https://bugzilla.redhat.com/show_bug.cgi?id=1833858
--- Comment #16 from Akira TAGOH <tagoh(a)redhat.com> ---
We do.
<fontconfig>
<match target="scan">
<test name="family">
<string>Droid Sans Fallback</string>
</test>
<edit name="lang" mode="assign"> <!-- 3. replace lang
property with the
result of ops -->
<minus> <!-- 2. subtract the given langset from lang property -->
<name>lang</name>
<langset> <!-- 1. convert string to langset -->
<string>ko</string>
</langset>
</minus>
</edit>
</match>
</fontconfig>
If the decomposed glyphs are really required to satisfy Hangul rendering, it
should be reflected to ko.orth instead of the sort of this workaround. Though,
needing some feedback about it from native speakers or hangul experts.
Please note that constructing strict and complicated orthograhy file isn't what
it is intended for. it should be enumerated for minimal requirements.
Current ko.orth was built based on KS X 1001 character set. that would means
Droid Sans Fallback satisfies KS X 1001 but not good enough somehow, for
harfbuzz.
--
You are receiving this mail because:
You are on the CC list for the bug.