youtube_dl/extractor/cspan.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     unescapeHTML,
   8     find_xpath_attr,
   9 )
  10
  11
  12 class CSpanIE(InfoExtractor):
  13     _VALID_URL = r'http://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
  14     IE_DESC = 'C-SPAN'
  15     _TESTS = [{
  16         'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
  17         'md5': '8e44ce11f0f725527daccc453f553eb0',
  18         'info_dict': {
  19             'id': '315139',
  20             'ext': 'mp4',
  21             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
  22             '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.',
  23         },
  24         'skip': 'Regularly fails on travis, for unknown reasons',
  25     }, {
  26         'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
  27         # For whatever reason, the served video alternates between
  28         # two different ones
  29         #'md5': 'dbb0f047376d457f2ab8b3929cbb2d0c',
  30         'info_dict': {
  31             'id': '340723',
  32             'ext': 'mp4',
  33             'title': 'International Health Care Models',
  34             'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
  35         }
  36     }]
  37
  38     def _real_extract(self, url):
  39         mobj = re.match(self._VALID_URL, url)
  40         page_id = mobj.group('id')
  41         webpage = self._download_webpage(url, page_id)
  42         video_id = self._search_regex(r'progid=\'?([0-9]+)\'?>', webpage, 'video id')
  43
  44         description = self._html_search_regex(
  45             [
  46                 # The full description
  47                 r'<div class=\'expandable\'>(.*?)<a href=\'#\'',
  48                 # If the description is small enough the other div is not
  49                 # present, otherwise this is a stripped version
  50                 r'<p class=\'initial\'>(.*?)</p>'
  51             ],
  52             webpage, 'description', flags=re.DOTALL)
  53
  54         info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id
  55         data = self._download_json(info_url, video_id)
  56
  57         url = unescapeHTML(data['video']['files'][0]['path']['#text'])
  58
  59         doc = self._download_xml('http://www.c-span.org/common/services/flashXml.php?programid=' + video_id + '&version=2014-01-23',
  60             video_id)
  61
  62         formats = [
  63             {
  64                 'url': url,
  65             }
  66         ]
  67
  68         def find_string(node, s):
  69             return find_xpath_attr(node, './/string', 'name', s).text
  70
  71         def find_number(node, s):
  72             return int(find_xpath_attr(node, './/number', 'name', s).text)
  73
  74         def find_array(node, s):
  75             return find_xpath_attr(node, './/array', 'name', s)
  76
  77         def process_files(files, url, formats):
  78             for file in files:
  79                 path = find_string(file, 'path')
  80                 #duration = find_number(file, './number', 'name', 'length')
  81                 hd = find_number(file, 'hd')
  82                 formats.append({
  83                     'url': url,
  84                     'play_path': path,
  85                     'ext': 'flv',
  86                     'quality': hd,
  87                 })
  88
  89         def process_node(node, formats):
  90             url = find_xpath_attr(node, './string', 'name', 'url')
  91             if url is None:
  92                 url = find_xpath_attr(node, './string', 'name', 'URL')
  93                 if url is None:
  94                     return
  95             url = url.text.replace('$(protocol)', 'rtmp').replace('$(port)', '1935')
  96             files = find_array(node, 'files')
  97             if files is None:
  98                 return
  99             process_files(files, url, formats)
 100
 101         process_node(doc.find('./media-link'), formats)
 102
 103         streams = find_array(doc, 'streams')
 104         if streams is not None:
 105             for stream in streams:
 106                 if find_string(stream, 'name') != 'vod':
 107                     continue
 108                 process_node(stream, formats)
 109
 110         return {
 111             'id': video_id,
 112             'title': find_string(doc, 'title'),
 113             'description': description,
 114             'thumbnail': find_string(doc, 'poster'),
 115             'formats': formats,
 116         }