[GRASS-dev] Re: g.region bugs [was: GRASS 6.3.0 release preparation]

[Maciej Sieczka]

Martin Landa wrote:

> hm, I disabled G_aling_window(), seems to work... (in CVS)

The test case from in my last email works OK now. But, a
different one still fails.

A different vector line, same intial region settings as in
last case:

$ echo "L 2
571600 5722275
571610 5722275" | v.in.ascii -n out=line format=standard --o

$ g.region n=5680980 s=5680960 w=602440 e=602480 res=20 -ag
n=5680980
s=5680960
w=602440
e=602480
nsres=20
ewres=20
rows=1
cols=2
cells=2

$ g.region vect=line res=1 -ag n=5722285
s=5722265
w=571600
e=571610
nsres=1
ewres=1
rows=20
cols=10
cells=200

Wrong. Should be:

n=5722276
s=5722274
w=571600
e=571610
nsres=1
ewres=1
rows=2
cols=10
cells=20

The above happens in GRASS 6.3. In 6.2 (CVS), the procedure
also leads to wrong region settings, but different:

$ g.region vect=line res=1 -ag
n=5722300
s=5722260
w=571600
e=571620
nsres=1
ewres=1
rows=40
cols=20

It'd be great if you can solve it.

Maciek