Mark O'Donohue schrieb:
>
> Patrick J. P. Griffin wrote:
>
> > Mark,
> >
> > Thanks, I've checked the output you've sent me and the results are exactly
> > what I hoped to see.
>
> Thats pleasant.
>
> > I'll admit that I've avoided uploading result sets because I had assumed
> > that the different database paths would cause them to be marked as failed.
> > I now see, even looking at my own result sets, that TCS must contain coding
> > to account for those differences without failing the test.
> >
>
> I must admit my knowledge is limited to doing some simple runs, from the
> little bit I looked at it seemed partially up to the test itself to
> determine if it was successful. I presume we'll all know how to do it
> pretty soon.
The compare routine in the tcs sorts out some things like
database names, paths and other "nuisances" . So the initializations
can be shared for different setups (and I hope platforms).
After I have fixed this thing to work with blobs of any segment size
I will surely know more :-)
Frank
--
"Fascinating creatures, phoenixes. They can carry immensely heavy loads,
their tears have healing powers and they make highly faithful pets."
- J.K. Rowling
http://sourceforge.net/projects/firebird
|
