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

On 6/23/25 7:40 PM, Arne Vajhøj wrote:

On 6/23/2025 8:37 PM, Lawrence D'Oliveiro wrote:

On Mon, 23 Jun 2025 20:17:57 -0400, Arne Vajhøj wrote:

Unfortunately that means that the entire fallback algorithm
in tds_vasprintf (write to null device to find output size,
allocate buffer of correct size and then write to buffer)
is broken.

Thanks for chasing that down.

What a dumb way to do it. Using the null device means making OS I/O calls
-- needless overhead just to find out how long a string -- which is being
put together entirely in userland -- needs to be.

 Fallback algorithms are often primitive.
 But if you have a simple better way of achieving it then please send
the code to Craig so he can send it upstream.

This one looks pretty simple and depends only on vsnprintf, which has
been in VMS since 7.3-2, so it was probably in some ancient standard,
though I haven't looked yet:
https://github.com/jkaivo/asprintf/blob/asprintf/asprintf.h
It's MIT which I think can be included in a GPL project like FreeTDS. I
will try to remember to look at this when I get some other things sorted
out.