-
Bug
-
Resolution: Fixed
-
P4
-
21
-
None
In JDK 21, the contents of make/jdk/src/classes/build/tools/cldrconverter/OtherCommonLocales.properties are invalid UTF-8:
$ grep --text mgo ./make/jdk/src/classes/build/tools/cldrconverter/OtherCommonLocales.properties
mgo=Meta��
The file was adapted from CLDR data that I think is now located here: https://raw.githubusercontent.com/unicode-org/cldr-staging/release-42/docs/charts/42/tsv/locale-coverage.tsv
From
curl https://raw.githubusercontent.com/unicode-org/cldr-staging/release-42/docs/charts/42/tsv/locale-coverage.tsv | grep Meta | xxd
it looks like that was supposed to be the character U+02BC (cabc in the UTF-8 encoding). The bytes in the corresponding range of the checked-in file are `a0ba`, which is not a valid UTF-8 encoding. The contents of OtherCommonLocales.properties in the latest version of the JDK repo match the CLDR data.
$ grep --text mgo ./make/jdk/src/classes/build/tools/cldrconverter/OtherCommonLocales.properties
mgo=Meta��
The file was adapted from CLDR data that I think is now located here: https://raw.githubusercontent.com/unicode-org/cldr-staging/release-42/docs/charts/42/tsv/locale-coverage.tsv
From
curl https://raw.githubusercontent.com/unicode-org/cldr-staging/release-42/docs/charts/42/tsv/locale-coverage.tsv | grep Meta | xxd
it looks like that was supposed to be the character U+02BC (cabc in the UTF-8 encoding). The bytes in the corresponding range of the checked-in file are `a0ba`, which is not a valid UTF-8 encoding. The contents of OtherCommonLocales.properties in the latest version of the JDK repo match the CLDR data.
- relates to
-
JDK-8303039 Utilize `coverageLevels.txt`
- Resolved