[icu-design] ICU4J break Thai character at an unexpected position

[Nattapong S. <na...@th...>]

Hi all,

Per this bug ticket:http://bugs.icu-project.org/trac/ticket/6317
My concern is, if this's really what ICU design, it's not an expected for
Thai.

Thai leading vowels(U+0E40-U+0E44) is expect to break immediately at
position after it when using CharacterBreakIterator.
Given this string: "กูกินกุ้งปิ้งอยู่ในถ้ำ"
Expecting outcome: "กู", "กิ", "น", "กุ้", "ง", "ปิ้", "ง", "อ", "ยู่", "ใ", "น",
"ถ้ำ" ICU4J outcome: "กู", "กิ", "น", "กุ้", "ง", "ปิ้", "ง", "อ", "ยู่", "ใน", "ถ้ำ"

Note: The different is "ใ" and "ใน" which "ใ" is Thai leading vowel.

Best regards,
Nattapong Sirilappanich
Engineer, National Language Development - GCoC & TSC Thailand
Software Group, IBM Thailand Co., Ltd.
National Phone: 02-273-4715, Fax: 02-619-2030
International Phone: +66 2 273 4715, Fax: +66 2 619 2030
e-mail: na...@th...