bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#64977: 29.1; `auth-source-macos-keychain-search' handles :user key i


From: LdBeth
Subject: bug#64977: 29.1; `auth-source-macos-keychain-search' handles :user key incorrectly
Date: Sat, 12 Aug 2023 14:02:33 -0500
User-agent: Wanderlust/2.15.9 (Almost Unreal) SEMI-EPG/1.14.7 (Harue) FLIM-LB/1.14.9 (Gojō) APEL-LB/10.8 EasyPG/1.0.0 Emacs/29.1 (x86_64-apple-darwin20.6.0) MULE/6.0 (HANACHIRUSATO)

Sure. I have made the test. I also found a issue not covered in my
previous patch, that if the protocol is longer than 4 characters it
should be truncated. Please have a look at them.

ldb

Attachment: patch1
Description: Text document


Attachment: patch2
Description: Text document


>>>>> In <87r0o8z1sc.fsf@gmx.de> 
>>>>>   Michael Albinus <michael.albinus@gmx.de> wrote:
> LdBeth <andpuke@foxmail.com> writes:

> Hi,

>> I think it is possible to use cl-left to redefine
>> `auth-source-macos-keychain-search-items' to avoid calling external
>> security command and just check the argument passed to
>> `call-process'.
>>
>> Something like:
>>
>> (cl-letf (((symbol-function 'call-process) (lambda (&rest r) (print r))))
>>   (auth-source-search :user "ldb" :port "irc"))
>>
>> ("/usr/bin/security" nil t nil "find-internet-password" "-g" "-a" "ldb" "-r" 
>> "irc ")
>>
>> ("/usr/bin/security" nil t nil "find-generic-password" "-g" "-a" "ldb" "-s" 
>> "irc")

> Yes, like this. And in the lambda function, you could check the expected
> arguments of the "/usr/bin/security" call by `should' and friends.

> Would you like to add such test(s)? This would give us more confidence
> that nothing is or will be broken, because people using macOS would test
> this by default when running "make check", even if they don't care the
> macOS keychain.

>> ldb

> Best regards, Michael.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]