spelling changes during transcoding

[funderburkjim]

mw_AB needs to be transcoded to slp1, in order for current display systems to function properly.

During the course of this conversion, a few changes were made (in a temporary copy) of mw_AB.txt. These changes are shown in mwtranscode/change_ab_1.txt.

Request @Andhrabharati to make these changes in mw_AB.txt.

[Andhrabharati]

While I have changed 8 lines [2667, 230411, 102414, 141321, 815036, 108492, 239075 and 239076] as per the observations by @funderburkjim , I hesitate changing the other 5 lines [ă change to a (37436); ă change to a (116756); ŭ́ -> ú (118589); ŭ́ -> ú (369500); ŭ́ -> ú <s>°rŭ́</s> (425594)].

I do understand that the breve accent mark has no equivalent in slp1 or other encodings, it does have a significance (esp. the pronunciation etc.). https://www.dictionary.com/browse/breve

Also in line 116756 specifically, "(causing the change of {ā} and a to a before ṛ)" will have no 'proper sense' as compared to the printed text "(causing the change of {ā} and a to ă before ṛ)"

Request Jim to find a suitable solution to retain this breve accent mark and still 'handle' the transliteration to various schemes. [Probably he could add an 'exception' entry, equating the 'breve'-ed letter ('ă' and 'ŭ') to a plain small-letter ('a' and 'u') internally.]

[funderburkjim]

Use <s1> tag.

For those 5 with a breve-accented vowel, let's change <s>X</s> to <s1>X</s1>. This will get avoid the difficulty of displaying such an X in Devanagari, slp1, etc. See mwtranscode/change_ab_1a.txt.

Maybe @Andhrabharati can make these few changes in mw_AB.txt ?

[Andhrabharati]

Yes, this indeed is a simple solution, @funderburkjim !

I will change those 5 lines and push again now.

[Andhrabharati]

Luckily these are not at the HW position, so that the 'problem' could be resolved by putting them under s1-tag.

Pushed the latest file, with the 5 lines modified.

[funderburkjim]

Nothing else to mention in this issue. No current problems in transcoding.