[GRASS-dev] Re: [grass-code I][394] scripts using g.message fail if message contains "="

[Glynn Clements]

Markus Neteler wrote:

> On Wed, May 09, 2007 at 07:43:27AM +0200, grass-codei at wald.intevation.org wrote:
> > code I item #394, was opened at 2007-05-09 17:43
> > Status: Open
> > Priority: 5
> > Submitted By: Hamish Bowman (hamish)
> > Assigned to: Nobody (None)
> > Summary: scripts using g.message fail if message contains "=" 
> > Issue type: module bad feature
> > Issue status: None
> > GRASS version: CVS HEAD
> > GRASS component: None
> > Operating system: None
> > Operating system version: 
> > GRASS CVS checkout date, if applies (YYMMDD): 
> > 
> > 
> > Initial Comment:
> > Hi,
> > 
> > this may be "user error", but g.message will fail if the message contains a "=" char and message= was not given. With message= the parser understands it ok.
> > 
> > This is causing an error in a number of scripts. (I found it in wms.request)
> > 
> > 
> > 
> > G63 > g.message "foo="
> > Sorry, <foo> is not a valid parameter
> 
> I observed a similar problem in my new v.in.wfs script
> (get from http://mpa.itc.it/markus/grass63/): The WFS URL is
> disliked by g.parser. I worked around this with ugly hacks but
> I think that we need a new parameter type which accepts chars
> such as '=', '&' and so forth. I used file,file,file in v.in.wfs
> but would prefer something else.

If you don't set a gisprompt, the parser will accept any string.

BTW, r.in.wfs suffers from the same bug which Hamish mentions: failure
to specify message= explicitly.

I don't know why you think that you need the "$2" hack.

-- 
Glynn Clements <glynn at gclements.plus.com>