[GRASSLIST:102] Re: [GRASS5] Re: New GIS Manager adds transparency to layers

Tue Mar 14 14:35:12 EST 2006

Carlos,

You are actually seeing 2 issues. One is a 'feature' not a bug. The other is a bug. 

First, I have a question. Are you using GRASS binaries for Mac OS X? If not, which OS are  you using?

The 'feature' first. Because the gism2 has region settings for EACH  map display, if you change the region using g.region, you will have to click the 'zoom to current region' button rather than simply redisplay.

The bug is a problem with map layer compositing somehow linked with g.pnmcomp. I'm working with Glynn Clements to debug. So far I've only seen  it on a Mac with the most recent binaries from Lorenzo Moretti. So if  you are using a Mac, this matches what I've seen so far. If you are using a different OS, then it is more widespread.

Here is what I know at that moment. If you have multiple layers to display, the first time you display them, they will all composite correctly. However, if you change a layer setting and redisplay, it will look just like before. Here are a couple of work arounds.

First, it seems like this doesn't happen if you only have 2 layers. Second, if  you have more than 2 layers, deselect all but 2 layers, redisplay, reselect the other layers, redisplay. All should be updated correctly.

I hope that this is cleared up soon.

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

-----Original Message-----
From: Carlos "Guâno" Grohmann [mailto:carlos.grohmann at gmail.com]
Sent: Mon 3/13/2006 11:45 AM
To: GRASS mailing list
Subject: [GRASSLIST:50] Re: [GRASS5] Re: New GIS Manager adds transparency to layers

Hello Michael,

I've been playing with gism and there are some points:

When I make the display window too big (like, the size of my monitor),
some times transparency won't work, some times it wil....

If I change region in terminal (say, g.region rast=myrast), and then
just ask for a redraw, nothing happens! I have to unselect the layers,
"draw" a blank screen and then select the layers and draw again.
Something similar was happening when I was testing some shaded relief
maps. I was changing vertical exaggeration, but the name of the map
was the same, so if I just asked for a redraw, nothing would happen.
I guess this is because the _name_ of the map being displayed hasn't changed.

About the region thing, how about two mor buttons: one for "region to
rast" and one for "region to vector" ?

BTW, you said that the ppm way of displaying would make things
faster... personally, I felt it slower.

all the best

Carlos

On 3/13/06, Thomas Adams <Thomas.Adams at noaa.gov> wrote:
> Michael,
>
> Sorry; I should have tried this first. no problem when I startup from my
> $HOME (login) directory. Typically I like to startup GRASS from my GRASS
> database directory. Is this the anticipated behavior or should it be
> considered a 'bug'?
>
> Thanks for your feedback and great work!
>
> Tom
>
>
> Thomas Adams wrote:
> > Michael,
> >
> > Where is GRASS/new display manager going to look for the .tmp
> > directory? In the directory GRASS is started from; one's $HOME
> > directory; other?
> >
> > Thanks,
> > Tom
> >
> >
> > Michael Barton wrote:
> >> Something similar has just been reported on some lab computers
> >> running GRASS
> >> under Cygwin. Maybe it's a different issue, though because it only
> >> happened
> >> with a version of init.sh that automatically runs gism2. It seems to
> >> run OK
> >> if launched manually.
> >>
> >> Your GRASS in a non-standard location might be an issue. I'm not sure
> >> where
> >> the working directory error comes from. Is it a TclTk error (buttons
> >> run,
> >> cancel, details) or another kind of error (xwindows, bash, etc..)?
> >>
> >> The GIS Manager display works by creating PPM files in a temp directory
> >> within your working GRASS mapset--set by g.tempfile. If you don't have
> >> access to such a .tmp directory, you will have problems.
> >>
> >> Let me know if any of these gives you a clue to what is wrong.
> >>
> >> Michael
> >>
> >>
> >> On 3/10/06 7:22 PM, "Thomas Adams" <Thomas.Adams at noaa.gov> wrote:
> >>
> >>
> >>> Michael,
> >>>
> >>> I have installed the GRASS 6.1 CVS release dated 2006_03_04
> >>>
> >>> and followed your instructions in the README in
> >>> _http://www.public.asu.edu/~cmbarton/files/grass_gismgr/
> >>> <http://www.public.asu.edu/%7Ecmbarton/files/grass_gismgr/>_gism_rc4_20060307.
> >>>
> >>> tgz.
> >>>
> >>> First off, the installation using the old d.m works fine < no errors.
> >>> When I try the new gis.m, the interface tries to come up, but it stalls
> >>> with an error Dialog that reads "Error getting working directory name:
> >>> no such file or directory". If I select the Error Dialog 'OK' button,
> >>> the GIS manager completes its startup. However, if I select a raster
> >>> map
> >>> to display and click on the refresh button (labeled: "Display active
> >>> layers in the current region") in the "Map Display 1" window, I get the
> >>> same Error Dialog as I got when I first started up the new GIS Manager
> >>> (gis.m). Also, from the time of the startup of the GIS Manager, the
> >>> 'status indicator' in the lower left of the "Map Display 1" window
> >>> shows: "please wait..." < this never goes away.
> >>>
> >>> Interestingly, if I select the NVIZ button, there is no problem
> >>> displaying the raster I selected; so, NVIZ seems OK, but the raster map
> >>> will never display in 2-D in the "Map Display 1" windowS
> >>>
> >>> Also, when I try to use the Init.sh file from gism_rc4_20060307.tgz, I
> >>> get shared library problems: "libgrass_gis.so cannot open shared object
> >>> file: No such file or directory" < even though the library is in the
> >>> GRASS lib directory. I should mention that due to system constraints,
> >>> the entire GRASS distribution and other libraries (GDAL, PROJ, etc.)
> >>> are
> >>> in a non-standard location.
> >>>
> >>> A final question: should I be using the gis.m file from
> >>> gism_rc4_20060307.tgz? I have tried both with and withoutS
> >>>
> >>> I'm running on a Redhat Linux system, FWIWS
> >>>
> >>> Tom
> >>>
> >>> Michael Barton wrote:
> >>>
> >>>> I¹ve just finished phase 2 of the GIS Manager update and it¹s pretty
> >>>> exciting. At first glance, it looks a lot like the one I¹ve been
> >>>> working on and have in the CVS now, but with a couple of important
> >>>> exceptions. One is that transparency is now supported for all GRASS
> >>>> layers.
> >>>>
> >>>> The second will probably generate the most discussion, so I¹m ready to
> >>>> duck and run for cover. Now GRASS is consistent with other GIS
> >>>> programs and more intuitive (as much as I am skeptical about that
> >>>> term) in that the uppermost layer in the layer tree is also the top
> >>>> map in the display. Related to this, new layers insert directly above
> >>>> the currently selected layer instead of at the bottom of the layer
> >>>> tree.
> >>>>
> >>>> The main changes come Ounder the hood¹ of this version of the GIS
> >>>> Manager. It uses g.pnmcomp to composite a set of maps (each rendered
> >>>> as *.PPM files) into a visual display. One of the results is that now
> >>>> GRASS is significantly faster in adding new layers, turning layers on
> >>>> and off, and changing layer attributes<only the new or changed layer
> >>>> needs to be rendered; the others will just be re-composited into a
> >>>> display from existing PPM files. This all happens in a GRASS .tmp
> >>>> directory, so that it doesn¹t clutter up your home directories. The
> >>>> other important thing about this new architecture is that it makes
> >>>> GRASS even less dependent on the xdriver and PNGdriver for
> >>>> visualization. GRASS can become more easily portable to different OS¹s
> >>>> and GUI frameworks. It is also more scriptable. You can write a script
> >>>> that will create a complete, multilayered map, and output it to a
> >>>> graphics file.
> >>>>
> >>>> One consequencer of this new architecture is that the GIS Manager does
> >>>> not currently support display output to PNG files. With GDAL (required
> >>>> for many GRASS features) and ghostscript installed, it does support
> >>>> output to PPM/PNM, TIF, JPG, BMP, EPS, and PDF.
> >>>>
> >>>> Command line usage is still supported in several ways<the old
> >>>> xterminal is still there, there is the new command console, and you
> >>>> can add command layers to the layer tree and display them.
> >>>>
> >>>> You can see a screen shot of semi-transparent soil vector areas
> >>>> overlaying a raster shaded relief map at
> >>>>
> >>>> _http://www.public.asu.edu/~cmbarton/files/grass_screenshots/gism2-2.jpg
> >>>>
> >>>> <http://www.public.asu.edu/%7Ecmbarton/files/grass_screenshots/gism2-2.jpg>
> >>>>
> >>>> _
> >>>> I¹ve committed all changes to the CVS. For those of you with binary
> >>>> versions of GRASS who would like to try it, I¹ve posted packages you
> >>>> can use to replace your current GIS Manager on my website. For full
> >>>> functionality, you need a version of grass that is dated
> >>>> 21-February-2006 or later. For those of you with a slightly earlier
> >>>> version, I made a second package with all functionality minus
> >>>> transparency. Here are the links.
> >>>>
> >>>> New GIS Manager files
> >>>>
> >>>> <_http://www.public.asu.edu/~cmbarton/files/grass_gismgr/
> >>>> <http://www.public.asu.edu/%7Ecmbarton/files/grass_gismgr/>_gism_rc4_20060307
> >>>>
> >>>> .tgz>
> >>>>
> >>>> New GIS Manager files for people with recent, but pre-21 February 2006
> >>>> binaries.
> >>>>
> >>>> <_http://www.public.asu.edu/~cmbarton/files/grass_gismgr/
> >>>> <http://www.public.asu.edu/%7Ecmbarton/files/grass_gismgr/>_gism_rc4_20060307
> >>>>
> >>>> pre_02-21.tgz>
> >>>>
> >>>> Enjoy
> >>>> 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
> >>>> <http://www.public.asu.edu/%7Ecmbarton>
> >>>>
> >>
> >> ___________________________
> >> Michael Barton, Professor of Anthropology
> >> School of Human Evolution & Social Change
> >> Arizona State University
> >> Tempe, AZ  85287
> >>
> >> WWW - http://www.public.asu.edu/~cmbarton
> >> Phone: 480-965-6262
> >> Fax: 480-965-7671
> >>
> >>
> >
> >
>
>
> --
> Thomas E Adams
> National Weather Service
> Ohio River Forecast Center
> 1901 South State Route 134
> Wilmington, OH 45177
>
> EMAIL:  thomas.adams at noaa.gov
>
> VOICE:  937-383-0528
> FAX:    937-383-0033
>
>

--
+-----------------------------------------------------------+
              Carlos Henrique Grohmann - Guano
  Geologist M.Sc  - Doctorate Student at IGc-USP - Brazil
Linux User #89721  - carlos dot grohmann at gmail dot com
+-----------------------------------------------------------+
_________________
"Good morning, doctors. I have taken the liberty of removing Windows
95 from my hard drive."
--The winning entry in a "What were HAL's first words" contest judged
by 2001: A SPACE ODYSSEY creator Arthur C. Clarke