[Dcmlib] new DicomDir syntax: aka Deprecation pass

[Jean-Pierre Roux]

Mathieu Malaterre wrote:

> Luca,
>
>     Hum this is pretty bad. I thought the old functionanlities where 
> being kept back...

The functionalities are kept.
I flaged them as 'deprecated', because 'old style' was a bad idea. 
(Unaware user was allowed to do stupid things : remember  the 'cascade' 
Loads)
Luca wanted to use 'new style' -he was right to do that-, and get 
problemes on a wery short and simple program sequence.
(Probabely on 'new style' line was missing)
JP

>     I would like then to introduce a deprecation mechanism that will 
> prevent event like this. A number of compilers (well in fact only 
> gcc/and visual studio but there are the main one), support a notion of 
> 'deprecated functions'. Therefore at *compiler* time the code produces 
> warning like:
>
> ...this function is deprecated...

Does it make a sense if I add a run time Warning message to tell 
user/debuger to update his code ?

> But the at run time the code still works ok.
>     JP can we keep the old API or is this too much troubles ? Is this 
> already broken ?
>
> Thanks,
> Mathieu
>     
> Luca Picello wrote:
>
>> Hi all,
>> Syncronizing with today's CVS my code is not working anymore.
>> Its code is very easy and I wonder if you can help me on make it 
>> working again.
>>
>> gdcm::DicomDir *e1;
>> gdcm::DicomDirPatient *pa;
>>
>> e1 = new gdcm::DicomDir( fold.c_str() , true ); // DEPRECATED OLD 
>> GDCM SYNTAX
>>
>> // NEW SYNTAX AS DESCRIBED IN ML BY JP:
>> e1 = new gdcm::DicomDir();
>> e1->SetDirectoryName(fold.c_str());
>>
>> ---
>>
>> if( e1->IsReadable() )
>> or
>> pa = e1->GetFirstPatient();
>> now both fails!!
>>
>> I do have also access violation error doing (because of failure):
>>  delete e1;
>>
>> at: gdcmDicomDir.cxx at line 163 ( delete MetaElems; in 
>> DicomDir::~DicomDir() )
>>
>> thank you all,
>> Luca
>>
>> ps: does anybody encoured my latest reported bug on halt on loading 2 
>> datasets?
>> I can prepare a simple test when gdcm will be functional to me again...
>>
>>
>> _______________________________________________
>> Dcmlib mailing list
>> Dcmlib at creatis.insa-lyon.fr
>> http://www.creatis.insa-lyon.fr/mailman/listinfo/dcmlib
>>
>
> _______________________________________________
> Dcmlib mailing list
> Dcmlib at creatis.insa-lyon.fr
> http://www.creatis.insa-lyon.fr/mailman/listinfo/dcmlib
>