Angus Leeming wrote:
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
Well, I think I communicate a lot already.
LOL! Email is a notoriously bad way to communicate. I'm forever writing emails
that are clear and lucid and which are taken to mean something else by the
eejit reading them. Often I seem to offend them. Honestly! (Who'd have thought
it!) I suspect that they misconstrue my words on purpose.
I have described a number of times the flaws in the controller.
If someone is opposing some argument to a patch, he should at
the minimum read the code in question. Opposing a patch because it
removes some pseudo-feature is wrong.
I've read the patch. I understand the rationale. I think that you're doing the
correct thing. Here's how I think you should proceed:
* strip out Ispell and Pspell support. Do it completely. That means configure
tests, src/ispell.C and src/pspell.C. Fix up the Preference dialog.
Yes, I forgot that in my patch as Georg did point out.
* re-write the ControlSpellchecker code. The logic is currently upside down.
* Add support for Enchant. Do it completely. That means configure tests and
src/enchant.C. Fix up the Preferences dialog. I'd assume that the dialog would
enable the user to pass options through to Enchant.
* Get people to test it out. If Enchant provides us with everything that
Aspell currently provides us with, start a "let's remove Aspell" debate. Don't
get upset when you meet resistance. (Grow a thicker skin ;-))
* Assuming that you win that argument you'll be able to get rid of SpellBase
(or whatever it's called).
Now that's a lot of work, but if your serious in your desire to clean up LyX,
it's the only sensible way to proceed.
You've just described my very plan ;-). I just thought it was obvious so
I didn't care to write it so explicitly. The problem is just that Lars
won't allow step 1.
Abdel.
