http://llvm.org/bugs/show_bug.cgi?id=22362
Bug ID: 22362
Summary: tuple constructor accepts less arguments than the
tuple has values
Product: libc++
Version: unspecified
Hardware: PC
OS: All
Status: NEW
Severity: normal
Priority: P
Component: All Bugs
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected], [email protected]
Classification: Unclassified
I'm expecting this code to not compile:
std::tuple<unsigned, unsigned> t(1);
but it does (and it behaves as if I had provided 1 and 0). It was suggested to
me that this stems from an implementation of N4064 as detailed in
http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2014/n4064.html - but I'm
not spotting it there, and I find it rather surprising that this should work.
It caused a bug where I emplace()d a tuple into a vector and provided one
argument too few. This should imho be at least a warning if it cannot be an
error.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
LLVMbugs mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
