Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ccc.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     int_or_none,
   8     qualities,
   9     unified_strdate,
  10 )
  11
  12
  13 class CCCIE(InfoExtractor):
  14     IE_NAME = 'media.ccc.de'
  15     _VALID_URL = r'https?://(?:www\.)?media\.ccc\.de/[^?#]+/[^?#/]*?_(?P<id>[0-9]{8,})._[^?#/]*\.html'
  16
  17     _TEST = {
  18         'url': 'http://media.ccc.de/browse/congress/2013/30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor.html#video',
  19         'md5': '205a365d0d57c0b1e43a12c9ffe8f9be',
  20         'info_dict': {
  21             'id': '20131228183',
  22             'ext': 'mp4',
  23             'title': 'Introduction to Processor Design',
  24             'description': 'md5:5ddbf8c734800267f2cee4eab187bc1b',
  25             'thumbnail': 're:^https?://.*\.jpg$',
  26             'view_count': int,
  27             'upload_date': '20131229',
  28         }
  29     }
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33         webpage = self._download_webpage(url, video_id)
  34
  35         if self._downloader.params.get('prefer_free_formats'):
  36             preference = qualities(['mp3', 'opus', 'mp4-lq', 'webm-lq', 'h264-sd', 'mp4-sd', 'webm-sd', 'mp4', 'webm', 'mp4-hd', 'h264-hd', 'webm-hd'])
  37         else:
  38             preference = qualities(['opus', 'mp3', 'webm-lq', 'mp4-lq', 'webm-sd', 'h264-sd', 'mp4-sd', 'webm', 'mp4', 'webm-hd', 'mp4-hd', 'h264-hd'])
  39
  40         title = self._html_search_regex(
  41             r'(?s)<h1>(.*?)</h1>', webpage, 'title')
  42         description = self._html_search_regex(
  43             r"(?s)<p class='description'>(.*?)</p>",
  44             webpage, 'description', fatal=False)
  45         upload_date = unified_strdate(self._html_search_regex(
  46             r"(?s)<span class='[^']*fa-calendar-o'></span>(.*?)</li>",
  47             webpage, 'upload date', fatal=False))
  48         view_count = int_or_none(self._html_search_regex(
  49             r"(?s)<span class='[^']*fa-eye'></span>(.*?)</li>",
  50             webpage, 'view count', fatal=False))
  51
  52         matches = re.finditer(r'''(?xs)
  53             <(?:span|div)\s+class='label\s+filetype'>(?P<format>.*?)</(?:span|div)>\s*
  54             <a\s+href='(?P<http_url>[^']+)'>\s*
  55             (?:
  56                 .*?
  57                 <a\s+href='(?P<torrent_url>[^']+\.torrent)'
  58             )?''', webpage)
  59         formats = []
  60         for m in matches:
  61             format = m.group('format')
  62             format_id = self._search_regex(
  63                 r'.*/([a-z0-9_-]+)/[^/]*$',
  64                 m.group('http_url'), 'format id', default=None)
  65             vcodec = 'h264' if 'h264' in format_id else (
  66                 'none' if format_id in ('mp3', 'opus') else None
  67             )
  68             formats.append({
  69                 'format_id': format_id,
  70                 'format': format,
  71                 'url': m.group('http_url'),
  72                 'vcodec': vcodec,
  73                 'preference': preference(format_id),
  74             })
  75
  76             if m.group('torrent_url'):
  77                 formats.append({
  78                     'format_id': 'torrent-%s' % (format if format_id is None else format_id),
  79                     'format': '%s (torrent)' % format,
  80                     'proto': 'torrent',
  81                     'format_note': '(unsupported; will just download the .torrent file)',
  82                     'vcodec': vcodec,
  83                     'preference': -100 + preference(format_id),
  84                     'url': m.group('torrent_url'),
  85                 })
  86         self._sort_formats(formats)
  87
  88         thumbnail = self._html_search_regex(
  89             r"<video.*?poster='([^']+)'", webpage, 'thumbnail', fatal=False)
  90
  91         return {
  92             'id': video_id,
  93             'title': title,
  94             'description': description,
  95             'thumbnail': thumbnail,
  96             'view_count': view_count,
  97             'upload_date': upload_date,
  98             'formats': formats,
  99         }