This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Mangle question
- To: dan at cgsoftware dot com
- Subject: Re: Mangle question
- From: Mike Stump <mrs at windriver dot com>
- Date: Mon, 28 Aug 2000 20:24:48 -0700 (PDT)
- Cc: gcc at gcc dot gnu dot org
> Date: Mon, 28 Aug 2000 18:50:35 -0700 (PDT)
> From: Daniel Berlin <dan@cgsoftware.com>
> To: Mike Stump <mrs@windriver.com>
> Except it's been proven it's ambiguous in the present of underscored names.
In the case that I added mine to, names that came before it, had a
length that said exactly when they were finished. No maximal munch.
So, it was symbols like
Q_20_3Max1A_20_aaaaaaaaaaaaaaaaaaaa...
:-( Sorry to hear about the current problems.
Sounds like you need to find the:
OB_PUTID (name);
of the thing that is ambiguous, and make it conform more to how the
rest of the mangler works:
icat (IDENTIFIER_LENGTH (tname));
OB_PUTID (tname);
to get rid of the ambiguity, it is always wrong to fix ambiguity in
the demangler by just trying more demangles :-(.
Since IDs never begin with numbers, should even be possible to
demangle older things...
I'd say it was a flaw that the icat was missing...