Emacs CVS keymaps (was my bug)

Emacs CVS keymaps (was my bug)

[T. V. Raman]

Stephane,

thanks for confirming that the problem was not in emacs itself,
it was originating from a bad use of where-is-internal in my code
that until now did not trigger a bug.

Explicitly:

I had code that removed edit commands from read-only mode and
made them undefined -- I believe this goes  all the way back to
the time of  Emacs 19  --- and has therefore been in emacspeak
since about 1996.

The incorrect use was:
doing
(where-is-internal command keymap)
rather than
(where-is-internal command (list keymap))

Not sure whether this subtle difference always existed -- but it
started biting only now.

Essentially my code was stomping on some (but not all ) bindings,
and the problem went away when I used the second usage form shown
above.

Checking further, I dont need to do the shadowing of edit
commands in read-only buffers because the way modern emacsuns
signal the read-only state does not pose a problem for
emacspeak. So I went ahead and deleted that aspect of my
code. But the subtlety in where-is-internal is worth noting which
is why I'm sending this -- I dont have any change to propose --
the documentation for the function is clear -- once you spot it;-)

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: Emacs CVS keymaps

[Stefan Monnier]

> (where-is-internal command keymap)
> rather than
> (where-is-internal command (list keymap))

> Not sure whether this subtle difference always existed -- but it
> started biting only now.

IIRC, in Emacs-20, the (list keymap) form was not allowed, but the other
form worked just as it does now.

In any case the above con only be a trigger for the bug, since the above
should not modify any keymap.  The bug (that causes malformed keymaps)
must still be elsewhere.


        Stefan

Re: Emacs CVS keymaps

[T. V. Raman]

HMM.

in my environment the effect of the bug, ie some keys not getting
bound -- is now gone.




>>>>> "Stefan" == Stefan Monnier <monnier@iro.umontreal.ca> writes:
    >> (where-is-internal command keymap) rather than
    >> (where-is-internal command (list keymap))
    Stefan> 
    >> Not sure whether this subtle difference always existed --
    >> but it started biting only now.
    Stefan> 
    Stefan> IIRC, in Emacs-20, the (list keymap) form was not
    Stefan> allowed, but the other form worked just as it does
    Stefan> now.
    Stefan> 
    Stefan> In any case the above con only be a trigger for the
    Stefan> bug, since the above should not modify any keymap.
    Stefan> The bug (that causes malformed keymaps) must still be
    Stefan> elsewhere.
    Stefan> 
    Stefan> 
    Stefan>         Stefan
    Stefan> 

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs