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
  17     _TESTS = [{
  18         'url': 'http://play.iprima.cz/gondici-s-r-o-33',
  19         'info_dict': {
  20             'id': 'p136534',
  21             'ext': 'mp4',
  22             'title': 'Gondíci s. r. o. (34)',
  23             'description': 'md5:16577c629d006aa91f59ca8d8e7f99bd',
  24         },
  25         'params': {
  26             'skip_download': True,  # m3u8 download
  27         },
  28     }, {
  29         'url': 'http://play.iprima.cz/particka/particka-92',
  30         'only_matching': True,
  31     }]
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35
  36         webpage = self._download_webpage(url, video_id)
  37
  38         video_id = self._search_regex(r'data-product="([^"]+)">', webpage, 'real id')
  39
  40         playerpage = self._download_webpage(
  41             'http://play.iprima.cz/prehravac/init',
  42             video_id, note='Downloading player', query={
  43                 '_infuse': 1,
  44                 '_ts': round(time.time()),
  45                 'productId': video_id,
  46             }, headers={'Referer': url})
  47
  48         formats = []
  49
  50         def extract_formats(format_url, format_key=None, lang=None):
  51             ext = determine_ext(format_url)
  52             new_formats = []
  53             if format_key == 'hls' or ext == 'm3u8':
  54                 new_formats = self._extract_m3u8_formats(
  55                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  56                     m3u8_id='hls', fatal=False)
  57             elif format_key == 'dash' or ext == 'mpd':
  58                 return
  59                 new_formats = self._extract_mpd_formats(
  60                     format_url, video_id, mpd_id='dash', fatal=False)
  61             if lang:
  62                 for f in new_formats:
  63                     if not f.get('language'):
  64                         f['language'] = lang
  65             formats.extend(new_formats)
  66
  67         options = self._parse_json(
  68             self._search_regex(
  69                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
  70                 playerpage, 'player options', default='{}'),
  71             video_id, transform_source=js_to_json, fatal=False)
  72         if options:
  73             for key, tracks in options.get('tracks', {}).items():
  74                 if not isinstance(tracks, list):
  75                     continue
  76                 for track in tracks:
  77                     src = track.get('src')
  78                     if src:
  79                         extract_formats(src, key.lower(), track.get('lang'))
  80
  81         if not formats:
  82             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
  83                 extract_formats(src)
  84
  85         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
  86             self.raise_geo_restricted()
  87
  88         self._sort_formats(formats)
  89
  90         return {
  91             'id': video_id,
  92             'title': self._og_search_title(webpage),
  93             'thumbnail': self._og_search_thumbnail(webpage),
  94             'formats': formats,
  95             'description': self._og_search_description(webpage),
  96         }