[GRASS-user] Joining vectors by location

Daniel Lee lee at isi-solutions.org
Thu Nov 24 11:18:35 EST 2011


>
> IIRC, geography_columns is a view, not sure why you would even want to
> access that...
>
> I don't have time right now to try to find out where the error message
> comes from. Can you give us the exact v.in.ogr command you use ?
>
Gotcha. I think it is a table, actually:

test=# \dt
             List of relations
 Schema |       Name       | Type  | Owner
--------+------------------+-------+-------
 public | geometry_columns | table | lee
 public | spatial_ref_sys  | table | lee
(2 rows)

This is my test DB that I'm using to kind of experiment and those are the
two tables that were produced as I was setting up the PostGIS DB.
Nonetheless, I don't necessarily want to access those tables, I just want
to upload my vectors. This is the v.out.ogr command I used:
v.out.ogr in=Gebaeude_Globalstrahlung dsn=’PG:host=localhost dbname=test
user=lee password=XXX type=area format=PostgreSQL’


>
Could it be that you're confusing two things:
> 1- Configure GRASS to use a PostgreSQL database for attribute data
> vs
> 2- Move your vector layers out to a PostGIS spatial database
>
> These are not the same. If you want to export a GRASS layer out to a
> PostGIS DB then v.out.ogr is the right module.
> If, on the other hand, you want to keep your data as GRASS vectors, but
> move the *attributes* to PostgreSQL then do:
> (straight from the man page...)
>     db.connect driver=pg database="host=myserver.itc.it,dbname=mydb"
>     db.login user=myname [pass=secret]
>     db.connect -p
> This will set your default database connection to PostgreSQL, instead of
> dbf. Now you should copy your vectors to a new name in the mapset, and let
> GRASS make the Postgresql attribute tables.
> g.copy <vector1>,<newvector1>
> The new copies will still be GRASS vectors, but with their attribs in
> PostgreSQL database.
>
Okay, I hadn't understood the difference so well, but I think I get it a
little bit more. The table join would also work in PostgreSQL, that's true,
although I would rather have the data as PostGIS vectors. The reason I'd
like PostGIS vectors is because I want to show them with a Geoserver that
accesses PostGIS layers.


>
>
>
> ERROR 1: ERROR: permission denied for relation geography_columns
>
>
> This is a permissions prob within PostGIS... You need to GRANT ALL ON
> geography_columns TO <username>;
>
Just tried it. As my username (lee) I got:
test=> GRANT ALL ON geography_columns TO lee;
ERROR:  permission denied for relation geography_columns

I tried it again then as postgres:
test=# GRANT ALL ON geography_columns TO lee;
GRANT

Looked all right, but GRASS still gives me the same error that I don't have
permission for the relation geography_columns.

Any ideas? I also considered exporting everything as a shapefile and then
using shp2pgsql to import everything, but that's not a great solutions...
And plus, shapefiles have a restricted size, which could make things
difficult with large datasets (some of the data I work with is very large).

Thanks a lot for the help, I really appreciate it!

Daniel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/grass-user/attachments/20111124/64a8fdcd/attachment.html


More information about the grass-user mailing list