[7u-dev] Request for review: 8037502: build.tools.javazic.Zoneinfo incorrectly calculates raw GMT offset change time

[Masayoshi Okutsu]

Hi Aleksej,

The bug report says, "The JDK7 code incorrectly calculates the raw GMT 
offset." Is there any example of the symptom in JDK 7? And there should 
be a regression test for the symptom.

Thanks,
Masayoshi

On 1/15/2015 6:59 AM, Aleksej Efimov wrote:
> Hi,
>
> Can I have a quick review please for the bug [1] inspired by already 
> resolved JDK8/9 test failure [2]: this test uses a copy of JDK7 
> classes and check if the JSR310 generated data is the same as data 
> generated by JDK7 code. The proposed fix [3] solves the possible 
> incorrect calculations of raw GMT offset change time in JDK7 code.
> Testing results: All JTREG tests somehow related to tzdata were 
> executed and JPRT shows no issues with the tz related tests.
> Adding the i18n-dev mail-list and JDK-8037180 reviewers.
>
> With Best Regards,
> Aleksej
>
> [1] JBS bug: https://bugs.openjdk.java.net/browse/JDK-8037502
> [2] JDK8/9 test failure: https://bugs.openjdk.java.net/browse/JDK-8037180
> [3] Webrev: http://cr.openjdk.java.net/~aefimov/8037502/7/webrev.00/