[GRASS5] Re: [GRASSLIST:10454] Re: gis.m run failure

GRASS GIS GRASS GIS developer
1

I've been working with 8.4.x for quite awhile with no problems on the Mac
platform.

Is there any reason now to stay with 8.3? (nviz? Other?)

The current version of TclTk is 8.4.12, with 8.5 nearing release.

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: Tom Russo <russo@bogodyn.org>
Date: Mon, 20 Feb 2006 18:39:56 -0700
To: <grasslist@baylor.edu>
Subject: [GRASSLIST:10454] Re: gis.m run failure

On Mon, Feb 20, 2006 at 06:25:45PM -0700, we recorded a bogon-computron
collision of the <russo@bogodyn.org> flavor, containing:

I have been a command-line bigot forever, and rarely use the d.m or gis.m
features of GRASS --- I once tried gis.m and liked what I saw, but don't
use it routinely, and indeed haven't tried it in a few weeks. As for
d.m, my first command on starting GRASS is typically "File->Exit" in d.m.

After all this talk of revamping GRASS so that the X displays can go away, I
decided to try out gis.m again lest I fall so far behind the times that
GRASS becomes unusable for me overnight. Using grass-6.1 cvs updated last
night, I get:

gis.m

Error in startup script: bad event type or keysym "ConfigureRequest"
    while executing
"bind .mapcan($mon) <ConfigureRequest> {

[...]

Any hints on where to look for what's wrong here? Could this be a tcl/tk
versioning issue? (I'm using tcl/tk 8.3, although I do have tcl 8.4
installed
--- just not set up so that configure finds it, coz once upon a time 8.4 was
unusable with grass.)

Google to the rescue. Yes, ConfigureRequest is something that is only present
in tcl 8.4.

If this incompatibility with tcl/tk 8.3 is necessary for proper function of
gis.m, the various install documents should ultimately be updated, and
perhaps configure should not accept tcl/tk 8.3 anymore.

Re-adjusting my tcl/tk 8.3 vs. 8.4 install and letting grass use 8.4 fixed
gis.m.

--
Tom Russo KM5VY SAR502 DM64ux You still here?
Tijeras, NM QRPL#1592 K2#398 SOC#236 AHTB#1
"And, isn't sanity really just a one-trick pony anyway? I mean all you get is
one trick, rational thinking, but when you're good and crazy, oooh, oooh,
oooh, the sky is the limit." --- The Tick

2

Michael Barton wrote:

I've been working with 8.4.x for quite awhile with no problems on the Mac
platform.

Is there any reason now to stay with 8.3? (nviz? Other?)

Historically, some versions of 8.4 have had compatibility problems
with NVIZ. AFAIK, the problems are related to the use of threads.

Does 8.4 have any additional functionality which is actually useful?
If not, Tcl code should work with either 8.3 or 8.4.

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

3

On Tue, 21 Feb 2006 02:16:18 -0700
Tom Russo <russo@bogodyn.org> wrote:

On Tue, Feb 21, 2006 at 01:10:17AM -0700, we recorded a
bogon-computron collision of the <michael.barton@asu.edu> flavor,
containing:
> I've been working with 8.4.x for quite awhile with no problems on
> the Mac platform.
>
> Is there any reason now to stay with 8.3? (nviz? Other?)

The one thing I ever heard of as a problem was NVIZ, and it was the
first thing I tried once switching to 8.4 --- as far as I can tell on
cursory examination, it does work on my system. For me, staying with
8.3 was mostly superstition --- some time ago there were strong
warnings not to try to use tcl/tk 8.4, and I never tried. There are
still warnings in the REQUIREMENTS.html file that some versions of
Linux still need 8.3.

It is about Debian based distros. 8.4 used to work fine for me in
Mandrake with NVIZ. On Ubuntu Hoary I had problems - 8.3 was required
for NVIZ to run. Same as for other users (see grasslist archives). I
haven't tried 8.4 with NVIZ on my current Ubuntu Breezy yet being
aware of problems in Ubuntu Hoary and Debian.

I did see a few odd behaviors in gis.m (at one point I "cut" the last
layer out of the list and the thing went into some sort of infinite
loop) but I didn't play around enough to see if they were
reproducible.

There are reports in bugtracker regarding infinite loops in tcl/tk
GUIs:
http://intevation.de/rt/webrt?serial_num=2937
http://intevation.de/rt/webrt?serial_num=3117

I used to get this error from time to time on Mandrake using tcl/tk
8.4 with different commands. And there are at least 2 reports more of
this bug by other users in grasslist archives (both for v.in.ogr) and in
grass5 archive: 1 for v.in.ogr and 1 r.flow. So it seems pretty common.

Interestingly, since I switched to Ubuntu and tcl/tk 8.3, as it is
required for a Debian fork to be able to run NVIZ, I've never had this
issue anymore. This could be a coincidence, but it might seem that 8.3
is less prone to this problem than 8.4. Can somebody confirm or deny?

I just fiddled around a little to see what was there,
but didn't really try to do anything for real.

Maciek

--------------------
W polskim Internecie s? setki milion?w stron. My przekazujemy Tobie tylko najlepsze z nich!
http://katalog.panoramainternetu.pl/