Re: [Pgpool-general] backend mismatch

Le jeudi 16 juillet 2009 à 15:48:58, Leonardo Carvalho a écrit :
> Unfortunately, PgAdmin always sends this commands , trying to use
> ISO date format.
> If you discover how to fix that in PgAdmin, tell us!
>

There's no way to fix that without recompiling. But you can try the patch
attached.

I CC-ed pgadmin-hackers list so Dave can grab and check my patch :)

Regards.

--
Guillaume.
http://www.postgresqlfr.org
http://dalibo.com

On Thu, Jul 16, 2009 at 3:38 PM, Guillaume
Lelarge<guillaume(at)lelarge(dot)info> wrote:
> Le jeudi 16 juillet 2009 à 15:48:58, Leonardo Carvalho a écrit :
>>    Unfortunately, PgAdmin always sends this commands , trying to use
>> ISO date format.
>>    If you discover how to fix that in PgAdmin, tell us!
>>
>
> There's no way to fix that without recompiling. But you can try the patch
> attached.
>
> I CC-ed pgadmin-hackers list so Dave can grab and check my patch :)

Looks OK to me. What's the problem though - the SET, or having
multiple command in one SQL query? If it's the latter, then that
happens in a bunch of other places where it can't be fixed so easily.

--
Dave Page
EnterpriseDB UK: http://www.enterprisedb.com

Le lundi 20 juillet 2009 à 17:50:53, Dave Page a écrit :
> On Thu, Jul 16, 2009 at 3:38 PM, Guillaume
>
> Lelarge<guillaume(at)lelarge(dot)info> wrote:
> > Le jeudi 16 juillet 2009 à 15:48:58, Leonardo Carvalho a écrit :
> >> Unfortunately, PgAdmin always sends this commands , trying to use
> >> ISO date format.
> >> If you discover how to fix that in PgAdmin, tell us!
> >
> > There's no way to fix that without recompiling. But you can try the patch
> > attached.
> >
> > I CC-ed pgadmin-hackers list so Dave can grab and check my patch :)
>
> Looks OK to me. What's the problem though - the SET, or having
> multiple command in one SQL query? If it's the latter, then that
> happens in a bunch of other places where it can't be fixed so easily.

The latter. I don't see a way to catch them all easily. I'll commit this, and
try to find a way to catch the others.

Thanks.

(and welcome back :) )

--
Guillaume.
http://www.postgresqlfr.org
http://dalibo.com

On Mon, Jul 20, 2009 at 5:38 PM, Guillaume
Lelarge<guillaume(at)lelarge(dot)info> wrote:

> The latter. I don't see a way to catch them all easily. I'll commit this, and
> try to find a way to catch the others.

Good luck with that. We use multi-sentence statements in practically
every SQL dialogue to group together commands like "CREATE xxx; ALTER
xxx; COMMENT ON xxx;" to run everything as one transaction without
explicitly starting or ending one. Changing that would require some
fairly invasive changes.

I would suggest fixing pgpool so it supports multi-sentence statements
as PostgreSQL has for as long as I can recall.

> Thanks.
>
> (and welcome back :) )

Thanks :-)

--
Dave Page
EnterpriseDB UK: http://www.enterprisedb.com

Le lundi 20 juillet 2009 à 19:06:50, Dave Page a écrit :
> On Mon, Jul 20, 2009 at 5:38 PM, Guillaume
>
> Lelarge<guillaume(at)lelarge(dot)info> wrote:
> > The latter. I don't see a way to catch them all easily. I'll commit this,
> > and try to find a way to catch the others.
>
> Good luck with that. We use multi-sentence statements in practically
> every SQL dialogue to group together commands like "CREATE xxx; ALTER
> xxx; COMMENT ON xxx;" to run everything as one transaction without
> explicitly starting or ending one. Changing that would require some
> fairly invasive changes.
>

Hmmmm. You're right. I don't know why I didn't think about this. Should I
revert my patch? it doesn't do any good (and no bad AFAICT).

--
Guillaume.
http://www.postgresqlfr.org
http://dalibo.com

On Mon, Jul 20, 2009 at 10:32 PM, Guillaume
Lelarge<guillaume(at)lelarge(dot)info> wrote:
> Le lundi 20 juillet 2009 à 19:06:50, Dave Page a écrit :
>> On Mon, Jul 20, 2009 at 5:38 PM, Guillaume
>>
>> Lelarge<guillaume(at)lelarge(dot)info> wrote:
>> > The latter. I don't see a way to catch them all easily. I'll commit this,
>> > and try to find a way to catch the others.
>>
>> Good luck with that. We use multi-sentence statements in practically
>> every SQL dialogue to group together commands like "CREATE xxx; ALTER
>> xxx; COMMENT ON xxx;" to run everything as one transaction without
>> explicitly starting or ending one. Changing that would require some
>> fairly invasive changes.
>>
>
> Hmmmm. You're right. I don't know why I didn't think about this. Should I
> revert my patch? it doesn't do any good (and no bad AFAICT).

The downside is an extra roundtrip. If it doesn't help, let's back it out.

--
Dave Page
EnterpriseDB UK: http://www.enterprisedb.com

Le mardi 21 juillet 2009 à 10:12:37, Dave Page a écrit :
> On Mon, Jul 20, 2009 at 10:32 PM, Guillaume
>
> Lelarge<guillaume(at)lelarge(dot)info> wrote:
> > Le lundi 20 juillet 2009 à 19:06:50, Dave Page a écrit :
> >> On Mon, Jul 20, 2009 at 5:38 PM, Guillaume
> >>
> >> Lelarge<guillaume(at)lelarge(dot)info> wrote:
> >> > The latter. I don't see a way to catch them all easily. I'll commit
> >> > this, and try to find a way to catch the others.
> >>
> >> Good luck with that. We use multi-sentence statements in practically
> >> every SQL dialogue to group together commands like "CREATE xxx; ALTER
> >> xxx; COMMENT ON xxx;" to run everything as one transaction without
> >> explicitly starting or ending one. Changing that would require some
> >> fairly invasive changes.
> >
> > Hmmmm. You're right. I don't know why I didn't think about this. Should I
> > revert my patch? it doesn't do any good (and no bad AFAICT).
>
> The downside is an extra roundtrip. If it doesn't help, let's back it out.

OK, I will do.

--
Guillaume.
http://www.postgresqlfr.org
http://dalibo.com