Re: [rfc] python API exposing inferior's frame stack.

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Thiago" == Thiago Jung Bauermann <bauerman@br.ibm.com> writes:

Tom> I think symbols are assumed to use the host encoding.

Thiago> Are they? This really surprises me, since symbols are obtained from the
Thiago> executable file. Or the symbol reading mechanism converts them from
Thiago> target to host encoding?

I think this is a bit of a weird area in gdb.  From what I can tell,
symbol readers don't convert between encodings.  So, I think if
symbols are actually encoded in some other encoding, we'll do the
wrong thing eventually.

Anyway, gdb definitely does currently assume that symbols are in the
host encoding.  E.g., the other caller of find_frame_funname calls
fprintf_symbol_filtered, which may print the argument string directly.

Tom