Re: Update: i3-4.12

[Josh Grosse]

On Tue, Mar 08, 2016 at 01:53:55PM +0100, David Coppa wrote:
> On Tue, 08 Mar 2016, Dmitrij D. Czarkoff wrote:
> 
> > David Coppa said:
> > > Here's the update to i3-4.12.
> > > 
> > > Please test it and report back.
> > 
> > I report that your patch does not apply cleanly.
> 
> Try with the attached one, i3-4.12_v2.diff.gz.
 
Working fine here on amd64.  portcheck(1) notes trailing whitespace
in pkg/DESCR, which was not part of the diff.  

Re: Update: i3-4.12

[David Coppa]

On Tue, 08 Mar 2016, Dmitrij D. Czarkoff wrote:

> David Coppa said:
> > Here's the update to i3-4.12.
> > 
> > Please test it and report back.
> 
> I report that your patch does not apply cleanly.

Try with the attached one, i3-4.12_v2.diff.gz.



i3-4.12_v2.diff.gz
Description: application/gunzip

Re: Update: i3-4.12

[Theo Buehler]

On Tue, Mar 08, 2016 at 10:15:55AM +0100, David Coppa wrote:
> On Tue, 08 Mar 2016, David Coppa wrote:
> 
> > 
> > Here's the update to i3-4.12.
> > 
> > Please test it and report back.
> 
> As attachment.

Thank you very much, this port is so much cleaner now!

Initial impression is good, the usual share of bogus warnings about
uninitialized variables and control reaches end of non-void functions. I
checked that none of them points to a real problem. It would be nice if
upstream could silence them, so we can spend our time focussing on
actual problems...

One question: is there a reason that halex@'s window hints fix is not
included?

The ssh-askpass problem is still there and it will continue to be
annoying for him and possibly others.  Here's a rebased version of the
patch to apply against -current. Attached is a naked patch that you
could apply directly in WRKSRC, if that's more convenient.

Index: patches/patch-src_manage_c
===
RCS file: /var/cvs/ports/x11/i3/patches/patch-src_manage_c,v
retrieving revision 1.17
diff -u -p -r1.17 patch-src_manage_c
--- patches/patch-src_manage_c  18 Dec 2015 15:53:09 -  1.17
+++ patches/patch-src_manage_c  8 Mar 2016 10:05:13 -
@@ -1,44 +1,26 @@
-$OpenBSD: patch-src_manage_c,v 1.17 2015/12/18 15:53:09 dcoppa Exp $
+$OpenBSD$
 
-Suppress no_focus for first window on a workspace.
+Fix size hints for ssh-askpass
 
-Fix multiple memory leaks with regular expressions.
-
 src/manage.c.orig  Wed Sep 30 08:55:10 2015
-+++ src/manage.c   Fri Dec 18 13:58:58 2015
-@@ -294,6 +294,7 @@ void manage_window(xcb_window_t window, xcb_get_window
- if (match != NULL && match->insert_where != M_BELOW) {
- DLOG("Removing match %p from container %p\n", match, nc);
- TAILQ_REMOVE(&(nc->swallow_head), match, matches);
-+match_free(match);
- }
- }
- 
-@@ -524,13 +525,23 @@ void manage_window(xcb_window_t window, xcb_get_window
- /* Send an event about window creation */
- ipc_send_window_event("new", nc);
- 
-+if (set_focus && assignment_for(cwindow, A_NO_FOCUS) != NULL) {
-+/* The first window on a workspace should always be focused. We have 
to
-+ * compare with == 1 because the container has already been inserted 
at
-+ * this point. */
-+if (con_num_children(ws) == 1) {
-+DLOG("This is the first window on this workspace, ignoring 
no_focus.\n");
-+} else {
-+DLOG("no_focus was set for con = %p, not setting focus.\n", nc);
-+set_focus = false;
-+}
-+}
-+
- /* Defer setting focus after the 'new' event has been sent to ensure the
-  * proper window event sequence. */
- if (set_focus && !nc->window->doesnt_accept_focus && nc->mapped) {
--if (assignment_for(cwindow, A_NO_FOCUS) == NULL) {
--DLOG("Now setting focus.\n");
--con_focus(nc);
--}
-+DLOG("Now setting focus.\n");
-+con_focus(nc);
- }
+--- src/manage.c.orig  Tue Mar  8 10:59:28 2016
 src/manage.c   Tue Mar  8 10:59:35 2016
+@@ -468,10 +468,16 @@ void manage_window(xcb_window_t window, xcb_get_window
+ if (cwindow->dock)
+ want_floating = false;
  
- tree_render();
+-/* Plasma windows set their geometry in WM_SIZE_HINTS. */
++/*
++ * Plasma windows set their geometry in WM_SIZE_HINTS.  Note that
++ * these fields are obsolete, so ignore unless there is a width and
++ * a height specified.
++ */
+ if ((wm_size_hints.flags & XCB_ICCCM_SIZE_HINT_US_POSITION || 
wm_size_hints.flags & XCB_ICCCM_SIZE_HINT_P_POSITION) &&
+-(wm_size_hints.flags & XCB_ICCCM_SIZE_HINT_US_SIZE || 
wm_size_hints.flags & XCB_ICCCM_SIZE_HINT_P_SIZE)) {
+-DLOG("We are setting geometry according to wm_size_hints x=%d y=%d 
w=%d h=%d\n",
++(wm_size_hints.flags & XCB_ICCCM_SIZE_HINT_US_SIZE || 
wm_size_hints.flags & XCB_ICCCM_SIZE_HINT_P_SIZE) &&
++wm_size_hints.width != 0 && wm_size_hints.height != 0) {
++DLOG("We are setting geometry according to the obsolete wm_size_hints 
"
++ "x=%d y=%d w=%d h=%d\n",
+  wm_size_hints.x, wm_size_hints.y, wm_size_hints.width, 
wm_size_hints.height);
+ geom->x = wm_size_hints.x;
+ geom->y = wm_size_hints.y;
+++ src/manage.cFri Dec 18 13:58:58 2015
+++ src/manage.cMon Jan  4 22:04:00 2016
@@ -468,10 +468,16 @@ void manage_window(xcb_window_t window, xcb_get_window
 if (cwindow->dock)
 want_floating = false;
 
-/* Plasma windows set their geometry in WM_SIZE_HINTS. */
+/*
+ * Plasma windows set their geometry in WM_SIZE_HINTS.  Note that
+ * these fields are obsolete, so ignore unless there is a width and
+ * a height specified.
+ */
 if ((wm_size_hints.flags & XCB_ICCCM_SIZE_HINT_US_POSITION || 
wm_size_hints.flags & 

Re: Update: i3-4.12

[Theo Buehler]

On Tue, Mar 08, 2016 at 01:39:27PM +0100, David Coppa wrote:
> On Tue, 08 Mar 2016, David Coppa wrote:
> 
> > On Tue, Mar 8, 2016 at 12:26 PM, Theo Buehler  wrote:
> > 
> > > One question: is there a reason that halex@'s window hints fix is not
> > > included?
> > 
> > No, I've just forgotten it ;)
> > 
> > > The ssh-askpass problem is still there and it will continue to be
> > > annoying for him and possibly others.  Here's a rebased version of the
> > > patch to apply against -current. Attached is a naked patch that you
> > > could apply directly in WRKSRC, if that's more convenient.
> > 
> > Thanks, I'll apply it and make a new diff.
> 
> Here it is, attached.

So I've been working with this for three hours with my ordinary
workflow, no issues.

I think this should be ok to go in.

Thanks once more,

Theo

Re: Update: i3-4.12

[David Coppa]

On Tue, 08 Mar 2016, David Coppa wrote:

> On Tue, Mar 8, 2016 at 12:26 PM, Theo Buehler  wrote:
> 
> > One question: is there a reason that halex@'s window hints fix is not
> > included?
> 
> No, I've just forgotten it ;)
> 
> > The ssh-askpass problem is still there and it will continue to be
> > annoying for him and possibly others.  Here's a rebased version of the
> > patch to apply against -current. Attached is a naked patch that you
> > could apply directly in WRKSRC, if that's more convenient.
> 
> Thanks, I'll apply it and make a new diff.

Here it is, attached.

Thanks!
David



i3-4.12_v2.diff.gz
Description: application/gunzip

Re: Update: i3-4.12

[David Coppa]

On Tue, 08 Mar 2016, David Coppa wrote:

> 
> Here's the update to i3-4.12.
> 
> Please test it and report back.

As attachment.

Ciao!
David



i3-4.12.diff.gz
Description: application/gunzip

Update: i3-4.12

[David Coppa]

Here's the update to i3-4.12.

Please test it and report back.

Cheers!
David

Index: Makefile
===
RCS file: /cvs/ports/x11/i3/Makefile,v
retrieving revision 1.103
diff -u -p -u -p -r1.103 Makefile
--- Makefile21 Jan 2016 13:22:58 -  1.103
+++ Makefile8 Mar 2016 09:08:20 -
@@ -2,8 +2,7 @@
 
 COMMENT =  improved dynamic tiling window manager
 
-DISTNAME = i3-4.11
-REVISION = 3
+DISTNAME = i3-4.12
 CATEGORIES =   x11
 
 EXTRACT_SUFX = .tar.bz2
Index: distinfo
===
RCS file: /cvs/ports/x11/i3/distinfo,v
retrieving revision 1.24
diff -u -p -u -p -r1.24 distinfo
--- distinfo4 Oct 2015 08:48:12 -   1.24
+++ distinfo8 Mar 2016 09:08:20 -
@@ -1,2 +1,2 @@
-SHA256 (i3-4.11.tar.bz2) = eM4eBvvZL9Y3ZbviP6p7j5KcF/me1iP3q/LlaBadmC8=
-SIZE (i3-4.11.tar.bz2) = 972929
+SHA256 (i3-4.12.tar.bz2) = 4Z4c4IwlScuoPgg8x2jUhyAsQXYNXCg/Z3UueR8deLQ=
+SIZE (i3-4.12.tar.bz2) = 988084
Index: patches/patch-common_mk
===
RCS file: /cvs/ports/x11/i3/patches/patch-common_mk,v
retrieving revision 1.23
diff -u -p -u -p -r1.23 patch-common_mk
--- patches/patch-common_mk 4 Oct 2015 08:48:12 -   1.23
+++ patches/patch-common_mk 8 Mar 2016 09:08:20 -
@@ -1,7 +1,7 @@
 $OpenBSD: patch-common_mk,v 1.23 2015/10/04 08:48:12 dcoppa Exp $
 common.mk.orig Wed Sep 30 08:55:10 2015
-+++ common.mk  Fri Oct  2 16:58:54 2015
-@@ -162,9 +162,9 @@ I3_LDFLAGS += -Wl,-rpath,/usr/local/lib -Wl,-rpath,/us
+--- common.mk.orig Sun Mar  6 16:17:18 2016
 common.mk  Mon Mar  7 15:15:37 2016
+@@ -171,9 +171,9 @@ I3_LDFLAGS += -Wl,-rpath,/usr/local/lib -Wl,-rpath,/us
  endif
  
  ifeq ($(UNAME),OpenBSD)
Index: patches/patch-i3-dump-log_main_c
===
RCS file: /cvs/ports/x11/i3/patches/patch-i3-dump-log_main_c,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 patch-i3-dump-log_main_c
--- patches/patch-i3-dump-log_main_c30 Mar 2015 12:33:46 -  1.3
+++ patches/patch-i3-dump-log_main_c8 Mar 2016 09:08:20 -
@@ -2,8 +2,8 @@ $OpenBSD: patch-i3-dump-log_main_c,v 1.3
 
 OpenBSD lacks pthread_condattr_setpshared()
 
 i3-dump-log/main.c.origSun Mar 29 19:07:08 2015
-+++ i3-dump-log/main.c Mon Mar 30 14:21:46 2015
+--- i3-dump-log/main.c.origSun Mar  6 16:17:18 2016
 i3-dump-log/main.c Mon Mar  7 15:15:37 2016
 @@ -28,8 +28,12 @@
  #include "shmlog.h"
  #include 
@@ -64,7 +64,7 @@ OpenBSD lacks pthread_condattr_setpshare
  return 0;
  }
  }
-@@ -161,6 +181,7 @@ int main(int argc, char *argv[]) {
+@@ -162,6 +182,7 @@ int main(int argc, char *argv[]) {
  walk = logbuffer + sizeof(i3_shmlog_header);
  print_till_end();
  
@@ -72,7 +72,7 @@ OpenBSD lacks pthread_condattr_setpshare
  if (follow) {
  /* Since pthread_cond_wait() expects a mutex, we need to provide one.
   * To not lock i3 (that???s bad, mhkay?) we just define one outside of
-@@ -176,6 +197,7 @@ int main(int argc, char *argv[]) {
+@@ -177,6 +198,7 @@ int main(int argc, char *argv[]) {
  }
  }
  }
Index: patches/patch-i3-msg_main_c
===
RCS file: patches/patch-i3-msg_main_c
diff -N patches/patch-i3-msg_main_c
--- patches/patch-i3-msg_main_c 26 Jan 2016 09:42:35 -  1.4
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,38 +0,0 @@
-$OpenBSD: patch-i3-msg_main_c,v 1.4 2016/01/26 09:42:35 dcoppa Exp $
-
-commit a1d1f456a1fbf44bfa5c6a9d041023ac498b0c46
-Author: Theo Buehler 
-Date:   Thu Jan 14 10:06:34 2016 +0100
-
-Add pledge(2) support for OpenBSD
-
-pledges for i3:
-"stdio rpath unix" for talking to the i3 socket usually in /tmp
-"proc exec" for executing programs
-"wpath cpath" are needed for the restart-in-place functionality
-
-To make this work, @semarie pointed out that it is sufficient to ensure
-that we get physical_mem_bytes only once, namely in init_logging().
-
-pledges for i3-msg:
-"stdio rpath unix" are needed for talking to the i3-socket
-
-pledges for i3-nagbar
-"rpath getpw" to find the home directory
-"wpath cpath" to write the script
-"proc exec" to execute the script
-"unix" for talking to the i3 socket
-
 i3-msg/main.c.orig Wed Sep 30 08:55:10 2015
-+++ i3-msg/main.c  Sun Dec 20 14:42:51 2015
-@@ -119,6 +119,10 @@ static yajl_callbacks reply_callbacks = {
- };
- 
- int main(int argc, char *argv[]) {
-+#if defined(__OpenBSD__)
-+if (pledge("stdio rpath unix", NULL) == -1)
-+err(EXIT_FAILURE, "pledge");
-+#endif
- char *env_socket_path = getenv("I3SOCK");
- if (env_socket_path)
- socket_path = sstrdup(env_socket_path);
Index: patches/patch-i3-nagbar_main_c
===
RCS file: