github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-21 20:46:36 -05:00
Code Issues Projects Releases Packages Wiki Activity

[youtube:api] Do not use seek when reading HTTPError response

Browse source 
Authored-by: coletdjnz
This commit is contained in:
coletdjnz 2022-01-23 19:11:32 +13:00
parent e600a5c908
commit 87e8e8a7d0
No known key found for this signature in database
GPG key ID: 91984263BB39894A
1 changed files with 9 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
yt_dlp/extractor/youtube.py
View file

@ -760,13 +760,15 @@ def _extract_response(self, item_id, query, note='Downloading API JSON', headers
note='%s%s' % (note, ' (retry #%d)' % count if count else '')) note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
except ExtractorError as e: except ExtractorError as e:
if isinstance(e.cause, network_exceptions): if isinstance(e.cause, network_exceptions):
if isinstance(e.cause, compat_HTTPError) and not is_html(e.cause.read(512)): if isinstance(e.cause, compat_HTTPError):
e.cause.seek(0) first_bytes = e.cause.read(512)
yt_error = try_get( if not is_html(first_bytes):
self._parse_json(e.cause.read().decode(), item_id, fatal=False), yt_error = try_get(
lambda x: x['error']['message'], compat_str) self._parse_json(
if yt_error: self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
self._report_alerts([('ERROR', yt_error)], fatal=False) lambda x: x['error']['message'], compat_str)
if yt_error:
self._report_alerts([('ERROR', yt_error)], fatal=False)
# Downloading page may result in intermittent 5xx HTTP error # Downloading page may result in intermittent 5xx HTTP error
# Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
# We also want to catch all other network exceptions since errors in later pages can be troublesome # We also want to catch all other network exceptions since errors in later pages can be troublesome