On Wed, 11 May 2022 18:25:09 GMT, Phil Race <p...@openjdk.org> wrote:
> This replaces the PR from Toshio https://git.openjdk.java.net/jdk/pull/8329 > It is similar in the idea from what we converged on towards the end there but > 1) I'd like to preserve all the support for the old encodings since JEP-400 > explicitly supports using -Dfile.encoding=windows-31j (for example) > 2) I wanted to have the look up for the font to use in the windows text > control succeed on the first font it finds not the last. In practice I expect > this to be a no-op but if someone did have two that match .. don't you think > they mean the first one to be the primary ? > 3) I needed to update the encoding match to ensure that some optional > re-ordering for Japanese logical fonts works in the UTF-8.ja locale as that > is now the default. > > I verified this fix by changing regional settings as well as passing > -Duser.language=ja -Duser.country=JP and logical fonts on lightweights > (verified in Font2DTest) are as expected and the "fonttest.java" now behaves > as it did in JDK 17 and earlier. Since you need to switch regional settings > automating this test was not possible. > > @toshiona you very likely want to take a look at it. src/java.desktop/windows/classes/sun/awt/windows/WFontConfiguration.java line 267: > 265: case "ru" -> "RUSSIAN_CHARSET"; > 266: case "el" -> "GREEK_CHARSET"; > 267: case "iw" -> "HEBREW_CHARSET"; It would be safe to specify both old(iw)/new(he) language codes for Hebrew, as "he" is now returned for Hebrew by default (https://bugs.openjdk.java.net/browse/JDK-8263202) ------------- PR: https://git.openjdk.java.net/jdk/pull/8662
