[streamcz] Fix extractor (#1616)

Closes #1329, closes #1731
Authored by: arkamar, pukkandan
This commit is contained in:
Petr Vaněk 2022-01-11 17:56:18 +01:00 committed by GitHub
parent 80fa6e5327
commit 7d449fff53
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23

View file

@ -1,105 +1,108 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals import json
import hashlib
import time
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
float_or_none,
int_or_none, int_or_none,
sanitized_Request, parse_codecs,
traverse_obj,
urljoin,
) )
def _get_api_key(api_path):
if api_path.endswith('?'):
api_path = api_path[:-1]
api_key = 'fb5f58a820353bd7095de526253c14fd'
a = '{0:}{1:}{2:}'.format(api_key, api_path, int(round(time.time() / 24 / 3600)))
return hashlib.md5(a.encode('ascii')).hexdigest()
class StreamCZIE(InfoExtractor): class StreamCZIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?stream\.cz/.+/(?P<id>[0-9]+)' _VALID_URL = r'https?://(?:www\.)?(?:stream|televizeseznam)\.cz/[^?#]+/(?P<display_id>[^?#]+)-(?P<id>[0-9]+)'
_API_URL = 'http://www.stream.cz/API'
_TESTS = [{ _TESTS = [{
'url': 'http://www.stream.cz/peklonataliri/765767-ecka-pro-deti', 'url': 'https://www.televizeseznam.cz/video/lajna/buh-57953890',
'md5': '934bb6a6d220d99c010783c9719960d5', 'md5': '40c41ade1464a390a0b447e333df4239',
'info_dict': { 'info_dict': {
'id': '765767', 'id': '57953890',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Peklo na talíři: Éčka pro děti', 'title': 'Bůh',
'description': 'Taška s grónskou pomazánkou a další pekelnosti ZDE', 'display_id': 'buh',
'thumbnail': 're:^http://im.stream.cz/episode/52961d7e19d423f8f06f0100', 'description': 'md5:8f5f09b9b7bc67df910486cdd88f7165',
'duration': 256, }
},
}, { }, {
'url': 'http://www.stream.cz/blanik/10002447-tri-roky-pro-mazanka', 'url': 'https://www.stream.cz/tajemno/znicehonic-jim-skrz-strechu-prolitnul-zahadny-predmet-badatele-vse-objasnili-64147267',
'md5': '849a88c1e1ca47d41403c2ba5e59e261', 'md5': '3ee4d0be040e8f4a543e67e509d55e3f',
'info_dict': { 'info_dict': {
'id': '10002447', 'id': '64147267',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Kancelář Blaník: Tři roky pro Mazánka', 'title': 'Zničehonic jim skrz střechu prolítnul záhadný předmět. Badatelé vše objasnili',
'description': 'md5:3862a00ba7bf0b3e44806b544032c859', 'display_id': 'znicehonic-jim-skrz-strechu-prolitnul-zahadny-predmet-badatele-vse-objasnili',
'thumbnail': 're:^http://im.stream.cz/episode/537f838c50c11f8d21320000', 'description': 'md5:1dcb5e010eb697dedc5942f76c5b3744',
'duration': 368, }
},
}] }]
def _extract_formats(self, spl_url, video):
for ext, pref, streams in (
('ts', -1, traverse_obj(video, ('http_stream', 'qualities'))),
('mp4', 1, video.get('mp4'))):
for format_id, stream in streams.items():
if not stream.get('url'):
continue
yield {
'format_id': f'{format_id}-{ext}',
'ext': ext,
'source_preference': pref,
'url': urljoin(spl_url, stream['url']),
'tbr': float_or_none(stream.get('bandwidth'), scale=1000),
'duration': float_or_none(stream.get('duration'), scale=1000),
'width': traverse_obj(stream, ('resolution', 0)),
'height': traverse_obj(stream, ('resolution', 1)) or int_or_none(format_id.replace('p', '')),
**parse_codecs(stream.get('codec')),
}
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) display_id, video_id = self._match_valid_url(url).groups()
api_path = '/episode/%s' % video_id
req = sanitized_Request(self._API_URL + api_path) data = self._download_json(
req.add_header('Api-Password', _get_api_key(api_path)) 'https://www.televizeseznam.cz/api/graphql', video_id, 'Downloading GraphQL result',
data = self._download_json(req, video_id) data=json.dumps({
'variables': {'urlName': video_id},
'query': '''
query LoadEpisode($urlName : String){ episode(urlName: $urlName){ ...VideoDetailFragmentOnEpisode } }
fragment VideoDetailFragmentOnEpisode on Episode {
id
spl
urlName
name
perex
duration
views
}'''
}).encode('utf-8'),
headers={'Content-Type': 'application/json;charset=UTF-8'}
)['data']['episode']
formats = [] spl_url = data['spl'] + 'spl2,3'
for quality, video in enumerate(data['video_qualities']): metadata = self._download_json(spl_url, video_id, 'Downloading playlist')
for f in video['formats']: if 'Location' in metadata and 'data' not in metadata:
typ = f['type'].partition('/')[2] spl_url = metadata['Location']
qlabel = video.get('quality_label') metadata = self._download_json(spl_url, video_id, 'Downloading redirected playlist')
formats.append({ video = metadata['data']
'format_note': '%s-%s' % (qlabel, typ) if qlabel else typ,
'format_id': '%s-%s' % (typ, f['quality']),
'url': f['source'],
'height': int_or_none(f['quality'].rstrip('p')),
'quality': quality,
})
self._sort_formats(formats)
image = data.get('image')
if image:
thumbnail = self._proto_relative_url(
image.replace('{width}', '1240').replace('{height}', '697'),
scheme='http:',
)
else:
thumbnail = None
stream = data.get('_embedded', {}).get('stream:show', {}).get('name')
if stream:
title = '%s: %s' % (stream, data['name'])
else:
title = data['name']
subtitles = {} subtitles = {}
srt_url = data.get('subtitles_srt') for subs in video.get('subtitles', {}).values():
if srt_url: if not subs.get('language'):
subtitles['cs'] = [{ continue
'ext': 'srt', for ext, sub_url in subs.get('urls').items():
'url': srt_url, subtitles.setdefault(subs['language'], []).append({
}] 'ext': ext,
'url': urljoin(spl_url, sub_url)
})
formats = list(self._extract_formats(spl_url, video))
self._sort_formats(formats)
return { return {
'id': video_id, 'id': video_id,
'title': title, 'display_id': display_id,
'thumbnail': thumbnail, 'title': data.get('name'),
'formats': formats, 'description': data.get('perex'),
'description': data.get('web_site_text'), 'duration': float_or_none(data.get('duration')),
'duration': int_or_none(data.get('duration')),
'view_count': int_or_none(data.get('views')), 'view_count': int_or_none(data.get('views')),
'formats': formats,
'subtitles': subtitles, 'subtitles': subtitles,
} }