Fix stubtest errors (#5063)

A security fix added a "separator" argument to several URL parsing
functions and method in point releases:

* 3.6.13
* 3.7.10
* 3.8.8
* 3.9.2

Until all these versions are available on the GitHub Actions runners,
we need to whitelist the functions in the stubtests.
This commit is contained in:
Sebastian Rittau
2021-02-24 14:37:38 +01:00
committed by GitHub
parent c21329b68a
commit c152e050f5
3 changed files with 34 additions and 27 deletions

View File

@@ -76,6 +76,13 @@ bz2.BZ2File.readinto
bz2.BZ2File.readlines
bz2.BZ2File.write
bz2.BZ2File.writelines
# The following methods were changed in point releases from Python 3.6 to 3.9
# as part of a security fix. These excludes can be removed when the GitHub
# action workflow uses Python versions that include the fix (adding a
# separator argument).
cgi.FieldStorage.__init__
cgi.parse
cgi.parse_multipart
codecs.BufferedIncrementalDecoder.decode
codecs.CodecInfo.decode
codecs.CodecInfo.encode
@@ -406,6 +413,12 @@ unittest.mock.patch # It's a complicated overload and I haven't been able to fi
urllib.error.ContentTooShortError.__init__
urllib.error.URLError.__init__
urllib.parse._DefragResultBase.__new__
# The following methods were changed in point releases from Python 3.6 to 3.9
# as part of a security fix. These excludes can be removed when the GitHub
# action workflow uses Python versions that include the fix (adding a
# separator argument).
urllib.parse.parse_qs
urllib.parse.parse_qsl
urllib.request.BaseHandler.http_error_nnn
urllib.request.HTTPPasswordMgrWithPriorAuth.__init__
urllib.robotparser.RobotFileParser.can_fetch