No big enhancements like adding support for symbols, but it now has
better error handling, debug info and will survive d.redraw if input was
given from a file. Added an updated version of grass.logo.sh from the
GRASS 5 source into the module's source dir as an example. Also added a
command to set the text rotation.
I didn't bother to do the same updates to d.mapgraph as I plan on
merging it into the d.graph module as the "-m" flag. In regards to that
effort, if anyone wants to merge the malloc -> G_malloc changes, please do!
How easy/hard would it be to add different line widths at some point?
Michael
__________________________________________
Michael Barton, Professor of Anthropology
School of Human Evolution and Social Change
Arizona State University
Tempe, AZ 85287-2402
From: Hamish <hamish_nospam@yahoo.com>
Date: Wed, 14 Dec 2005 15:22:42 +1300
To: grass5 <grass5@grass.itc.it>
Subject: [GRASS5] d.graph updates
Hi,
I've done a little cleanup on the d.graph module.
No big enhancements like adding support for symbols, but it now has
better error handling, debug info and will survive d.redraw if input was
given from a file. Added an updated version of grass.logo.sh from the
GRASS 5 source into the module's source dir as an example. Also added a
command to set the text rotation.
I didn't bother to do the same updates to d.mapgraph as I plan on
merging it into the d.graph module as the "-m" flag. In regards to that
effort, if anyone wants to merge the malloc -> G_malloc changes, please do!
How easy/hard would it be to add different line widths at some point?
Easy & done.
Also added a -m flag to use geographic coordinates (the d.mapgraph merge).
TODO: (preferably by someone else)
a) merge in G_malloc() fixes from d.mapgraph, any other bug fixes?
After this is done we can add a
G_warning("Module superseded; use 'd.graph -m' instead");
to d.mapgraph. Perhaps it is preferable to remove d.mapgraph from CVS
and have init scripts set up `alias d.mapgraph='d.graph -m'` ?
b) update "icon" instruction to use GRASS 6 symbols. This requires
copying out the icon display code from d.vect and making a new display
lib routine. After that we would modify d.graph to use the new
D_symbol() fn.
-- do we want to call them symbols or icons? I find it confusing to use both.
Please read the updated help page. Pointed out a previously undocumented
feature (polyline) among other things.
It may be useful to run at DEBUG level = 3.
I think the module is pretty useful now. It will be even beter when the
icons are updated.