"Ansley, Michael" <Michael.Ansley@intec.co.za> writes:
> Why are these limits even an issue? Surely I should be able to adjust
> certain constants in the code (and this would include constants to take care
> of the tuple length) in order to adjust the performance of the system. I
> realise that life is not that simple, but perhaps it could be made simpler
> than it is now. At the moment, some of the constants used are pretty
> useless, because they can't really be adjusted.
> Can't we come up with a way to make this more amenable to adjustment?
If it bothers you, step right up and fix it ...
Where I was actually hoping to end up is with no fixed limit at all,
ie make all buffers for text be dynamically resizable to handle the
longest query presented in a given session. But it'd be a useful
intermediate step to find all the places that are dependent on max query
length and drive them all from a single config.h constant. All that
code would eventually be replaced, if we do dynamic sizing, but doing
this would still be helpful as a way of identifying everything that
needs to change. Just finding all the places that depend on max query
length is not as trivial as you might think.
regards, tom lane