Re: [RFA] Fix C++ typedefs parsing errors

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Doug" == Doug Evans <dje@google.com> writes:

Doug> It looks fine to me, with one proviso.
Doug> Do we want the typedef-stripped type passed to write_exp_elt_type?
Doug> It kinda feels like we don't, but I don't honestly know.

I was wondering that too.

What would the failure mode look like?  I can't think of one.  It seems
to me that any actual definition would have to appear in the original
type, not solely in any typedef.

Also, if we do find some failure mode, it isn't like this patch can make
anything worse.  The current code will just error out early for any
typedef.

Therefore, this patch is ok.

thanks,
Tom