Here are my comments.
- Looks like this change removed the 8055088 fix for BreakIteratorInfo
optimization.
- The langtags field in each *Impl class should be volatile.
- DateFormatProviderImpl has static langtags to be shared by some other
*Impl. But JRE and CLDR have different sets of language tags.
- Now language tags are hard coded, but that will be error prone when
adding or changing locale data. There should be a test case for checking
consistency between existing resource bundles and the hard-coded tags
(to detect any missing locales in the hard-coded lists).
- JREENLocaleDataMetaInfo: the first upper case part looks like one word
JREEN. JreEnLocaleDataMetaInfo may be better. It's not consistent with
others, though.
Masayoshi
On 8/23/2014 3:46 AM, Naoto Sato wrote:
Hello,
Please review the changes for the following issue:
https://bugs.openjdk.java.net/browse/JDK-8038436
The proposed changes are located at:
http://cr.openjdk.java.net/~naoto/8038436/webrev.3/
The idea is to introduce an SPI so that supported locales are
dynamically searched at runtime, not depending on the existence of
physical jar files.
I'd appreciate it if build folks could review the makefile related
changes, i18n forks to review locale framework files, and Mandy from
modularization point of view.
Naoto
