Re: MinGW version binary gobbledygook error messages?

[Paul Kienzle]

On Feb 28, 2006, at 9:10 AM, Sebastien Loisel wrote:

> > If you are persistent, you could probably make things work with
> > eval_string, but I think you will end up duplicating a fair amount of
> > Octave in one way or another.  I think it would be best to avoid doing
> Yes, I am a significant distance along that path. Which is slightly 
> perverse. Indeed, Octave uses readline to compute the width of the 
> terminal to decide how to print matrices. In Workshop, readline isn't 
> used for the reading of the input, the history buffer nor the 
> automatic completion. The output is handled by capturing 
> octave_stdout. So it seems like I could cut readline out of the loop 
> completely, except that octave still uses readline to get the terminal 
> width.
>
> So in Workshop, I call rl_set_screen_size(), a readline function, not 
> because we use readline in any meaningful way, but because that's the 
> only way I know of for telling Octave how many characters to use. So 
> in octave workshop, there is a dependency on readline, which is a 
> glorified int pair storing device.

Probably too difficult for now, but it would be nice if the first few 
elements of the array were displayed, but you could click on it to get 
a spreadsheet representation.

Easy enough to do for ans because that is returned from eval_string.  
You could do it for intermediate results too, by hacking into the 
matrix output functions, but you probably don't want to ... it would 
require keeping a copy of the displayed intermediate data around until 
it falls out of the session, which could be expensive.

- Paul