v.in.ogr key flag does not change category values and you loose connection between features and attributes

[mlennert]

Currently, v.in.ogr internally creates cat values for an imported vector. It would be great if one could tell v.in.ogr to use an existing attribute column as cat value, including when importing using the -t flag to not create any attribute table within grass.

[martinl]

v.in.ogr in GRASS 7 has key parameter which allows you to define key column,

[mlennert]

Replying to martinl:

v.in.ogr in GRASS 7 has key parameter which allows you to define key column,

Yes, but apparently it doesn't work:

g.copy vect=boundary_county,bc
v.db.addcol bc col="newcat int"
v.db.addcolumn bc col="newcat int"
v.db.update bc col=newcat qcol=cat+1000
v.out.ogr bc out=bc
v.in.ogr bc/bc.shp out=bc_test key=newcat
v.db.connect -p bc_test
v.db.univar bc_test col=newcat
v.category bc_test op=report

You will see that the db connection is correctly defined with the 'newcat' column as the key, but the features' category values are still automatically generated from 1. You thus loose the connection between features and attributes !

Redefining this as a bug and reopening.

Moritz

[martinl]

In 69241:

v.in.ogr key flag does not change category values and you loose connection between features and attributes (see #1494)

[martinl]

Replying to mlennert:

Redefining this as a bug and reopening.

Please try r69241.

[mlennert]

Replying to martinl:

Replying to mlennert:

Redefining this as a bug and reopening.

Please try r69241.

Seems to work perfectly. Thanks !

Leaving open until backport.

[martinl]

In 69250:

v.in.ogr key flag does not change category values and you loose connection between features and attributes (see #1494) (merge r69241 from trunk)

[martinl]

In 69251:

v.in.ogr key flag does not change category values and you loose connection between features and attributes (fix #1494) (merge r69241 from trunk)