🚀 go-pugleaf

A day or 2 ago, I asked about emacs' X interface and its handling of
keysyms. Since that time, it has become clear that *one* way to handle
this issue is to follow the lead of other X clients, and use
resources. This in turn leads to another question, below. Before I go 
on, however, this should be considered an open call for all those
who have and/or are working on the v19 X interface, since I do not
want to spend much time working on this if an attempt to do something
similar is already in place. If so, I will use a quick an dirty fix
rather than on of the ideas mapped out below.

First, a premise: emacs should not allow dynamic remapping of the
string returned by XLookupString() from the lisp level. You may
disagree with me about this - if so, I'd like to hear your reasons.

Second, my initial guess at the solution is for emacs to use the same
interface as everyone else: resources. Hence, the following might be
a legal .Xresources (or .Xdefaults) entry:

Emacs.Translations: #override \
<Key>F1: string ("^[[1") \
<Key>Insert: string ("^[xoverwrite-mode) \
<Key>KP_1: string ("^[[123)

Emacs would use the relevant X calls to get the mappings, and then
call XRebindKeysym to install them into the area accessed by
XLookupString. 

Donn Cave <donn@milton.u.washington.edu>suggested that it would be
even better to allow:

	<Key>Select: set-mark()

but noted that this would involve a lot more work, which is perhaps
an understatement. Personnaly, I don't favor this approach. Emacs
already has extensive facilities to remap particular key sequences
to functions, and there's little point duplicating them in an
X-specific layer. Also, the first approach would still allow the
easy provision of an elisp primitive like x-rebind-keysym, if people
really insisted on having access to this layer from elisp.

So, emacs X hackers and users and designers ? Is something like this
already being done ? Is this the right way ? Should I start yesterday
or next year ? Thoughts please ...



-- 
Paul Barton-Davis			<pauld@cs.washington.edu>
UW Computer Science Lab		``to shatter tradition makes us feel free''