Re: I think I am finished with syncing with https://github.com/indic-transliteration/m17n-db-indic

[Mike FABIAN]

I had no time to investigate the things that do not yet work in
hi-itrans according to what you want, for example:

>> However, after a dependent or independent vowel,
>> LLa LLi LLu  ळ्ळ ळ्ळि ळ्ळु  
>> etc.. should be produced.
>> Eg. kaLLi कळ्ळि
>> aLLi अळ्ळि
>
> That does not seem to work at the moment:
>
> kaLLi yields कऌ
> aLLi yields  अऌ

But considering that it didn’t work in the past either, it is not a
regression and should not stop me from releasing m17n-db-1.8.4 now, I
think. Maybe I should only add the 

+  ("LL" "ळ्ळ्")

and then release m17n-db-1.8.4. When we have more improvements, I can
quickly make a new release.


Mike FABIAN <mfabian@redhat.com> さんはかきました:

> "विश्वासो वासुकिजः (Vishvas Vasuki)" <vishvas.vasuki@gmail.com> さんはかきました:
>
>> On Thu, 31 Aug 2023 at 18:45, Mike FABIAN <mfabian@redhat.com> wrote:
>>
>>> "विश्वासो वासुकिजः (Vishvas Vasuki)" <vishvas.vasuki@gmail.com> さんはかきました:
>>>
>>> > On Thu, 31 Aug 2023 at 14:07, Mike FABIAN <mfabian@redhat.com> wrote:
>>> >
>>> >> "विश्वासो वासुकिजः (Vishvas Vasuki)" <vishvas.vasuki@gmail.com>
>>> さんはかきました:
>>> >>
>>> >> >
>>> >> > coming to think of it, with the exception of LLi and LLI, it would be
>>> >> good
>>> >> > to map LLa, LLA, LLaa etc.. to ळ्ळ ळ्ळा ळ्ळा etc..
>>> >>
>>> >> What is "etc.."? There is more to do than LLa, LLA, LLaa?
>>> >>
>>> >>
>>> >>
>>> > Everything in the dependent map, in the context explained in the previous
>>> > email:
>>> >
>>> >  (dependent
>>> >   ("a" (delete @-) "")
>>> >   ("aa" (delete @-) "ा")
>>> >   ("A" (delete @-) "ा") .....
>>>
>>> I think I am still confused. I think I get that already. When typing
>>> with hi-itrans and ibus-m17n, I get:
>>>
>>> LLa yields ळ्ळ  (still in preedit)
>>> LLA yields ळ्ळा  (committed)
>>> LLaa yields ळ्ळा  (committed)
>>>
>>> What am I misunderstanding?
>>>
>>>
>>
>> You saw my email starting with "kLLi should yield कॢ "?
>>
>> If that works as well, we're good.
>
> Not everything seems to work as we want. I am currently using a
> hi-itrans with this difference to your version in
> https://github.com/indic-transliteration/m17n-db-indic:
>
>
> $ diff -u /local/mfabian/src/m17n-db-indic/hi-itrans.mim  MIM/hi-itrans.mim 
> --- /local/mfabian/src/m17n-db-indic/hi-itrans.mim      2023-08-08 
> 18:39:41.882632048 +0200
> +++ MIM/hi-itrans.mim   2023-08-31 09:21:55.764133981 +0200
> @@ -158,6 +158,9 @@
>    ("x" "क्ष्")
>    ("ks" "क्स्")
>    ("ksh" "क्ष्")
> +  ("LL" "ळ्ळ्")
> +  ;; As a simple unambiguous option, it would be good to retain ("LLL" 
> "ळ्ळ्")
> +  ;; (See 
> https://lists.nongnu.org/archive/html/m17n-list/2023-08/msg00027.html)
>    ("LLL" "ळ्ळ्")
>    ("gH"  ॻ्)
>    ("jH"  ॼ्)
>
> I.e. I am adding the ("LL" "ळ्ळ्"). I get the following results:
>
>> kLLi should yield कॢ
>
> OK.
>
>> kLLI should yield कॣ
>
> OK.
>
>> similarly for any other consonant.
>
> OK for everything besides "LL" and "LLL": खॢ खॣ गॢ गॣ घॢ घॣ घॢ घॣ क़ॢ क़ॣ क़ॢ 
> क़ॣ ख़ॢ ख़ॣ ख़ॢ ख़ॣ ग़ॢ ग़ॣ ङ्कॢ ङ्कॣ ङ्क्षॢ ङ्क्षॣ ...
>
> I am not sure how LLi shoujld behave after LL or LLL. Currently:
>
> LLLLi yields ळ्ळ्ळि 
> LLLLLi yields  ळ्ळॢ
>
>> However, after a dependent or independent vowel,
>> LLa LLi LLu  ळ्ळ ळ्ळि ळ्ळु  
>> etc.. should be produced.
>> Eg. kaLLi कळ्ळि
>> aLLi अळ्ळि
>
> That does not seem to work at the moment:
>
> kaLLi yields कऌ
> aLLi yields  अऌ
>
> I wonder how to fix that ...
>
>> coming to think of it, with the exception of LLi and LLI, it would be good
>> to map LLa, LLA, LLaa etc.. to ळ्ळ ळ्ळा ळ्ळा etc..
>
> That seems to work already.

-- 
Mike FABIAN <mfabian@redhat.com>
睡眠不足はいい仕事の敵だ。