youtube_dl/extractor/cda.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     decode_packed_codes,
   9     ExtractorError,
  10     parse_duration
  11 )
  12
  13
  14 class CDAIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:(?:www|ebd)\.)?cda\.pl/(?:video|[0-9]+x[0-9]+)/(?P<id>[0-9a-z]+)'
  16     _TESTS = [
  17         {
  18             'url': 'http://www.cda.pl/video/5749950c',
  19             'md5': '6f844bf51b15f31fae165365707ae970',
  20             'info_dict': {
  21                 'id': '5749950c',
  22                 'ext': 'mp4',
  23                 'height': 720,
  24                 'title': 'Oto dlaczego przed zakrętem należy zwolnić.',
  25                 'duration': 39
  26             }
  27         },
  28         {
  29             'url': 'http://www.cda.pl/video/57413289',
  30             'md5': 'a88828770a8310fc00be6c95faf7f4d5',
  31             'info_dict': {
  32                 'id': '57413289',
  33                 'ext': 'mp4',
  34                 'title': 'Lądowanie na lotnisku na Maderze',
  35                 'duration': 137
  36             }
  37         }
  38     ]
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42         webpage = self._download_webpage('http://ebd.cda.pl/0x0/' + video_id, video_id)
  43
  44         if 'Ten film jest dostępny dla użytkowników premium' in webpage:
  45             raise ExtractorError('This video is only available for premium users.', expected=True)
  46
  47         title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title', fatal=False)
  48
  49         def _get_format(page, version=''):
  50             unpacked = decode_packed_codes(page)
  51             duration = self._search_regex(r"duration:\\'(.+?)\\'", unpacked, 'duration', fatal=False)
  52             format_id = None
  53             height = None
  54
  55             m = re.search(r'<a data-quality="(?P<format_id>[^"]+)" href="[^"]+" class="quality-btn quality-btn-active">(?P<height>[0-9]+)p<\/a>', page)
  56             if m:
  57                 format_id = m.group('format_id')
  58                 height = int(m.group('height'))
  59
  60             url = self._search_regex(r"url:\\'(.+?)\\'", unpacked, version + ' url', fatal=False)
  61             if url is None:
  62                 return None
  63
  64             return {
  65                 'format_id': format_id,
  66                 'height': height,
  67                 'url': url
  68             }, parse_duration(duration)
  69
  70         formats = []
  71
  72         format_desc, duration = _get_format(webpage) or (None, None)
  73         if format_desc is not None:
  74             formats.append(format_desc)
  75
  76         pattern = re.compile(r'<a data-quality="[^"]+" href="([^"]+)" class="quality-btn">([0-9]+p)<\/a>')
  77         for version in re.findall(pattern, webpage):
  78             webpage = self._download_webpage(version[0], video_id, 'Downloading %s version information' % version[1], fatal=False)
  79             if not webpage:
  80                 # Manually report warning because empty page is returned when invalid version is requested.
  81                 self.report_warning('Unable to download %s version information' % version[1])
  82                 continue
  83
  84             format_desc, duration_ = _get_format(webpage, version[1]) or (None, None)
  85             duration = duration or duration_
  86             if format_desc is not None:
  87                 formats.append(format_desc)
  88
  89         self._sort_formats(formats)
  90
  91         return {
  92             'id': video_id,
  93             'title': title,
  94             'formats': formats,
  95             'duration': duration
  96         }