]> jfr.im git - yt-dlp.git/blame - youtube_dl/extractor/iprima.py
[facebook] Fix for new handleServerJS syntax (closes #10846)
[yt-dlp.git] / youtube_dl / extractor / iprima.py
CommitLineData
369e7e3f 1# coding: utf-8
7881a644 2from __future__ import unicode_literals
3
4import re
f406c787 5import time
7881a644 6
7from .common import InfoExtractor
1cc79574 8from ..utils import (
369e7e3f
S
9 determine_ext,
10 js_to_json,
5c2266df 11 sanitized_Request,
82642235 12)
7881a644 13
14
15class IPrimaIE(InfoExtractor):
f406c787 16 _VALID_URL = r'https?://play\.iprima\.cz/(?:.+/)?(?P<id>[^?#]+)'
7881a644 17
18 _TESTS = [{
f406c787 19 'url': 'http://play.iprima.cz/gondici-s-r-o-33',
7881a644 20 'info_dict': {
f406c787 21 'id': 'p136534',
22 'ext': 'mp4',
23 'title': 'Gondíci s. r. o. (34)',
24 'description': 'md5:16577c629d006aa91f59ca8d8e7f99bd',
7881a644 25 },
26 'params': {
f406c787 27 'skip_download': True, # m3u8 download
7881a644 28 },
973f2532 29 }, {
f406c787 30 'url': 'http://play.iprima.cz/particka/particka-92',
bc03e585 31 'only_matching': True,
973f2532 32 }]
7881a644 33
34 def _real_extract(self, url):
369e7e3f 35 video_id = self._match_id(url)
7881a644 36
37 webpage = self._download_webpage(url, video_id)
38
f406c787 39 video_id = self._search_regex(r'data-product="([^"]+)">', webpage, 'real id')
82642235 40
70328330
JMF
41 req = sanitized_Request(
42 'http://play.iprima.cz/prehravac/init?_infuse=1'
59b35c67 43 '&_ts=%s&productId=%s' % (round(time.time()), video_id))
7881a644 44 req.add_header('Referer', url)
f406c787 45 playerpage = self._download_webpage(req, video_id, note='Downloading player')
7881a644 46
369e7e3f 47 formats = []
7881a644 48
369e7e3f
S
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)
91264ce5
PH
83
84 self._sort_formats(formats)
7881a644 85
86 return {
f406c787 87 'id': video_id,
88 'title': self._og_search_title(webpage),
7881a644 89 'thumbnail': self._og_search_thumbnail(webpage),
90 'formats': formats,
f406c787 91 'description': self._og_search_description(webpage),
91264ce5 92 }