[rfa/6.0] TUI patch for paginated backtraces

[rfa/6.0] TUI patch for paginated backtraces

[Daniel Jacobowitz]

This is the exact parallel to a CLI patch I checked in last month.  This
fixes the test in fputs_maybe_filtered, which causes backtraces to be
paginated again.  It should only affect the TUI in CLI mode.

[Now that we have interp-switching infrastructure, I think tui_old_uiout
should go away, and instead cli_out should be used - we can switch
interpreters, that's the whole point.]

OK?

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

2003-08-01  Daniel Jacobowitz  <drow@mvista.com>

	* tui-interp.c (tui_resume): Update tui_old_uiout's stream
	to gdb_stdout.

diff -urp gdb-5.3.20030801/gdb/tui.orig/tui-interp.c gdb-5.3.20030801/gdb/tui/tui-interp.c
--- gdb-5.3.20030801/gdb/tui.orig/tui-interp.c	2003-08-01 10:13:07.000000000 -0400
+++ gdb-5.3.20030801/gdb/tui/tui-interp.c	2003-08-01 14:10:02.000000000 -0400
@@ -63,7 +63,23 @@ tui_init (void)
 static int
 tui_resume (void *data)
 {
+  struct ui_file *stream;
+
+  /* gdb_setup_readline will change gdb_stdout.  If the TUI was previously
+     writing to gdb_stdout, then set it to the new gdb_stdout afterwards.  */
+
+  stream = cli_out_set_stream (tui_old_uiout, gdb_stdout);
+  if (stream != gdb_stdout)
+    {
+      cli_out_set_stream (tui_old_uiout, stream);
+      stream = NULL;
+    }
+
   gdb_setup_readline ();
+
+  if (stream != NULL)
+    cli_out_set_stream (tui_old_uiout, gdb_stdout);
+
   if (tui_start_enabled)
     tui_enable ();
   return 1;

Re: [rfa/6.0] TUI patch for paginated backtraces

[Stephane Carrez]

Hi!

Daniel Jacobowitz wrote:
> This is the exact parallel to a CLI patch I checked in last month.  This
> fixes the test in fputs_maybe_filtered, which causes backtraces to be
> paginated again.  It should only affect the TUI in CLI mode.
> 
> [Now that we have interp-switching infrastructure, I think tui_old_uiout
> should go away, and instead cli_out should be used - we can switch
> interpreters, that's the whole point.]
> 
> OK?
> 
Yes.

I seem to get a warning that 'cli_out_set_stream' is not defined.
Could you add the 'cli-out.h' include?

	Stephane

Re: [rfa/6.0] TUI patch for paginated backtraces

[Daniel Jacobowitz]

On Fri, Aug 01, 2003 at 11:24:27PM +0200, Stephane Carrez wrote:
> Hi!
> 
> Daniel Jacobowitz wrote:
> >This is the exact parallel to a CLI patch I checked in last month.  This
> >fixes the test in fputs_maybe_filtered, which causes backtraces to be
> >paginated again.  It should only affect the TUI in CLI mode.
> >
> >[Now that we have interp-switching infrastructure, I think tui_old_uiout
> >should go away, and instead cli_out should be used - we can switch
> >interpreters, that's the whole point.]
> >
> >OK?
> >
> Yes.
> 
> I seem to get a warning that 'cli_out_set_stream' is not defined.
> Could you add the 'cli-out.h' include?

Thanks for noticing my mistake!  Committed with that fix.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer