[GRASS-user] help with testing the location wizard for the upcoming release

[Hamish]

Hi, this is a general call for help.

if anyone/everyone is able to help test, we have made some fixes
to the wxGUI new location wizard since 6.4.3rc3, which really
need to be tested well before the final release. We think all is
ok now, but a messed up location projection definition and ruin
everything downstream in the location, so it's an important
thing to be totally sure of.


since there are so many projections and datums and ways to
select them, it's pretty hard to test for "working/not working"
as we usally could do. Please try throwing your local odd-ball
CRS at it and see what happens. (check all the way through to
'g.proj -p' and 'g.proj -j' in the final location session. make
sure it does what you'd expect it to do.

specifically how datum transforms are handled, and CRSs without
a specificed datum or specified datum terms; usuing all methods
of defining the projection system. see trac issues #1849 and
#1967 for further details.


known issues:
users of proj 4.7.0 and 4.8.0 will get different results for
the default datum transform terms. (PROJ.4 now decides differently,
beyond our control)
when reading from a .prj or WKT file, or creating from a geo-
referenced file, with unspecified datum transform opts, you will
not be prompted to enter them. the location will be created not
mentioning them, but in the background PROJ.4 will assign some
default value (of its choosing) silently.


for those who use Subversion and compile, you know what to do;
for those who don't, nightly test builds are available for Windows
and Ubuntu.


thanks a bunch,
Hamish