> We are currently running tmux version 1.8
That's too old. We've fixed a tonne of bugs since then. Please try and
reproduce this with tmux 2.0 or better yet tmux from git.
-- Thomas Adam
--
One dashb
t found: 8
> tmux.conf:16: couldn't set 'default-terminal'
You can't set options for sessions which don't yet exist.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I
=),
> if mouse flag is set (mouse_any_flag) then well I have trouble with
> parsing at this stage. I think it would help greatly if the double-quote
> escapement scheme could be explained. I see 3 quotes in a row, and I
> suspect that that is significant.
What is it you want to know?
As of automake-1.11, it's now possible to tell automake that the compiler
output is "hidden", producing output almost identical to compiling on
OpenBSD. The effects of this are such that this can be disabled with
'--disable-silent-rules', or 'make V=1' to enable "verbose" output.
Even with silent
On Fri, May 15, 2015 at 09:52:06PM -0700, Lander Brandt wrote:
> When upgrading from tmux 1.9a to 2.0, shell commands are no longer
> executed in title strings. For example I have in my .tmux.conf:
Yes, it will be replaced eventually.
-- Thomas Adam
--
"Deep in my heart I wish
nd who are using the current rate-limting options, please give tmux from
git a try to see if this helps? Ensure you adapt your .tmux.conf to remove
the c0-* options if you'd set them there.
Any problems, let us know.
Kind
hat tmux was "using" the drive. I assume this means that the tmux
It's possible the FD is open to the directory or something. Can you
send through some logs?
-- Thomas Adam
--
One dashboard for servers and a
lf of everyone involved in this release, we wish you happy
tmuxing.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("Girl Least Likely To&
erm tmux - -Ltest -f/dev/null new
-- Thomas Adam
--
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, sta
has been removed,
allowing expansion of that through formats instead.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("Girl Least Likely To" -- off of Viva Hate.)
>From 6fd7702571fcc64b26d184a9b790e4f3fafdf952 M
sessions and also in grouped sessions, if you don't
> want to make it exact a 0 or 1 would be better.
Done. Now, the flag window_linked considered only unique windows
linked across sessions (via linkw)---even in the case of a grouped
session.
-- Thomas Adam
--
"Deep in my heart I
On Thu, Apr 30, 2015 at 08:42:49AM -0700, jungle Boogie wrote:
> Yes, of course but I seem to recall having a hard time with the
> dependency on ARM.
There is no such thing. What are you referring to?
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart
ling up with the mouse, this was useful to emulate intuitively expected
> behaviour of scrolling in the terminal.
I haven't tested this, but I don't think this patch is needed, given
something like:
bind -n WheelDownPane if -Ft=
"#{?pane_in_mode,#{?scroll_region_lower},,}
This adds support for indicating in the status line whether a given window
has been referenced more than once. In the case of grouped sessions this is
likely always going to be the case, but this also helps in showing which
windows in a session have been included via linkw, for example. Currently
When expanding/toggling a given session in the choose-tree list, ensure we
start from the select item, and not the index of the item's position, since
this ends up producing odd screen redraw problems since the index doesn't
match the array index of the items on the screen.
---
window-choose.c | 4
Hmm,
That's not the screen shot I saw originally, sorry. The image you sent is
either vim or nvim (but who cares about the distinction?). As for the
colour scheme, pass. Desert256?
Thomas Adam
On 30 Apr 2015 17:58, "Suraj N. Kurapati" wrote:
> On Thu, 30 Apr 2015 15:58:12
On Thu, Apr 30, 2015 at 08:24:38AM -0700, jungle Boogie wrote:
> On 30 April 2015 at 08:04, Thomas Adam wrote:
> > The portable version uses the same code as this, but slightly adapted to run
> > on platforms which are not OpenBSD, but tmux is always developed against
> >
ur question?
See: http://undeadly.org/cgi?action=article&sid=20090707041154
-- Thomas Adam
--
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ appl
ent this
email.
-- Thomas Adam
--
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give
On Wed, Apr 29, 2015 at 12:06:42AM +0100, Nicholas Marriott wrote:
> Actually I think I already fixed this?
Yes, I think you rolled this into a larger fix for something else. It
doesn't crash for me anymore.
Cheers,
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. B
you actually tried it? The point here is for this library to
monitor the output from control mode and act accordingly.
-- Thomas Adam
--
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest
When checking a client's tty path, don't segfault if it's NULL.
---
cmd-find.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/cmd-find.c b/cmd-find.c
index fab3849..3ac9936 100644
--- a/cmd-find.c
+++ b/cmd-find.c
@@ -243,6 +243,8 @@ cmd_find_current_session_with_client(struct cmd_find_stat
Rather than blindly call session_renumber_session when the option wasn't
requested (and hence breaking the case of renumbering a winlink manually),
instead only renumber the src session, since that session is the one which
will have "gaps".
---
cmd-move-window.c | 12 +++-
server-fn.c
s one here looks a lot less obtrusive, since it doesn't just
comment out the pane redrawing stuff, etc.
-- Thomas Adam
--
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
Develop your own process in accor
Remove 'set-but-not-used' variables, to shut GCC up.
---
window-copy.c | 4
1 file changed, 4 deletions(-)
diff --git a/window-copy.c b/window-copy.c
index afa609e..11cef71 100644
--- a/window-copy.c
+++ b/window-copy.c
@@ -2217,13 +2217,11 @@ void
window_copy_start_drag(struct client *c, u
When moving winlinks between sessions, honour the renumber-windows session
option if it is set.
---
server-fn.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/server-fn.c b/server-fn.c
index 71745d7..984caca 100644
--- a/server-fn.c
+++ b/server-fn.c
@@ -351,6 +351,11 @@ server_unlink_wi
I really wouldn't go conflating ECONNREFUSED with "no sessions" when
there's any number of other reasons for that.
-- Thomas Adam
--
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM ED
you
like. Submitting patches can be done using git-format-patch, just
like a number of other projects do; git itself, for instance.
So, given the above, what does moving gain us at the moment? Nothing.
-- Thomas Adam
--
On Wed, Apr 01, 2015 at 04:01:56PM +1300, Jan Larres wrote:
> On 01/04/15 12:53, Thomas Adam wrote:
> > Renumbering of winlinks only happens when the winlink is
> > killed/destroyed (assuming renumber-windows session option). The point
> > of 'movew -r' is solely
date-environment");
I think you want &s->options here, no?
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("Girl Least Likely To" -- off of Viva Hate.)
--
rrent window to
session 3, you can just do:
movew -t3:
And tmux can figure that out itself. What you can do is this:
movew -t 3:50 \; movew -r -t3
Which would move the current window to session 3, window index 50 and
then renumber the winlinks in session 3.
-- Thomas Adam
--
"Deep
Hi,
This is because -r is a separate operation to how the rest of movew works.
Thomas Adam
On 31 Mar 2015 04:26, "Jan Larres" wrote:
> Hi,
>
> there seems to be a bug in the move-window command when using the -r
> option.
>
> Let's assume I have a session &quo
gular user
>
> The configure step seems ok, but I am in trouble with make. Essentially I get
> multiple definition errors
>
> strtonum.c:(.text+0x40): multiple definition of `gnu_dev_makedev'
As discussed on IRC, you
nable-debug
Note that you should try installing git HEAD because there's been many fixes
in there since tmux 1.9a was released.
-- Thomas Adam
--
Dive into the World of Parallel Programming The Go Parallel Website, spon
o extract a backtrace and send it to this mailing list.
> I'm doing so in hopes of having someone more knowledgeable look at it.
What version of libevent?
-- Thomas Adam
--
Dive into the World of Parallel Progra
or it is still running?
Use either monitor-silence or monitor-activity depending on how the
command works. Or you could do something like:
program && echo -e "\a"
Assuming you use terminal bells.
There's a few more ways, I'm sure.
-- Thomas Adam
---
On Tue, Mar 10, 2015 at 03:12:05PM +, David Chanters wrote:
> On 16 February 2015 at 15:59, Thomas Adam wrote:
> > % git log | egrep -o '(from |by )[A-Z][A-Za-z0-9 ]*\.'|sort|uniq -c|sort
> > -nr| head -n10
> > 40 from Thomas Adam.
> > 39 from G
gt;
> Has this better way been implemented? If so, can you describe how I can
> enable it?
Nope, nothing yet.
-- Thomas Adam
--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by In
ple panes.
>
> Any idea of what the problem is and how I can solve it?
Unset the set-clipboard option.
-- Thomas Adam
--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and de
ll patches still
> flow through CVS?
Only portable fixes which aren't needed by OpenBSD are committed directly.
This is documented in the SYNCING file in the tmux git repo.
-- Thomas Adam
--
Dive into the World
On 23 February 2015 at 12:18, Daniel wrote:
> resize-pane -Z is pretty neat. Though, is it possible to somehow have an
> indicator of a pane is zoomed, and there are currently panes hidden?
The window flags already contain 'Z' to indicate this.
f I
> don't know about it.
I mentioned it to you. You said you'd incorporate those things into
subsequent patches. The diff is here:
https://github.com/ThomasAdam/tmux/commit/f85a0fcfe9cc54aecdca205e2f2c93cce2235d47.patch
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But
ders(wp->window);
wp = wp->window->active; /* may have changed */
}
-- Thomas Adam
--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your
n which those statistics are using
don't take into account the fact that the real contributor is part of the
commit message itself. See for example:
% git log | egrep -o '(from |by )[A-Z][A-Za-z0-9 ]*\.'|sort|uniq -c|sort -nr|
head -n10
40 from Thomas Adam.
39 from G
all sessions.
Thanks, but we don't need this as another command. You can still use
"move-window -r", and for now, script the rest.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I kno
asAdam/tmux/commits/jr/pane-colours
Have a look, tell me what you think. The cleanups are pretty trivial, but
reduces some of the copy/paste code churn.
-- Thomas Adam
--
Dive into the World of Parallel Programming. Th
7;)
> + gc = NULL;
> + else {
> + gc = xmalloc(sizeof *gc);
> + memcpy(gc, &grid_default_cell, sizeof *gc);
> + ret = style_parse(&grid_default_cell, gc, str);
I wouldn
ve, this isn't the case? What are
you expecting to see in status-right which you're not seeing?
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("Girl Least Likely To" -- off of Viva Hate.)
---
r instance with your config, hence:
% tmux -Ltest -v
Let the status update, then kill the server, and send through the
tmux-server-*.log which will be in the directory you invoked the above
comamnd in.
-- Thomas Adam
--
"Deep in my heart I wish I
Hi,
Yes it is. Nicholas and I are working on this, so I'd hold off for now.
Thomas Adam
On 30 Jan 2015 14:06, "Kaushal" wrote:
> I forgot to mention that when I started the new tmux server using *tmux
> -Ltemp -f/home/$USER/.tmux-2.0.conf* , it printed the "can
Switch over all tmux commands to make use of prepare.
---
cmd-attach-session.c | 11 +++---
cmd-bind-key.c | 3 ++-
cmd-break-pane.c | 9 +---
cmd-capture-pane.c | 7 +++---
cmd-choose-buffer.c| 9
cmd-choose-client.c| 10 +
cmd-choose-tree.
lly_ appreciate user testing on this, so please, shunt this
towards your friends or people who've been interested in this, and if it
breaks, send me a corefile, please!
Any questions, do please ask.
Kindly,
-- Thomas Adam
Thomas Adam (3):
Introduce prepare mechanism for commands
Ad
/cmd-set-hook.c b/cmd-set-hook.c
new file mode 100644
index 000..75aebb9
--- /dev/null
+++ b/cmd-set-hook.c
@@ -0,0 +1,95 @@
+/* $Id$ */
+
+/*
+ * Copyright (c) 2012 Thomas Adam
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is
Commands now set their running context separately from executing. The
state is used at execution time of commands, and by separating this out, the
command state can be queried for any hooked commands.
---
cmd.c | 32
tmux.h | 32
it's not needed or wanted within tmux.
> and stay in the foreground, like screen -D -m? This will greatly
> simplify running tmux from a process manager that do not support
> forking services like supervisord.
That would be the wrong so
s related to this: https://sourceforge.net/p/tmux/tickets/66/
Hmm. I thought the steps to reproduce that were clear; perhaps not,
and perhaps my patch is effectively useless.
-- Thomas Adam
--
New Year. New Location. New
On 18 January 2015 at 12:19, İsmail Dönmez wrote:
> Hi,
>
> I am trying to find a shortcut to create a window at the end of the current
> windows. Currently new-window -a just appends the new window after the
How is this *not* just 'tmux neww' -- that is, without the '
On 15 January 2015 at 21:49, Kaushal wrote:
> But to begin with, I am unable to find a way to copy the last few lines of
> the current pane history using a script.
capture-pane
-- Thomas Adam
--
New Year. New Lo
On Fri, Jan 09, 2015 at 09:07:07PM -0500, Helmut K. C. Tessarek wrote:
> On 09.01.15 20:48 , Thomas Adam wrote:
> > Can we stop with these rather silly suggestions for other software we
> > don't need?
>
> Apparently nobody understands that I am the maintainer for the t
On 10 January 2015 at 00:26, Helmut K. C. Tessarek wrote:
> On 09.01.15 5:14 , Thomas Adam wrote:
>> There's already a repo there if you want to use that.
>
> It s not the main repo, is it? It is also not a mirror, with automatic pushes
> to it when the sf repo is updated.
On Thu, Jan 08, 2015 at 08:13:40PM -0500, Helmut K. C. Tessarek wrote:
> On 08.01.15 18:51 , Thomas Adam wrote:
> > https://github.com/ThomasAdam/tmux-ut
>
> Have you ever thought of moving the sf repo to github?
There's already a repo there if you want to use that.
> O
u can use either, and having the
redundancy on Github as a mirror of the SF one is a good thing.
-- Thomas Adam
--
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in
homasAdam/tmux-ut
It's under the ISC license. Patches welcome!
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("Girl Least Likely To" -- off of Viva Hate.)
ut a specific patch from git that I
> can apply to the stable 1.9_a codebase to get this functionality?
Why? If you can already use the Git version, just use that. That's easier
than patching 1.9a.
-- Thomas Adam
ll check for the utempter header file automatically).
> logname still errors out. Is there a corresponding config option I
> need to tweak or?
No. No config option.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("
nux, install libutempter, this will then allow
the utmp database to be queried, and commands like w, who, and logname
should work.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("Gi
If the window that's being killed is in a zoomed state, unzoom it first,
otherwise there's no chance of finding the correct winlink, causing tmux to
exit.
---
server-fn.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/server-fn.c b/server-fn.c
index 6cab6bd..1ed3a01 100644
--- a/server-fn.c
+
TMPDIR is not set. See also TMUX_TMPDIR. The
default won't change.
-- Thomas Adam
--
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slas
when I switch to version 1.9a or
> 1.9?
Possibly. Why don't you just try it? But you've still not adequately
explained what the problem is, and despite repeated attempts to get that
information out of you, it's not been forth-coming.
I wish you luck solving your problem,
een.
Which I fixed sometime ago. Likely after 1.8 in this case; tmux git
will confirm.
> And I found that the problem tmux cmd is “choose-tree”, when I switch to
> "choose-session" cmd, it works.
You've still not _explained_ w
rks fine for me. Which tmux version is this? What size is your terminal
window? Mine's 80x24. Can you _PLEASE_ describe in more detail what
"doesn't work" mean because I'm not a bloody psychic. A scr
is small, but not when the session number is large.
Can you be specific about what you mean?
-- Thomas Adam
--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge
draw_screen(wp);
> + window_copy_for_selection(wp, old_cy);
Rather than having a new function for this, why not put the logic into
window_copy_redraw_screen? The mode for the wp is already known
On 1 December 2014 at 23:57, Thomas Adam wrote:
> In the case of creating a new window, tmux would end up checking the window
> name
> before the index. This is usually OK, save for the fact that if a window ever
> has as its title, the same number as the index, then the check
In the case of creating a new window, tmux would end up checking the window name
before the index. This is usually OK, save for the fact that if a window ever
has as its title, the same number as the index, then the check is in error.
Instead, make cmd_lookup_index() check the index and then the
Patch to fix this sent to the mailing list.
-- Thomas Adam
On 30 November 2014 at 20:16, rorytmux wrote:
>
>
> [tickets:#170] window_zoomed_flag not working
>
> Status: open
> Created: Sun Nov 30, 2014 08:16 PM UTC by rorytmux
> Last Updated
Don't test setting the WINDOW_ZOOM flag on a winlink, as it's set on the
window.
---
format.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/format.c b/format.c
index e7fd3e1..47b8b79 100644
--- a/format.c
+++ b/format.c
@@ -471,6 +471,8 @@ format_window(struct format_tree
window_choose_add_item() was only ever used by choose-item, and hence is no
longer used by anything else.
---
tmux.h | 3 ---
window-choose.c | 30 --
2 files changed, 33 deletions(-)
diff --git a/tmux.h b/tmux.h
index 7c0a7b5..317d1d1 100644
--- a/tmux.h
+++
> frame after switch, this annoys me a bit.
Just set an appropriate repeat-time value.
-- Thomas Adam
--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Busi
M UTC
> Owner: nobody
>
> I'm a previous screen user and one feature that I miss is having 2 terminals
> sharing the same session but while being on different tmux windows.
>
> Is this already possible?
> Or is this a feature request?
$ tm
to the file later.
What is it you're asking us (tmux developers) to do, exactly? I'd
rather just point people to style(9)...
-- Thomas Adam
--
___
tmux-users maili
On 21 Oct 2014 03:16, "Helmut Tessarek" wrote:
> Do you think that would make sense, or should I update MacPorts only when
a
> new release of tmux is cut?
Why not have both? We appreciate on-going testing of tmux git until we
decide to release a new version. In fact, we rely on
On 21 Oct 2014 01:50, "Helmut Tessarek" wrote:
>
> On 20.10.14 4:41 , Thomas Adam wrote:
> > Whenever Nicholas and I decide it appropriate, based on on-going
patches,
> > number of features, etc.
>
> Ok, so basically there is no release cycle. Thx for the info.
On Mon, Oct 20, 2014 at 01:27:10AM -0400, Helmut Tessarek wrote:
> What is the release cycle of tmux? There is quite some activity in the repo,
> but releases are cut rather seldomly.
Whenever Nicholas and I decide it appropriate, based on on-going patches,
number of features, etc.
--
On 20 October 2014 00:53, Mark Volkmann wrote:
> That also opened a new pane, but not in the current directory.
Which platform?
-- Thomas Adam
--
Comprehensive Server Monitoring with Site24x7.
Monitor 10 servers for
r the old way. Is there a setting for this?
I write release descriptions for a reason:
http://sourceforge.net/p/tmux/tmux-code/ci/master/tree/CHANGES
Note the "Incompatible Changes" section for 1.9
-- Thomas Adam
red
is. If you're using something like bash, check to see if it's being
changed from PROMPT_COMMAND.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morrissey ("Girl
On 16 September 2014 15:18, Thomas Stüfe wrote:
> Is this intentional or a graphical bug? If intentional, is there an option
Intentional.
> to switch this off?
No. It's useful to see which pane on which side is currently focused.
--
is a great opportunity for tmux to get some nice visual
> identity!
>
> Thomas, Nicholas, what do you think? Can I include you via direct email in a
> discussion with this guy?
If you want, although I'm only interested in updating files, etc. I
don't want to decide or
l.
> Have anyone encountered this?
>
> I'm using tmux 1.9a.
Correct. I believe this is what xterm does.
-- Thomas Adam
--
Slashdot TV.
Video for Nerds. Stuff that matters.
http://tv.slashdot.org/
___
Updated:** Tue Sep 02, 2014 10:20 PM UTC
> **Owner:** nobody
>
> When the status line is too full it currently displays '<' and '>' on the
> left and right edges respectively. Please make those characters
> configurable so more descriptive characters could
ing to
> you
> all. I thought a window-option with the current behaviour as default would be
> the best approach to this.
We discussed something like this earlier this year:
http://sourceforge.net/p/tmux/tickets/122/
I can't say I'm in favour of your patches.
-- Thomas Ada
but I've got some patches
under review with Nicholas that I'm waiting to see merged first.
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I know I am
not." -- Morriss
ost server]
> $ tmux -L bug attach-session
> no sessions
This patch shouldn't be necessary, I've synced from OpenBSD, try the
Git version again.
-- Thomas Adam
--
___
clients, i);
if (c == NULL || c->session == NULL)
continue;
- if (c->session == s)
+ if (c->session == s) {
server_redraw_client(c);
+ session_ren
nt for the strlen() call, please.
-- Thomas Adam
--
Want fast and easy access to all the code in your enterprise? Index and
search up to 200,000 lines of code with a free copy of Black Duck
Code Sight - the same software that
# not ok, should be "foobar"
When NicM and I were discussing this, we didn't think appending for @
variables was a good idea; it's currently not supported.
Patches welcome, perhaps.
-- Thomas Adam
--
"Deep in my h
, as in the
> previous command
Why? This looks correct to me. And the code agrees; return
CMD_RETURN_NORMAL in the case of quiet given. What are you using this
for where this is damaging?
-- Thomas Adam
--
"Deep in my heart I wish I was wrong. But deep in my heart I k
l 31, 2014 12:48 PM UTC
> **Owner:** nobody
>
> When creating new windows, current working directory is not passed to the new
> window, leaving it in the working directory of the tmux process.
Read the CHANGES file for tmux 1.9
-- Thomas Adam
--
"Deep in my heart I wish
e(const char *name) {
> + struct wait_channel *wc = wait_channel_find_by_name(name);
> +
> + if(!wc)
if (wc != NULL)
> + wc = wait_channel_alloc(name);
Suggest you call this wait_channel_new();
> +
&
1 - 100 of 659 matches
Mail list logo
