Chrome Version: 71.0.3556.0
OS: 11042.0.0

"Vietnamese keyboard (Telex)" currently implements a flexible diacritic trigger mode where diacritic triggers can be typed "later" (towards the end) or "earlier" (towards the front). 

The IME is supposed to place the diacritic on the most linguistically correct letter as per the orthography at every step. However, it's not always the case.

E.g. (non-exhaustive):

type "tufaan"
--> expect "tù" -> "tùa" -> "tuầ" -> finally "tuần"
--> actual "tù" -> "tùa" --> "tuàa" --> finally "tuaàn"

type "cusooi" --> expect "cuối", currently get "cuoói"
--> expect "cú" -> "cúo" -> "cuố" --> finally "cuối"
--> actual "cú" -> "cúo" -> "cuóo" --> finally "cuoói"

type "thurow"
--> expect "thủ" -> "thủo" -> finally "thuở"
--> actual "thủ" -> "thủo" -> finally "thủơ"

type "giseeng"
--> expect "gí" -> "gié" -> "giế" -> finally "giếng"
--> actual "gí" -> "gíe" -> "giée" -> finally "giếng"

type "nguxyeen"
--> expect "ngũy" -> then "nguyẽ" -> "nguyễ" -> "nguyễn"
--> actual "ngũy" -> then "nguỹe" -> "nguyẽe" -> "nguyeẽn"

etc.

NOTE: This is expected to work correctly for *all* cases as per Vietnamese orthography, with regard to diacritic placement and syllable structure, so that it's consistent, predictable, and reliable.

 

Chrome Version: 71.0.3556.0
OS: 11042.0.0

"Vietnamese keyboard (Telex)" currently implements a flexible diacritic trigger mode where diacritic triggers can be typed "later" (towards the end) or "earlier" (towards the front). 

The IME is supposed to place the diacritic on the most linguistically correct letter as per the orthography at every step. However, it's not always the case.

E.g. (non-exhaustive):

type "tufaan"
--> expect "tù" -> "tùa" -> "tuầ" -> finally "tuần"
--> actual "tù" -> "tùa" --> "tuàa" --> finally "tuaàn"

type "cusooi"
--> expect "cú" -> "cúo" -> "cuố" --> finally "cuối"
--> actual "cú" -> "cúo" -> "cuóo" --> finally "cuoói"

type "thurow"
--> expect "thủ" -> "thủo" -> finally "thuở"
--> actual "thủ" -> "thủo" -> finally "thủơ"

type "giseeng"
--> expect "gí" -> "gié" -> "giế" -> finally "giếng"
--> actual "gí" -> "gíe" -> "giée" -> finally "gieéng"

type "nguxyeen"
--> expect "ngũy" -> then "nguyẽ" -> "nguyễ" -> "nguyễn"
--> actual "ngũy" -> then "nguỹe" -> "nguyẽe" -> "nguyeẽn"

etc.

NOTE: This is expected to work correctly for *all* cases as per Vietnamese orthography, with regard to diacritic placement and syllable structure, so that it's consistent, predictable, and reliable.