Re: [Gzz] New PEG: prefix_argument--benja

[Tuomas Lukka]

On Sat, Nov 16, 2002 at 08:30:38PM +0100, Benja Fallenstein wrote:
> Tuomas Lukka wrote:
> 
> >E.g. if "GOTO" function takes prefix parameter, or other window's cell.
> >
> >Now, if you're starting to mark cells for some operation, suddenly
> >doing 'goto' takes you to a different place from where you intended
> >to go.
> >
> >Marks should only be prefix args if explicitly asked for.
> >
> 
> Hmm. That's a pretty fundamental issue... I've understood Ted so that 
> when there are marks, all operations should automatically operate on 
> *them*; do you think I'm wrong there?
> 
> To some degree I agree, but there's also the problem that this can be a 
> slowdown: additional keys to be pressed. For example, with connect, it 
> seems natural to just mark, then hit 'connect <dir>', not another key in 
> between. As for goto, when cycling through the cells, it doesn't seem 
> nice to have to hit two keys each time we make a step.
> 
> I'm inclined to think that having to switch the mark set (so that the 
> cells aren't the *currently* marked ones any more) before pressing 
> 'g'/'-', for using the other window as a prefix, is better.

Ok, that could work if it's really easy to cycle to an empty mark set.

> Can you suggest a key binding for indicating a 'marked cells' prefix 
> argument, in your proposal?

No, since we don't have all proposed key bindings in one place ;)

Maybe we should have pegboard.py generate a table of where a given key
has a suggested binding, from tables in PEGs like

Proposed bindings
------------------

a
        (in X mode) lblah blah

        Tuomas