Re: [libredwg] Fwd: please help transform of coords for Prishtina DXF

[Felipe Sanches]

the issue is not clear to me, how does that involve LibreDWG?

On Sun, Jan 24, 2010 at 10:16 AM, address@hidden
<address@hidden> wrote:
> here is another test case
>
>
> ---------- Forwarded message ----------
> From: address@hidden <address@hidden>
> Date: Sun, Jan 24, 2010 at 1:12 PM
> Subject: please help transform of coords for Prishtina DXF
> To: address@hidden, address@hidden, "PROJ.4 and general
> Projections Discussions" <address@hidden>
>
>
> I have here a dxf file of Prishtina in high quality.
> http://www.archive.org/details/PrishtinaStreets
>
> I have been able to convert it to lat/lon like this
> cs2cs +proj=utm  +zone=34T +ellps=WGS84 +datum=WGS84 +units=m -f \"%.7f\"
>
> Problem is that most of the points are in relative position, there is a grid.
>
> I have pushed a version of two nickels that can process alot of this :
> To address@hidden:h4ck3rm1k3/TwoNickels.git
>   c8e4009..41f23f1  master -> master
>
> It produces osm files with data in three groups :
>
> 1. points that are exact on the mark, (few)
> 2. points on the equator (0,0) relative to something
> 3. points that seem to be shifted over from utm34T
>
>         720496.25, 506444.9375 -> 22.987450699423831679, 4.5791053696411907481
>
> I have created averages of good points versus the shifted ones and
> came up with these constant shifts:
>        $n -= 187380;
>        $e += 4228672;
>
> They are still 15km off of prishtina..
>
>
> So I would like one of you to please help with the autocad and
> transform all coordinates into some single form so I can convert them.
> Or please advice on these points.
>
> thanks,
> mike
>
>
>