public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Eli Zaretskii <eliz@gnu.org>
To: Pedro Alves <palves@redhat.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCHSET] [2/4] Fix various issue in TUI
Date: Mon, 05 Jan 2015 19:40:00 -0000	[thread overview]
Message-ID: <83h9w5819b.fsf@gnu.org> (raw)
In-Reply-To: <54AAE1D9.9000409@redhat.com>

> Date: Mon, 05 Jan 2015 19:11:21 +0000
> From: Pedro Alves <palves@redhat.com>
> 
> On 12/31/2014 05:45 PM, Eli Zaretskii wrote:
> > This patch fixes the problem whereby setting the border attributes had
> > no effect whatsoever.  
> 
> Rather, you need to switch out of the TUI and back, with c-x,a.

Yeah, well... not really friendly.

> > While at that, this also fixes a copy/paste error in a comment to a
> > function.
> 
> Could you please push this part separately?

Will do.

> > 2014-12-31  Eli Zaretskii  <eliz@gnu.org>
> > 
> > 	* tui/tui-win.c (tui_rehighlight_all): New function.
> > 
> 
> No empty lines between ChangeLog entries, please.

Why are we using a format that is different from how Emacs formats
ChangeLog entries?  It's just annoying extra manual work.

> > --- gdb/tui/tui-command.c~0	2014-10-29 21:45:50 +0200
> > +++ gdb/tui/tui-command.c	2014-12-31 13:49:43 +0200
> > @@ -50,7 +50,12 @@
> 
> Seems like you're not using git diff to generate the diffs
> for some reason.

This wasn't done in the git repo, I did it with the GDB 7.8.1 sources.

> If using GNU diff, could you make sure to use the "-p" switch?  It
> makes review a lot easier.

I'll try to remember.  (The ChangeLog entries state the function
names, don't they?)

> > --- gdb/tui/tui-command.c~0	2014-10-29 21:45:50 +0200
> > +++ gdb/tui/tui-command.c	2014-12-31 13:49:43 +0200
> > @@ -50,7 +50,12 @@
> >
> >    /* Handle the CTRL-L refresh for each window.  */
> >    if (ch == '\f')
> > -    tui_refresh_all_win ();
> > +    {
> > +      if (tui_update_variables ())
> > +	tui_rehighlight_all ();
> > +
> > +      tui_refresh_all_win ();
> 
> 
> tui_refresh_all_win already calls tui_check_and_display_highlight_if_needed,
> so why do we need to call tui_rehighlight_all?  Is it the tui_update_variables
> call that is really missing?

Yes, the call to tui_update_variables is required to take notice of
the changes.  Also, I'm not sure tui_refresh_all_win does that for all
windows, it has a switch that handles on 3 types.

> But why would we need to call tui_update_variables on CTRL-L, if we
> already call it when the user changes the variables?

I thought it'd be better to ensure this is called directly on refresh,
since C-l can be used in all kinds of weird situations where the
screen is messed up e.g. by the program being debugged.

> > --- gdb/tui/tui-hooks.c~0	2014-06-11 18:34:41 +0300
> > +++ gdb/tui/tui-hooks.c	2014-12-31 14:41:11 +0200
> > @@ -247,12 +247,23 @@
> >    tui_display_main ();
> >  }
> >
> > +/* Refresh the display when settings important to us change.  */
> > +static void
> > +tui_note_setting_change (const char *param, const char *value)
> > +{
> > +  if (tui_active
> > +      && strncmp (param, "tui ", sizeof ("tui ") - 1) == 0
> > +      && tui_update_variables ())
> > +    tui_rehighlight_all ();
> > +}
> > +
> 
> Please do this from the "set" hook of the relevant commands instead.
> IOW, replace NULL below (and in the other commands):
> 
>   add_setshow_enum_cmd ("active-border-mode", no_class, tui_border_mode_enums,
> 			&tui_active_border_mode, _("\
> ...
> 			NULL,
> 			show_tui_active_border_mode,
> 			&tui_setlist, &tui_showlist);

OK.

> > @@ -985,11 +995,14 @@
> >    /* Make sure the curses mode is enabled.  */
> >    tui_enable ();
> >
> > +  if (tui_update_variables ())
> > +    tui_rehighlight_all ();
> > +
> >    tui_refresh_all_win ();
> >  }
> >
> 
> I don't understand this one.  When is it ever necessary?

Same as Ctrl-l: this is the "refresh" command.

  reply	other threads:[~2015-01-05 19:40 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-12-31 17:45 Eli Zaretskii
2015-01-05 19:11 ` Pedro Alves
2015-01-05 19:40   ` Eli Zaretskii [this message]
2015-01-06 16:03     ` Pedro Alves
2015-01-06 15:58   ` Eli Zaretskii
2015-01-07 13:48     ` Pedro Alves
2015-01-16 16:29       ` Eli Zaretskii
2015-01-16 16:59         ` Sergio Durigan Junior
2015-01-16 17:59           ` Eli Zaretskii
2015-01-16 16:34   ` Eli Zaretskii

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=83h9w5819b.fsf@gnu.org \
    --to=eliz@gnu.org \
    --cc=gdb-patches@sourceware.org \
    --cc=palves@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).