Re: [Texmacs-dev] Logical cursor movement

[Joris van der Hoeven]

> > Yes, although I am not yet sure about which modifier keys to use.
> > Indeed, there are many types of directional actions,
> > such as inserting arguments, rows or columns, alignment of cells,
> > or redimensioning sizes of pictures and white space.
> > We have to find out which ones are used most frequently,
> > because these will get the C-, A- and M- prefixes.
> > Others will have to do with A-C-, M-C-, M-A-, etc.
> > I personally think that logical cursor movement of
> > the first type will not be used very much.
> 
> I currently have no willingness to get in that discussion, but I just
> want to drop a word to say "do not use A-C and M-C modifiers"!

We will use them for *less* important tasks, while providing alternative
bindings. This is why it is important to find out which kind of 
directional commands are most useful.

> Personally, I use these modifiers combinations for global bindings in
> XFree. For example I use C-A arrows to change workspaces, C-A-m to
> minimize a window, not even speaking of the system meaning of C-A-Fx
> keys. I am sure that many people use these keys in a similar fashion.

I am aware of this, but unfortunately, we often have little choice.
I now also provide "A-C-letter" for typing Greek letters.
This is better than "F5 letter" for most purposes, but if your system
undertakes some strange action on "A-C-letter" for some reason,
then you will have to fall back on "F5 letter" or map the hyper
modifier to a suitable key.

> And for M-C, I just need to recall the generalized confusion between
> Alt and Meta key.

I am aware of this too, but I think that much of the confusion has been
removed as far as TeXmacs is concerned: if you have an alt key (as found
by xmodmap) on your keyboard, then it will be mapped to A-, as expected,
and if you have a meta key or a super (microsoft) key, then it will be
mapped to meta. I think that this is clean.

However, Emacs itself is not clean: I have tested "emacs &" and
"emacs -nw" on several systems and noticed that there is no clear
rule of what modifier key is mapped to M- (and which one to alt).
On my PC, the alt key is used for meta, which is not very clean.
Nevertheless, I have been able to make things compatible by mapping
the microsoft key to meta (instead of super) in my .bash_profile.
So I now have a TeXmacs and an Emacs which have a similar behavior
as far as C- and M- combinations are concerned.

> You have to be more careful in letting TeXmacs cooperate with its
> environment. The obnoxious wait window is a typical example.

I am still waiting for a clean fix for this,
which displays the message in the active window.