On 16/09/17 18:22, Barry Rowlingson wrote: > > > On Sat, Sep 16, 2017 at 3:45 PM, Edzer Pebesma > <edzer.pebe...@uni-muenster.de <mailto:edzer.pebe...@uni-muenster.de>> > wrote: > > > > On 16/09/17 15:20, Eduardo Diez wrote: > > Dear list, > > I have a MULTIPOLYGON geometry set where the rings are in the wrong > > direction and would like to correct that with *st_make_valid*. As I'm in > > Windows and use the binary version of the *sf* package it gives me the > > error: > > > >> sp2 = st_make_valid(sp1) > > Error in CPL_make_valid(x) : > > st_make_valid requires compilation against liblwgeom > > > > Is there a series of steps I can follow for compiling *sf* in Windows > > linked to liblwgeom? > > No, and I also don't think that st_make_valid does this. > > > st_make_valid certainly did something: > > https://gis.stackexchange.com/questions/254836/sf-lines-to-polygons-with-holes
I only tried a single outer ring in both directions, and it didn't revert them. Maybe it only reverts if if outer and inner ring have the same direction. > > I've just added an edit to my answer there with the sp-and-back round > trip, which seems to work, but its not exactly the most satisfying solution. It would be fairly trivial to move some of the code to sf - at a performance penalty if we'd do the check always, like sp does. > > What's the problem with liblwgeom on Windows? It's not part of the rwinlib collection at https://github.com/rwinlib ; I'll bring it up with Jeroen next time I see him. lwgeom has more good things than what sf uses. > > Barry > -- Edzer Pebesma Institute for Geoinformatics (ifgi), University of Münster Heisenbergstraße 2, 48149 Münster, Germany; +49 251 83 33081 Journal of Statistical Software: http://www.jstatsoft.org/ Computers & Geosciences: http://elsevier.com/locate/cageo/
signature.asc
Description: OpenPGP digital signature
_______________________________________________ R-sig-Geo mailing list R-sig-Geo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo