Re: New problem with Firebird dialect 3

Geoff Wass [dBVIPS] schrieb:
> ? 2^16			//  65,536
> ? 800*150		// 120,000

in Firebird, it is fortunately a much bigger limit, but it is 
unfortunately a bit more complicated.

in Firebird, INTEGER is a 32-Bit, and it is a signed type (negative 
values allowed).

That leads to a value ranging from -2.147.483.648 to 2.147.483.647.

Jaime's problem was somewhere else.... only where... ??
Maybe in the too strict definition of numerics.

There was a time when it was STANDARD to define a numeric as numeric(9,2).

This would probably cause problems in such situations.

In article <yYw0vASfIHA.1880@news-server>, nobody@nowhere.de says...
> Geoff Wass [dBVIPS] schrieb:
> > ? 2^16			//  65,536
> > ? 800*150		// 120,000
> 
> in Firebird, it is fortunately a much bigger limit, but it is 
> unfortunately a bit more complicated.
> 
> in Firebird, INTEGER is a 32-Bit, and it is a signed type (negative 
> values allowed).
> 
> That leads to a value ranging from -2.147.483.648 to 2.147.483.647.
> 
> Jaime's problem was somewhere else.... only where... ??
> Maybe in the too strict definition of numerics.
> 
> There was a time when it was STANDARD to define a numeric as numeric(9,2).
> 
> This would probably cause problems in such situations.
> If you make numeric(15,2) instead of numeric(9,2) it should be OK.
> 

André,

Could it be related to differences in the dialect he had? It seems to me 
there were differences in the datatypes between versions (not just 
BIGINT, but other things) but I can't recall where I saw that (or maybe 
I"m confusing it with something else).

In any case, it sounds like he has it solved though he probably should 
do some testing that stresses the limits of his calculations.

-- 
Geoff Wass [dBVIPS]
Montréal, Québec, Canada

.|.|.|        dBASE info at http://geocities.com/geoff_wass       |.|.|.
.|.|.| ---------------------------------------------------------- |.|.|.

Geoff Wass [dBVIPS] schrieb:
> Could it be related to differences in the dialect he had?  

probably.
If for example the numerics were defined as "numeric(9,2)" like it was

quite standardized in earlier versions, this format would be outdated