Misbehavior in Jove

[Casey Leedom]

In article <5584 at xanth.cs.odu.edu> kent at cs.odu.edu (Kent Paul Dolan) writes:
> 
> [Description of what happens in emacs and jove when you type
> case-word-capitalize in the middle of a word.]
> 
> Hmmm.  I've seen two notes now claiming this isn't a bug.  OK.  Let's
> call it a horrid human factors design flaw instead, then.  One time in
> 1000 I need to capitalize some middle letter of a word ...

  If you want this behavior when using jove, just define ^[c and ^[C
to do ^[-case-word-capitalize.  This can be done under jove 4.8 and later
via:

	define-macro my-case-word-capitalize ^F^[-^[xcase-word-capitalize^M
	bind-macro-to-key my-case-word-capitalize ^[c
	bind-macro-to-key my-case-word-capitalize ^[C

  Under earlier versions of jove you'll have to define the macro by typing
it once and then saving all of your macros to a file.  Not being a GNU
emacs user, I don't know how to do this there, but I'm sure it's possible.
By the way, a nicer way to do the above is:

	define-macro my-case-word-capitalize ^F^[-^[C
	bind-macro-to-key my-case-word-capitalize ^[c

This binds the new my-case-word-capitalize to ^[c (lower case c) and
leaves the standard case-word-capitalize bound to ^[C (upper case C).
It's faster, doesn't leave the "case-word-capitalize" command cluttering
up the command line, and leaves the old case-word-capitalize conveniently
available if you do find a case where you need it.

> While we're speaking of design flaws, why in the world does ^N insert
> newlines at the end of a file, when the return key is available for that
> job?

  It should be noted that this is a ``feature'' of GNU emacs.  Jove
doesn't do that and quite properly stops at the end of the buffer.  I have
absolutely no idea why GNU does this.  After all, ^F doesn't automatically
extend lines when you reach the end.  Likewise ^P and ^B don't insert
blank lines at the beginning of the buffer or a line.

  It's very frustrating since one gets it into the mode of thinking that
^N is a cursor movement command, and therefore save to use, only to
discover that under special circumstances it will cause buffer
modification.

  If Richard is listening, I'd vote to remove this ``feature''.

Casey