--- In ib-support@y..., "Claudio Valderrama C." <cvalde@u...> wrote:

> I personally won't make the internal code a mess just to please you,

sorry.

> There would be too many places to mangle. But you may want to

sponsor

> someone that wants to do that, of course. Assignments are checked at
> run-time, but maximum capability is checked at compile time or we

would need

> lots of internal tests.

Claudio, largest varchar personally I ever used is 16K :) and I can
use FB if I just know it's strong and weak peculiarities. My intention
is via discussions to help you to make FB extremely functional and
suitable for wide circle of developers. Of course, it is'nt altruism,
I'm "slave of oil lamp" like Aladdin's genie :) IB/FB is most known
SQL server for me and I don't want to become depopulating dinosaur if
it will not be popular.

> > Note that FB allows to
> > have in table chars in size more than declared (after alter column

to

> > less size).
>
> Example, please, using normal DDL.

You are right, I was mistaken. Can't remember was it feature of
early IB 6 or I'm got into the habit of the tool I use. Most probably
the second.

> Do you want the engine to include full Artificial Intelligence

Well, most of programming languages operate with strings of varying
length limited in max size (is'nt it alike varchar definition?) and
haven't problems on check of real strings length against limit on
concatenation. Is it Artificial Intelligence? It is habitual behavior
for most programmers and they want it in FB too.

> Well, if you know that your
> concatenation will produce a value that's not representable in

system tables

> and it's not representable in the external DSQL API, then it's your
> responsability to manage your Frankenstein properly.

WILL produce or CAN produce? That's the point.

Best regards, Alexander V.Nevsky.