http://d.puremagic.com/issues/show_bug.cgi?id=9071
--- Comment #6 from monarchdo...@gmail.com 2012-12-27 06:00:26 PST --- (In reply to comment #1) > This is actually a problem with "hasSlicing": It is currently being tightened > so that the result of a slice operation *must* be convertible back to the > original type. If this is not the case, then the range will not be considered > sliceable. > > Once this is deployed, you will (should) be forwarded to the "correct" > overload > of sortImpl: The one that doesn't use slicing. > > Still, could you please provide the code example that fails? This will help us > figure out which of the following is the culprit: > * The range? > * The definition of hasSlicing? > * The implementation of sort? > And correctly address the issue. OK: According to the new tests, the new definition of has slicing partially fixes the problem. However, there are currently no restraints for "sort", so the compilation error is: /usr/local/include/dmd2git/std/algorithm.d(7357): Error: template instance SortedRange!(Deque, "a < b") SortedRange!(Deque, "a < b") does not match template declaration SortedRange(Range, alias pred = "a < b") if (isRandomAccessRange!(Range) && hasLength!(Range)) I'll add a new restraint then. -- Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email ------- You are receiving this mail because: -------
