Re: [PATCH] A hook for session start

From: Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
To: Fabrízio Mello <fabriziomello(at)gmail(dot)com>
Cc: Aleksandr Parfenov <a(dot)parfenov(at)postgrespro(dot)ru>, Pgsql Hackers <pgsql-hackers(at)postgresql(dot)org>, Yugo Nagata <nagata(at)sraoss(dot)co(dot)jp>
Subject: Re: [PATCH] A hook for session start
Date: 2017-11-09 02:09:20
Message-ID: CAB7nPqQKXzL2gtDso2a+VgdGFTRoA5gLxVz2RaCLZfzV7NhWTw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Thu, Nov 9, 2017 at 2:42 AM, Fabrízio de Royes Mello
<fabriziomello(at)gmail(dot)com> wrote:
> On Wed, Nov 8, 2017 at 12:47 AM, Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
> wrote:
>> - Let's restrict the logging to a role name instead of a database
>> name, and let's parametrize it with a setting in the temporary
>> configuration file. Let's not bother about multiple role support with
>> a list, for the sake of tests and simplicity only defining one role
>> looks fine to me. Comments in the code should be clear about the
>> dependency.
>
> Makes sense and simplify the test code. Fixed.

+ if (!strcmp(username, "regress_sess_hook_usr2"))
+ {
+ const char *dbname;
[...]
+++ b/src/test/modules/test_session_hooks/session_hooks.conf
@@ -0,0 +1 @@
+shared_preload_libraries = 'test_session_hooks'
Don't you think that this should be a GUC? My previous comment
outlined that. I won't fight hard on that point in any case, don't
worry. I just want to make things clear :)
--
Michael

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Paquier 2017-11-09 02:12:49 Re: [PATCH] Assert that the correct locks are held when calling PageGetLSN()
Previous Message Michael Paquier 2017-11-09 01:33:19 Re: Simplify ACL handling for large objects and removal of superuser() checks