[GRASS5] Re: [GRASSLIST:6657] Re: map or input

> > Is there a practical reason why some commands require "input="
> > while others require "map="?

..

- If a command comes with a single map parameter, it is map (v.info
etc) [1] - If a command comes with input and output, it is input and
output
  (r.neighbors etc)
- If a command supports raster and vector and ... the type is
specified
  (g.region etc)

[1] e.g., r.univar [-qg] input=name is *not* following the logic.

Sorry, this discussion evaded me. But AFAIU now we have to wait until
some unknown time in the future (beyond 6.1) to fix any of these due to
functionality freeze?

Unfortunately it's still inconsistent. See discussion here:

http://freegis.org/cgi-bin/viewcvs.cgi/grass/documents/parameter_proposal.txt?rev=HEAD&content-type=text/vnd.viewcvs-markup

shall that be added to grass6/doc/ in cvs, and pointed to in the
SUBMITTING file? In a large project consistency is king...

Hamish

On Mon, May 02, 2005 at 11:51:24AM +1200, Hamish wrote:
...

> Unfortunately it's still inconsistent. See discussion here:
>
> http://freegis.org/cgi-bin/viewcvs.cgi/grass/documents/parameter_proposal.txt?rev=HEAD&content-type=text/vnd.viewcvs-markup

shall that be added to grass6/doc/ in cvs, and pointed to in the
SUBMITTING file? In a large project consistency is king...

Please go ahead :slight_smile:

Markus