xZero707/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-27 06:10:12 +01:00
Code Issues Packages Projects Releases Wiki Activity

[iprima] Improve extraction (closes #25138)

Browse Source
This commit is contained in:
Sergey M․ 2020-05-06 23:20:14 +07:00
parent 6c907eb33f
commit 30fa5c6087
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 24 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
youtube_dl/extractor/iprima.py
View File

@ -16,12 +16,22 @@ class IPrimaIE(InfoExtractor):
_GEO_BYPASS = False
_TESTS = [{
'url': 'http://play.iprima.cz/gondici-s-r-o-33',
'url': 'https://prima.iprima.cz/particka/92-epizoda',
'info_dict': {
'id': 'p136534',
'id': 'p51388',
'ext': 'mp4',
'title': 'Gondíci s. r. o. (34)',
'description': 'md5:16577c629d006aa91f59ca8d8e7f99bd',
'title': 'Partička (92)',
'description': 'md5:859d53beae4609e6dd7796413f1b6cac',
},
'params': {
'skip_download': True, # m3u8 download
},
}, {
'url': 'https://cnn.iprima.cz/videa/70-epizoda',
'info_dict': {
'id': 'p681554',
'ext': 'mp4',
'title': 'HLAVNÍ ZPRÁVY 3.5.2020',
},
'params': {
'skip_download': True, # m3u8 download
@ -68,9 +78,15 @@ def _real_extract(self, url):
webpage = self._download_webpage(url, video_id)
title = self._og_search_title(
webpage, default=None) or self._search_regex(
r'<h1>([^<]+)', webpage, 'title')
video_id = self._search_regex(
(r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
r'data-product="([^"]+)">'),
r'data-product="([^"]+)">',
r'id=["\']player-(p\d+)"',
r'playerId\s*:\s*["\']player-(p\d+)'),
webpage, 'real id')
playerpage = self._download_webpage(
@ -125,8 +141,8 @@ def extract_formats(format_url, format_key=None, lang=None):
return {
'id': video_id,
'title': self._og_search_title(webpage),
'thumbnail': self._og_search_thumbnail(webpage),
'title': title,
'thumbnail': self._og_search_thumbnail(webpage, default=None),
'formats': formats,
'description': self._og_search_description(webpage),
'description': self._og_search_description(webpage, default=None),
}