On Tue, Sep 18, 2001 at 12:45:35PM -0500, Dimitri Maziuk wrote: | * dman ([EMAIL PROTECTED]) spake thusly: | > On Mon, Sep 17, 2001 at 10:20:33AM -0500, Dimitri Maziuk wrote: | > | * dman ([EMAIL PROTECTED]) spake thusly: | > | ... | > | > | > | > The difference bewteen server and workstation is a couple registry | > | > keys and the price tag. :-) | > | | > | I used to think that, too, until a (shudder) VB/MSSQL app I wrote | > | refused to run on NT swerver (developed on WS). And yes, I made sure | > | all relevant DLLs had the same version etc... | > | > Hmm, maybe that app (or it's MS-developed libraries) knew which key to | > look for and didn't want to work on the other system. Other than | > that, I don't know. | | Heh. I wish. The problem was that idjit who created the database used | non-standard [var]char fields -- M$ extensions for multibyte characters. | On NTws "select *" worked fine, on NTsrv it b0rked on the first | non-standard field and returned only some of the columns... The | not-so-surprising part is that M$ themselves apparently have no idea | what's causing it, or why "select cast(badfield as varchar)" works.
Heh. Wow. Interesting. I originally thought that it didn't run or something. -D
