On 2025/03/24 18:00, Ted Unangst wrote:
> On 2025-03-23, Ted Unangst wrote:
> > On 2025-03-23, Job Snijders wrote:
> > > Did you consider cranking it up to 7 digits (to better accommodate
> > > chubby instances of for example chromium, mount_mfs, and mutt)?

how on earth do you get mutt to 7 digits?!

mount_mfs, clamd, mariadb, squid, and many java-based programs can
easily get there though.

> > I think a better fix is to dynamically size some columns, but this
> > is a bit more involved giving the way the printing code works. Will
> > think about it for a bit.
> 
> So I'm not wedded to this idea, but if the terminal is wider, we can
> make a few adjustments. You want to see big processes, you need to buy
> a bigger monitor.

an additional 1 or 2 columns if the conditions are right would certainly
give cleaner output on many systems.

> 
> Index: ps.c
> ===================================================================
> RCS file: /cvs/src/bin/ps/ps.c,v
> diff -u -p -r1.81 ps.c
> --- ps.c	18 May 2024 13:08:09 -0000	1.81
> +++ ps.c	24 Mar 2025 21:57:20 -0000
> @@ -412,6 +412,9 @@ scanvars(void)
>  
>  	for (vent = vhead; vent; vent = vent->next) {
>  		v = vent->var;
> +		if (termwidth >= 84 && (strcmp(v->name, "rss") == 0 ||
> +		    strcmp(v->name, "vsz") == 0))
> +			v->width += 2;
>  		i = strlen(v->header);
>  		if (v->width < i)
>  			v->width = i;
>