Hallo,
when I try to use gis.m for displaying maps, the maps do not use the entire area of the gis.m canvas
see figures http://les-ejk.cz/tmp/fig1.png and http://les-ejk.cz/tmp/fig2.png
what is wrong?
jachym
Jachym Cepicky wrote:
when I try to use gis.m for displaying maps, the maps do not use the
entire area of the gis.m canvas
see figures http://les-ejk.cz/tmp/fig1.png and
http://les-ejk.cz/tmp/fig2.pngwhat is wrong?
My first guess would be that the canvas has a different aspect ratio
to the region. AFAIK, gis.m won't automatically enlarge the region to
match the aspect ratio used for display.
--
Glynn Clements <glynn@gclements.plus.com>
that could be it. thanks
jachym
--
Jachym Cepicky
e-mail: jachym.cepicky@centrum.cz
URL: http://les-ejk.cz
GPG: http://les-ejk.cz/gnupg_public_key/
-----------------------------------------
OFFICE:
Department of Geoinformation Technologies
LDF MZLU v Brnì
Zemìdìlská 3
613 00 Brno
e-mail: xcepicky@node.mendelu.cz
URL: http://mapserver.mendelu.cz
Tel.: +420 545 134 514
There is nothing wrong.
The geometry of the area displayed is determined by the region (g.region
and/or zoom), not the geometry of the canvas. gis.m will attempt to fill as
much of the canvas as possible given the geometry of the region, while
showing ALL of the region defined. I assume that people want to see all of a
defined region rather than have a filled canvas that cuts off part of a
region.
So, a tall, narrow canvas and a short, wide region will appear as a short
map extending across the width of the narrow canvas.
The xdisplays behaved in exactly the same way. The only difference now is
that the displayed map is aligned with the upper left of the canvas rather
than centered. This is for easier and more accurate coordinate conversions
between the screen and geographic coordinates.
Note that graphic output matches the *region*, not the canvas.
On a different note, the lack of zooming (recentering only) was a bug that
was fixed several weeks ago. You need to update.
Hope this is helpful.
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: Jachym Cepicky <jachym.cepicky@centrum.cz>
Date: Tue, 21 Mar 2006 14:41:14 +0100
To: <grass5@grass.itc.it>, <jachym.cepicky@centrum.cz>
Subject: [GRASS5] gis.m does not use whole areaHallo,
when I try to use gis.m for displaying maps, the maps do not use the
entire area of the gis.m canvas
see figures http://les-ejk.cz/tmp/fig1.png and
http://les-ejk.cz/tmp/fig2.pngwhat is wrong?
jachym
Note that graphic output matches the *region*, not the canvas.
can we try a grey & hatched background (black "X" diagonal lines
crossing) to indicate what's NULL and what's out of bounds? Or
alternating block pattern like GIMP uses? Or something like that.
This has been a confusing part of GRASS for a Long long time.
?
Hamish
On Wed, Mar 22, 2006 at 01:07:42PM +1200, Hamish wrote:
> Note that graphic output matches the *region*, not the canvas.
can we try a grey & hatched background (black "X" diagonal lines
crossing) to indicate what's NULL and what's out of bounds? Or
alternating block pattern like GIMP uses? Or something like that.
This has been a confusing part of GRASS for a Long long time.?
Hamish
That would be nice ... the new graphical interface looks so good, that
it is really a bit confusing... Idrisi shows new map always in window,
which is resised according to displayed map. In GRASS this could be done
for the default region. Would this be sollution?
Jachym
--
Jachym Cepicky
e-mail: jachym.cepicky@centrum.cz
URL: http://les-ejk.cz
GPG: http://les-ejk.cz/gnupg_public_key/
-----------------------------------------
OFFICE:
Department of Geoinformation Technologies
LDF MZLU v Brnì
Zemìdìlská 3
613 00 Brno
e-mail: xcepicky@node.mendelu.cz
URL: http://mapserver.mendelu.cz
Tel.: +420 545 134 514
I'll see what's possible. I guess I'm not all that confused. What is blank
white is not in the map.
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: Hamish <hamish_nospam@yahoo.com>
Date: Wed, 22 Mar 2006 13:07:42 +1200
To: Michael Barton <michael.barton@asu.edu>
Cc: <jachym.cepicky@centrum.cz>, <grass5@grass.itc.it>
Subject: Re: [GRASS5] gis.m does not use whole areaNote that graphic output matches the *region*, not the canvas.
can we try a grey & hatched background (black "X" diagonal lines
crossing) to indicate what's NULL and what's out of bounds? Or
alternating block pattern like GIMP uses? Or something like that.
This has been a confusing part of GRASS for a Long long time.?
Hamish
I haven't seen how Idrisi does this. I can imagine how to do it in TclTk,
but it seems that it could be potentially quite frustrating to a user trying
to resize a window. You set a window size and it would snap back in either
the horizontal or vertical dimension to something that fits the region
geometry. Remember, any output is determined by the region, not canvas
geometry.
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: Jachym Cepicky <jachym.cepicky@centrum.cz>
Date: Wed, 22 Mar 2006 10:09:15 +0100
To: Hamish <hamish_nospam@yahoo.com>
Cc: <grass5@grass.itc.it>
Subject: Re: [GRASS5] gis.m does not use whole areaOn Wed, Mar 22, 2006 at 01:07:42PM +1200, Hamish wrote:
Note that graphic output matches the *region*, not the canvas.
can we try a grey & hatched background (black "X" diagonal lines
crossing) to indicate what's NULL and what's out of bounds? Or
alternating block pattern like GIMP uses? Or something like that.
This has been a confusing part of GRASS for a Long long time.?
HamishThat would be nice ... the new graphical interface looks so good, that
it is really a bit confusing... Idrisi shows new map always in window,
which is resised according to displayed map. In GRASS this could be done
for the default region. Would this be sollution?Jachym
--
Jachym Cepicky
e-mail: jachym.cepicky@centrum.cz
URL: http://les-ejk.cz
GPG: http://les-ejk.cz/gnupg_public_key/
-----------------------------------------
OFFICE:
Department of Geoinformation Technologies
LDF MZLU v Brnì
Zemìdìlská 3
613 00 Brno
e-mail: xcepicky@node.mendelu.cz
URL: http://mapserver.mendelu.cz
Tel.: +420 545 134 514
Hamish wrote:
> Note that graphic output matches the *region*, not the canvas.
can we try a grey & hatched background (black "X" diagonal lines
crossing) to indicate what's NULL and what's out of bounds? Or
alternating block pattern like GIMP uses? Or something like that.
This has been a confusing part of GRASS for a Long long time.
Tk doesn't support background pixmaps, but it would be possible to add
a rectangle object with a -stipple option at the bottom of the canvas'
object stack; the rectangle would need to be resized (or re-created)
whenever the canvas was resized.
--
Glynn Clements <glynn@gclements.plus.com>
> can we try a grey & hatched background (black "X" diagonal lines
> crossing) to indicate what's NULL and what's out of bounds? Or
> alternating block pattern like GIMP uses? Or something like that.
> This has been a confusing part of GRASS for a Long long time.That would be nice ... the new graphical interface looks so good, that
it is really a bit confusing... Idrisi shows new map always in window,
which is resised according to displayed map. In GRASS this could be
done for the default region. Would this be sollution?
I always found the old NVIZ controls jumping around & resizing
themselves very distracting.
bonus feature: If checkered blocks are used, make their size as the
closest power of 10 to the map units. Sort of an undocumented barscale.
"cute" but useful.
Hamish
> This has been a confusing part of GRASS for a Long long time.
I'll see what's possible. I guess I'm not all that confused. What is
blank white is not in the map.
It isn't a problem when you are dealing with imagery data that covers
the full region, then it's pretty clear. When you are working with
isolated vectors or raster data as islands-in-the-NULL, then it isn't
always clear why the blob or line stops an inch from the edge of the
pane. Any edge-of-the-world hint need not be obtrusive, 10% alpha on
white background might be just enough .. This is especially a pain when
doing something interactive, e.g. historically with v.digit.
Hamish
OK. I see your problem. As Glynn, said, this is not as simple as I might
want. I will try making the canvas background a slightly different color and
see what happens.
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: Hamish <hamish_nospam@yahoo.com>
Date: Thu, 23 Mar 2006 11:40:45 +1200
To: Michael Barton <michael.barton@asu.edu>
Cc: <jachym.cepicky@centrum.cz>, <grass5@grass.itc.it>
Subject: Re: [GRASS5] gis.m does not use whole areaThis has been a confusing part of GRASS for a Long long time.
I'll see what's possible. I guess I'm not all that confused. What is
blank white is not in the map.It isn't a problem when you are dealing with imagery data that covers
the full region, then it's pretty clear. When you are working with
isolated vectors or raster data as islands-in-the-NULL, then it isn't
always clear why the blob or line stops an inch from the edge of the
pane. Any edge-of-the-world hint need not be obtrusive, 10% alpha on
white background might be just enough .. This is especially a pain when
doing something interactive, e.g. historically with v.digit.Hamish
Hamish wrote:
> > can we try a grey & hatched background (black "X" diagonal lines
> > crossing) to indicate what's NULL and what's out of bounds? Or
> > alternating block pattern like GIMP uses? Or something like that.
> > This has been a confusing part of GRASS for a Long long time.
>
> That would be nice ... the new graphical interface looks so good, that
> it is really a bit confusing... Idrisi shows new map always in window,
> which is resised according to displayed map. In GRASS this could be
> done for the default region. Would this be sollution?I always found the old NVIZ controls jumping around & resizing
themselves very distracting.bonus feature: If checkered blocks are used, make their size as the
closest power of 10 to the map units. Sort of an undocumented barscale.
"cute" but useful.
And awkward. If you use a stipple (tile), the repeat frequency is
constrained to be an integral number of pixels.
By far the easiest solution is to just change the canvas' background
from white to light grey.
--
Glynn Clements <glynn@gclements.plus.com>
From: Glynn Clements <glynn@gclements.plus.com>
Date: Thu, 23 Mar 2006 13:57:10 +0000
To: Hamish <hamish_nospam@yahoo.com>
Cc: Jachym Cepicky <jachym.cepicky@centrum.cz>, <grass5@grass.itc.it>
Subject: Re: [GRASS5] gis.m does not use whole areaHamish wrote:
can we try a grey & hatched background (black "X" diagonal lines
crossing) to indicate what's NULL and what's out of bounds? Or
alternating block pattern like GIMP uses? Or something like that.
This has been a confusing part of GRASS for a Long long time.That would be nice ... the new graphical interface looks so good, that
it is really a bit confusing... Idrisi shows new map always in window,
which is resised according to displayed map. In GRASS this could be
done for the default region. Would this be sollution?I always found the old NVIZ controls jumping around & resizing
themselves very distracting.bonus feature: If checkered blocks are used, make their size as the
closest power of 10 to the map units. Sort of an undocumented barscale.
"cute" but useful.And awkward. If you use a stipple (tile), the repeat frequency is
constrained to be an integral number of pixels.By far the easiest solution is to just change the canvas' background
from white to light grey.
This is what I've done.
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