FreeTDS Development Group

[Brian Bruns <camber AT umcc.ais.org>]

On Mon, 1 Feb 1999, Gregg Jensen wrote:

> Brian Bruns wrote:
> > Let me know how it works,
> 
> Well, initial tests are positive!  I have tested with a modified dbtest 
> (that
> failed before) and with sqsh-1.4.  So far most queries have worked.  I do
> have one problem that persists with sqsh-1.4.  I run a query (sp_help)
> and it spits out the right results, but then I turn around and make simple
> select and no output.  I run another similar query and the output for the
> previous query spits out.
>
I'll try this out on this end (maybe we should open a bug? I'd really like
to start using the bug tracking system...). Actually, I just built
sqsh-1.4 against the latest on Linux and it works against a Sybase 11.0.1
server. Is yours not working only on Sun or is it that you're hitting a
MS-SQL database?

[snip] 
> 
> Brian, is this what you meant by changes going in different functions for
> different TDS versions?  I changed tds_process_col_info specific for 4.2
> and tds_process_result specific for 5.0.  tds_process_compute_result is
> changed for all.
> 
Exactly, go ahead and check it in, and I'll do the platform specific
checks afterwards (it won't hurt anything, just waste a couple bytes on
intel 'til it's corrected). BTW, does anyone know if word aligning intel
helps performance as it does on Alpha, 68k, etc..? (that is, does a
non-aligned access take two cycles instead of one on intel?).

Brian