Add Current ICU Word-breaking for Khmer?

l10n
#1

Hello,

I was wondering if someone could help me to add current ICU word-breaking
capabilities for Khmer (Cambodian) into LibreOffice. Word-break support for
Khmer has been added to ICU (http://bugs.icu-project.org/trac/ticket/8329)
but I am not sure how to get it added into LibreOffice for spelling checker
and line-break ease of use (Khmer has no "real" spaces between words like
Thai, but we add zero-width spaces between words in order to use a spelling
checker and for line-breaks).

I helped add Khmer word-breaking to ICU, but I am not much of a programmer,
so if anyone would be willing to help, that would be wonderful.

Thank you,
Nathan Wells

#2

Hi Nathan,

I was wondering if someone could help me to add current ICU word-breaking
capabilities for Khmer (Cambodian) into LibreOffice. Word-break support for
Khmer has been added to ICU (http://bugs.icu-project.org/trac/ticket/8329)

That states it was added to ICU 4.8, we recently upgraded our internal
ICU to 49 so it should already be available in LibO master and 3-6 at
least on Windows where we ship the internal ICU. Whether it is available
in a specific Linux distribution depends on the version shipped with the
distribution, some are still on ICU 4.6

Unless we override something with i18npool's breakiterator data it
should work out of the box with ICU >= 4.8, would be nice if you could
verify that somehow.

  Eike

#3

Thanks for your reply Eike,

I did a test on LibO master 3.6 in Windows 7 64-bit and confirm that
line-breaking for Khmer is working correctly! This is wonderful! It is the
first confirmed project that has used the new code in ICU.

But for spell checking, the word-breaks are not being used. Is there a way
to enable the ICU word breakiterator word boundary for spell checking in
LibreOffice? If not, does anyone know who we can contact to point us in the
right direction (line-breaks are a great addition, but automatic
word-breaks for spell checking would be absolutely a great thing for the
Cambodian people).

Thanks,
Nathan