debian/NEWS: Add link to my webpage to give more details.

[youtubedl] / youtube_dl / extractor / cspan.py

diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py
index a4853279bbfc0bce6517a2b1d032ea1eafe07482..d65046f588d0bf4481ec6f8d8de7e031e6bdb2f9 100644 (file)
--- a/youtube_dl/extractor/cspan.py
+++ b/youtube_dl/extractor/cspan.py
@@ -1,53 +1,60 @@
+from __future__ import unicode_literals
+

 import re
 
 from .common import InfoExtractor
 from ..utils import (
 import re
 
 from .common import InfoExtractor
 from ..utils import (

-    compat_urllib_parse,
+    unescapeHTML,
+    find_xpath_attr,

 )
 
 )
 

+

 class CSpanIE(InfoExtractor):
 class CSpanIE(InfoExtractor):

-    _VALID_URL = r'http://www.c-spanvideo.org/program/(.*)'
+    _VALID_URL = r'http://(?:www\.)?c-span\.org/video/\?(?P<id>\d+)'
+    IE_DESC = 'C-SPAN'

     _TEST = {
     _TEST = {

-        u'url': u'http://www.c-spanvideo.org/program/HolderonV',
-        u'file': u'315139.flv',
-        u'md5': u'74a623266956f69e4df0068ab6c80fe4',
-        u'info_dict': {
-            u"title": u"Attorney General Eric Holder on Voting Rights Act Decision"
+        'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
+        'md5': '8e44ce11f0f725527daccc453f553eb0',
+        'info_dict': {
+            'id': '315139',
+            'ext': 'mp4',
+            'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
+            'description': 'Attorney General Eric Holder spoke to reporters following the Supreme Court decision in Shelby County v. Holder in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced until Congress established new guidelines for review.',

         },
         },

-        u'skip': u'Requires rtmpdump'
+        'skip': 'Regularly fails on travis, for unknown reasons',

     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        prog_name = mobj.group(1)
-        webpage = self._download_webpage(url, prog_name)
-        video_id = self._search_regex(r'programid=(.*?)&', webpage, 'video id')
-        data = compat_urllib_parse.urlencode({'programid': video_id,
-                                              'dynamic':'1'})
-        info_url = 'http://www.c-spanvideo.org/common/services/flashXml.php?' + data
-        video_info = self._download_webpage(info_url, video_id, u'Downloading video info')
-
-        self.report_extraction(video_id)
-
-        title = self._html_search_regex(r'<string name="title">(.*?)</string>',
-                                        video_info, 'title')
-        description = self._html_search_regex(r'<meta (?:property="og:|name=")description" content="(.*?)"',
-                                              webpage, 'description',
-                                              flags=re.MULTILINE|re.DOTALL)
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.*?)"',
-                                            webpage, 'thumbnail')
-
-        url = self._search_regex(r'<string name="URL">(.*?)</string>',
-                                 video_info, 'video url')
-        url = url.replace('$(protocol)', 'rtmp').replace('$(port)', '443')
-        path = self._search_regex(r'<string name="path">(.*?)</string>',
-                            video_info, 'rtmp play path')
-
-        return {'id': video_id,
-                'title': title,
-                'ext': 'flv',
-                'url': url,
-                'play_path': path,
-                'description': description,
-                'thumbnail': thumbnail,
-                }
+        page_id = mobj.group('id')
+        webpage = self._download_webpage(url, page_id)
+        video_id = self._search_regex(r'data-progid=\'(\d+)\'>', webpage, 'video id')
+
+        description = self._html_search_regex(
+            [
+                # The full description
+                r'<div class=\'expandable\'>(.*?)<a href=\'#\'',
+                # If the description is small enough the other div is not
+                # present, otherwise this is a stripped version
+                r'<p class=\'initial\'>(.*?)</p>'
+            ],
+            webpage, 'description', flags=re.DOTALL)
+
+        info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id
+        data = self._download_json(info_url, video_id)
+
+        url = unescapeHTML(data['video']['files'][0]['path']['#text'])
+
+        doc = self._download_xml('http://www.c-span.org/common/services/flashXml.php?programid=' + video_id,
+            video_id)
+
+        def find_string(s):
+            return find_xpath_attr(doc, './/string', 'name', s).text
+
+        return {
+            'id': video_id,
+            'title': find_string('title'),
+            'url': url,
+            'description': description,
+            'thumbnail': find_string('poster'),
+        }