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 )
  12
  13
  14 class IPrimaIE(InfoExtractor):
  15     _VALID_URL = r'https?://play\.iprima\.cz/(?:.+/)?(?P<id>[^?#]+)'
  16     _GEO_BYPASS = False
  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         # geo restricted
  34         'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
  35         'only_matching': True,
  36     }]
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40
  41         webpage = self._download_webpage(url, video_id)
  42
  43         video_id = self._search_regex(r'data-product="([^"]+)">', webpage, 'real id')
  44
  45         playerpage = self._download_webpage(
  46             'http://play.iprima.cz/prehravac/init',
  47             video_id, note='Downloading player', query={
  48                 '_infuse': 1,
  49                 '_ts': round(time.time()),
  50                 'productId': video_id,
  51             }, headers={'Referer': url})
  52
  53         formats = []
  54
  55         def extract_formats(format_url, format_key=None, lang=None):
  56             ext = determine_ext(format_url)
  57             new_formats = []
  58             if format_key == 'hls' or ext == 'm3u8':
  59                 new_formats = self._extract_m3u8_formats(
  60                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  61                     m3u8_id='hls', fatal=False)
  62             elif format_key == 'dash' or ext == 'mpd':
  63                 return
  64                 new_formats = self._extract_mpd_formats(
  65                     format_url, video_id, mpd_id='dash', fatal=False)
  66             if lang:
  67                 for f in new_formats:
  68                     if not f.get('language'):
  69                         f['language'] = lang
  70             formats.extend(new_formats)
  71
  72         options = self._parse_json(
  73             self._search_regex(
  74                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
  75                 playerpage, 'player options', default='{}'),
  76             video_id, transform_source=js_to_json, fatal=False)
  77         if options:
  78             for key, tracks in options.get('tracks', {}).items():
  79                 if not isinstance(tracks, list):
  80                     continue
  81                 for track in tracks:
  82                     src = track.get('src')
  83                     if src:
  84                         extract_formats(src, key.lower(), track.get('lang'))
  85
  86         if not formats:
  87             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
  88                 extract_formats(src)
  89
  90         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
  91             self.raise_geo_restricted(countries=['CZ'])
  92
  93         self._sort_formats(formats)
  94
  95         return {
  96             'id': video_id,
  97             'title': self._og_search_title(webpage),
  98             'thumbnail': self._og_search_thumbnail(webpage),
  99             'formats': formats,
 100             'description': self._og_search_description(webpage),
 101         }