<i18n dev> RFR: 8274544: Langtools command's usage were garbled on Japanese Windows

Jan Lahoda jlahoda at openjdk.java.net
Fri Oct 1 12:24:27 UTC 2021


On Fri, 1 Oct 2021 11:56:03 GMT, Jan Lahoda <jlahoda at openjdk.org> wrote:

> Regarding javac, the patch to `Log.java` seems to be in a reasonable direction: the write is to the physical `System.out/err` which should be done(?) using the native encoding. The order of the changed lines should be fixed, so that the javadoc is kept above the `initWriters` method. (As a secondary comment, it maybe a matter of discussion on whether keeping the native encoding in a static field is warranted here, but I don't mind it much.)

I've forgot to write a note on the test, sorry: simply add `native.encoding` into `noResourceRequired` set in the test. The test checks that there are not hardcoded string that should be part of the resource bundle (and the resource bundle does not have unused keys), but names of system properties should be excluded, which is what the `noResourceRequired` set does.

-------------

PR: https://git.openjdk.java.net/jdk/pull/5771


More information about the i18n-dev mailing list