]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/cspan.py
d/p/remove-autoupdate-mechanism.patch: Update patch with metadata at the top.
[youtubedl] / youtube_dl / extractor / cspan.py
index d4576160b4489e599e4ca7dabc1e18c9d685610f..67d6df4b0eb285f588b1352eae4a46e0c4b220fd 100644 (file)
@@ -4,12 +4,14 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    int_or_none,
-    unescapeHTML,
-    find_xpath_attr,
-    smuggle_url,
     determine_ext,
     ExtractorError,
+    extract_attributes,
+    find_xpath_attr,
+    get_element_by_class,
+    int_or_none,
+    smuggle_url,
+    unescapeHTML,
 )
 from .senateisvp import SenateISVPIE
 from .ustream import UstreamIE
@@ -67,7 +69,12 @@ class CSpanIE(InfoExtractor):
             'uploader': 'HouseCommittee',
             'uploader_id': '12987475',
         },
+    }, {
+        # Audio Only
+        'url': 'https://www.c-span.org/video/?437336-1/judiciary-antitrust-competition-policy-consumer-rights',
+        'only_matching': True,
     }]
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -78,6 +85,19 @@ class CSpanIE(InfoExtractor):
         if ustream_url:
             return self.url_result(ustream_url, UstreamIE.ie_key())
 
+        if '&vod' not in url:
+            bc = self._search_regex(
+                r"(<[^>]+id='brightcove-player-embed'[^>]+>)",
+                webpage, 'brightcove embed', default=None)
+            if bc:
+                bc_attr = extract_attributes(bc)
+                bc_url = self.BRIGHTCOVE_URL_TEMPLATE % (
+                    bc_attr.get('data-bcaccountid', '3162030207001'),
+                    bc_attr.get('data-noprebcplayerid', 'SyGGpuJy3g'),
+                    bc_attr.get('data-newbcplayerid', 'default'),
+                    bc_attr['data-bcid'])
+                return self.url_result(smuggle_url(bc_url, {'source_url': url}))
+
         # We first look for clipid, because clipprog always appears before
         patterns = [r'id=\'clip(%s)\'\s*value=\'([0-9]+)\'' % t for t in ('id', 'prog')]
         results = list(filter(None, (re.search(p, webpage) for p in patterns)))
@@ -96,7 +116,15 @@ class CSpanIE(InfoExtractor):
                     title = self._og_search_title(webpage)
                     surl = smuggle_url(senate_isvp_url, {'force_title': title})
                     return self.url_result(surl, 'SenateISVP', video_id, title)
+                video_id = self._search_regex(
+                    r'jwsetup\.clipprog\s*=\s*(\d+);',
+                    webpage, 'jwsetup program id', default=None)
+                if video_id:
+                    video_type = 'program'
         if video_type is None or video_id is None:
+            error_message = get_element_by_class('VLplayer-error-message', webpage)
+            if error_message:
+                raise ExtractorError(error_message)
             raise ExtractorError('unable to find video id and type')
 
         def get_text_attr(d, attr):
@@ -123,7 +151,7 @@ class CSpanIE(InfoExtractor):
         entries = []
         for partnum, f in enumerate(files):
             formats = []
-            for quality in f['qualities']:
+            for quality in f.get('qualities', []):
                 formats.append({
                     'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')),
                     'url': unescapeHTML(get_text_attr(quality, 'file')),