This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: RFA: Make cli-out follow gdb_stdout
On Mon, Jul 22, 2002 at 04:11:15PM -0400, Andrew Cagney wrote:
>
> [...]
>
> >Well, that's not helpful in the short term :P
> >
> >What do you envision by explicitly parametrized i/o object? I really
> >don't think that's a good idea. It would have to be passed down all
> >the way to every target function which might want to print some kind of
> >output. That's a lot of bulk for no visible gain; somewhere down the
> >road if we supported multiple instantiations of the gdb library, maybe,
> >but we're so far away from that that this sort of direction seems
> >futile, until we are at least passing an object cookie everywhere.
>
> As they say, I don't yet know. I think it is pretty clear that the I/O
> is just one part of this state problem. Since of thread, frame,
> register cache, ... are heavily dependant on state implemented with
> globals a guess is an object (or object relationship) that contains that
> state. In the mean time we need to ensure that we're not entrenching
> the problem (eg by adding another bit of code assuming global state).
>
> For the immediate problem, the intent is for catch_exceptions() to be
> used when overriding gdb's I/O. Given, looking at the code, it too
> fails to keep that assertion (oops):
>
> > global uiout->stream->ui_file == global gdb_stdout
>
> I think the thing to do is fix catch-exceptions and then use that.
catch_exceptions is for executing a particular command. These commands
redirect the output of multiple commands; possibly an entire session.
catch_exceptions isn't built for that sort of use.
The only way that I could see to fix catch_exceptions in this regard
would involve some way to get gdb_stdout etc. from the uiout object.
In that case I'm just going to restrict my commands to CLI only, and
do:
uiout = cli_out_new (new_gdb_stdout)
in the function.
Patch withdrawn until we figure out a better model here. I bet you
aren't going to like my introduction of another wrapper much either...
but I fail to see an option.
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer