Merge branch 'vgtv' of https://github.com/mrkolby/youtube-dl into mrkolby-vgtv

[youtube-dl] / youtube_dl / extractor / cspan.py

diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py
index a2cbd4d8d592fd2dd6899361a3ab8f4bb3b8a8ed..5411066846eb94b9c9295bae4f8860e07112b2d1 100644 (file)
--- a/youtube_dl/extractor/cspan.py
+++ b/youtube_dl/extractor/cspan.py
@@ -1,48 +1,90 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
 from ..utils import (
+    int_or_none,
     unescapeHTML,
+    find_xpath_attr,
 )
 
 
 class CSpanIE(InfoExtractor):
-    _VALID_URL = r'http://www\.c-spanvideo\.org/program/(.*)'
+    _VALID_URL = r'http://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
     IE_DESC = 'C-SPAN'
-    _TEST = {
-        'url': 'http://www.c-spanvideo.org/program/HolderonV',
-        'file': '315139.mp4',
+    _TESTS = [{
+        '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.',
+            '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.',
         },
-    }
+        'skip': 'Regularly fails on travis, for unknown reasons',
+    }, {
+        'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
+        # For whatever reason, the served video alternates between
+        # two different ones
+        #'md5': 'dbb0f047376d457f2ab8b3929cbb2d0c',
+        'info_dict': {
+            'id': '340723',
+            'ext': 'mp4',
+            'title': 'International Health Care Models',
+            'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
+        }
+    }, {
+        'url': 'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall',
+        'info_dict': {
+            'id': '342759',
+            'title': 'General Motors Ignition Switch Recall',
+        },
+        'playlist_duration_sum': 14855,
+    }]
 
     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')
+        page_id = mobj.group('id')
+        webpage = self._download_webpage(url, page_id)
+        video_id = self._search_regex(r'progid=\'?([0-9]+)\'?>', webpage, 'video id')
 
-        title = self._html_search_regex(
-            r'<!-- title -->\n\s*<h1[^>]*>(.*?)</h1>', webpage, 'title')
-        description = self._og_search_description(webpage)
+        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_json = self._download_webpage(
-            info_url, video_id, 'Downloading video info')
-        data = json.loads(data_json)
+        data = self._download_json(info_url, video_id)
+
+        doc = self._download_xml(
+            'http://www.c-span.org/common/services/flashXml.php?programid=' + video_id,
+            video_id)
+
+        title = find_xpath_attr(doc, './/string', 'name', 'title').text
+        thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
 
-        url = unescapeHTML(data['video']['files'][0]['path']['#text'])
+        files = data['video']['files']
+
+        entries = [{
+            'id': '%s_%d' % (video_id, partnum + 1),
+            'title': (
+                title if len(files) == 1 else
+                '%s part %d' % (title, partnum + 1)),
+            'url': unescapeHTML(f['path']['#text']),
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': int_or_none(f.get('length', {}).get('#text')),
+        } for partnum, f in enumerate(files)]
 
         return {
-            'id': video_id,
+            '_type': 'playlist',
+            'entries': entries,
             'title': title,
-            'url': url,
-            'description': description,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'id': video_id,
         }