Обсуждение: Re: trap for any exception

>
> I read 37.7.5. Trapping Errors, but the syntax requires to write 
> explicitly the exception condition, and not a generic one.
>
> Is it possible to build a generic trap or do you know a workaround for 
> that ?
>
Sorry, I found the OTHERS condition that trap any error.

But the question still remains to find how could I interpret it and found 
the error code and message ?

Sabin 

Hi there,

I'd like to build a PL/pgSQL function which is able to generic trap any 
error, and interpret it.

I read 37.7.5. Trapping Errors, but the syntax requires to write explicitly 
the exception condition, and not a generic one.

Is it possible to build a generic trap or do you know a workaround for that 
?

TIA,
Sabin 

That's ok, I found SQLSTATE and SQLERRM. 

On Thu, 15 Nov 2007, Sabin Coanda wrote:

> Hi there,
>
> I'd like to build a PL/pgSQL function which is able to generic trap any
> error, and interpret it.
>
> I read 37.7.5. Trapping Errors, but the syntax requires to write explicitly
> the exception condition, and not a generic one.
>
> Is it possible to build a generic trap or do you know a workaround for that
> ?

Won't OTHERS catch most of what you want?

From the 8.2 docs:
"The special condition name OTHERS matches every error type except
QUERY_CANCELED. (It is possible, but often unwise, to trap QUERY_CANCELED
by name.)"

On Thu, 15 Nov 2007, Sabin Coanda wrote:

> >
> > I read 37.7.5. Trapping Errors, but the syntax requires to write
> > explicitly the exception condition, and not a generic one.
> >
> > Is it possible to build a generic trap or do you know a workaround for
> > that ?
> >
> Sorry, I found the OTHERS condition that trap any error.
>
> But the question still remains to find how could I interpret it and found
> the error code and message ?

I think you'll want to look at SQLSTATE and SQLERRM inside the exception
handler.