Eli Zaretskii | 1 Sep 2002 07:34
Picon

Re: [PATCH]: Provide TUI fast key action with specific readline keymap


On Sat, 31 Aug 2002, Stephane Carrez wrote:

> > Thanks for working on this, but you never responded to my request to
> > change the manual accordingly.  I really don't think we should have
> > first-class user-level features that are not documented.
> 
> I agree with you; I responded and I sent the following patch few days ago:
> 
> [RFC]: Document TUI breakpoints and SingleKey mode
> 	http://sources.redhat.com/ml/gdb-patches/2002-08/msg00907.html

Sorry, I somehow missed that.

Those patches are approved with the following comments:

  - " <at> table  <at> emph" is IMHO not a good idea for markers, since  <at> emph is 
    typeset in a slanted font in the printed manual.  I suggest to use 
    " <at> table code" instead.

  - In the description of the SingleKey keybindings, please make the 
    index entries indicate that those are SingleKey bindings, like this:

     <at> kindex c <at> r{, SingleKey TUI key}

    Otherwise, the user looking at the index will not be able to 
    distinguish between, say, the entry for the normal GDB `c' (continue) 
    command and the SingleKey binding of `c'.

Thanks!
(Continue reading)

Eli Zaretskii | 1 Sep 2002 07:37
Picon

Re: [RFA]: Document TUI SingleKey


On Sat, 31 Aug 2002, Stephane Carrez wrote:

> Can you approve this patch?

Approved; see my comments in my other mail.

Eli Zaretskii | 1 Sep 2002 07:38
Picon

Re: [RFA]: Document TUI breakpoint markers


On Sat, 31 Aug 2002, Stephane Carrez wrote:

> Can you approve this patch?

Approved with the comments I sent earlier today.

Thanks.

Stephane Carrez | 1 Sep 2002 12:45
Picon
Favicon

Re: [RFA]: Document TUI SingleKey

Hi Eli!

Eli Zaretskii wrote:
>
 >  <at> kindex c <at> r{, SingleKey TUI key}
 >
Ok! The index looks better.

To be homogeneous with other index entries I've used the following:

 <at> kindex c  <at> r{(SingleKey TUI key)}

so we now see

c (continue)
c (Single Key TUI Key)

I've committed the following patch and verified on pdf (well, is it acroread or
a better texinfo, it's the first time I see a good/outstanding PDF for a GNU manual).

	Stephane

2002-09-01  Stephane Carrez  <stcarrez <at> nerim.fr>

	* gdb.texinfo (TUI Single Key Mode): Document new SingleKey mode.
	(TUI Keys): Likewise.
Index: gdb.texinfo
===================================================================
(Continue reading)

Stephane Carrez | 1 Sep 2002 14:24
Picon
Favicon

[PATCH]: Improve TUI status line

Hi!

The TUI displays a status line above the command window to give various
information on the program.  This patch improves that by:

   - avoiding to display the file name (because it's displayed in window border)
   - display the current target (if enough width)
   - display the current pid (if enough width)
   - use print_address_numeric() to print the PC (instead of an sprintf; so
     that it now handles PC masking & 64-bit PCs)
   - print a (SingleKey) marker to know we are in this mode.

Committed on mainline.

Note: The documentation says nothing about the status line...  I'll try to write something...

	Stephane

2002-09-01  Stephane Carrez  <stcarrez <at> nerim.fr>

	* tuiStack.c (tui_make_status_line): New function to create the
	status line.
	(tuiShowLocatorContent): Use it instead of displayableWinContentAt.
	* tuiData.h (PROC_PREFIX): Use "In:" to reduce length of prefix.
	(PC_PREFIX): Use upper case.
	(SINGLE_KEY, MIN_LINE_WIDTH, MIN_PROC_WIDTH): Define.
	(MAX_TARGET_WIDTH, MAX_PID_WIDTH): Define.
Index: tuiData.h
(Continue reading)

Stephane Carrez | 1 Sep 2002 14:49
Picon
Favicon

[PATCH]: Cleanup tuiData

Hi!

I've committed this patch to remove several functions in tuiData.c
that are no longer used (some were not used for a while...).

	Stephane

2002-09-01  Stephane Carrez  <stcarrez <at> nerim.fr>

	* tuiData.h (FILE_PREFIX): Don't define.
	(blankStr, locationStr, breakStr): Don't declare.
	(breakLocationStr, nullStr, historyLimit, setHistoryLimitTo): Likewise.
	(displayableWinContentOf, displayableWinContentAt): Likewise.
	(winElementHeight, winByName, freeAllWindows): Likewise.

	* tuiData.c (blankStr, locationStr, breakStr): Remove.
	(breakLocationStr, nullStr, historyLimit, setHistoryLimitTo): Remove.
	(displayableWinContentOf, displayableWinContentAt): Remove.
	(winElementHeight, winByName, freeAllWindows): Remove.

Index: tuiData.c
===================================================================
RCS file: /cvs/src/src/gdb/tui/tuiData.c,v
retrieving revision 1.10
diff -u -p -r1.10 tuiData.c
--- tuiData.c	30 Aug 2002 20:07:05 -0000	1.10
+++ tuiData.c	1 Sep 2002 10:43:21 -0000
 <at>  <at>  -40,6 +40,7  <at>  <at> 
(Continue reading)

Stephane Carrez | 1 Sep 2002 16:49
Picon
Favicon

[PATCH]: Add detach_hook for TUI

Hi!

This patch adds a hook for TUI to know when a process is detached or dies.
The status line is refreshed.

Committed on mainline.

	Stephane

2002-09-01  Stephane Carrez  <stcarrez <at> nerim.fr>

	* tui-hooks.c (tui_detach_hook): New hook to know when a process dies.
	(tui_install_hooks): Install it.
	(tui_remove_hooks): Remove it.
Index: tui-hooks.c
===================================================================
RCS file: /cvs/src/src/gdb/tui/tui-hooks.c,v
retrieving revision 1.6
diff -u -p -r1.6 tui-hooks.c
--- tui-hooks.c	30 Aug 2002 20:07:05 -0000	1.6
+++ tui-hooks.c	1 Sep 2002 12:44:23 -0000
 <at>  <at>  -277,6 +277,15  <at>  <at>  tui_print_frame_info_listing_hook (struc
   tuiShowFrameInfo (selected_frame);
 }

+/* Called when the target process died or is detached.
+   Update the status line.  */
+static void
(Continue reading)

Eli Zaretskii | 1 Sep 2002 15:07
Picon

Re: [RFA]: Document TUI SingleKey


On Sun, 1 Sep 2002, Stephane Carrez wrote:

> To be homogeneous with other index entries I've used the following:
> 
>  <at> kindex c  <at> r{(SingleKey TUI key)}
> 
> so we now see
> 
> c (continue)
> c (Single Key TUI Key)

Yes, thanks.

Eli Zaretskii | 1 Sep 2002 15:08
Picon

Re: [PATCH]: Improve TUI status line


On Sun, 1 Sep 2002, Stephane Carrez wrote:

> Note: The documentation says nothing about the status line...  I'll try to write something...

Thanks!

Stephane Carrez | 1 Sep 2002 20:14
Picon
Favicon

[PATCH]: Fix spurious 'type return to continue' when scrolling TUI source window

Hi!

When we scroll TUI source window (page up/down for example), we obtain a
spurious 'type return to continue'.  It is due to a gdb annotation that
is printed and later ignored by tui-out.  But the \n increments the
gdb lines_printed.

I've committed this patch to fix that by using print_source_lines() instead
of identify_source_line() which is normally used only when producing gdb annotations.
The tui_out_new() fix is necessary to fix the recognition of output produced by
print_source_lines().

Committed on mainline.

	Stephane

2002-09-01  Stephane Carrez  <stcarrez <at> nerim.fr>

	* tui-out.c (tui_out_new): Clear start_of_line.
	* tuiSource.c (tuiVerticalSourceScroll): Use print_source_lines
	to update the current source line.
Index: tui-out.c
===================================================================
RCS file: /cvs/src/src/gdb/tui/tui-out.c,v
retrieving revision 1.2
diff -u -p -r1.2 tui-out.c
--- tui-out.c	19 Mar 2002 02:51:09 -0000	1.2
+++ tui-out.c	1 Sep 2002 15:52:59 -0000
(Continue reading)


Gmane