nhyatt/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp synced 2025-04-05 06:30:17 -05:00
Code Issues Projects Releases Wiki Activity

[ie/mitele] Fix extractor (#12689)

Browse Source 
Closes #12655
Authored by: bashonly
This commit is contained in:
bashonly 2025-03-21 18:14:46 -05:00 committed by GitHub
parent f5fb2229e6
commit 7223d29569
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 14 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
yt_dlp/extractor/mitele.py
View File

@ -1,5 +1,7 @@
from .telecinco import TelecincoBaseIE from .telecinco import TelecincoBaseIE
from ..networking.exceptions import HTTPError
from ..utils import ( from ..utils import (
ExtractorError,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
) )
@ -79,7 +81,17 @@ class MiTeleIE(TelecincoBaseIE):
def _real_extract(self, url): def _real_extract(self, url):
display_id = self._match_id(url) display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
try: # yt-dlp's default user-agents are too old and blocked by akamai
webpage = self._download_webpage(url, display_id, headers={
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; rv:136.0) Gecko/20100101 Firefox/136.0',
})
except ExtractorError as e:
if not isinstance(e.cause, HTTPError) or e.cause.status != 403:
raise
# Retry with impersonation if hardcoded UA is insufficient to bypass akamai
webpage = self._download_webpage(url, display_id, impersonate=True)
pre_player = self._search_json( pre_player = self._search_json(
r'window\.\$REACTBASE_STATE\.prePlayer_mtweb\s*=', r'window\.\$REACTBASE_STATE\.prePlayer_mtweb\s*=',
webpage, 'Pre Player', display_id)['prePlayer'] webpage, 'Pre Player', display_id)['prePlayer']

2
yt_dlp/extractor/telecinco.py
View File

@ -46,7 +46,7 @@ class TelecincoBaseIE(InfoExtractor):
error_code = traverse_obj( error_code = traverse_obj(
self._webpage_read_content(error.cause.response, caronte['cerbero'], video_id, fatal=False), self._webpage_read_content(error.cause.response, caronte['cerbero'], video_id, fatal=False),
({json.loads}, 'code', {int})) ({json.loads}, 'code', {int}))
if error_code == 4038: if error_code in (4038, 40313):
self.raise_geo_restricted(countries=['ES']) self.raise_geo_restricted(countries=['ES'])
raise raise