[threeten-dev] A potential bug in JapaneseDate.of(Era era, int yearOfEra, int month, int dayOfMonth)

[Frank Ding]

Hi Roger,
   A full test is available @ 
http://cr.openjdk.java.net/~dingxmin/jsr310/JpDateTest.java. Please take 
a look at it.

Best regards,
Frank

On 4/30/2013 5:56 AM, roger riggs wrote:
> Hi Frank,
>
> Thanks for the reports, we will followup.  If the test case code is 
> available
> it would be useful to integrate it into the existing java.time tests.
>
> Thanks, Roger
>
> On 4/27/2013 3:00 AM, Frank Ding wrote:
>> Hi Threeten guys,
>>   Our Japanese test team found out a bug when calling 
>> JapaneseDate.of(Era era, int yearOfEra, int month, int dayOfMonth). 
>> See the code below.
>>
>>   JapaneseDate date = 
>> JapaneseDate.of(java.time.chrono.JapaneseEra.HEISEI, 1, 1, 7);
>>
>>   Heisei era starts in Heisei 1/1/8 which implies Heisei 1/1/7 is 
>> invalid.  According to its spec [1],
>>   It throws DateTimeException - if the value of any field is out of 
>> range, or if the day-of-month is invalid for the month-year, or if 
>> the date is not a Japanese era.
>>   So the correct behavior is observing DateTimeException for the code 
>> above.  What's your opinion of it?
>>
>> Best regards,
>> Frank
>>
>> [1] 
>> http://download.java.net/jdk8/docs/api/java/time/chrono/JapaneseDate.html#of%28java.time.chrono.Era,%20int,%20int,%20int%29
>>
>