Re: small patch to crypt.c - Mailing list pgsql-hackers

From Stephen Frost
Subject Re: small patch to crypt.c
Date
Msg-id 20130609034744.GN7200@tamriel.snowman.net
Whole thread Raw
In response to small patch to crypt.c  ("Joshua D. Drake" <jd@commandprompt.com>)
Responses Re: small patch to crypt.c
List pgsql-hackers
JD,

* Joshua D. Drake (jd@commandprompt.com) wrote:
> In my quest to understand how all the logging etc works with
> authentication I came across the area of crypt.c that checks for
> valid_until but it seems like it has an extraneous check.
>
> If I am wrong I apologize for the noise but wouldn't mind an explanation.

Alright, there probably aren't too many people out there running with
their clock set to pre-2000, but wouldn't this end up giving the wrong
result in those cases, as GetCurrentTimestamp() would end up returning a
negative value, which would make it less than vuntil's default of zero?

Perhaps we could change what vuntil is set to by default, but I think
it's probably better to keep things as-is; we should really be checking
for null cases explicitly in general.
Thanks,
    Stephen

pgsql-hackers by date:

Previous
From: Stephen Frost
Date:
Subject: Re: Proposed patch: remove hard-coded limit MAX_ALLOCATED_DESCS
Next
From: Andrew Dunstan
Date:
Subject: Re: [COMMITTERS] pgsql: Don't downcase non-ascii identifier chars in multi-byte encoding