<i18n dev> RFR: 8031145: Re-examine closed i18n tests to see it they can be moved to the jdk repository.

Yuka Kamiya yuka.kamiya at oracle.com
Thu May 26 10:27:43 UTC 2016


The fix looks okay to me.


On 2016/05/26 18:41, Masayoshi Okutsu wrote:
> Naoto pointed out that test/java/text/Format/common/*Format.props 
> should have the copyright header. Unfortunately, 
> test/java/text/Format/common/PParser.java, which parses the .props 
> files, doesn't support any comment lines. So, PParser has been changed 
> to be able to support comment lines. 
> test/java/text/Format/common/FormatIteratorTest.java, PParser.java, 
> and *Format.props have been changed. No other changes.
> Webrev:
> http://cr.openjdk.java.net/~okutsu/9/8031145/webrev.02
> Thanks,
> Masayoshi
> p.s. The URL for webrev.01 was wrong. It should be 
> http://cr.openjdk.java.net/~okutsu/9/8031145/webrev.01.
> On 5/24/2016 11:49 PM, Masayoshi Okutsu wrote:
>> Changed all binary files (serialized objects) into hex dump text 
>> files (with ".ser.txt" suffix) and added a utility class to read it 
>> as an InputStream.
>> Updated webrev:
>> http://javasoft.jp.oracle.com/~mokutsu/9/8031145/webrev.01/
>> Thanks,
>> Masayoshi
>> On 5/23/2016 7:02 PM, Alan Bateman wrote:
>>> On 23/05/2016 10:12, Masayoshi Okutsu wrote:
>>>> On 5/23/2016 5:10 PM, Alan Bateman wrote:
>>>>> Is there anything that we can do with the binary files? In the 
>>>>> case of the .ser file then could it be a byte[] in the test with 
>>>>> an execution mode that re-generates it? There is also at least one 
>>>>> .data file that looks to be the serialized version of a bad 
>>>>> ChoiceFormat and maybe that could be moved into the source too.
>>>> I think it's OK for these tests to use ByteArrayInputStream to test 
>>>> serialization compatibility. I will convert the binary files to 
>>>> source files.
>>> Thanks, that would be great and should avoid needing to checking in 
>>> any binary files.
>>> -Alan

More information about the i18n-dev mailing list