Re: [Texmacs-dev] Keyboard handling

[Joris van der Hoeven]

> I am having a look at the online documentation, and I happened to
> noticed some problems with keyboard handling. At first I assumed the
> document was just outdated, but then I examined kbd-emacs.scm and
> saw that its intent matched the documentation.
> 
> The problems are:
> 
>     Multiple [escape] are not bound to [A-] and [H-] modifiers, and
>     generally all those definitions seem to be ignored:
> 
>     (set-post-wildcards
>       ("escape" "M-" #t)
>       ("escape escape" "A-" #t)
>       ("escape escape escape" "H-" #t)
>       ("escape escape escape escape" "noop" #t)
>       ("A-C-" "H-" #t)
>       ("H-" "F5" #t)
>       ("H-C-" "F6" #t)
>       ("H-C-S-" "S-F6" #t)
>       ("H-A-" "F7" #t)
>       ("H-M-" "F8" #t)

I do not have this impression: on my keyboard escape yields M,
escape escape yields A and escape escape escape yields H
(which starts emphasized text in text mode).

>     math:x prefix defined to be F5..F8 are not applied. That is even
>     visible from the tooltips of the greek, bold, etc. menus.
> 
>     [A-] shortcuts for accents are disabled, yet [M-] shortcuts for
>     accents do work. I made sure that was not a pb of modifier
>     configuration by looking at the feedback in the window footer.

I did not notice this either; please be more specific.

> Well... there are probably more, but that is all I have found for the
> moment.
> 
> Other complaints:
> 
>     I have not redefined caps-lock (it keeps its std behaviour) and I
>     noticed it still behave like SHIFT w.r.t. to shortcuts, which is
>     BAD!

How do you want caps-lock to behave?

>     Also, there are still shortcuts which read "M-C" instead of
>     "M-S-c" which is just confusing.

I prefer M-C, which may be confusing, yet it is unambiguous.

> Kudos for the new kbd handling:
> 
>     Well... there are still some good things with the new kbd handling
>     code.
> 
>     For non alphabetic character I agree that "S" must not be
>     explicit, because the layout may change and uppercase is
>     ill-defined. And for action keys (tab, return, bacspace,
>     arrows...) the S modifier is explicit.
> 
>     Alt-Gr seems to be handled properly now. Esp. it now has the
>     expected compose functionality on my dvorak layout :-)
> 
>     The numeric keypad now works proprely, and is properly modified by
>     num-lock and shift. That feels good.
> 
>     It is now possible to rebind the unmodified [space] key.
> 
> There a probably more goodies, too bad changes in the code are so
> badly advertised.

You volunteer for writing documentation?