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|prima)\.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         # iframe api.play-backend.iprima.cz
  38         'url': 'https://prima.iprima.cz/my-little-pony/mapa-znameni-2-2',
  39         'only_matching': True,
  40     }, {
  41         # iframe prima.iprima.cz
  42         'url': 'https://prima.iprima.cz/porady/jak-se-stavi-sen/rodina-rathousova-praha',
  43         'only_matching': True,
  44     }]
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48
  49         self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
  50
  51         webpage = self._download_webpage(url, video_id)
  52
  53         video_id = self._search_regex(
  54             (r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
  55              r'data-product="([^"]+)">'),
  56             webpage, 'real id')
  57
  58         playerpage = self._download_webpage(
  59             'http://play.iprima.cz/prehravac/init',
  60             video_id, note='Downloading player', query={
  61                 '_infuse': 1,
  62                 '_ts': round(time.time()),
  63                 'productId': video_id,
  64             }, headers={'Referer': url})
  65
  66         formats = []
  67
  68         def extract_formats(format_url, format_key=None, lang=None):
  69             ext = determine_ext(format_url)
  70             new_formats = []
  71             if format_key == 'hls' or ext == 'm3u8':
  72                 new_formats = self._extract_m3u8_formats(
  73                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  74                     m3u8_id='hls', fatal=False)
  75             elif format_key == 'dash' or ext == 'mpd':
  76                 return
  77                 new_formats = self._extract_mpd_formats(
  78                     format_url, video_id, mpd_id='dash', fatal=False)
  79             if lang:
  80                 for f in new_formats:
  81                     if not f.get('language'):
  82                         f['language'] = lang
  83             formats.extend(new_formats)
  84
  85         options = self._parse_json(
  86             self._search_regex(
  87                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
  88                 playerpage, 'player options', default='{}'),
  89             video_id, transform_source=js_to_json, fatal=False)
  90         if options:
  91             for key, tracks in options.get('tracks', {}).items():
  92                 if not isinstance(tracks, list):
  93                     continue
  94                 for track in tracks:
  95                     src = track.get('src')
  96                     if src:
  97                         extract_formats(src, key.lower(), track.get('lang'))
  98
  99         if not formats:
 100             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
 101                 extract_formats(src)
 102
 103         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
 104             self.raise_geo_restricted(countries=['CZ'])
 105
 106         self._sort_formats(formats)
 107
 108         return {
 109             'id': video_id,
 110             'title': self._og_search_title(webpage),
 111             'thumbnail': self._og_search_thumbnail(webpage),
 112             'formats': formats,
 113             'description': self._og_search_description(webpage),
 114         }