'))
 
         return self._parse_brightcove_metadata(
-            json_data, video_id, headers=headers)
+            json_data, video_id, headers=headers, options=options)

youtube_dl/extractor/cspanlive.py

@@ -0,0 +1,26 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .brightcove import BrightcoveNewIE
+from .common import InfoExtractor
+from ..utils import smuggle_url
+
+
+class CSpanLiveIE(InfoExtractor):
+    IE_NAME = 'cspanlive'
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/3162030207001/2B2qWQJYYM_default/index.html?videoId=%s'
+
+    _VALID_URL = r'^https?://(?:www\.)?c-span\.org/networks'
+
+    def _real_extract(self, url):
+        webpage = self._download_webpage(url, 'stream')
+
+        akamai_token = self._html_search_regex(r'data-akamaitoken="([^"]+)"', webpage, 'akamai_token')
+        video_id = self._html_search_regex(r'data-bcid="([^"]+)"', webpage, 'video_id')
+
+        brightcove_url = smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, {
+            'akamai_token': akamai_token,
+            'source_url': url
+        })
+
+        return self.url_result(brightcove_url, ie=BrightcoveNewIE.ie_key(), video_id=video_id)

youtube_dl/extractor/extractors.py

@@ -238,6 +238,7 @@ from .crunchyroll import (
     CrunchyrollShowPlaylistIE
 )
 from .cspan import CSpanIE
+from .cspanlive import CSpanLiveIE
 from .ctsnews import CtsNewsIE
 from .ctvnews import CTVNewsIE
 from .cultureunplugged import CultureUnpluggedIE