[Edbsilon-list] Error during conversation

[Stephan Holl]

Hi Jan-Oliver,

Am Dienstag, 12. Dezember 2006 01:58 schrieb "Jan-Oliver Wagner" 
<jan-oliver.wagner at intevation.de>:
> Hallo Stephan,
>
> On Monday 11 December 2006 10:29, Stephan Holl wrote:
> > I have checked out the edbsilon-code from CVS and run it against my
> > edbs-file.
> >
> > edbsilon died with the following error:
> > Lese UeberseeStadt.edb
> > Traceback (most recent call last):
> >   File "./edbsilon/Konverter/edbsilon.py", line 184, in ?
> >     parsen(fc.ONS, quelle, edbsformat)
> >   File "./edbsilon/Konverter/parse.py", line 204, in parsen
> >     return second_parsing(ONS, liste, edbsformat)
> >   File "./edbsilon/Konverter/parse.py", line 82, in second_parsing
> >     II_daten = e.get_II_data(edbsformat)
> >   File "./edbsilon/Konverter/edbsclasses.py", line 989, in get_II_data
> >     ainfo.paralist[0])
> >   File "./edbsilon/Konverter/edbsclasses.py", line 1157, in arc2wkt
> >     beta = getwinkel((bx/r, by/r))
> >   File "./edbsilon/Konverter/edbsclasses.py", line 1131, in getwinkel
> >     return acos(vx)
> > ValueError: math domain error
> >
> > Perhaps someone has a pointer where to look at?
>
> well, I just modified edbsclasses.py to ensure vx is <= 1 -- which
> is not the case for line 18474 where the error occurs in your
> file (which I have access to, fortunately ;-)
>
> The patch is attached - but be aware it is probably not
> the right solution.

Well it gives another error.

> The next problem I then observe is that no definition for Folie 066
> is found. Seems, Brandenburg does not use it.

Right. In a first step I do not want to adopt the config just to see what 
comes out.

> Then this error occurs:
>
> Traceback (most recent call last):
>   File "edbsilon.py", line 185, in ?
>     fc.export()
>   File "/powerhome/jan/project/edbsilon/Konverter/exportOGR.py", line 124,
> in export
>     self.convertONS()
>   File "/powerhome/jan/project/edbsilon/Konverter/exportOGR.py", line 62,
> in convertONS
>     self.readObj(self.ONS[Objnr])
>   File "/powerhome/jan/project/edbsilon/Konverter/exportOGR.py", line 94,
> in readObj
>     self.exportObjekt(obj)
>   File "/powerhome/jan/project/edbsilon/Konverter/exportOGR.py", line 205,
> in exportObjekt
>     geom = getLinie(obj)
>   File "/powerhome/jan/project/edbsilon/Konverter/exportOGR.py", line 318,
> in getLinie
>     return ogr.CreateGeometryFromWkt('MULTILINESTRING(%s)'
> %join(multiline)) File "/usr/lib/python2.3/site-packages/ogr.py", line 970,
> in
> CreateGeometryFromWkt
>     _obj = _gdal.OGR_G_CreateFromWkt( string, srs_o )
> ValueError: Corrupt WKT geometry passed to OGR_G_CreateFromWkt.
>
>
> which is caused by this multiline:
> ['(3485710.497 5883739.401,3485710.997 5883739.399)', '(3485721.265
> 5883736.618,3485710.994 5883739.669)']
>
> which is the first one to be in fact a multiline! All other multlines so
> far were just lines.

There seems to be something wrong within the detection/parsing/formating of 
multilinestrings.

I temporarily commented out line 
318 (  multiline.append('(%s)' % join(coordAsList,','))  ) in 
edbsilon/Konverter/exportOGR.py. This results in wrong line and point shapes, 
but the polygone-shapes get build. Please see this only as a fast workarround 
to see at least some results.

Best
	Stephan


-- 
Stephan Holl   : www.intevation.de/~stephan | GISpatcher: www.gispatcher.de
Intevation GmbH: www.intevation.de          | GAV e.V.: www.grass-verein.de
Georgstr.4     : 49074 Osnabrück            | Telefon:   +49(0)541. 3350832