[freetds] CS_INT size bug

entropy at freetds.org entropy at freetds.org
Tue Sep 5 10:02:53 EDT 2006


Norbert Sendetzky wrote:
> On Monday 04 September 2006 23:25, freetds-request at lists.ibiblio.org wrote:
>>> It's not that I want to make assumptions about types but printf forces me
>>> to do so. Your example above works as long as it doesn't hit a plattform
>>> where "long" is defined as 64 bit type (e.g. Alpha, AIX). On these
>>> plattforms, it will fail miserably.
>> Err, no, it works just fine on Alpha where I just tested it, and I see
>> no reason why it shouldn't work on any other platform with 64-bit longs.
>>   Can you explain why you think it should "fail miserably?"  If so,
>> perhaps I can help you see where you're going wrong.
> 
> I assume, you've used gcc which probably don't define long to 64 bit. Maybe 
> it's only a problem when using the native plattform compiler - but who does 
> this today.
> 
> On the other hand, did you get warnings in -Wall -pedantic mode?

It doesn't get warnings with gcc.  It doesn't get warnings with gcc 
-Wall -pedantic.  It doesn't get warnings with the native platform compiler.

In fact, there is no reason why it should get any warnings with any 
compiler with any switches on any platform where long is at least 32 
bits (and I guarantee you that that applies to every host you're ever 
likely to care about), which is precisely why I suggested the code to 
you in the first place.

The code I suggested is *correct*.  Instead of speculating endlessly why 
don't you just try it?

-- 
Cheers,
entropy



More information about the FreeTDS mailing list