Subject: Re: Proposal: Save user's original authenticated identity for logging
To: Peter Eisentraut <peter.eisentraut@enterprisedb.com>,
 Michael Paquier <michael@paquier.xyz>, Jacob Champion <pchampion@vmware.com>
Cc: "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
References: <a4fddbcd2f99c293991b7669bd0da3e5a7d3f93b.camel@vmware.com>
 <YGQk74rdCdb8F9gZ@paquier.xyz> <YGUgHMF9zxNvY323@paquier.xyz>
 <d17b919e27474abfa55d97786cb9cfadfe2b59e9.camel@vmware.com>
 <YGaha/euiRjZ820X@paquier.xyz> <YGhf4TN6ukOZTUcd@paquier.xyz>
 <ae1143fb-455c-c80f-ed66-78d45bd93303@enterprisedb.com>
 <YUrdnoEc4NPfBZfI@paquier.xyz>
 <2dba8b35-c607-ae30-07d2-610886aff897@enterprisedb.com>
 <814ec049ab4085d9efd3a7a654337ac5008cdaf1.camel@vmware.com>
 <YUxYPBLOLrXWulDh@paquier.xyz>
 <846a497e-e020-76c9-b268-103332872d87@enterprisedb.com>
From: Andrew Dunstan <andrew@dunslane.net>
Message-ID: <aa4e1273-7b29-cdbf-c9b4-e36b44dc78e9@dunslane.net>
Date: Fri, 24 Sep 2021 17:37:48 -0400
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.4.0
MIME-Version: 1.0
In-Reply-To: <846a497e-e020-76c9-b268-103332872d87@enterprisedb.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Archived-At: 
 <https://www.postgresql.org/message-id/aa4e1273-7b29-cdbf-c9b4-e36b44dc78e9%40dunslane.net>
Precedence: bulk


On 9/23/21 5:20 PM, Peter Eisentraut wrote:
> On 23.09.21 12:34, Michael Paquier wrote:
>> On Wed, Sep 22, 2021 at 03:18:43PM +0000, Jacob Champion wrote:
>>> On Wed, 2021-09-22 at 10:20 +0200, Peter Eisentraut wrote:
>>>> This should be added to each level of a function call that
>>>> represents a
>>>> test.  This ensures that when a test fails, the line number points to
>>>> the top-level location of the test_role() call.  Otherwise it would
>>>> point to the connect_ok() call inside test_role().
>>>
>>> Patch LGTM, sorry about that. Thanks!
>>
>> For the places of the patch, that seems fine then.  Thanks!
>
> committed
>
>> Do we need to care about that in other places?  We have tests in
>> src/bin/ using subroutines that call things from PostgresNode.pm or
>> TestLib.pm, like pg_checksums, pg_ctl or pg_verifybackup, just to name
>> three.
>
> Yeah, at first glance, there is probably more that could be done. 
> Here, I was just looking at a place where it was already and was
> accidentally removed.


It probably wouldn't be a bad thing to have something somewhere
(src/test/perl/README ?) that explains when and why we need to bump
$Test::Builder::Level.


cheers


andrew

--
Andrew Dunstan
EDB: https://www.enterprisedb.com