[LLVMbugs] [Bug 19460] New: Unable to detect operator>> (extraction operator) using libc++

bugzilla-daemon at llvm.org bugzilla-daemon at llvm.org
Wed Apr 16 22:10:12 PDT 2014


http://llvm.org/bugs/show_bug.cgi?id=19460

            Bug ID: 19460
           Summary: Unable to detect operator>> (extraction operator)
                    using libc++
           Product: libc++
           Version: 3.4
          Hardware: Macintosh
                OS: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: All Bugs
          Assignee: unassignedclangbugs at nondot.org
          Reporter: lachlan.hetherton at csiro.au
                CC: llvmbugs at cs.uiuc.edu, mclow.lists at gmail.com
    Classification: Unclassified

Created attachment 12390
  --> http://llvm.org/bugs/attachment.cgi?id=12390&action=edit
Test file: compile this to reproduce the issue.

We have been using some code for a number of years now to detect whether the
streaming operators (operator<< and operator>>) exist on a given class. I've
recently updated one of our machines to OSX 10.9, which uses Clang 3.4 and
libc++ by default, and now our operator>> detection code no longer compiles. If
I explicitly enable libstdc++ however, it still compiles.

It may well be that our code no longer meets a new stringent requirement of the
c++11 standard (keep in mind that we're not yet targeting this: -std=c++11 is
not enabled), and if so, please feel free to point out my glaring error!

I've attached a very cut-down example of the affected code. To reproduce the
issue:

1) Compile the attached cpp file with clang++ v3.4 as follows:

clang++ extraction_test.cpp

2) You'll notice that it fails. To confirm that it works with the stdc++
library, compile as follows:

clang++ -stdlib=libstdc++ extraction_test.cpp

I apologise if this is something new and / or obvious - I'm new to compiling on
anything other than GCC / MSVC, but I thought it best to raise this issue in
case there was a genuine problem in the new libc++ library that may affect
others trying to implement this kind of functionality. 

Regards,
Lachlan Hetherton

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20140417/a6cb0fcf/attachment.html>


More information about the llvm-bugs mailing list