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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     determine_ext,
  10     js_to_json,
  11     sanitized_Request,
  12 )
  13
  14
  15 class IPrimaIE(InfoExtractor):
  16     _VALID_URL = r'https?://play\.iprima\.cz/(?:.+/)?(?P<id>[^?#]+)'
  17
  18     _TESTS = [{
  19         'url': 'http://play.iprima.cz/gondici-s-r-o-33',
  20         'info_dict': {
  21             'id': 'p136534',
  22             'ext': 'mp4',
  23             'title': 'Gondíci s. r. o. (34)',
  24             'description': 'md5:16577c629d006aa91f59ca8d8e7f99bd',
  25         },
  26         'params': {
  27             'skip_download': True,  # m3u8 download
  28         },
  29     }, {
  30         'url': 'http://play.iprima.cz/particka/particka-92',
  31         'only_matching': True,
  32     }]
  33
  34     def _real_extract(self, url):
  35         video_id = self._match_id(url)
  36
  37         webpage = self._download_webpage(url, video_id)
  38
  39         video_id = self._search_regex(r'data-product="([^"]+)">', webpage, 'real id')
  40
  41         req = sanitized_Request(
  42             'http://play.iprima.cz/prehravac/init?_infuse=1'
  43             '&_ts=%s&productId=%s' % (round(time.time()), video_id))
  44         req.add_header('Referer', url)
  45         playerpage = self._download_webpage(req, video_id, note='Downloading player')
  46
  47         formats = []
  48
  49         def extract_formats(format_url, format_key=None, lang=None):
  50             ext = determine_ext(format_url)
  51             new_formats = []
  52             if format_key == 'hls' or ext == 'm3u8':
  53                 new_formats = self._extract_m3u8_formats(
  54                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  55                     m3u8_id='hls', fatal=False)
  56             elif format_key == 'dash' or ext == 'mpd':
  57                 return
  58                 new_formats = self._extract_mpd_formats(
  59                     format_url, video_id, mpd_id='dash', fatal=False)
  60             if lang:
  61                 for f in new_formats:
  62                     if not f.get('language'):
  63                         f['language'] = lang
  64             formats.extend(new_formats)
  65
  66         options = self._parse_json(
  67             self._search_regex(
  68                 r'(?s)var\s+playerOptions\s*=\s*({.+?});',
  69                 playerpage, 'player options', default='{}'),
  70             video_id, transform_source=js_to_json, fatal=False)
  71         if options:
  72             for key, tracks in options.get('tracks', {}).items():
  73                 if not isinstance(tracks, list):
  74                     continue
  75                 for track in tracks:
  76                     src = track.get('src')
  77                     if src:
  78                         extract_formats(src, key.lower(), track.get('lang'))
  79
  80         if not formats:
  81             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
  82                 extract_formats(src)
  83
  84         self._sort_formats(formats)
  85
  86         return {
  87             'id': video_id,
  88             'title': self._og_search_title(webpage),
  89             'thumbnail': self._og_search_thumbnail(webpage),
  90             'formats': formats,
  91             'description': self._og_search_description(webpage),
  92         }