[CLAM] SDIF in SMSTools...

[Francois Thibault]

Hello,

I was aware of this solution, thanks anyhow :-). One thing that I 
haven't figured out yet is the following crash:

##########################################################
################### ASSERTION FAILED #####################
##########################################################
At file ../../../../src/Processing/ArithOps/SpectrumAdder2.cxx line 165
SpectrumAdder2::SetPrototypes:Size mismatch in spectrum sum
Abort trap

This occurs when storing analysis as sdif or xml file, reload the 
analysis file and then try to synthesize it. Any help regarding this 
one will be much appreciated.

Bye
Francois

On Dec 3, 2003, at 9:14 AM, Xavier Amatriain wrote:

> Hi,
>
> This bug was found and fixed yesterday on our development branch and 
> will be available in next release. In the meantime if you want a 
> workaround, just take the mSDIFWriter.Configure(cfg) out of the 
> CLAM_DEBUG_ASSERT. The problem is that the SDIFWriter is only being 
> configured in debug mode but not in release :( Thanks for your 
> feedback.
>
> Francois Thibault wrote:
>
>> Hi,
>>
>> In version 0.5.5 and 0.5.4, there is a possible problem preventing 
>> the storage of analysis data in SDIF format (the file never gets 
>> created) if ASSERTS are chosen not to be considered. It happens in 
>> examples/SMS/SDIFSerializer.cxx, here is the diff between version 
>> 0.5.3 and 0.5.5 (it worked under 0.5.3):
>>
>> 76,78c76,77
>> <
>> <       CLAM_DEBUG_ASSERT(mSDIFWriter.Configure( cfg ), "Error 
>> storing temporal SDIF for Morph" );
>> <
>> ---
>> >       mSDIFWriter.Configure( cfg );
>> >
>>
>> Also while we are on the topic of SDIF, making the followin change in 
>> /src/Processing/SDIFOut.cxx will allow other software like IRCAM SDIF 
>> software utilities to use the produced SDIF file, otherwise it 
>> crashes as it gets an empty string for the matrix type by default. A 
>> similar change for "1TRC" is also required.
>>
>> <               pMatrix=new 
>> SDIF::ConcreteMatrix<TFloat32>(SDIF::TypeId::sDefault,1,1);
>> ---
>> >               pMatrix=new 
>> SDIF::ConcreteMatrix<TFloat32>("1FQ0",1,1);
>>
>> Hope this is helpful for some people...
>>
>> Bye
>> Francois
>>
>>
>> _______________________________________________
>> CLAM mailing list
>> CLAM at iua.upf.es
>> http://www.iua.upf.es/mtg/clam
>>
>>
>
> -- 
> /**
> *     Xavier Amatriain
> *   Music Technology Group
> *  Universitat Pompeu Fabra
> * http://iua.upf.es/mtg/~xamat
> *
> */
>
>