Gunnar Evermann <ge204(a)eng.cam.ac.uk> writes:
Hrvoje Niksic <hniksic(a)arsdigita.com> writes:
> Use:
> (define-key ... [(meta control h)] '...)
>
> rather than
>
> (define-key ... "\M-\C-h" '...)
>
> The latter affects BS, while the former just meta-control-h.
maybe we should fix the following cases as well:
ge204@parthenope$ find . -name \*.el | xargs grep '\\M-\\C-h'
./comm/gnus/gnus/lisp/gnus-sum.el: "\M-\C-h" gnus-summary-hide-thread
./os/ilisp/extra/hyperspec-carney.el:;;;; "\M-\C-h")
./prog/idlwave/idlwave.el:(define-key idlwave-mode-map "\M-\C-h"
'idlwave-mark-subprogram)
./prog/prog-modes/pascal.el: (define-key pascal-mode-map "\M-\C-h"
'pascal-mark-defun)
./prog/prog-modes/verilog-mode.el: (define-key verilog-mode-map "\M-\C-h"
'verilog-mark-defun)
./prog/vhdl/vhdl-mode.el: (define-key vhdl-mode-map "\M-\C-h"
'vhdl-mark-defun))
FWIW, I think \M-\C-h work fine (at least in Gnus and VHDL mode), it's
\e\C-h that's failing.
Also, if occurances of \e\C-h are being changed into \M-\C-h en masse,
perhaps it should be stated that that \e\C-h does not work in the
documentation. Or perhaps fix the problem.