In general, I agree with Stephen that the tests should be testing an
algorithm against facts.
Embedding an algorithm in the test increases the risk that the test will
just replicate the
implementation code and therefor not be much of a test.
Though in this case, the specification of aligned day of week is of a
If the test were to independently compute the correct answer, it would
be valid as a 'tck' test.
Since the Hijrah calendar is data driven, the tests should be in
Tests in java/time/tck/... should correspond directly to specified
In this case, the algorithm is specified but the test is data dependent.
(Perhaps a gray area).
On 10/12/2016 11:03 AM, Anubhav Meena wrote:
Have incorporated the changes you suggested. Updated webrev is
Please review and suggest if anymore changes are required.
On Oct 12, 2016, at 3:21 PM, Anubhav Meena <anubhav.me...@oracle.com
On Oct 7, 2016, at 2:12 PM, Anubhav Meena <anubhav.me...@oracle.com
Bug Id :https://bugs.openjdk.java.net/browse/JDK-8163330
Issue:The HijrahDate class incorrectly calculates the aligned-day-of-week
field. It based the calculation on the day-of-week, when it should
be based on the day-of-month.
Thanks and Regards,