This is the mail archive of the archer@sourceware.org mailing list for the Archer project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [python] Anonymous namespaces prevent some Python pretty-printing


>>>>> "Daniel" == Daniel Jacobowitz <drow@false.org> writes:

>> That's because
>> cp_demangled_name_to_comp("std::map<int, <unnamed>::Foo, std::less<int>, std::allocator<std::pair<const int, <unnamed>::Foo> > >", ...)
>> returns NULL, with error set to "syntax error, near ..."

Daniel> FWIW, there's already support for anonymous namespaces in the parser -
Daniel> but it sounds like this is something that has changed in GCC
Daniel> w.r.t. how anonymous namespaces are labeled.  The demangler produces
Daniel> (anonymous namespace)::Foo.

I see this in the dwarf from a simple example:

    <41>   DW_AT_name        : (indirect string, offset: 0x33): map<int, <unnamed>::Foo*>

I think that <unnamed> comes from g++, there are a couple places where
that appears.

I suppose I can try to make gdb recognize <unnamed> as well.

Or, we could fix it in g++ -- it seems strange that g++ and the
demangler would disagree.

Tom


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]