[josm-dev] JOSM extension for orthogonalizing ways

[Harald Kucharek]

-------- Original-Nachricht --------
> Datum: Fri, 31 Oct 2008 09:11:46 +0000
> Von: "Brian Quinion" <brian at brian.quinion.co.uk>
> An: "Frederik Ramm" <frederik at remote.org>
> CC: harald.kucharek at gmx.net, josm-dev at openstreetmap.org
> Betreff: Re: [josm-dev] JOSM extension for orthogonalizing ways

> Hi,
> 
> Sounds good - I keep hitting problems where I have a rectangle that
> has an extra point on on of the edges and so can't easily use the
> existing functionality.

Hmmm, I guess, this case would also not be accepted by my function, as one of the prerequisites is, that every segment must have a distinct direction change compared to the segment before. Otherwise I had to handle the case of collinear segments or segments where direction changes only slightly. Would be difficult to decide what to do with those. But I'm open for any ideas.

> 
> One questions (sorry - I've not ready the code!).  Does your version
> align to what's on the ground or in the current projection? This is my
> big problem with the current implementation - it creates a rectangle
> in the current projection with the result that (as you go further
> north / south) the rectangle become more and more misaligned to
> physical reality.
> 
> (http://josm.openstreetmap.de/ticket/1597)

I hadn't given thought to this. I work with the coordinates data I find in the EastNorth-nodes and don't care about pojection issues.
> 
> --
>  Brian
> 
> On Fri, Oct 31, 2008 at 7:28 AM, Frederik Ramm <frederik at remote.org>
> wrote:
> > Hi,
> >
> > Harald Kucharek wrote:
> >> Now, my question is: Is there interest by the maintainers of JOSM to
> put
> >> my function into JOSM, maybe replacing the current "aligning nodes in a
> >> rectangle" function of JOSM?
> >
> > Sure, sounds great. I assume if used on a 5-node way it will roughly do
> > the same as the current align-in-rectangle, right?
> >
> > I did the align-in-recangle initially, and I have been asked by users
> > whether it would be possible to, instead of have the algorithm compute
> > the general direction, pre-select one edge somehow for all others to
> > follow. I haven't had the time to look into that though, maybe it is
> > something you want to tackle on your next holiday ;-)
> >
> > I'll look at the patch and apply it if I don't see a major problem.
> >
> > Bye
> > Frederik
> >
> > --
> > Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09"
> E008°23'33"
> >
> > _______________________________________________
> > josm-dev mailing list
> > josm-dev at openstreetmap.org
> > http://lists.openstreetmap.org/listinfo/josm-dev
> >

-- 
Harald Kucharek  | harald.kucharek at gmx.net
Werderstrasse 24 | FON:+49(721)389566
76137 Karlsruhe  | FAX/VoiceMail:+49(12120) 211259
Germany          | Mobile:+49(174)2458743