Eric Patton wrote: > > Just a clarification...so it's the case that the column parameter > > in v.dissolve must always be of type integer, and *only* integer?
Hamish: >The v.dissolve script with dissolve based on a column currently >requires that column to be of type integer. We should probably add a >check in the script for that (v.info -c). Ok, that's what I thought; thanks. I wanted to make sure before I added a hint in the docs. Actually, I think Nikos mentioned something about writing up a hint about it, so I'll hold off. >It would be possible to add any type of column to v.dissolve with unix >tools 'db.select | sort | uniq | nl' then create a new table with the >unique attributes + sequential integer column, then v.reclass as >before. >or add a new column to (a copy of) the existing table (v.db.addcol) and >populate it with a 'db.select | sort | uniq' loop and v.db.update, or >SQL + db.execute << "set DISOLVEKEY = $i where ATTRIBUTE = '$VAL'" >(sorry my SQL is not good but you may be the idea of what I mean) Cool. It's always amazing how powerful these little command line tools can be. ~ Eric. _______________________________________________ grass-user mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-user
