<i18n dev> [9] RFR: 8141243: Unexpected timezone returned after parsing a date

Masayoshi Okutsu masayoshi.okutsu at oracle.com
Thu Dec 3 04:48:23 UTC 2015


Naoto and I discussed the preferred ordered map. We decided to add the 
following change to the fix. This additional change doesn't affect the 
behavior, though.

diff --git a/make/src/classes/build/tools/cldrconverter/ResourceBundleGenerator.java b/make/src/classes/build/tools/cldrconverter/ResourceBundleGenerator.java
--- a/make/src/classes/build/tools/cldrconverter/ResourceBundleGenerator.java
+++ b/make/src/classes/build/tools/cldrconverter/ResourceBundleGenerator.java
...
@@ -136,7 +137,7 @@
              for (String preferred : preferredTZIDs) {
                  if (map.containsKey(preferred)) {
                      newMap.put(preferred, map.remove(preferred));
-                } else if ("GMT".equals(preferred) &&
+                } else if (("GMT".equals(preferred) || "UTC".equals(preferred)) &&
                             metaKeys.contains(CLDRConverter.METAZONE_ID_PREFIX+preferred)) {
                      newMap.put(preferred, preferred);
                  }

Thanks,
Masayoshi

On 11/23/2015 12:45 PM, Masayoshi Okutsu wrote:
> That part doesn't seem to affect UTC because it comes from the legacy 
> JRE locale data.
>
> Masayoshi
>
> On 11/21/2015 2:57 AM, Naoto Sato wrote:
>> In ResourceBundleGenerator.java, if we move "UTC" into the preferred 
>> names list, should we also treat it when we create the preferred 
>> ordered map? In other words, should we do the same for "UTC" at the 
>> line 140?
>>
>> Naoto
>>
>> On 11/20/15 12:24 AM, Masayoshi Okutsu wrote:
>>> Hello,
>>>
>>> This is a follow up of the fix for JDK-8141243 done by Ramanand. I've
>>> changed resources so that time zone ID "UTC" is picked up by
>>> SimpleDateFormat before Antarctica/Troll.
>>>
>>> issue: https://bugs.openjdk.java.net/browse/JDK-8141243
>>>
>>> webrev: http://cr.openjdk.java.net/~okutsu/9/8141243/webrev.00/
>>>
>>> Thanks,
>>> Masayoshi
>>>
>



More information about the i18n-dev mailing list