[GRASS5] Mac OS X beta 10

Markus Neteler neteler at geog.uni-hannover.de
Fri Dec 15 06:49:30 EST 2000


Hi Jeshua,

don't feel being ignored! You will have realized that we try our
very best (limitated by our jobs and some private life).
:-)

I encourage you to find another Mac enthusiast to join the
team. For the non-Mac users it's rather impossible to
reproduce platform specific errors (no surprise). So
another tester may help you sometimes more than the other
current developers can do.

On Fri, Dec 15, 2000 at 12:24:35AM +0100, Andreas Lange wrote:
> Hi Jeshua,
> 
> Jeshua Lacock wrote:
> > 
> > Greetings,
> > 
> > I have not heard any advice or suggestions regarding the Tcl/Tk GUI
> > bugs that I am experiencing, and I am at lost if I should post the
> > binaries in their state or if I should wait until the GUI is usable.
> 
> I read your posting on the errors. There are some misconceptions about
> the working in it.
> First: tcltkgrass does not depend on any compiled binary except of the
> wish shell from tcl/tk. So recompiling GRASS will not help.
> Second: tcltkgrass works with tcl/tk 8.0, 8.2 and 8.3 on Linux and on
> IRIX. I tested myself and found not problems.
On solaris is is o.k. as well (beside that static library detection problem 
in configure).

> Mixing tcltkgrass and commandline to start/stop monitors is no good
> idea, as tcltkgrass makes the assumption that e.g. a monitor started
> with tcltkgrass is still running. It may hang if you stop the monitor by
> hand and try to stop it with tcltkgrass menue then. 
Up to now the monitors are very sensitive. I run regularly into this
trap while teaching GRASS to new users...

> The message you recieve if you kill the monitor with the window manager
> [x] button or kill option i experience on Linux too. We are working on
> this, but this requires a lot of knowlegde on system programming and the
> internals of Unix. And Unix=Unix is never true. Mac OS X may behave
> differently that Linux. 
> 
> Could you test the following with a standard installation on Mac OS X:
> remove both .grassrc5 and .tcltkgrass (or rename both files) on the
> account of an ordinary user (not root) and start a grass session and
> open a monitor, quit tcltkgrass and restart it. Then try to draw
> something on the monitor e. g. with d.rast or d.vect. Report the errors.
> 
> > I know everyone is busy, but I have devoted a major percentage of my
> > time for the Mac/GRASS community and I feel ignored at this time.
> 
> Please notice that most of us work part time or spare time on grass, so
> we can not fix anything in the speed some commercial services can offer.
> But you will get a feedback, which is different from my personal
> expirience with some commercial support services...
However I am convinced that the overall GRASS development speed and update
frequency is quite high(er?) comparing to some proprietry software companies.

> > So the options are basically: post the build in it's current state,
> > fix and then post the build, or don't post the build at.  What should
> > I do?
> 
> I think that tcltkgrass is so important that you should not post the
> binaries unless it is working.
I agree. Let's get this stable first. Even on other platforms there are
still problems with GRASS startup etc.

> Please note too that tcl/tk is nowadays some sort of moving target, with
> several different versions in broad use. 

Yours

 Markus

---------------------------------------- 
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo at geog.uni-hannover.de with
subject 'unsubscribe grass5'



More information about the grass-dev mailing list