Re: FreeTDS port to VMS V9.x on x86?

On 6/26/2025 8:03 PM, Lawrence D'Oliveiro wrote:

On Tue, 24 Jun 2025 12:25:50 -0400, Arne Vajhøj wrote:

The algorithm is basically to try vsnprintf with buffers increasing in
size 512, 1024, 1536, ... until there is space.
>
That is very inefficient compared to just use the return value.

 It’s a good choice in the absence of any other clues as to how much to
allocate. I have used it myself. (Of course having an explicit number
returned is best of all.)
 One may conduct statistical analyses, I suppose to decide on the optimum
multiplication factor. E.g. would going up in powers of 3 better than
powers of 2? I suspect larger factors (too far from e = 2.71828 ...) would
not be worth using.

The specific factor is not as important as using multiplication instead
of addition.
Starting at 512 going towards a 1 GB CLOB:
+512 => 2 million
*1.5 => 36
*2 => 21
*3 => 13

And don’t forget to do a realloc(3) at the end, to trim the unused part of
the last allocation.

That seems very natural.
But for FreeTDS dbfcmd then it has:
vasprintf call to format buffer
test of return value
dbcmd call with buffer
free call for buffer
No point in realloc.
And I suspect that the pattern is common: format, use
once and free immediately.
So I don't know if it makes sense.
Arne