paradizelost/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[malltv] fix extraction(closes #27035)

Browse Source
This commit is contained in:
Remita Amine 2020-11-17 19:13:38 +01:00
parent cb2b9a22a5
commit aa613ef7e1
1 changed files with 46 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

60
youtube_dl/extractor/malltv.py
View File

@ -1,10 +1,16 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import merge_dicts from ..utils import (
clean_html,
dict_get,
float_or_none,
int_or_none,
merge_dicts,
parse_duration,
try_get,
)
class MallTVIE(InfoExtractor): class MallTVIE(InfoExtractor):
@ -17,7 +23,7 @@ class MallTVIE(InfoExtractor):
'display_id': '18-miliard-pro-neziskovky-opravdu-jsou-sportovci-nebo-clovek-v-tisni-pijavice', 'display_id': '18-miliard-pro-neziskovky-opravdu-jsou-sportovci-nebo-clovek-v-tisni-pijavice',
'ext': 'mp4', 'ext': 'mp4',
'title': '18 miliard pro neziskovky. Opravdu jsou sportovci nebo Člověk v tísni pijavice?', 'title': '18 miliard pro neziskovky. Opravdu jsou sportovci nebo Člověk v tísni pijavice?',
'description': 'md5:25fc0ec42a72ba602b602c683fa29deb', 'description': 'md5:db7d5744a4bd4043d9d98324aa72ab35',
'duration': 216, 'duration': 216,
'timestamp': 1538870400, 'timestamp': 1538870400,
'upload_date': '20181007', 'upload_date': '20181007',
@ -37,20 +43,46 @@ class MallTVIE(InfoExtractor):
webpage = self._download_webpage( webpage = self._download_webpage(
url, display_id, headers=self.geo_verification_headers()) url, display_id, headers=self.geo_verification_headers())
SOURCE_RE = r'(<source[^>]+\bsrc=(?:(["\'])(?:(?!\2).)+|[^\s]+)/(?P<id>[\da-z]+)/index)\b' video = self._parse_json(self._search_regex(
r'videoObject\s*=\s*JSON\.parse\(JSON\.stringify\(({.+?})\)\);',
webpage, 'video object'), display_id)
video_source = video['VideoSource']
video_id = self._search_regex( video_id = self._search_regex(
SOURCE_RE, webpage, 'video id', group='id') r'/([\da-z]+)/index\b', video_source, 'video id')
media = self._parse_html5_media_entries( formats = self._extract_m3u8_formats(
url, re.sub(SOURCE_RE, r'\1.m3u8', webpage), video_id, video_source + '.m3u8', video_id, 'mp4', 'm3u8_native')
m3u8_id='hls', m3u8_entry_protocol='m3u8_native')[0] self._sort_formats(formats)
subtitles = {}
for s in (video.get('Subtitles') or {}):
s_url = s.get('Url')
if not s_url:
continue
subtitles.setdefault(s.get('Language') or 'cz', []).append({
'url': s_url,
})
entity_counts = video.get('EntityCounts') or {}
def get_count(k):
v = entity_counts.get(k + 's') or {}
return int_or_none(dict_get(v, ('Count', 'StrCount')))
info = self._search_json_ld(webpage, video_id, default={}) info = self._search_json_ld(webpage, video_id, default={})
return merge_dicts(media, info, { return merge_dicts({
'id': video_id, 'id': video_id,
'display_id': display_id, 'display_id': display_id,
'title': self._og_search_title(webpage, default=None) or display_id, 'title': video.get('Title'),
'description': self._og_search_description(webpage, default=None), 'description': clean_html(video.get('Description')),
'thumbnail': self._og_search_thumbnail(webpage, default=None), 'thumbnail': video.get('ThumbnailUrl'),
}) 'formats': formats,
'subtitles': subtitles,
'duration': int_or_none(video.get('DurationSeconds')) or parse_duration(video.get('Duration')),
'view_count': get_count('View'),
'like_count': get_count('Like'),
'dislike_count': get_count('Dislike'),
'average_rating': float_or_none(try_get(video, lambda x: x['EntityRating']['AvarageRate'])),
'comment_count': get_count('Comment'),
}, info)