> It would appear that when data in VARCHAR's are returned over the network
> that they are fixed width, is this true?

I proved this to be true some time ago. Using the API as closely as I do I
could put together a simple demonstration of it if anyone would be
interested. (It has been a while now since I've bothered investigating.)

> I.e. If I have a VARCHAR(1024) in a table with 'ABC' in one of the rows

then

> the TCP/IP packet contains ABC and a whole load of nil's.
>
> I've posted this here as I believe it to be true (i.e. not support) and if
> it is, then it represents a major network traffic problem.
>
> Is it true? Am I missing something?

Yes, I believe that it is. I have not tested every version of InterBase in
every possible configuration so I can't speak at that level of accuracy.

I suggested some changes some time ago about "suppressing whitespace" from
the network transmission buffers so that more records could be wadded up in
an outbound network packet (ignoring any inside or outside compression
applied). My thinking was, regardless of compression, if more records were
batched together for processing queries that the whole query would run MUCH
faster.

This is one area I plan to look into as soon as I get my grubbies on the
source. I think it is a potential area to really improve InterBase's network
friendliness.

Not everyone liked my thinking here and I met quite some opposition so
hopefully this time around we can discuss this more openly and get to a
consensus. I recall just dropping the thread to avoid more tension...

FWIW,
Jason Wharton
CPS - Mesa AZ
http://www.ibobjects.com