Bug 7365 – [Regression after 2.057] AAs broken for Object keys and values with opEquals

Didn't fully track it down yet, but the following program asserts in getImpl (mtype.c, line 4371) because AssociativeArray!(Object, TickDuration) can't be instantiated (DMD 2.058 Git): --- struct TickDuration { bool opEquals(ref const TickDuration rhs) const { return true; } } void main() { TickDuration[Object] aa; aa.keys; } --- Having a look at the gagged errors, it seems like the cause is related to the opEquals const correctness issue: --- Error: function object.opEquals (Object lhs, Object rhs) is not callable using argument types (const(Object),const(Object)) Error: cannot implicitly convert expression (this.key) of type const(Object) to object.Object Error: cannot implicitly convert expression (p.key) of type const(Object) to object.Object --- The error is encountered while generating the default opEquals for the Slot type added in https://github.com/D-Programming-Language/druntime/commit/cd7b477b83cc8a75a90258bd2ab63fa8c92ba10b.

Forgot to mark as regression.

Commit pushed to master at https://github.com/D-Programming-Language/dmd https://github.com/D-Programming-Language/dmd/commit/ba943d54d4f97ec98261d5d00631aecbea7cdbea add test case for issue 7365

It works for me, but I added a test case for it anyway to the test suite.

Seems to WORKSFORME now as well, not sure when it was fixed or what the problem was before…

Ah, screw it, now the auto-tester fails because of it as well: http://d.puremagic.com/test-results/test_data.ghtml?dataid=146136. Must have inadvertently introduced a workaround in my original client code, sorry for the noise.

Commit pushed to master at https://github.com/D-Programming-Language/druntime https://github.com/D-Programming-Language/druntime/commit/048caceac7c678b16ccd0f6555502a6a622a5ba3 fix Issue 7365 - [Regression after 2.057] AAs broken for Object keys and values with opEquals