[GRASS5] gis.m

On Feb 27, 2006, at 6:19 PM, Michael Barton wrote:

The new gism does not use d.what.vect, but the new v.what. I’m not sure if

it will show information in different layers or not. But that is something

that should be fixable.

I am looking at the v.what and it may be just me, but I don’t see that it is a replacement for d.what.vect in that d.what.vect -e allows the user to interactively click on a polygon and have a form open with that polygon’s attributes visible in an editable form. If an attribute has not not been assigned yet (as in the case of a new attribute column), d.what.vect -e can be used to “fill in” attributes. I’m just not seeing this kind of interactive control in v.what. Am I missing something here?

BTW, I’ve now fixed the show columns and show data buttons in the vector,

chart, and thematic panels so that they respect current database connections

and layer settings. So you you can see the columns and data associated with

your vector file.

Also the “show attribute data” button does not seem to be working for me. While the “show attibue colmns” butotn returns the output of v.info -c map=t61v1_working layer=1to the output window:

INTEGER|cat
INTEGER|id
CHARACTER|type
CHARACTER|class

The “show attribute data” button returns:
GRASS 6.1.cvs (minnesota_utm):~ > table=t61v1_working database=host=localhost,dbname=ts2325 driver=pg
2

It seems that the problem is the 2 at the end of the return. I get the expected output if I send the first part of the command from the command line:
GRASS 6.1.cvs (minnesota_utm):~ > db.select table=t61v1_working database=host=localhost,dbname=ts2325 driver=pg

Michael, this may be due to the fact that I am using pg for attibute management. Just thought you should know

Kirk

Kirk,

Thanks for continuing to test the new GIS Manager. V.what is a replacement for d.what vect because d.what.vect REQUIRES you to interact with a xdisplay (tektronic emulation) to work. V.what does not require this. It is like r.what. It takes input xy coordinates and returns the attributes linked to the vector feature at that coordinate. V.what is a bit more flexible than d.what.vect in that you can set the ‘snapping’ distance. More importantly, v.what can be used in any GUI and is not restricted to xdisplays, but returns the same kind of information as d.what.vect. We need to have GRASS modules that work with interfaces other than the old xmonitors if we want to build a modern interface for GRASS. V.what does this.

The –e flag in d.what.vect invokes a separate program, the “form library” which includes a TclTk form (and maybe associated C-code? I’m not sure on this). This is what allows you to edit the associated attributes for a vector object. Theoretically, v.what—or TclTk or other GUI code that calls v.what—could also invoke the form library (or something like it) to edit a record. This would be very handy, though perhaps other methods might be better—such as doing a query that selected the associated attribute record in a GUI table manager like SQLite Browser. IMHO, GRASS needs these kind of attribute management tools. Moving to a new display model that does not depend on a specific, antiquated display technology, will give us much more flexibility to develop such tools.

Michael


Michael Barton, Professor of Anthropology
School of Human Evolution and Social Change
Arizona State University
Tempe, AZ 85287-2402

phone: 480-965-6213
fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton


From: “Kirk R. Wythers” kwythers@umn.edu
Date: Wed, 1 Mar 2006 09:09:31 -0600
To: Michael Barton michael.barton@asu.edu
Cc: devel grass grass5@grass.itc.it
Subject: gis.m

On Feb 27, 2006, at 6:19 PM, Michael Barton wrote:

The new gism does not use d.what.vect, but the new v.what. I’m not sure if

it will show information in different layers or not. But that is something

that should be fixable.

I am looking at the v.what and it may be just me, but I don’t see that it is a replacement for d.what.vect in that d.what.vect -e allows the user to interactively click on a polygon and have a form open with that polygon’s attributes visible in an editable form. If an attribute has not not been assigned yet (as in the case of a new attribute column), d.what.vect -e can be used to “fill in” attributes. I’m just not seeing this kind of interactive control in v.what. Am I missing something here?

BTW, I’ve now fixed the show columns and show data buttons in the vector,

chart, and thematic panels so that they respect current database connections

and layer settings. So you you can see the columns and data associated with

your vector file.

Also the “show attribute data” button does not seem to be working for me. While the “show attibue colmns” butotn returns the output of v.info -c map=t61v1_working layer=1to the output window:

INTEGER|cat
INTEGER|id
CHARACTER|type
CHARACTER|class

The “show attribute data” button returns:
GRASS 6.1.cvs (minnesota_utm):~ > table=t61v1_working database=host=localhost,dbname=ts2325 driver=pg
2

It seems that the problem is the 2 at the end of the return. I get the expected output if I send the first part of the command from the command line:
GRASS 6.1.cvs (minnesota_utm):~ > db.select table=t61v1_working database=host=localhost,dbname=ts2325 driver=pg

Michael, this may be due to the fact that I am using pg for attibute management. Just thought you should know

Kirk

Kirk,

Just saw the part of your message below. The issue probably is how to correctly parse database option for your pg database. Maybe I can try a couple options and you can test, as I don’t have anything set up in pg at the moment.

Michael


Michael Barton, Professor of Anthropology
School of Human Evolution and Social Change
Arizona State University
Tempe, AZ 85287-2402

phone: 480-965-6213
fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton

From: “Kirk R. Wythers” kwythers@umn.edu
Date: Wed, 1 Mar 2006 09:09:31 -0600
To: Michael Barton michael.barton@asu.edu
Cc: devel grass grass5@grass.itc.it
Subject: [GRASS5] gis.m

Also the “show attribute data” button does not seem to be working for me. While the “show attibue colmns” butotn returns the output of v.info -c map=t61v1_working layer=1to the output window:

INTEGER|cat
INTEGER|id
CHARACTER|type
CHARACTER|class

The “show attribute data” button returns:
GRASS 6.1.cvs (minnesota_utm):~ > table=t61v1_working database=host=localhost,dbname=ts2325 driver=pg
2

It seems that the problem is the 2 at the end of the return. I get the expected output if I send the first part of the command from the command line:
GRASS 6.1.cvs (minnesota_utm):~ > db.select table=t61v1_working database=host=localhost,dbname=ts2325 driver=pg

Michael, this may be due to the fact that I am using pg for attibute management. Just thought you should know

Kirk

On Mar 1, 2006, at 11:08 AM, Michael Barton wrote:

Kirk,

Just saw the part of your message below. The issue probably is how to correctly parse database option for your pg database. Maybe I can try a couple options and you can test, as I don’t have anything set up in pg at the moment.

I’d be happy to test. Just let me know.

Kirk

Michael Barton wrote:

The ­e flag in d.what.vect invokes a separate program, the ³form library²
which includes a TclTk form (and maybe associated C-code? I¹m not sure on
this).

The form utility ($GISBASE/etc/form/form) is a custom Tcl/Tk
interpreter, i.e. a mix of C and Tcl/Tk. The C code mostly relates to
interfacing with the GRASS database libraries.

The form library includes functions for starting and stopping the form
utility, as well as for generating HTML forms from database tables.

It may be possible to run the form utility directly from gis.m to
provide functionality similar to "d.what.vect -e", although it would
probably be better to bypass it altogether and do the database access
from Tcl via the db.* utilities.

--
Glynn Clements <glynn@gclements.plus.com>

This is what I was thinking too.

Michael
______________________________
Michael Barton, Professor of Anthropology
School of Human Evolution and Social Change
Arizona State University
Tempe, AZ 85287-2402
USA

voice: 480-965-6262; fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton

From: Glynn Clements <glynn@gclements.plus.com>
Date: Wed, 1 Mar 2006 18:33:06 +0000
To: Michael Barton <michael.barton@asu.edu>
Cc: "Kirk R. Wythers" <kwythers@umn.edu>, devel grass <grass5@grass.itc.it>
Subject: Re: [GRASS5] Re: gis.m

Michael Barton wrote:

The ­e flag in d.what.vect invokes a separate program, the ³form library²
which includes a TclTk form (and maybe associated C-code? I¹m not sure on
this).

The form utility ($GISBASE/etc/form/form) is a custom Tcl/Tk
interpreter, i.e. a mix of C and Tcl/Tk. The C code mostly relates to
interfacing with the GRASS database libraries.

The form library includes functions for starting and stopping the form
utility, as well as for generating HTML forms from database tables.

It may be possible to run the form utility directly from gis.m to
provide functionality similar to "d.what.vect -e", although it would
probably be better to bypass it altogether and do the database access
from Tcl via the db.* utilities.

--
Glynn Clements <glynn@gclements.plus.com>