On Fri, Mar 23, 2012 at 11:54 PM, Rafael Laboissiere
<raf...@laboissiere.net> wrote:
> * Juan Pablo Carbajal <carba...@ifi.uzh.ch> [2012-03-22 23:39]:
>
>> I just updated the OF repository.
>>
>> - I added licenses to
>> inst/io/ <at> svg/parsePath.py
>> inst/io/ <at> svg/parseSVGData.py
>>
>> - I removed deprecated functions such as
>> inst/io/private/parsePath.py
>>
>> - The bug in shapecentriod is fixed. It was due to ccw and cw
>> parametrization of the shapes. Warnings are now produced. In the
>> future I will provide ccw2cw and cw2ccw functions for piece-wise
>> polynomial planar shapes.
>>
>> - I moved the tests in shapetransfrom to shapecentriod, since these
>> test were testing shapecentriod under homogenous transformations.
>>
>> - I improved some docstring.
>>
>> -I all tests using assert[a-zA-Z]+ functions add and remove the
>> private folder from path.
>>
>> I will release the packages in a few days. I give time to detect further 
>> bugs.
>
> Good news, thanks for the update.
>
> The Debian package builds cleanly now.  There are some issues, though:
>
> 1) At several places, "centroid" is wrongly spelled "centriod".
>
> 2) In the 1.4.0 tarball, there are empty doc/ and devel/.  I guess you
>   can remove them in the forthcoming release tarball.
>
> 3) The Debian packaging system looks for test blocks in all *.m and
>   *.cc files and try to run them.  There is a test in file
>   io/private/getSVGPaths_py.m which, evidently, fails.  What is the
>   meaning of having a test in a private function?  Could it be moved
>   elsewhere?
>
> 4) When doing the unit testing in lineDraw.m, a plot in ASCII is output.
>   The relevant part of the build log with the ascii-art figure is
>   attached below.  Do you now where it comes from?  Is this plot really
>   necessary for the test?  I am asking because I will need to add a
>   build-dependency on gnuplot-nox if the plot is really needed.
>
> Rafael
>
> ------------------------------------------------------------------------------
> This SF email is sponsosred by:
> Try Windows Azure free for 90 days Click Here
> http://p.sf.net/sfu/sfd2d-msazure
> _______________________________________________
> Octave-dev mailing list
> Octave-dev@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/octave-dev
>

Thanks again Rafael!

1) Sed fixed the recurrent typo!

2) I am porting graphs algorithms and the devel folder there is code
under development. I will remove it from the released version but not
from svn (I will release the package as soon as everything seems
quiet). I removed the doc folder

3) I removed all the files that are now implemented as methods of the
@svg class. I am sorry I was using private as "do not use this files".
They are in inst/io/deprecated. Can the package system skip
"deprecated" folders?

4) This one I am not able to solve/answer right now. The graphical
output is not necessary, but the handle to the graphical object is,
since the tests are checking whether the line is drawn correctly by
checking the data in that object. I will try to find if there is a way
to generate the handle to the graphic object without producing any
plot. I do not think we need to add a dependency because of this.

I will release geometry 1.4.1 now. You can use this one, I hope
without further problems.

Thanks again

-- 
M. Sc. Juan Pablo Carbajal
-----
PhD Student
University of Zürich
http://ailab.ifi.uzh.ch/carbajal/

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
_______________________________________________
Octave-dev mailing list
Octave-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to