Hi again,
small drawback concerning the backingstore: while the "common"
GRASS monitor does backingstore well now, it's not yet done in
v.digit if the menu text window hides the monitor.
I would like to help but have no idea how to do that.
Regards
Markus
----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'
On Tue, Feb 13, 2001 at 04:09:33PM +0000, Markus Neteler wrote:
Hi again,
small drawback concerning the backingstore: while the "common"
GRASS monitor does backingstore well now, it's not yet done in
v.digit if the menu text window hides the monitor.I would like to help but have no idea how to do that.
Yea, this is the backingstore problem I reported about. I imagine it
affects all the long running modules, and will be expecially bothersome
if the module presents some kind of user interaction at the terminal
after starting (most???). Looks like all I've managed to do is move the
read block from one place in the code to another ;).
--
Eric G. Miller <egm2@jps.net>
----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'
On Tue, Feb 13, 2001 at 04:09:33PM +0000, Markus Neteler wrote:
Hi again,
small drawback concerning the backingstore: while the "common"
GRASS monitor does backingstore well now, it's not yet done in
v.digit if the menu text window hides the monitor.I would like to help but have no idea how to do that.
Okay, I have a fix for multiple monitors syndrome, and I think I've
isolated the backingstore problem for long running processes (it wasn't
where I thought). I've been unable to use the CVS all night (connection
refused) -- don't know if others are having problems (might be on my
end).
For the BackingStore thing, it looks like the problem is in calls like
Get_location_with_pointer() and such, as these functions seem to use
blocking calls. I did a little experiment, and it bears out this
conclusion (debugging is a pain with detached processes...). Anyway,
I'll try to work up a solution (the test fix I did sucks CPU cycles...).
--
Eric G. Miller <egm2@jps.net>
----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'
Hi Bernhard,
the CVS is down...
Secure connection to intevation.de on port 2402 refused.
cvs [commit aborted]: end of file from server (consult above messages if
any)
Please check this,
thanks in advance
Markus
----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'
Okay, I made a few more changes to the X driver to try to handle Expose
events better. I tested it briefly with v.digit (among others) and it
seems to handle the repaint in most cases. There are a few places where
the expose event handling doesn't seem to work right (seen when v.digit
first draws screen -- when an action is chosen, the screen gets
repainted; also seen when drawing lines if you switch to a different
window and back in the middle of the operation).
The law of unintended consequences says these changes should produce
several other problems.... Let me know...
--
Eric G. Miller <egm2@jps.net>
----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'
Hi Eric,
On Sun, Feb 18, 2001 at 08:04:16PM -0800, Eric G. Miller wrote:
Okay, I made a few more changes to the X driver to try to handle Expose
events better. I tested it briefly with v.digit (among others) and it
seems to handle the repaint in most cases. There are a few places where
the expose event handling doesn't seem to work right (seen when v.digit
first draws screen -- when an action is chosen, the screen gets
repainted; also seen when drawing lines if you switch to a different
window and back in the middle of the operation).
If v.digit displays the map and menu, I still can wipe out the map
(as you state above).
However, after leaving v.digit (the vector map remains on screen),
it comes back again (that's better than in past :-).
So far my report.
The law of unintended consequences says these changes should produce
several other problems.... Let me know...
[Disclaimer for Rich: The sockets driver we are testing here is optional]
Markus
----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'