On Wednesday 14 October 2009, Dylan Beaudette wrote: > Great! Thanks for the tips. I was able to accomplish this (in PostGresql) > with the following steps: > > 1. generate unique list of text categories as a temp table > 2. add a serial column to the temp table (auto-incrementing integer) > 3. add new column to original table > 4. UPDATE original table with the new serial number, assigned by text > category 5. v.to.rast > > Cheers, > Dylan
One small note-- I noticed that the operation takes MUCH longer to complete when the labelcolumn argument is specified. So long in fact, that I canceled the process, and converted only the integer representation to raster. I then added the category labels with r.category. Something in v.to.rast was causing the postgresql instance to peg the CPU at 100% when specifying a labelcolumn. This was for a rather large vector conversion, about 410520 features. Cheers, Dylan > On Wednesday 14 October 2009, Markus Neteler wrote: > > IMHO ticket 175 has been fixed. I can now run (thanks to Martin): > > > > v.db.select lsat7_training > > cat|name|id > > 1|water|1 > > 3|forest|2 > > 4|forest|2 > > 5|asphalt|3 > > 6|asphalt|3 > > 7|water|1 > > ... > > v.to.rast in=lsat7_training out=lsat7_training use=attr col=id > > labelcol=name > > > > As Achim says: col must be numerical and should be the numerical > > representation of the text attributes (see example above). > > > > Markus > > > > On Wed, Oct 14, 2009 at 1:39 PM, Hamish <[email protected]> wrote: > > > see also https://trac.osgeo.org/grass/ticket/175 > > > > > > Achim wrote: > > >> v.to.rast has the option: > > >> > > >> --%<-- > > >> labelcolumn=name > > >> Name of column used as raster category > > >> labels > > >> --%<-- > > >> > > >> And the raster values have to be numeric: > > >> > > >> --%<-- > > >> column=name > > >> Name of column for attr parameter (data type > > >> must be numeric) > > >> --%<-- > > >> > > >> (http://grass.itc.it/grass64/manuals/html64_user/v.to.rast.html) > > >> > > >> Try to make unique numbers from your string-column and > > >> convert them with > > >> column=... and labelcolumn... > > >> > > >> Achim > > >> > > >> Dylan Beaudette schrieb: > > >> > Hi, > > >> > > > >> > v.to.rast is letting me know that it does not accept > > >> > > >> text columns in an > > >> > > >> > attribute table. Has anyone figured out a simple > > >> > > >> approach to making this > > >> > > >> > work? I suppose that I could import the data into R, > > >> > > >> convert to factors, save > > >> > > >> > the levels back to the table and use those... but > > >> > > >> there may be something out > > >> > > >> > there that is simpler and can be done within GRASS. > > >> > > >> Also, would this be > > >> > > >> > something worth adding to the v.to.rast code? > > >> > > > >> > Cheers, > > >> > Dylan > > > > > > _______________________________________________ > > > grass-user mailing list > > > [email protected] > > > http://lists.osgeo.org/mailman/listinfo/grass-user -- Dylan Beaudette Soil Resource Laboratory http://casoilresource.lawr.ucdavis.edu/ University of California at Davis 530.754.7341 _______________________________________________ grass-user mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-user
