Vladimir Prus wrote:
Previously, we decided on this list that on a certain date, all remaining failures will be marked as expected.
Freeze was scheduled to be at 11:00 UTC on Mar 2nd; the regression was only introduced at 23:08 UTC on Mar 1st. And the file in question (<boost/none.hpp>) has been modified twice since then to fix other regressions introduced by that commit. Don't get me wrong -- I'm not against that commit having gone in at the last minute -- other parts of the commit fixed several important issues, but it seems unreasonable to expect any problems with it to have been fixed in the twelve hours between commit and freeze. And indeed the file in question has been edited in the last week to fix other regressions.
Now, you're pointing out that we have a regression, for which we don't even have a test.
Yes you do. The patch I attached to my email this morning added a test to the test suite.
In light of that decision, we probably can add a test for that problem, and immediately mark all failures of said test as expected. But that would not be very helpful. I don't think we should give this problem any bonus points just because it's not discovered by the tests yet.
Indeed, and I'm not suggesting it should. But if the test suites were indiciating a failure that would silently change legitimate user code across all platforms, I would hope that too would be fixed rather than marked 'expected'. Richard