Re: [RFC/RFA] do not call decode_compound with Ada expressions.

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Joel" == Joel Brobecker <brobecker@adacore.com> writes:

Tom> There is a comment saying this is for C++ and Java.
Tom> How about checking explicitly for those and nothing else?

Joel> That would work for me, but I'm concerned that we might be breaking
Joel> other languages which expect this to help (somehow) even when they
Joel> are using language C. For instance, I think obj-C developers rely on
Joel> being able to use obj-C linespecs even when the current language
Joel> is C.

Yeah, but decode_compound is really just about inheritance.
At least, that is the only reason I can think of for it to exist.

That is, "break derivedclass::method" will also set breakpoints in
"baseclass::method".

The Objective C stuff, IIUC, is handled by decode_objc, along a
different patch.

Joel> So, at a minimum, I would probably include c, c++, objc, java.
Joel> But why not minimal and asm? If the language is auto,asm, I think
Joel> users might expect the debugger to be able to understand C++ linespecs.

I tend to think that "minimal" would be best if it just looked at
linkage names, or maybe linkage names and demangled names, but no funny
treatment.

I see your point about "auto;asm" -- but if we carry that argument to
its logical conclusion, we should not check current_language at all in
linespec.  That maybe would have been a nice idea at the beginning, but
I think now it is far too late to change.

Tom