On 30 Aug 2001 21:35:42 -0400, you wrote:
>"Christopher Kings-Lynne" <[EMAIL PROTECTED]> writes:
>> Are you sure? I thought all that autocommit meant was that a statement that
>> is not enclosed within a begin/commit is automatically committed after it is
>> run. So, in the this case all three q
Christopher Kings-Lynne writes:
> Are you sure?
Yes.
> I thought all that autocommit meant was that a statement that
> is not enclosed within a begin/commit is automatically committed after it is
> run. So, in the this case all three queries will be independent, unless the
> first statements i
"Christopher Kings-Lynne" <[EMAIL PROTECTED]> writes:
> Are you sure? I thought all that autocommit meant was that a statement that
> is not enclosed within a begin/commit is automatically committed after it is
> run. So, in the this case all three queries will be independent, unless the
> firs
a 'commit;'...
Chris
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Peter Eisentraut
> Sent: Friday, 31 August 2001 1:57 AM
> To: Rene Pijlman
> Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
> Subject: Re: [HACKERS] Multiple semico
On Thu, 30 Aug 2001 19:56:53 +0200 (CEST), you wrote:
>Rene Pijlman writes:
>> If autocommit is _enabled_ and S1;S2;S3 is send to the database,
>> what exactly is the behaviour of the backend? For example, what
>> happens if S1 succeeds, S2 fails and S3 would succeed?
>
>All three commands are exe
Rene Pijlman writes:
> If autocommit is _enabled_ and S1;S2;S3 is send to the database,
> what exactly is the behaviour of the backend? For example, what
> happens if S1 succeeds, S2 fails and S3 would succeed?
All three commands are executed in a single transaction. So if S2 fails,
S3 would no
