On Thu, 22 Oct 2020 22:29:16 GMT, Joe Wang <jo...@openjdk.org> wrote:
>> Marked as reviewed by bchristi (Reviewer). > > Hi Naoto, > > Do we need a test similar to ISO3166 where display names of Locale.ROOT and > Locale.US are compared? I see LocaleEnhanceTest.java has references to > Locale.ROOT and a few selected names were updated. But we don't seem to have > a test like ISO3166 that does a straight-forward comparison. > > The code you used to generate the updated LocaleNames.properties (in the > comment section), that seems to be valuable. I wonder if there's a better way > to preserve it, as part of the manual tool, or maybe put it back into the > repo as part of a test? Hi Joe, > Do we need a test similar to ISO3166 where display names of Locale.ROOT and > Locale.US are compared? I see LocaleEnhanceTest.java has references to > Locale.ROOT and a few selected names were updated. But we don't seem to have > a test like ISO3166 that does a straight-forward comparison. test/jdk/sun/text/resources/LocaleData validates all the changes with this changeset (added lines at the end), by checking the names in the resource bundle. > > The code you used to generate the updated LocaleNames.properties (in the > comment section), that seems to be valuable. I wonder if there's a better way > to preserve it, as part of the manual tool, or maybe put it back into the > repo as part of a test? Yeah, but I cannot think of a better location to preserve it. The reason is that it cannot be used as a test because CLDR's English names and COMPAT's root names can inherently be different. I just regard the code to be one-off to retrieve authentic English names for those language ids. Naoto ------------- PR: https://git.openjdk.java.net/jdk/pull/786
