[Dgtal-users] C++11 & release 0.9.1

Kacper Pluta kacper.pluta at gmail.com
Mon Nov 16 22:13:29 CET 2015


Hi David,

Well, I think we should think about C++11 finally as the default C++
standard and make it default in DGtal (BTW a nice example with thinning,
good way to make an advertisement of C++11).  On the other hand, if we
switch to C++11 I think we should review the code and see what was in C++11
marked as deprecated---for instance unary_function---and clean DGtal from
that futures.

It is good that you have mentioned builds for MS Windows. I have had an
impress that Windows is not treated sufficiently enough. Could not we test
builds for Windows as we do for OS X and Linux? All the other new
components looks nice. I will be happy to see CubicalComplex class in 0.9.1.


Best,
Kacper


On Sun, Nov 15, 2015 at 4:03 PM, David Coeurjolly <
david.coeurjolly at liris.cnrs.fr> wrote:

> Hi all,
>
> At some point, we would have to exchange about Roland’s request to require
> CPP11 in DGtal (PR #1029). In the meantime, I’ve played a bit with fancy
> c++11 features in some tools compiled on top of DGtal (auto, for loop on
> ranges, lambda expressions…). Please have look to this example
> https://github.com/dcoeurjo/MyDGtalTools/blob/master/dgtal-cpp11.cpp
>
> For example, ultimate homotopic thinning is reduced to few  lines of code
> (2 for_each with 2 lambdas):
>
> while(!stable)
>  {
>    stable = true;
>    std::for_each(S.begin(), S.end(),
>                  [&Q,&object](Point p){ if (object.isSimple(p))
> Q.push_back(p); });
>    std::for_each(Q.begin(), Q.end(),
>                  [&object,&stable,&S](Point p){ if (object.isSimple(p))
> {stable = false; S.erase(p);} });
>    Q.clear();
>   }
>
> ;)
>
> So the question: Should we restrict DGtal to C++11 enabled compilers only
> ?
>

> Beside that, we have to plan a 0.9.1 release of DGtal. As discussed during
> the 0.9 release preparation, I think that it could be nice to have a
> release announced during the Digital Geometry Workshop on Lyon (25/11). At
> this point, we already have plenty of new features in the master
> (Alphathick segments -pending but ok-, MS build, linux binary packages,
> statistics classes, various fixes…).
>
> What do you think about that?
>
> @Jacques-Olivier: could the CubicalComplex class be ready for 0.9.1 ?
>
> regards
> david
> ----------------
> David Coeurjolly - Directeur de recherche CNRS
> CNRS, LIRIS-UMR 5205
> Université de Lyon
> Batiment Nautibus
> 43 boulevard du 11 novembre 1918, 69622 Villeurbanne cedex, France
> Tel : (+33) [0]4.72.44.82.40  Fax : (+33) [0]4.72.43.15.36
> http://liris.cnrs.fr/david.coeurjolly
>
>
>
>
>
>
>
> _______________________________________________
> Dgtal-users mailing list
> Dgtal-users at lists.gforge.liris.cnrs.fr
> https://lists.gforge.liris.cnrs.fr/cgi-bin/mailman/listinfo/dgtal-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.gforge.liris.cnrs.fr/pipermail/dgtal-users/attachments/20151116/8cde7423/attachment.html>


More information about the Dgtal-users mailing list