[GRASS5] Re: [GRASSLIST:4932] Re: overwriting a raster layer with g.rename

[Markus Neteler]

On Mon, Nov 11, 2002 at 07:37:04PM +0000, Paul Kelly wrote:
> On Mon, 11 Nov 2002, Eric G. Miller wrote:
> > On Mon, Nov 11, 2002 at 11:50:51AM -0000, Thomas Dewez wrote:
> > > Dear all,
> > >
> > > I am puzzled by the behaviour of g.rename. The Note in the man pages say:
> > >
> > > "However, g.rename will allow the user to overwrite other existing files in the current
> > > mapset by making the new file name that of an already existing file."
> > [snip]
> > > Instead of working smoothly I get the answer:
> > > <map2> already exists in mapset <blah>
> > >
> > > Is there a particular command switch to overwrite the file?
> >
> > You want the '-o' flag, for overwrite...
> >
> > This is a bug against the man page.  Sometimes people forget to update
> > the manual... Using the "help" argument to most commands will give you
> > the definitive brief summary of what flags and arguments the command
> > takes.

The -o is added to the HTML page now.


> This also explains the strange behaviour of the v.line2area and
> v.area2line scripts, which use g.rename without the -o flag. When I first
> started using GRASS this puzzled me and I didn't know what to look for; it
> would be more obvious now that I'm a little bit more experienced but I
> haven't used them for a while.
> So somebody should update them (in src/scripts/contrib/v.line2area) to use
> the -o flag when they are overwriting the original file.

... also done. Thanks!

> And I suppose if
> someone changes the functionality of a command they should check which
> other programs and scripts use it (as well as updating the man page)?

Oh yes. All developers changing a program should also
 - update the HTML file
 - update the tcltkgrass entry if required
 - search for system calls in the GRASS code (with 'find' this can be done
   in a few minutes) and update them if required

A "hot" candidate for update is probably the r.mapcalc HTML page.

Markus