[GRASS5] nviz problem

[Glynn Clements]

Bob Covill wrote:

> The original reason I had changed nviz to run directly as nvwish was the 
> error handling. All errors seemed to generate the same unhelpful output 
> when passed through the original script. Somthing along the line of ...
> 
> .............
> child killed: segmentation violation
>      while executing
> "exec /usr/local/grass5/etc/nviz2.2/NVWISH2.2 -f
> /usr/local/grass5/etc/nviz2.2/scripts/nviz2.2_script -q -name NVIZ
>  >&@stdout"
>      ("eval" body line 1)

Right. The easiest solution to that issue is to rewrite the "nviz"
script as a Bourne-shell script instead of a Tcl script.

Bourne shell doesn't print anything if an external command fails.

> Interestingly, when I started looking into this I notice that v.digit is 
> written as a modified wish (I think?).

v.digit uses the "form" library (as does d.what.vect). Most of the
functions in that library just send data to an external program, which
is a modified wish (sort of; it tries to do too much in C, which is
probably why it isn't particularly reliable).

> What would be the best front-end to nviz? It needs to satisfy G_gui(), 
> allow errors to handled more cleanly, and allow by-passing G_parser when 
> in script mode.

The error messages can be fixed by using /bin/sh instead of Tcl.

The G_gui() issues can be fixed by either:

a) having NVWISH2.2 force the program name to "nviz" (this may require
the addition of e.g. G_set_program_name() to parser.c, but that's
trivial), or

b) making the "nviz" script use g.parser (so, in the G_gui() case, the
re-execution is performed before NVWISH2.2 gets involved).

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