On 23 May 2011, at 22:08, Shane W wrote:
> Hello list,
>
> I have a table with double precision columns and update
> queries which multiply and divide these values. I am
> wondering if it's possible to catch overflow and underflow
> errors to set the column to 0 in the case of an underflow
> and
UPDATE tbl SET score = divide_double_default(score, s, 1e-200) ...
UPDATE tbl SET score = multiply_double_default(score, s, 9) ...
Code the divide_double_default/multiply_double_default functions with error
handling that will return the desired value (either zero or the supplied
paramete
On Sunday 01 July 2007 21:50, Tom Lane wrote:
> Why would you expect it to? The raise_exception SQLSTATE applies
> specifically and solely to the plpgsql RAISE command. The entire
> point of those identifiers is to match fairly narrow classes of
> exceptions, not anything thrown by anyone.
>
>
On Sun, Jul 01, 2007 at 03:50:09PM -0400, Tom Lane wrote:
> IMHO the real problem with both RAISE and the plperl elog command
> is there's no way to specify which SQLSTATE to throw. In the case
> of the elog command I think you just get a default.
That default is XX000 (internal_error):
test=> c
Wiebe Cazemier <[EMAIL PROTECTED]> writes:
> When I do something on the table which the trigger function prevents, I get a
> message saying ERROR: blablabla. When such an error is generated by a pl/pgsql
> trigger function, I can trap the error with WHEN raise_exception. This does
> not work for th
On Sunday 01 July 2007 21:16, Michael Fuhr wrote:
> What have you tried and how did the outcome differ from your
> expectations?
The pl/perl trigger function in question generates an exception by
elog(ERROR, "message"). I also tried die(), which didn't make a difference.
When I do something on t
On Sat, Jun 30, 2007 at 10:30:32PM +0200, Wiebe Cazemier wrote:
> I have a pl/perl trigger function which can give an error, and I would like to
> catch it in a pl/pgsql function, but I can't seem to trap it.
What have you tried and how did the outcome differ from your
expectations?
> Is it possi
CSN wrote:
I have a table like so:
id|username|email
with unique indices on username and email. In a
plpgsql function if an insert fails because of a
duplicate on one of those fields, is it possible to
trap the error, figure out which unique fields it
applies to, and raise a custom error messag
