yade-dev team mailing list archive

[Janek Kozicki <janek_listy@xxxxx>]

Very good!

I was about to check that :) I should have done that in the first
place, my mistake :)

Also, indeed even these extra `.point()` could be eliminated if
several of those functions were redeclared as overload with another
signature. I just feel a little uneasy about cross_product of two
(sphere-sphere) pairs ;)

best regards
Janek


Bruno Chareyre said:     (by the date of Tue, 7 Nov 2017 18:19:44 +0100)

> I can confirm my previous message: all the changes should compile with 
> older CGAL.
> The only place where an ifdef is needed maybe is when including 
> "constructions_on_weighted_points_cartesian_3.h".
> Bruno
> 
> On 11/07/2017 11:04 AM, Bruno Chareyre wrote:
> > Awesome.
> > Whenever implicit conversion from wp (weighted point) to bp (bare 
> > point) is used in current code it is just because there is no reason 
> > to type v->point().point() when we can write simply v->point(). 
> > However, what we need is really a bp, so the v->point().point() syntax 
> > reflects more accurately what is needed and using it should be 
> > compatible with the older versions of CGAL, hence no reason to make 
> > the fix depend on CGAL_VERSION.
> > Or did you see clear reasons why a fixed version would not compile 
> > with older CGAL?
> > Obviously it needs tedious search/replace and changes of some 
> > functions signatures. It sounds like you did a good part of it, I'll 
> > check your diff.
> > Cheers
> > Bruno
> >
> >
> >
> >
> >
> >
> > On 11/06/2017 08:31 PM, Janek Kozicki wrote:
> >> In hindsight there has to be a better way, instead of all those
> >> defines. If we cannot have an implicit conversion we could create a
> >> function which takes argument type of Point_3 or a type of
> >> Weighted_point_3 and always returns Point_3.
> >>
> >> Then the fixes scattered around the code would not depend on
> >> CGAL_VERSION, just the implementation of that function.
> >>
> >> The fixes though still will be scattered through the code, because
> >> that function would have to be called everywhere. Just those (really)
> >> annoying #defines will disappear.
> >>
> >> So next step would be to figure out how to make this conversion
> >> implicit back like in older CGAL. I'm afraid that CGAL will not agree
> >> easily on that.
> >>
> >> best regards
> >> Janek
> >>
> >>
> >>
> >>
> >>
> >>
> >> _______________________________________________
> >> Mailing list: https://launchpad.net/~yade-dev
> >> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> >> Unsubscribe : https://launchpad.net/~yade-dev
> >> More help   : https://help.launchpad.net/ListHelp
> >>
> >
> >
> >
> > _______________________________________________
> > Mailing list: https://launchpad.net/~yade-dev
> > Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> > Unsubscribe : https://launchpad.net/~yade-dev
> > More help   : https://help.launchpad.net/ListHelp
> >
> 
> 
> 
> _______________________________________________
> Mailing list: https://launchpad.net/~yade-dev
> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~yade-dev
> More help   : https://help.launchpad.net/ListHelp


-- 
Janek Kozicki                               http://janek.kozicki.pl/  |