FreeTDS Development Group

["James K. Lowden" <jklowden AT freetds.org>]

Frediano Ziglio wrote:
> 
> Time ago I added two scripts to CVS:
> - misc/prepare_win32.sh: this script cross compile freetds for win32 and
> win64 and package all test/libraries ready to test or manually deploy on
> a windows system. There is also an option to compile dblib against ms
> library.
> - misc/sybase_tests: this script alter dblib Makefiles in order to
> compile dblib tests again Sybase libraries
> Feel free to use to compare the three dblib implementations (MS, Sybase
> and FreeTDS) or ODBC ones (MS and FreeTDS)

These are very nice contributions, Freddy.  I myself haven't been able to
use them yet, but having a framework for comparing the results of
different libraries is very helpful.  

I noticed many differences between FreeTDS and Microsoft while replacing
Microsoft's db-lib during a Win64 port.  You showed the unit tests match
Sybase pretty well, but many don't even work with Microsoft.  I would like
to reduce and isolate those differences, but that's a work in progress.  I
don't think we need to resolve them before the next release.  For me it's
a 1.0 goal now.  

> About stopover I remember:
> - connect/login cleanup. Finished ??

Yes.  I'm satisfied for now, although I do think "tsql -L" needs reworking
to be easier to use.  

> - bcp file encoding property

Important, but I'm not going to have time to work on it for the forseeable
future.  I'm thinking of nominating it for a SoC project next year.  

> - fix "seems dblib can get TDS protocol wrong in unusual case" report

Still no progress.  It's *very* important -- any result-set processing
error is important -- but it's also hard to isolate.  We don't have a test
case, just a problem description.  

> - documentation updates... I don't remember if we updated it...

IIRC, done.  

For me, the bcp file-encoding and the db-lib result-set processing issue
are both important.  But I don't have much time these days to work on
FreeTDS.  If we can't make any real progress on them, perhaps it's better
to release what we have.  

--jkl