bug#33024: 26.1.50; the ethiopic input method does not allow the insertion of a very frequent character

bug#33024: 26.1.50; the ethiopic input method does not allow the insertion of a very frequent character

[yilkal argaw]

I tried to use emacs for typing in Amharic the largest language that currently
uses the Geez-script characters and some Geez to (the latin of
Ethiopia). The ethipic input method does not allow for the insertion of
አ character when pressing a . It rather inserts ኣ. The two letters
sound the same but they are used differently. አ is very frequent in
Amharic and Geez (I think it is one of the top 10 frequent
characters). ኣ on the other hand is less frequent and not often use.
Some language like Tigrigna and Tigre use the ኣ character more often
than Amharic but the frequency of አ in those languages is very high
too. Being forced to insert አ in its unicode number renders the
ethiopic input method unusable for regular operations any of the four
language. So this
behavoiour should be altered to allow insertion of both አ and ኣ
through variations in of the A key as:
a:(1/1) 1.አ
a2:(1/1) 1.ኣ
a3:(1/1) 1.ዓ
ae:(1/1) 1.ዐ
aaa:(1/1) 1.ዓ
aa:(1/1) 1.ዐ
aaaa(1/1) 1. ዓ

bug#33024: 26.1.50; the ethiopic input method does not allow the insertion of a very frequent character

[Eli Zaretskii]

> From: yilkal argaw <yilkalargawworkneh@gmail.com>
> Date: Fri, 12 Oct 2018 06:13:21 +0300
> 
> I tried to use emacs for typing in Amharic the largest language that currently
> uses the Geez-script characters and some Geez to (the latin of
> Ethiopia). The ethipic input method does not allow for the insertion of
> አ character when pressing a . It rather inserts ኣ. The two letters
> sound the same but they are used differently. አ is very frequent in
> Amharic and Geez (I think it is one of the top 10 frequent
> characters). ኣ on the other hand is less frequent and not often use.
> Some language like Tigrigna and Tigre use the ኣ character more often
> than Amharic but the frequency of አ in those languages is very high
> too.

I think you can have what you want if you do this:

 (setq ethio-primary-language 'amharic)
 (ethio-select-a-translation)

Without this, the default primary language is tigrigna.

I don't know anything about Ethiopic languages, so I don't know how
reasonable is the default we use.  (I also think the defvar's at the
beginning of ethio-util.el should be defcustoms.)

> So this behavoiour should be altered to allow insertion of both አ
> and ኣ through variations in of the A key as:
> a:(1/1) 1.አ
> a2:(1/1) 1.ኣ
> a3:(1/1) 1.ዓ
> ae:(1/1) 1.ዐ
> aaa:(1/1) 1.ዓ
> aa:(1/1) 1.ዐ
> aaaa(1/1) 1. ዓ

Maybe we could have an alternative input method that worked like
above, since this is an incompatible change.

Thanks.

bug#33024: 26.1.50; the ethiopic input method does not allow the insertion of a very frequent character

[yilkal argaw]

[-- Attachment #1: Type: text/plain, Size: 1686 bytes --]

Thanks, I'll take a look.

On Sat, Oct 13, 2018, 11:14 Eli Zaretskii <eliz@gnu.org> wrote:

> > From: yilkal argaw <yilkalargawworkneh@gmail.com>
> > Date: Fri, 12 Oct 2018 06:13:21 +0300
> >
> > I tried to use emacs for typing in Amharic the largest language that
> currently
> > uses the Geez-script characters and some Geez to (the latin of
> > Ethiopia). The ethipic input method does not allow for the insertion of
> > አ character when pressing a . It rather inserts ኣ. The two letters
> > sound the same but they are used differently. አ is very frequent in
> > Amharic and Geez (I think it is one of the top 10 frequent
> > characters). ኣ on the other hand is less frequent and not often use.
> > Some language like Tigrigna and Tigre use the ኣ character more often
> > than Amharic but the frequency of አ in those languages is very high
> > too.
>
> I think you can have what you want if you do this:
>
>  (setq ethio-primary-language 'amharic)
>  (ethio-select-a-translation)
>
> Without this, the default primary language is tigrigna.
>
> I don't know anything about Ethiopic languages, so I don't know how
> reasonable is the default we use.  (I also think the defvar's at the
> beginning of ethio-util.el should be defcustoms.)
>
> > So this behavoiour should be altered to allow insertion of both አ
> > and ኣ through variations in of the A key as:
> > a:(1/1) 1.አ
> > a2:(1/1) 1.ኣ
> > a3:(1/1) 1.ዓ
> > ae:(1/1) 1.ዐ
> > aaa:(1/1) 1.ዓ
> > aa:(1/1) 1.ዐ
> > aaaa(1/1) 1. ዓ
>
> Maybe we could have an alternative input method that worked like
> above, since this is an incompatible change.
>
> Thanks.
>

[-- Attachment #2: Type: text/html, Size: 2177 bytes --]

bug#33024: 26.1.50; the ethiopic input method does not allow the insertion of a very frequent character

[Lars Ingebrigtsen]

Eli Zaretskii <eliz@gnu.org> writes:

> I think you can have what you want if you do this:
>
>  (setq ethio-primary-language 'amharic)
>  (ethio-select-a-translation)
>
> Without this, the default primary language is tigrigna.
>
> I don't know anything about Ethiopic languages, so I don't know how
> reasonable is the default we use.  (I also think the defvar's at the
> beginning of ethio-util.el should be defcustoms.)

I've now converted the defvars to defcustom.

If I understand this bug report correctly, Emacs has the requested
functionality, but what default to use is unclear.  But making these
variables into user options should presumably help, so I'm closing this
bug report.  If there's more to be done here, please respond to the
debbugs address and we'll reopen.

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