[Insight-developers] Wrapping - WrapITK

[Dan Mueller]

Hi Alex, (Gaëtan, others),

> We decided at washington that TCL would not be a "first class"
> citizens of itk wrapping, i.e. that the mechanism to wrap in tcl would
> be here, but maintenance should be done by elements of the community
> that need it (as opposed to python, java and C#).

Interestingly, C# did not feature strongly in the SimpleITK survey:
    http://www.itk.org/Wiki/ITK_Release_4/SimpleITK/Tcon_2010_11_3
    http://www.itk.org/Wiki/File:SimpleITK-R1.pdf

That said, I am still very keen on C# wrapping, and will continue to
help realize it in both WrapITK and SimpleITK.

Cheers, Dan

On 4 November 2010 15:40, Alexandre GOUAILLARD <agouaillard at gmail.com> wrote:
> hi kent,
>
> I think moving to python makes a lot of sense and is definitely the
> way of the future.
>
> WrapITK was a one man show. The mechanism for tcl wrapping is in
> place. I'm going to run it and see if we're lucky or not. I'm neither
> a tcl expert nor (yet) a wrapitk expert, so I might need your help to
> read / guess through the errors though. In one hour together in iowa
> we should be able to assess the needed effort (or actually make it
> work, who knows)
>
> We decided at washington that TCL would not be a "first class"
> citizens of itk wrapping, i.e. that the mechanism to wrap in tcl would
> be here, but maintenance should be done by elements of the community
> that need it (as opposed to python, java and C#).
>
> alex.
>
> On Thu, Nov 4, 2010 at 10:26 PM, kent williams
> <norman-k-williams at uiowa.edu> wrote:
>> Alexandre -- Tcl wrapping in the abstract would be A Good Thing.  But for
>> BRAINS development at Iowa the current thinking is to abandon Tcl as a
>> scripting language and move to Python when we go to ITKV4.  At this point --
>> no offense -- Tcl hasn't gotten much love from WrapITK, and what Hans told
>> me is that it wasn't going to be supported in ITKV4.
>>
>> On the other hand, most of the current BRAINS3 tests are in Tcl and exercise
>> Tcl wrapping.  Will the new WrapITK change the bindings enough to break
>> existing code?  If the bindings are intact, testing with BRAINS3 could be
>> done quickly.
>>
>> 30 Hours on a plane? I can't begin to tell you how much I'd dread such a
>> thing!
>>
>>
>> On 11/4/10 2:10 AM, "Alexandre GOUAILLARD" <agouaillard at gmail.com> wrote:
>>
>>> dear all,
>>>
>>> just a quick message to let you know that Gaetan merged the newest
>>> version of wrapitk in an itkv4 branch there:
>>> https://github.com/glehmann/ITK/tree/wrapitk
>>>
>>> and that I'm actively working on making it work.
>>>
>>> As of today, the python (2.x) version is fully operational and Im
>>> working on the java wrapping. I sent an experimental, but dashboard
>>> seems to be unstable still. The idea is to have a fully operational
>>> integration by next week to present at the meeting (along with a full
>>> implementation of PACS support from mathieu M., support for RT STRUCT,
>>> and most of the streaming for Dicom as well).
>>>
>>> kent: if java wrapping is working early enough, I will spend some of
>>> my 30 hours of plane this week end to take a look at the tcl wrapping.
>>> If you want, we can try to take some time to look into it, even though
>>> I do not commit to make it work.
>>>
>>> wanlin, wen li, luis and the few waiting for me to bring in some mesh
>>> filters, sorry, I won't be able to work on that until monday at best.
>>>
>>> alex.