testcase

I am using Xcode 2.4 Max4.6.2 and latest UBSDK

when trying to use C++ for objects i am having a problem with the values being passed to my objects new method.

here is an example that highlight the problem.

setup((t_messlist **)&myobj_class,(method)myobj_new, (method)dsp_free, (short)sizeof(t_myobj), 0L, 

    A_DEFLONG, A_DEFLONG, A_DEFLOAT, A_DEFFLOAT, 0 );

void *myobj_new(long fftsize, long hop, double maxblurtime, double blurtime)

post("%i %i %f %f", fftsize, hop, maxblurtime, blurtime);

if i create the object in the patcher as follows.

if i reverse the order of args expected to be doubles first i get similar results. anyone else experience this?? any hints?

here is an  example that highlight the problem.

extern "C" int main(void)
{
 setup((t_messlist **)&myobj_class,(method)myobj_new, (method)dsp_free, (short)sizeof(t_myobj), 0L,  
	A_DEFLONG,  A_DEFLONG, A_DEFLOAT, A_DEFFLOAT, 0 );
.....
}

void *myobj_new(long fftsize, long hop, double maxblurtime, double blurtime)
{

   post("%i %i %f %f", fftsize, hop, maxblurtime, blurtime);
....
}

c-and-object-arg-problem

Looks like you're getting uninitialized pointers. no idea why, the 

I usually use the more generic A_GIMME form:

void * myobj_new(t_symbol *s, short ac, t_atom *av)

This way you can allow integers in place of floats (with some arg 

On Feb 21, 2007, at 3:16 PM, testcase wrote:

> I am using Xcode 2.4 Max4.6.2 and latest UBSDK

> when trying to use C++ for objects i am having a problem with the 

> values being passed to my objects new method.

> here is an example that highlight the problem.

> setup((t_messlist **)&myobj_class,(method)myobj_new, (method) 

>     A_DEFLONG, A_DEFLONG, A_DEFLOAT, A_DEFFLOAT, 0 );

> void *myobj_new(long fftsize, long hop, double maxblurtime, double 

> post("%i %i %f %f", fftsize, hop, maxblurtime, blurtime);

> if i create the object in the patcher as follows.

> if i reverse the order of args expected to be doubles first i get 

> similar results. anyone else experience this?? any hints?

Looks like you're getting uninitialized pointers.  no idea why, the  
code looks OK.

I usually use the more generic A_GIMME form:
void * myobj_new(t_symbol *s, short ac, t_atom *av)

This way you can allow integers in place of floats (with some arg  
testing).

>
> hello,
>
> I am using Xcode 2.4 Max4.6.2 and latest UBSDK
>
> when trying to use C++ for objects i am having a problem with the  
> values being passed to my objects new method.
>
> here is an  example that highlight the problem.
>
> if i have an object like this
>
> extern "C" int main(void)
> {
>  setup((t_messlist **)&myobj_class,(method)myobj_new, (method) 
> dsp_free, (short)sizeof(t_myobj), 0L,
> 	A_DEFLONG,  A_DEFLONG, A_DEFLOAT, A_DEFFLOAT, 0 );
> .....
> }
>
> void *myobj_new(long fftsize, long hop, double maxblurtime, double  
> blurtime)
> {
>
>    post("%i %i %f %f", fftsize, hop, maxblurtime, blurtime);
> ....
> }
>
>
> if i create the object in the patcher as follows.
>
> myobj 1024 256 .6 .3
>
> this prints in the max window
>
> 1024 256 0.00000 4294967296.000000
>
> if i reverse the order of args expected to be doubles first i get  
> similar results. anyone else experience this?? any hints?
>
>
>


I would definitely go the A_GIMME route. I think there have been some

problems with signatures like A_DEFLONG, A_DEFLONG, A_DEFLOAT,

A_DEFFLOAT. For instance, things worked ok on macintel but not on PPC

when I was working with such functions a few months ago.

> Looks like you're getting uninitialized pointers. no idea why, the

> I usually use the more generic A_GIMME form:

> void * myobj_new(t_symbol *s, short ac, t_atom *av)

> This way you can allow integers in place of floats (with some arg

> On Feb 21, 2007, at 3:16 PM, testcase wrote:

> > I am using Xcode 2.4 Max4.6.2 and latest UBSDK

> > when trying to use C++ for objects i am having a problem with the

> > values being passed to my objects new method.

> > here is an example that highlight the problem.

> > setup((t_messlist **)&myobj_class,(method)myobj_new, (method)

> > dsp_free, (short)sizeof(t_myobj), 0L,

> > A_DEFLONG, A_DEFLONG, A_DEFLOAT, A_DEFFLOAT, 0 );

> > void *myobj_new(long fftsize, long hop, double maxblurtime, double

> > post("%i %i %f %f", fftsize, hop, maxblurtime, blurtime);

> > if i create the object in the patcher as follows.

> > if i reverse the order of args expected to be doubles first i get

I would definitely go the A_GIMME route.  I think there have been some
problems with signatures like A_DEFLONG,  A_DEFLONG, A_DEFLOAT,
A_DEFFLOAT.  For instance, things worked ok on macintel but not on PPC
when I was working with such functions a few months ago.

On 2/21/07, Graham Wakefield  wrote:
> Looks like you're getting uninitialized pointers.  no idea why, the
> code looks OK.
>
> I usually use the more generic A_GIMME form:
> void * myobj_new(t_symbol *s, short ac, t_atom *av)
>
> This way you can allow integers in place of floats (with some arg
> testing).
>
> On Feb 21, 2007, at 3:16 PM, testcase wrote:
>
> >
> > hello,
> >
> > I am using Xcode 2.4 Max4.6.2 and latest UBSDK
> >
> > when trying to use C++ for objects i am having a problem with the
> > values being passed to my objects new method.
> >
> > here is an  example that highlight the problem.
> >
> > if i have an object like this
> >
> > extern "C" int main(void)
> > {
> >  setup((t_messlist **)&myobj_class,(method)myobj_new, (method)
> > dsp_free, (short)sizeof(t_myobj), 0L,
> >       A_DEFLONG,  A_DEFLONG, A_DEFLOAT, A_DEFFLOAT, 0 );
> > .....
> > }
> >
> > void *myobj_new(long fftsize, long hop, double maxblurtime, double
> > blurtime)
> > {
> >
> >    post("%i %i %f %f", fftsize, hop, maxblurtime, blurtime);
> > ....
> > }
> >
> >
> > if i create the object in the patcher as follows.
> >
> > myobj 1024 256 .6 .3
> >
> > this prints in the max window
> >
> > 1024 256 0.00000 4294967296.000000
> >
> > if i reverse the order of args expected to be doubles first i get
> > similar results. anyone else experience this?? any hints?
> >
> >
> >
>
>


On 22-Feb-2007, at 6:47, Wesley Smith wrote:

> I would definitely go the A_GIMME route. I think there have been some

> problems with signatures like A_DEFLONG, A_DEFLONG, A_DEFLOAT,

Wes is correct. This has been confirmed on the mail list in the past 

by DDZ and JKC. Mixing floats and ints if there are 3 or more 

parameters breaks parameter set up. Search the archives for details.

It would be a Good Idea if this could information could find its way 

Peter Castine +--> Litter Power & Litter Bundle for Jitter

Interface Building for |home | chez nous|

Max/MSP Extremely cool |bei uns | i nostri|

> I would definitely go the A_GIMME route.  I think there have been some
> problems with signatures like A_DEFLONG,  A_DEFLONG, A_DEFLOAT,
> A_DEFFLOAT.

Wes is correct. This has been confirmed on the mail list in the past  
by DDZ and JKC. Mixing floats and ints if there are 3 or more  
parameters breaks parameter set up. Search the archives for details.

It would be a Good Idea if this could information could find its way  
into the SDK documention.

--------------    http://www.bek.no/~pcastine/Litter/    -------------
Peter Castine             +--> Litter Power & Litter Bundle for Jitter
                                Universal Binaries on the way
iCE:  Sequencing,  Recording &
       Interface  Building  for                   |home    | chez nous|
       Max/MSP   Extremely cool                   |bei uns |  i nostri|
       http://www.dspaudio.com/                   http://www.castine.de

thanks everyone. i had thought in the past i had mixed floats and ints in more than 3 args and everything worked fine. perhaps i am mistaken. anyway, a very good thing to know and certainly something that should make it into the sdk doc.

C++ and object arg problem