| From: | Michael Fuhr <mike(at)fuhr(dot)org> |
|---|---|
| To: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
| Cc: | Jim(dot)Gray(at)Bull(dot)com, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Ken(dot)Rosensteel(at)Bull(dot)com, pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: BUG #1862: ECPG Connect, host variable trailing blanks |
| Date: | 2005-09-16 16:31:53 |
| Message-ID: | 20050916163153.GA14785@winnie.fuhr.org |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
On Fri, Sep 16, 2005 at 12:00:59PM -0400, Bruce Momjian wrote:
> Jim(dot)Gray(at)Bull(dot)com wrote:
> > What do you think of an Postgres option that would enable stripping of
> > trailing blanks from connect host variables when turned ON?
>
> I would not support such an option. If we added flags for every single
> thing that someone wanted, the system would be unusable.
[snip]
> Just because Oracle does it doesn't mean we should.
Does Oracle really munge data on the client side? Or does it, like
PostgreSQL, pass the host variable's value as-is to the server, and
the server considers trailing spaces significant or not depending
on the context? Is it the client-side behavior or the server-side
behavior that's different between PostgreSQL and Oracle?
If Oracle strips trailing spaces on the client side, is that a
configurable option? How would you insert significant trailing
spaces into a VARCHAR column if the client library strips them?
--
Michael Fuhr
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2005-09-16 19:39:38 | Re: BUG #1871: operations with data types |
| Previous Message | Bruce Momjian | 2005-09-16 16:00:59 | Re: BUG #1862: ECPG Connect, host variable trailing blanks |