[GRASS-dev] Re: [GRASS GIS] #1300: WxGUI measure tool gives wrong
results
GRASS GIS
trac at osgeo.org
Sun Apr 10 15:52:09 EDT 2011
#1300: WxGUI measure tool gives wrong results
--------------------------+-------------------------------------------------
Reporter: marisn | Owner: martinl
Type: defect | Status: assigned
Priority: blocker | Milestone: 6.4.1
Component: wxGUI | Version: svn-releasebranch64
Keywords: measure tool | Platform: MSWindows Vista
Cpu: Unspecified |
--------------------------+-------------------------------------------------
Comment(by hamish):
Replying to [comment:10 martinl]:
> backported in r45891
cheers
> > lat/lon is still reporting meaningless results, in grass7 it should
use
> > m.measure to do the calc so geodesic distance is automatically used if
needed.
>
> probably we should warn user about that. Would be
>
{{{
> Warning: Measuring reports meaningless results for LL projection
(geodetic
> distance ignored). Check future versions.
}}}
>
> be OK?
sounds fine. I'd consider the additional step of disabling the tool after
showing that message. Maybe with "Geodesic distance is not yet supported
by this tool, try d.measure or m.measure.", or something. (apple usability
guide: an error message should say what the trouble was, and give a hint
about where the user should look next/for more info)
veering off-topic a little, but along the same reporting seemingly valid
(but in fact broken) numbers to the user, I've been trying to follow how
the mapscale setting in the drop down menu on the bottom bar of the Map
Display window does its job. Does .ppm[0], .ppm[1] refer to image
width,height? Map scale is fundamentally unknowable if you don't know the
correct pixels-per-mm ratio for the user's particular monitor.
X11,OSX,Window's reports for that have in the past not been very good,
maybe with modern versions it has improved? (being optimistic..)
We get away with it in ps.map because we know the exact size of the paper,
and PostScript exists in that reference frame.
GIMP provides a calibration tool (Prefs->Display->Calibrate) where the
user can press a ruler up to their screen and type in the distance covered
by x,y pixel bars, then hope the user doesn't change screen resolution.
Hamish
ps- At some point I'd like to add the v.in.geodesic addon to trunk, but
I'd like feedback on options & assumptions made before finalizing it and
porting it to python, or turning it into a new C module reusing proj4's
`geod` code if that is preferred.
--
Ticket URL: <http://trac.osgeo.org/grass/ticket/1300#comment:11>
GRASS GIS <http://grass.osgeo.org>
More information about the grass-dev
mailing list