[extractor/generic] Use `Accept-Encoding: identity` for initial request

The existing comment seems to imply this was the desired behavior from the beginning.

Partial fix for https://github.com/yt-dlp/yt-dlp/issues/5855, https://github.com/yt-dlp/yt-dlp/issues/5851, https://github.com/yt-dlp/yt-dlp/issues/4748
This commit is contained in:
coletdjnz 2023-01-01 18:40:26 +13:00
parent 8e40b9d1ec
commit 3e01ce744a
No known key found for this signature in database
GPG Key ID: 91984263BB39894A
1 changed files with 16 additions and 1 deletions

View File

@ -2154,6 +2154,21 @@ class GenericIE(InfoExtractor):
'age_limit': 0, 'age_limit': 0,
'direct': True, 'direct': True,
} }
}, {
'note': 'server returns data in brotli compression by default if `accept-encoding: *` is specified.',
'url': 'https://www.extra.cz/cauky-lidi-70-dil-babis-predstavil-pohadky-prymulanek-nebo-andrejovy-nove-saty-ac867',
'info_dict': {
'id': 'cauky-lidi-70-dil-babis-predstavil-pohadky-prymulanek-nebo-andrejovy-nove-saty-ac867',
'ext': 'mp4',
'title': 'čauky lidi 70 finall',
'description': 'čauky lidi 70 finall',
'thumbnail': 'h',
'upload_date': '20220606',
'timestamp': 1654513791,
'duration': 318.0,
'direct': True,
'age_limit': 0,
}
} }
] ]
@ -2312,7 +2327,7 @@ class GenericIE(InfoExtractor):
# It may probably better to solve this by checking Content-Type for application/octet-stream # It may probably better to solve this by checking Content-Type for application/octet-stream
# after a HEAD request, but not sure if we can rely on this. # after a HEAD request, but not sure if we can rely on this.
full_response = self._request_webpage(url, video_id, headers={ full_response = self._request_webpage(url, video_id, headers={
'Accept-Encoding': '*', 'Accept-Encoding': 'identity',
**smuggled_data.get('http_headers', {}) **smuggled_data.get('http_headers', {})
}) })
new_url = full_response.geturl() new_url = full_response.geturl()