Re: error: target of using declaration conflicts with declaration already in scope

[John W. Eaton]

On 05/31/2016 02:25 PM, Ben Abbott wrote:
> John/others,
>
> I’m seeing the error below on Mac OS X. I’m using Apple’s clang to build.
>
> Apple LLVM version 7.0.2 (clang-700.1.81)
> Target: x86_64-apple-darwin14.5.0
> Thread model: posix
>
> Bisecting produced the result below.
>
> The first bad revision is:
> changeset:   21782:2aef506f3fec
> user:        John W. Eaton <address@hidden>
> date:        Wed May 25 16:51:16 2016 -0400
> summary:     use namespace for lo-mappers.h functions
>
> I’m not proficient enough with c++ to understand what is happening, but am 
> happy to attempt a fix if it is easier for someone to explain to me what needs 
> to be done. if additional information is needed from my end, please let me know.
>
> Ben
>
>    CXX      libinterp/libinterp_liboctinterp_la-octave.lo
> In file included from libinterp/dldfcn/__delaunayn__.cc:50:
> In file included from libinterp/corefcn/Cell.h:31:
> In file included from ./liboctave/array/Array.h:38:
> In file included from ./liboctave/array/idx-vector.h:38:
> In file included from ./liboctave/util/oct-inttypes.h:36:
> liboctave/numeric/lo-mappers.h:61:16: error: target of using declaration 
> conflicts with declaration already in scope
>      using std::arg;
>                 ^
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/complex:907:1:
>  note: target of using declaration
> arg(double __re)
> ^
> liboctave/numeric/lo-mappers.h:58:19: note: conflicting declaration
>      inline double arg (double x) { return atan2 (0.0, x); }
>                    ^
> liboctave/numeric/lo-mappers.h:61:16: error: target of using declaration 
> conflicts with declaration already in scope
>      using std::arg;

I was hoping to import existing versions of some standard functions into 
the octave::math namespace without having to write wrappers.  The using 
declaration seemed to work with GCC.  But I see that for some of these, 
C++ now provides versions for complex<T> and T arguments.  I don't know 
how long that has been true, but I didn't notice until now.  Could you 
try the attached patch?

Thanks,

jwe

diffs.txt
Description: Text document