[GRASS5] GRASS 5.0.0stable to be released soonish

[Morten Hulden]

On Tue, 11 Jun 2002, Markus Neteler wrote:

> On Sat, Jun 08, 2002 at 10:12:37PM +0100, Glynn Clements wrote:
> > 
> > Glynn Clements wrote:
> > 
> > > For the remaining changes, I would appreciate opinions as to which
> > > should be included in the next release. I'll submit my opinions in a
> > > follow-up.
> > 
> > I recommend including these:
> > 
...
> > > src/raster/r.proj/cmd/bordwalk.c
> > > 	ignore invalid points
> 
> no idea - is the bordwalk issue clarified now? Morten?

Ignoring invalid points is a separate issue from the bordwalk() topic. It
can happen in any PROJ routine. For raster projections I think it's OK to
ignore invalid points in bordwalk(), and set cells to NULL if we hit them 
in the main() (even if this may result in NULL holes in the map, but it's 
better than aborting r.proj completely).

> > > src/raster/r.proj/cmd/main.c
> > > src/raster/r.proj/cmd/r.proj.h
> > > 	tidy
> > > 	add -n (nocrop) flag
> > > 	move "menu" array from r.proj.h to main.c
> 
> no idea - is the bordwalk issue clarified now? Morten?

My view is:

	-the new flag -n (nocrop) should inhibit both of the two calls to 
bordwalk() from main. without -n, both calls to boardwalk() should be 
enabled. 
	-r.proj should honour current region settings, not 
unconditionally project into default region.

Morten Hulden