On 12/27/2010 10:23 AM, Nina Jeliazkova wrote:
> Hi Tobias,
>
> On 27 December 2010 10:56, Tobias Girschick<tobias.girschick at in.tum.de>wrote:
>
>> Hi All,
>>
>>
>> On 12/20/2010 04:49 PM, Christoph Helma wrote:
>>
>>>    - POST/PUT for descriptor calculation
>>> I have checked: PUT operations for dataset subsets are already in the
>>> dataset API.
>>>
>> OK. Is there a rule or scheme when POST and when PUT is used? The default
>> should be POST, I guess, so we don't mess around with existing datasets. But
>> how can the user switch to PUT (if he has privileges for that)? An
>> additional parameter?
>>
> Possible options coming to my mind:
>
> 1) Additional parameter  (result_dataset - there was already such suggestion
> if I remember right)

Keeps flexibility. I prefer this option. The service provider could 
choose the default or we could fix a default in the API (clearer option 
maybe).

> 2) Up to the calculation service to select between POST or PUT for the
> results
That's how it is at the moment. The problem here is, that it's not easy 
to change the behavior and especially in development and test cases it 
might be wise not to use PUT although this could be the default option 
of a certain service.
Imagine a user wants to try out/play around a little bit with different 
descriptors ... If PUT is fixed he will mess up the dataset. If POST is 
fixed he will have to remember one dataset URI per step in his final 
workflow.

> 3) Introduce PUT for model/algorithm service, which will have the same
> meaning as POST (returning dataset results), but will assume the results
> will be send by PUT operation to dataset service.

Not sure about this option.

Best regards,
Tobias
> What do you think?
>
> Best regards,
> Nina
>
>
>> Best regards,
>> Tobias
>>
>> --
>>
>> Dipl.-Bioinf. Tobias Girschick
>>
>> Technische Universität München
>> Institut für Informatik
>> Lehrstuhl I12 - Bioinformatik
>> Bolzmannstr. 3
>> 85748 Garching b. München, Germany
>>
>> Room: MI 01.09.042
>> Phone: +49 (89) 289-18002
>> Email: tobias.girschick at in.tum.de<mailto:tobias.girschick at in.tum.de>
>> Web: http://wwwkramer.in.tum.de/girschick
>>
>>
>> _______________________________________________
>> Development mailing list
>> Development at opentox.org
>> http://www.opentox.org/mailman/listinfo/development
>>
> _______________________________________________
> Development mailing list
> Development at opentox.org
> http://www.opentox.org/mailman/listinfo/development

-- 

Dipl.-Bioinf. Tobias Girschick

Technische Universität München
Institut für Informatik
Lehrstuhl I12 - Bioinformatik
Bolzmannstr. 3
85748 Garching b. München, Germany

Room: MI 01.09.042
Phone: +49 (89) 289-18002
Email: tobias.girschick at in.tum.de <mailto:tobias.girschick at in.tum.de>
Web: http://wwwkramer.in.tum.de/girschick