bug#17643: 24.3; Please remove comma bindings from french-prefix

bug#17643: 24.3; Please remove comma bindings from french-prefix

[Juliusz Chroboczek]

The input method french-prefix binds comma to cedilla, which makes it
unusable.

Because of this binding, typing a single comma requires typing ", SPC",
which means that the common combination "comma space" requires typing
", SPC SPC".

Please move the cedilla binding to some less commonly used key, such
as "~" (as in latin-9-prefix).

-- Juliusz

bug#17643: 24.3; Please remove comma bindings from french-prefix

[Stefan Monnier]

> Because of this binding, typing a single comma requires typing ", SPC",
> which means that the common combination "comma space" requires typing
> ", SPC SPC".

I think the problem is not in binding , to cedilla, but in binding ",
SPC" to just a comma.  How 'bout the patch below instead?


        Stefan


=== modified file 'lisp/leim/quail/latin-pre.el'
--- lisp/leim/quail/latin-pre.el	2014-01-01 07:43:34 +0000
+++ lisp/leim/quail/latin-pre.el	2014-05-30 18:35:05 +0000
@@ -306,7 +306,7 @@
  ("~ " ?~)
  (",C" ?Ç)
  (",c" ?ç)
- (", " ?,)
+ (",," ?,)
 )
 
 (quail-define-package

bug#17643: 24.3; Please remove comma bindings from french-prefix

[Juliusz Chroboczek]

>> Because of this binding, typing a single comma requires typing ", SPC",
>> which means that the common combination "comma space" requires typing
>> ", SPC SPC".
>
> I think the problem is not in binding , to cedilla, but in binding 
> ", SPC" to just a comma.  How 'bout the patch below instead?

> - (", " ?,)
> + (",," ?,)

It's usable now, but still annoying, since commas get underlined as
you type, which I find distracting.  But perhaps it's a simple matter
of habit -- I'm much more tolerant of apostrophes getting underlined.

Stefan, cedilla is not that common in French, so I think that keeping
it out of the way on "~" is the reasonable thing to do.  Additionally,
it's consistent with latin-prefix and latin-9-prefix.

-- Juliusz (who really likes the polish-slash input method, and who
            likes to claim he's the one who invented it)

bug#17643: 24.3; Please remove comma bindings from french-prefix

[Stefan Monnier]

> Stefan, cedilla is not that common in French,

Yes, I'm quite familiar with that language ;-)

> so I think that keeping it out of the way on "~" is the reasonable
> thing to do.

The underlining is really not a problem, so there's no strong reason to
keep it "out of the way".

> Additionally, it's consistent with latin-prefix and latin-9-prefix.

Various input methods in latin-pre.el use various conventions,
e.g. others use ",c" as well (and ",," rather than ", " to insert
a single comma).

We could try to make them more consistent, but I think it's outside the
scope of this bug-report.  Also in order to make them all consistent,
"~c" will have to be used for "č" rather than for "ç".

So I installed the suggested patch into `emacs-24'.


        Stefan

bug#17643: 24.3; Please remove comma bindings from french-prefix

[Juliusz Chroboczek]

>> Stefan, cedilla is not that common in French,

> Yes, I'm quite familiar with that language ;-)

Heh.

> So I installed the suggested patch into `emacs-24'.

Ok, I'll see if I get used to it.  Thanks for your help.

-- Juliusz

bug#17654: latin-pre.el input-methods are not consistent

[Stefan Monnier]

>> So I installed the suggested patch into `emacs-24'.
> Ok, I'll see if I get used to it.  Thanks for your help.

You're welcome.  I'm just creating a new bug report here about the
inconsistent use of ,c vs ~c (and probably others) in the various input
methods of latin-pre.el.
Some of those input methods should probably be merged or deprecated,
I think, since they partly date back to the pre-unicode era where the
various latin-N charsets were all disjoint.


        Stefan

bug#17654: latin-pre.el input-methods are not consistent

[Lars Ingebrigtsen]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>>> So I installed the suggested patch into `emacs-24'.
>> Ok, I'll see if I get used to it.  Thanks for your help.
>
> You're welcome.  I'm just creating a new bug report here about the
> inconsistent use of ,c vs ~c (and probably others) in the various input
> methods of latin-pre.el.
> Some of those input methods should probably be merged or deprecated,
> I think, since they partly date back to the pre-unicode era where the
> various latin-N charsets were all disjoint.

It seems that this has gotten some more work in the passing years, for
instance:

commit 80c3b9aebaf99da73d10a949f475c5778e843cf8
Author:     Robert Pluim <rpluim@gmail.com>
AuthorDate: Fri Jan 11 17:45:25 2019 +0100
Commit:     Robert Pluim <rpluim@gmail.com>
CommitDate: Mon Jan 14 11:19:41 2019 +0100

    Improve some of the breve/caron/cedilla mappings
    
That adds a bunch of ~c/,c mappings.

In any case, I'm not sure deprecating/reworking the rest of this would
be helpful -- people are really fond of their key bindings.  Is there
more to be done in this bug report?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#17654: latin-pre.el input-methods are not consistent

[Lars Ingebrigtsen]

Lars Ingebrigtsen <larsi@gnus.org> writes:

> In any case, I'm not sure deprecating/reworking the rest of this would
> be helpful -- people are really fond of their key bindings.  Is there
> more to be done in this bug report?

There was no response in a few months, so I'm assuming "not", and
closing this bug report.  There's probably things that can be improved,
but perhaps more specific bug reports for those things would be nice.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no