* RFC: status icon support
@ 2008-01-11 23:28 Tom Tromey
2008-01-12 1:57 ` Dan Nicolaescu
2008-01-14 3:47 ` YAMAMOTO Mitsuharu
0 siblings, 2 replies; 55+ messages in thread
From: Tom Tromey @ 2008-01-11 23:28 UTC (permalink / raw)
To: emacs-devel
Here is a patch to add status icon support to Emacs.
I think this one is ready for review and, hopefully, checkin.
It includes documentation and the necessary configure support.
I omitted the generated files from the patch, so if you apply it you
must first run autoconf and autoheader.
The xmenu.c changes are mildly ugly, IMO. But, this was the simplest
way to make menus on status icons work, and the damage is reasonably
localized.
Note that the Gtk docs recommend using the activation time from the
GtkStatusIcon ::popup-menu signal as one of the argument to
gtk_menu_popup. I tried this, but it did not work, I think because we
send the popup event through the event loop, perhaps invalidating this
timestamp by the time we reach the actual popup call. So, I did what
the Gtk docs recommend as an alternative, namely using
gtk_get_current_event_time.
There are still a few libnotify features that we do not expose to the
user. These are not critical features (and "hints", in particular,
don't seem to be documented in a way that would make them useful for
anything), and anyway support for them can be added later if needed.
If Gtk is not available, this code is not built at all. Other
platforms (both Windows and MacOS) do provide similar functionality,
but I do not know those platforms, so someone else would have to write
that support. I didn't break out the guts into a platform-specific
file; that would be easy to do if someone wanted to do a port.
Once this goes in I will update my status-icon-using elisp and submit
it as well.
your comments appreciated,
Tom
ChangeLog:
2008-01-11 Tom Tromey <tromey@redhat.com>
* configure: Rebuild.
* configure.in: Check for GtkStatusIcon, libnotify.
doc/lispref/ChangeLog:
2008-01-11 Tom Tromey <tromey@redhat.com>
* frames.texi (Status Icons): New node.
(Frames): Update.
src/ChangeLog:
2008-01-11 Tom Tromey <tromey@redhat.com>
* xmenu.c: Include systray.h.
(Fx_popup_menu): Update documentation. Handle status-icon
position argument.
(create_and_show_popup_menu): Add status_icon and status_button
arguments. Update. Use gtk_get_current_event_time.
(xmenu_show): Add status_icon and status_button arguments.
Update.
* termhooks.h (enum event_kind) <STATUS_ICON_CLICK_EVENT>: New
constant.
* print.c (print_object): Handle status icons.
* lisp.h (enum pvec_type) <PVEC_STATUS_ICON> New constant.
<PVEC_TYPE_MASK>: Update.
(STATUS_ICONP): New macro.
(GC_STATUS_ICONP): Likewise.
(allocate_status_icon, syms_of_systray): Declare.
* keyboard.c (Qstatus_icon_click_event): New global.
(kbd_buffer_get_event): Handle STATUS_ICON_CLICK_EVENT.
(make_lispy_event): Likewise.
(syms_of_keyboard): Initialize Qstatus_icon_click_event. Define
"status-icon-click-event" event.
* emacs.c (main): Call syms_of_systray.
* alloc.c: Include systray.h.
(allocate_status_icon): New function.
* config.in: Rebuild.
* Makefile.in (LIBNOTIFY_CFLAGS): New variable.
(LIBNOTIFY_LIBS): Likewise.
(ALL_CFLAGS): Add LIBNOTIFY_CFLAGS.
(GTK_OBJ): Add systray.o
(LIBES): Add LIBNOTIFY_LIBS.
(systray.o): New target.
* systray.c: New file.
* systray.h: New file.
Index: configure.in
===================================================================
RCS file: /sources/emacs/emacs/configure.in,v
retrieving revision 1.490
diff -u -r1.490 configure.in
--- configure.in 6 Jan 2008 11:49:25 -0000 1.490
+++ configure.in 11 Jan 2008 23:47:08 -0000
@@ -4,7 +4,7 @@
dnl in the directory containing this script.
dnl
dnl Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2003,
-dnl 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+dnl 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
dnl
dnl This file is part of GNU Emacs.
dnl
@@ -1777,6 +1777,15 @@
[Define to 1 if GTK has both file selection and chooser dialog.])
fi
+ dnl Check if we have GtkStatusIcon.
+ HAVE_GTK_STATUS_ICON=no
+ AC_CHECK_FUNCS(gtk_status_icon_new, HAVE_GTK_STATUS_ICON=yes)
+
+ if test "$HAVE_GTK_STATUS_ICON" = yes; then
+ AC_DEFINE(HAVE_GTK_STATUS_ICON, 1,
+ [Define to 1 if GTK has GtkStatusIcon.])
+ fi
+
dnl Check if pthreads are available. Emacs only needs this when using
dnl gtk_file_chooser under Gnome.
if test "$HAVE_GTK_FILE_CHOOSER" = yes; then
@@ -1807,6 +1816,14 @@
fi
fi
+dnl Check if we have libnotify.
+PKG_CHECK_MODULES(LIBNOTIFY, libnotify >= 0.4.1, HAVE_LIBNOTIFY=yes,
+ HAVE_LIBNOTIFY=no)
+if test "$HAVE_LIBNOTIFY" = yes; then
+ AC_DEFINE(HAVE_LIBNOTIFY, 1,
+ [Define to 1 if the system has libnotify.])
+fi
+
### Link with -lXft if available to work around a bug.
HAVE_XFT=maybe
if test "${HAVE_GTK}" = "yes"; then
Index: doc/lispref/frames.texi
===================================================================
RCS file: /sources/emacs/emacs/doc/lispref/frames.texi,v
retrieving revision 1.5
diff -u -r1.5 frames.texi
--- doc/lispref/frames.texi 28 Dec 2007 22:26:13 -0000 1.5
+++ doc/lispref/frames.texi 11 Jan 2008 23:49:34 -0000
@@ -1,7 +1,7 @@
@c -*-texinfo-*-
@c This is part of the GNU Emacs Lisp Reference Manual.
@c Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1998, 1999, 2001,
-@c 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+@c 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
@c See the file elisp.texi for copying conditions.
@setfilename ../../info/frames
@node Frames, Positions, Windows, Top
@@ -75,6 +75,7 @@
* Text Terminal Colors:: Defining colors for text-only terminals.
* Resources:: Getting resource values from the server.
* Display Feature Testing:: Determining the features of a terminal.
+* Status Icons:: Status icons and notifications.
@end menu
@xref{Display}, for information about the related topic of
@@ -2218,6 +2219,128 @@
width and height of an X Window frame, measured in pixels.
@end ignore
+@node Status Icons
+@section Status Icons and Notifications
+@cindex status icons
+@cindex system tray
+
+Some platforms provide a ``system tray'', an area holding icons tied
+to particular applications. Applications use these icons to
+unobtrusively communicate information to the user. Status icons also
+support ``notifications'', which are pop-up windows displaying some
+text and perhaps having some buttons. A notification is less
+intrusive than a dialog because it does not steal the focus and need
+not be manually dismissed. Emacs can create and manipulate status
+icons, and send notifications.
+
+If status icons are available, the feature @code{systray} is provided.
+
+@defun make-status-icon &optional alist
+This function creates and returns a new status icon. The optional
+argument @var{alist} provides initial properties for the icon. Some
+properties are specially recognized; unrecognized properties are
+retained (for use by @code{status-icon-parameters}) but otherwise
+ignored.
+
+The recognized properties are:
+@table @code
+@item icon-name
+The value is the file name of the image to display on the icon.
+@code{nil} means that Emacs should display a default image.
+
+@item blinking
+If the value is non-@code{nil}, the icon will blink. The default is
+@code{nil}.
+
+@item visible
+If the value is @code{nil}, the icon will not be shown. The default
+is @code{t}.
+
+@item help-echo
+The value is a string which will displayed as a tooltip for the icon.
+
+@item click-callback
+The value is a no-argument function which will be called when the user
+clicks on the icon.
+
+@item menu
+The value is a menu (of any kind recognized by @code{popup-menu})
+which will be displayed when the user right-clicks on the icon.
+@end table
+@end defun
+
+@defun delete-status-icon icon
+Destroy @var{icon}.
+@end defun
+
+@defun modify-status-icon-parameters icon alist
+Modify the parameters of @var{icon} according. @var{alist} is an
+alist of parameters of the same kind accepted by
+@code{make-status-icon}.
+@end defun
+
+@defun status-iconp object
+Return non-@code{nil} if @var{object} is a status icon.
+@end defun
+
+@defun status-icon-live-p object
+Return non-@code{nil} if @var{object} is an undestroyed status icon.
+@end defun
+
+@defun status-icon-list
+Return a list of all undestroyed status icons.
+@end defun
+
+@defun show-status-icon-message icon summary &rest args
+This function creates and displays a new notification. Note that this
+function may not be available on all platforms, and can be missing
+even if the basic status icon support is available.
+
+@var{icon} is the status icon to which the notification should attach.
+If @var{icon} is @code{nil}, then the notification will not be
+attached to an icon, but instead will display in some default
+location.
+
+@var{summary} is the summary text of the notification, a string.
+
+@var{args}, if given, is a list of key-value pairs. The recognized
+keys and their meanings are:
+@table @code
+@item :body @var{string}
+Set the body text of the notification. This is ordinarily displayed
+in a different font than the summary text.
+
+@item :timeout @var{value}
+By default a notification will time out. The default timeout interval
+is system-specific. If @var{value} is @code{nil}, the default is
+used; if it is @code{t}, the notification will never time out.
+Otherwise, if @var{value} is an integer, then it specifies the amount
+of time in milliseconds.
+
+@item :urgency @var{value}
+Set the urgency level. @var{value} is a symbol, one of @samp{low},
+@samp{normal}, @samp{critical}. The default is @samp{normal}. This
+setting may affect how the notification is displayed; for instance a
+@samp{critical} notification may have a red background.
+
+@item :icon @var{value}
+Set the notification's icon. @samp{value} is a symbol, currently one
+of @samp{warning}, @samp{info}, @samp{question}, @samp{error}. The
+default is @samp{info}.
+
+@item :action (@var{label} . @var{function})
+Add an action to the notification, with label @var{label}. An action
+typically appears on the notification as a small button. Multiple
+actions can be added, to give the user a choice of how to react to a
+given event. @var{function} is called with no arguments if the action
+is chosen.
+
+@item :closed-callback @var{function}
+Call @var{function} with no arguments when the notification is closed.
+This is not called when the user clicks an action button.
+@end table
+@end defun
+
@ignore
arch-tag: 94977df6-3dca-4730-b57b-c6329e9282ba
@end ignore
Index: src/Makefile.in
===================================================================
RCS file: /sources/emacs/emacs/src/Makefile.in,v
retrieving revision 1.361
diff -u -r1.361 Makefile.in
--- src/Makefile.in 2 Dec 2007 16:23:40 -0000 1.361
+++ src/Makefile.in 11 Jan 2008 23:47:12 -0000
@@ -1,6 +1,6 @@
# Makefile for GNU Emacs.
# Copyright (C) 1985, 1987, 1988, 1993, 1994, 1995, 1999, 2000, 2001, 2002,
-# 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+# 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
# This file is part of GNU Emacs.
@@ -268,6 +268,11 @@
DBUS_OBJ = dbusbind.o
#endif
+#ifdef HAVE_LIBNOTIFY
+LIBNOTIFY_CFLAGS = @LIBNOTIFY_CFLAGS@
+LIBNOTIFY_LIBS = @LIBNOTIFY_LIBS@
+#endif
+
/* DO NOT use -R. There is a special hack described in lastfile.c
which is used instead. Some initialized data areas are modified
at initial startup, then labeled as part of the text area when
@@ -281,7 +286,7 @@
/* C_SWITCH_X_SITE must come before C_SWITCH_X_MACHINE and C_SWITCH_X_SYSTEM
since it may have -I options that should override those two. */
-ALL_CFLAGS=-Demacs -DHAVE_CONFIG_H $(TOOLKIT_DEFINES) $(MYCPPFLAGS) -I. -I${srcdir} C_SWITCH_MACHINE C_SWITCH_SYSTEM C_SWITCH_SITE C_SWITCH_X_SITE C_SWITCH_X_MACHINE C_SWITCH_X_SYSTEM C_SWITCH_SYSTEM_TEMACS ${CFLAGS_SOUND} ${RSVG_CFLAGS} ${DBUS_CFLAGS} ${CFLAGS}
+ALL_CFLAGS=-Demacs -DHAVE_CONFIG_H $(TOOLKIT_DEFINES) $(MYCPPFLAGS) -I. -I${srcdir} C_SWITCH_MACHINE C_SWITCH_SYSTEM C_SWITCH_SITE C_SWITCH_X_SITE C_SWITCH_X_MACHINE C_SWITCH_X_SYSTEM C_SWITCH_SYSTEM_TEMACS ${CFLAGS_SOUND} ${RSVG_CFLAGS} ${DBUS_CFLAGS} ${LIBNOTIFY_CFLAGS} ${CFLAGS}
.c.o:
$(CC) -c $(CPPFLAGS) $(ALL_CFLAGS) $<
@@ -309,7 +314,7 @@
#ifdef HAVE_MENUS
#ifdef USE_GTK
-GTK_OBJ= gtkutil.o
+GTK_OBJ= gtkutil.o systray.o
#endif
/* The X Menu stuff is present in the X10 distribution, but missing
@@ -945,6 +950,7 @@
duplicated symbols. If the standard libraries were compiled
with GCC, we might need gnulib again after them. */
LIBES = $(LOADLIBES) $(LIBS) $(LIBX) $(LIBSOUND) $(RSVG_LIBS) $(DBUS_LIBS) \
+ $(LIBNOTIFY_LIBS) \
LIBGPM LIBRESOLV LIBS_SYSTEM LIBS_MACHINE LIBS_TERMCAP \
LIBS_DEBUG $(GETLOADAVG_LIBS) $(GNULIB_VAR) LIB_MATH LIB_STANDARD \
$(GNULIB_VAR)
@@ -1190,6 +1196,7 @@
sysdep.o: sysdep.c syssignal.h systty.h systime.h syswait.h blockinput.h \
process.h dispextern.h termhooks.h termchar.h termopts.h \
frame.h atimer.h window.h msdos.h dosfns.h keyboard.h cm.h $(config_h)
+systray.o: systray.c $(config_h) lisp.h frame.h
term.o: term.c termchar.h termhooks.h termopts.h $(config_h) cm.h frame.h \
disptab.h dispextern.h keyboard.h charset.h coding.h ccl.h msdos.h \
window.h keymap.h blockinput.h atimer.h systime.h
Index: src/alloc.c
===================================================================
RCS file: /sources/emacs/emacs/src/alloc.c,v
retrieving revision 1.432
diff -u -r1.432 alloc.c
--- src/alloc.c 16 Nov 2007 21:24:59 -0000 1.432
+++ src/alloc.c 11 Jan 2008 23:47:12 -0000
@@ -1,6 +1,6 @@
/* Storage allocation and gc for GNU Emacs Lisp interpreter.
Copyright (C) 1985, 1986, 1988, 1993, 1994, 1995, 1997, 1998, 1999,
- 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+ 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Emacs.
@@ -56,6 +56,7 @@
#include "charset.h"
#include "syssignal.h"
#include "termhooks.h" /* For struct terminal. */
+#include "systray.h"
#include <setjmp.h>
/* GC_MALLOC_CHECK defined means perform validity checks of malloc'd
@@ -3021,6 +3022,17 @@
return f;
}
+struct status_icon *
+allocate_status_icon ()
+{
+ struct status_icon *s = ALLOCATE_PSEUDOVECTOR (struct status_icon,
+ notification_alist,
+ PVEC_STATUS_ICON);
+ /* Zero out the non-GC'd fields. FIXME: This should be made unnecessary. */
+ bzero (&(s->param_alist),
+ ((char*)(s+1)) - ((char*)&(s->notification_alist)));
+ return s;
+}
struct Lisp_Process *
allocate_process ()
Index: src/emacs.c
===================================================================
RCS file: /sources/emacs/emacs/src/emacs.c,v
retrieving revision 1.413
diff -u -r1.413 emacs.c
--- src/emacs.c 2 Dec 2007 16:23:40 -0000 1.413
+++ src/emacs.c 11 Jan 2008 23:47:12 -0000
@@ -1,6 +1,6 @@
/* Fully extensible Emacs, running on Unix, intended for GNU.
Copyright (C) 1985, 1986, 1987, 1993, 1994, 1995, 1997, 1998, 1999,
- 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+ 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Emacs.
@@ -1643,6 +1643,10 @@
syms_of_dbusbind ();
#endif /* HAVE_DBUS */
+#ifdef HAVE_GTK
+ syms_of_systray ();
+#endif
+
#ifdef SYMS_SYSTEM
SYMS_SYSTEM;
#endif
Index: src/keyboard.c
===================================================================
RCS file: /sources/emacs/emacs/src/keyboard.c,v
retrieving revision 1.937
diff -u -r1.937 keyboard.c
--- src/keyboard.c 6 Jan 2008 21:34:57 -0000 1.937
+++ src/keyboard.c 11 Jan 2008 23:47:13 -0000
@@ -500,6 +500,8 @@
#ifdef HAVE_DBUS
Lisp_Object Qdbus_event;
#endif
+Lisp_Object Qstatus_icon_click_event;
+
/* Lisp_Object Qmouse_movement; - also an event header */
/* Properties of event headers. */
@@ -4298,6 +4300,11 @@
kbd_fetch_ptr = event + 1;
}
#endif
+ else if (event->kind == STATUS_ICON_CLICK_EVENT)
+ {
+ obj = make_lispy_event (event);
+ kbd_fetch_ptr = event + 1;
+ }
else
{
/* If this event is on a different frame, return a switch-frame this
@@ -6174,6 +6181,9 @@
}
#endif /* HAVE_DBUS */
+ case STATUS_ICON_CLICK_EVENT:
+ return Fcons (Qstatus_icon_click_event, event->arg);
+
#ifdef HAVE_GPM
case GPM_CLICK_EVENT:
{
@@ -11781,6 +11791,9 @@
staticpro (&Qdbus_event);
#endif
+ Qstatus_icon_click_event = intern ("status-icon-click-event");
+ staticpro (&Qstatus_icon_click_event);
+
Qmenu_enable = intern ("menu-enable");
staticpro (&Qmenu_enable);
Qmenu_alias = intern ("menu-alias");
@@ -12517,6 +12530,9 @@
initial_define_lispy_key (Vspecial_event_map, "dbus-event",
"dbus-handle-event");
#endif
+
+ initial_define_lispy_key (Vspecial_event_map, "status-icon-click-event",
+ "status-icon-handle-click-event");
}
/* Mark the pointers in the kboard objects.
Index: src/lisp.h
===================================================================
RCS file: /sources/emacs/emacs/src/lisp.h,v
retrieving revision 1.603
diff -u -r1.603 lisp.h
--- src/lisp.h 22 Nov 2007 01:01:26 -0000 1.603
+++ src/lisp.h 11 Jan 2008 23:47:14 -0000
@@ -349,7 +349,8 @@
PVEC_HASH_TABLE = 0x40000,
PVEC_TERMINAL = 0x80000,
PVEC_OTHER = 0x100000,
- PVEC_TYPE_MASK = 0x1ffe00
+ PVEC_STATUS_ICON = 0x200000,
+ PVEC_TYPE_MASK = 0x2ffe00
#if 0 /* This is used to make the value of PSEUDOVECTOR_FLAG available to
GDB. It doesn't work on OS Alpha. Moved to a variable in
@@ -1552,6 +1553,8 @@
#define GC_BOOL_VECTOR_P(x) GC_PSEUDOVECTORP (x, PVEC_BOOL_VECTOR)
#define FRAMEP(x) PSEUDOVECTORP (x, PVEC_FRAME)
#define GC_FRAMEP(x) GC_PSEUDOVECTORP (x, PVEC_FRAME)
+#define STATUS_ICONP(x) PSEUDOVECTORP (x, PVEC_STATUS_ICON)
+#define GC_STATUS_ICONP(x) GC_PSEUDOVECTORP (x, PVEC_STATUS_ICON)
#define SUB_CHAR_TABLE_P(x) (CHAR_TABLE_P (x) && NILP (XCHAR_TABLE (x)->top))
@@ -2616,6 +2620,7 @@
extern struct Lisp_Hash_Table *allocate_hash_table P_ ((void));
extern struct window *allocate_window P_ ((void));
extern struct frame *allocate_frame P_ ((void));
+extern struct status_icon *allocate_status_icon P_ ((void));
extern struct Lisp_Process *allocate_process P_ ((void));
extern struct terminal *allocate_terminal P_ ((void));
extern int gc_in_progress;
@@ -3248,6 +3253,9 @@
extern void syms_of_sound P_ ((void));
extern void init_sound P_ ((void));
+/* Defined in systray.c */
+extern void syms_of_systray P_ ((void));
+
/* Defined in category.c */
extern void init_category_once P_ ((void));
extern void syms_of_category P_ ((void));
Index: src/print.c
===================================================================
RCS file: /sources/emacs/emacs/src/print.c,v
retrieving revision 1.243
diff -u -r1.243 print.c
--- src/print.c 22 Nov 2007 16:16:02 -0000 1.243
+++ src/print.c 11 Jan 2008 23:47:14 -0000
@@ -2031,6 +2031,8 @@
strout (buf, -1, -1, printcharfun, 0);
PRINTCHAR ('>');
}
+ else if (STATUS_ICONP (obj))
+ strout ("#<status-icon>", -1, -1, printcharfun, 0);
else
{
EMACS_INT size = XVECTOR (obj)->size;
Index: src/systray.c
===================================================================
RCS file: src/systray.c
diff -N src/systray.c
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ src/systray.c 11 Jan 2008 23:47:14 -0000
@@ -0,0 +1,680 @@
+/* Elisp bindings system tray icons
+ Copyright (C) 2007, 2008 Free Software Foundation, Inc.
+
+This file is part of GNU Emacs.
+
+GNU Emacs is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GNU Emacs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Emacs; see the file COPYING. If not, write to
+the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA. */
+
+#include "config.h"
+
+#include "lisp.h"
+#include "systray.h"
+#include "frame.h"
+#include "charset.h"
+#include "coding.h"
+#include "blockinput.h"
+#include "termhooks.h"
+#include "xterm.h"
+
+#include <gtk/gtk.h>
+
+#ifdef HAVE_LIBNOTIFY
+#include <libnotify/notify.h>
+#endif
+
+#ifdef HAVE_GTK_STATUS_ICON
+
+/* FIXME: should be in a header. */
+extern Lisp_Object Qhelp_echo;
+extern Lisp_Object Qmenu;
+extern Lisp_Object Qstatus_icon_click_event;
+
+Lisp_Object Vstatus_icon_list;
+Lisp_Object Vstandalone_icon;
+
+Lisp_Object Qblinking;
+Lisp_Object Qstatus_icon_live_p;
+Lisp_Object Qclick_callback;
+Lisp_Object QCbody, QCtimeout, QCurgency, QCicon, QCaction, QCclosed_callback;
+Lisp_Object Qlow, Qcritical, Qwarning, Qquestion, Qerror;
+Lisp_Object Qpopup_menu;
+
+#define GET_ICON(s) ((GtkStatusIcon *) ((s)->icon))
+
+/* Callback which is called when the user clicks on a status icon. */
+static void
+activate_icon (widget, data)
+ GtkWidget *widget;
+ gpointer data;
+{
+ Lisp_Object icon = (Lisp_Object) data;
+
+ /* Just ignore garbage. */
+ if (STATUS_ICON_LIVE_P (XSTATUS_ICON (icon)))
+ {
+ struct status_icon *sicon = XSTATUS_ICON (icon);
+ Lisp_Object elt = Fassq (Qmenu, sicon->param_alist);
+ if (CONSP (elt) && FUNCTIONP (XCDR (elt)))
+ {
+ struct input_event event;
+ EVENT_INIT (event);
+ event.kind = STATUS_ICON_CLICK_EVENT;
+ event.frame_or_window = Qnil;
+ event.arg = XCDR (elt);
+ kbd_buffer_store_event (&event);
+ }
+ }
+}
+
+static void
+pop_up_status_menu (gicon, button, activate_time, user_data)
+ GtkStatusIcon *gicon;
+ guint button;
+ guint activate_time;
+ gpointer user_data;
+{
+ Lisp_Object icon = (Lisp_Object) user_data;
+
+ /* Just ignore garbage. */
+ if (STATUS_ICON_LIVE_P (XSTATUS_ICON (icon)))
+ {
+ struct status_icon *sicon = XSTATUS_ICON (icon);
+ Lisp_Object elt = Fassq (Qmenu, sicon->param_alist);
+ if (CONSP (elt) && ! NILP (XCDR (elt)))
+ {
+ struct gcpro gcpro1;
+ struct input_event event;
+ Lisp_Object function;
+
+ GCPRO1 (function);
+ /* Construct the 'position' list. */
+ function = Fcons (icon,
+ Fcons (make_number (button), Qnil));
+ function = Fcons (Qquote,
+ Fcons (function, Qnil));
+ /* Construct the call to `popup-menu'. */
+ function = Fcons (Qpopup_menu,
+ Fcons (Fcons (Qquote,
+ Fcons (XCDR (elt), Qnil)),
+ Fcons (function, Qnil)));
+ /* Construct the function. */
+ function = Fcons (Qlambda,
+ Fcons (Qnil,
+ Fcons (function, Qnil)));
+
+ EVENT_INIT (event);
+ event.kind = STATUS_ICON_CLICK_EVENT;
+ event.frame_or_window = Qnil;
+ event.arg = function;
+ kbd_buffer_store_event (&event);
+ }
+ }
+}
+
+/* Apply property KEY with value VALUE to icon SICON. */
+static void
+apply_one_icon_property (sicon, key, value)
+ struct status_icon *sicon;
+ Lisp_Object key, value;
+{
+ if (EQ (key, Qicon_name))
+ {
+ struct gcpro gcpro1;
+ Lisp_Object filename = Qnil;
+ GCPRO1 (filename);
+ if (NILP (value))
+ {
+ /* Default to the GNU. */
+ filename = build_string ("emacs_48.png");
+ }
+ else
+ {
+ CHECK_STRING (value);
+ filename = value;
+ }
+ /* Not clear if x_find_image_file returns a string with the
+ correct encoding for Gtk. However, it is already used this
+ way in gtkutil.c. */
+ filename = x_find_image_file (filename);
+ if (NILP (filename))
+ {
+ /* No good choice, so choose a semi-appropriate stock
+ icon. */
+ gtk_status_icon_set_from_stock (GET_ICON (sicon),
+ GTK_STOCK_DIALOG_QUESTION);
+ }
+ else
+ gtk_status_icon_set_from_file (GET_ICON (sicon),
+ (char *) SDATA (filename));
+ UNGCPRO;
+ }
+ else if (EQ (key, Qblinking))
+ gtk_status_icon_set_blinking (GET_ICON (sicon), ! NILP (value));
+ else if (EQ (key, Qvisible))
+ gtk_status_icon_set_visible (GET_ICON (sicon), ! NILP (value));
+ else if (EQ (key, Qhelp_echo))
+ {
+ struct gcpro gcpro1;
+ Lisp_Object tooltip = Qnil;
+ GCPRO1 (tooltip);
+ if (NILP (value))
+ gtk_status_icon_set_tooltip (GET_ICON (sicon), NULL);
+ else
+ {
+ /* FIXME: should accept a sexp to eval here. */
+ CHECK_STRING (value);
+ tooltip = ENCODE_UTF_8 (value);
+ gtk_status_icon_set_tooltip (GET_ICON (sicon),
+ (char *) SDATA (tooltip));
+ }
+ UNGCPRO;
+ }
+ else
+ {
+ /* Ignore things we don't understand. */
+ }
+}
+
+/* Apply the ALIST of properties to status icon ICON. */
+static void
+apply_status_icon_alist (icon, alist)
+ Lisp_Object icon, alist;
+{
+ Lisp_Object tail;
+ struct status_icon *sicon;
+
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+
+ for (tail = alist; CONSP (tail); tail = XCDR (tail))
+ {
+ Lisp_Object key, value;
+ Lisp_Object elt = XCAR (tail);
+ if (! CONSP (elt))
+ continue;
+ key = XCAR (elt);
+ value = XCDR (elt);
+
+ apply_one_icon_property (sicon, key, value);
+
+ /* Update frame parameters. */
+ elt = Fassq (key, sicon->param_alist);
+ if (NILP (elt))
+ sicon->param_alist = Fcons (Fcons (key, value), sicon->param_alist);
+ else
+ Fsetcdr (elt, value);
+ }
+}
+
+DEFUN ("make-status-icon", Fmake_status_icon, Smake_status_icon,
+ 0, 1, 0,
+ doc: /* Return a newly created status icon.
+Optional argument ALIST is an alist of parameters for the new frame.
+See `modify-status-icon-parameters' for a list of recognized parameters.*/)
+ (alist)
+ Lisp_Object alist;
+{
+ struct status_icon *sicon;
+ Lisp_Object icon, deflist;
+ struct gcpro gcpro1;
+
+ check_x ();
+
+ BLOCK_INPUT;
+
+ sicon = allocate_status_icon ();
+ XSETSTATUS_ICON (icon, sicon);
+ sicon->param_alist = Qnil;
+ sicon->notification_alist = Qnil;
+ sicon->icon = gtk_status_icon_new ();
+
+ g_signal_connect (sicon->icon, "activate", G_CALLBACK (activate_icon),
+ (gpointer) icon);
+ g_signal_connect (sicon->icon, "popup-menu", G_CALLBACK (pop_up_status_menu),
+ (gpointer) icon);
+
+ Vstatus_icon_list = Fcons (icon, Vstatus_icon_list);
+
+ GCPRO1 (deflist);
+
+ /* Make some defaults. */
+ deflist = Fcons (Fcons (Qclick_callback, Qnil),
+ Fcons (Fcons (Qicon_name, Qnil),
+ Fcons (Fcons (Qblinking, Qnil),
+ Fcons (Fcons (Qvisible, Qt),
+ Fcons (Fcons (Qhelp_echo, Qnil),
+ alist)))));
+ apply_status_icon_alist (icon, deflist);
+
+ UNBLOCK_INPUT;
+
+ RETURN_UNGCPRO (icon);
+}
+
+DEFUN ("modify-status-icon-parameters", Fmodify_status_icon_parameters,
+ Smodify_status_icon_parameters,
+ 2, 2, 0,
+ doc: /* Modify the parameters of status icon ICON according to ALIST.
+Each element of alist has the form (PARM . VALUE), where PARM is a symbol.
+Undefined PARMs are ignored, but stored in the frame's parameter list
+so that `status-icon-parameters' will return them.
+
+Currently defined parameters and their values are:
+
+ icon-name The file name of an icon to display.
+ blinking If non-nil, the status icon will blink.
+ visible If nil, status icon is invisible.
+ help-echo A string which will be displayed as a tooltip for the
+ status icon.
+ click-callback A no-argument function which will be called when the
+ user clicks on the icon.
+ menu A menu which will pop up when the user clicks on the
+ icon. May not be available on all platforms.*/)
+ (icon, alist)
+ Lisp_Object icon, alist;
+{
+ BLOCK_INPUT;
+ apply_status_icon_alist (icon, alist);
+ UNBLOCK_INPUT;
+ return Qnil;
+}
+
+DEFUN ("status-icon-parameters", Fstatus_icon_parameters,
+ Sstatus_icon_parameters,
+ 1, 1, 0,
+ doc: /* Return the parameters-alist of the status icon ICON.
+It is a list of elements of the form (PARM . VALUE), where PARM is a symbol.*/)
+ (icon)
+ Lisp_Object icon;
+{
+ struct status_icon *sicon;
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+ return Fcopy_alist (sicon->param_alist);
+}
+
+DEFUN ("status-icon-list", Fstatus_icon_list, Sstatus_icon_list,
+ 0, 0, 0,
+ doc: /* Return a list of all status icons. */)
+ ()
+{
+ return Fcopy_sequence (Vstatus_icon_list);
+}
+
+DEFUN ("delete-status-icon", Fdelete_status_icon, Sdelete_status_icon,
+ 1, 1, 0,
+ doc: /* Delete the status icon ICON. */)
+ (icon)
+ Lisp_Object icon;
+{
+ struct status_icon *sicon;
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+ BLOCK_INPUT;
+ g_object_unref (sicon->icon);
+ UNBLOCK_INPUT;
+ sicon->icon = NULL;
+ Vstatus_icon_list = Fdelq (icon, Vstatus_icon_list);
+ return Qnil;
+}
+
+#ifdef HAVE_LIBNOTIFY
+
+/* Called in response to notification action clicks. */
+static void
+handle_notification_event (notification, action_number, callback_arg)
+ NotifyNotification *notification;
+ int action_number;
+ Lisp_Object callback_arg;
+{
+ Lisp_Object icon = XCAR (callback_arg);
+ struct status_icon *sicon = XSTATUS_ICON (icon);
+ Lisp_Object callback = Fnth (make_number (action_number + 2), callback_arg);
+ if (FUNCTIONP (callback))
+ {
+ struct input_event event;
+ EVENT_INIT (event);
+ event.kind = STATUS_ICON_CLICK_EVENT;
+ event.frame_or_window = Qnil;
+ event.arg = callback;
+ kbd_buffer_store_event (&event);
+ }
+ sicon->notification_alist = Fdelq (callback_arg,
+ sicon->notification_alist);
+ g_object_unref (notification);
+}
+
+static void
+action_click (notification, actionid, user_data)
+ NotifyNotification *notification;
+ gchar *actionid;
+ gpointer user_data;
+{
+ int id = atoi (actionid);
+ Lisp_Object callback_arg = (Lisp_Object) user_data;
+ handle_notification_event (notification, id, callback_arg);
+}
+
+static void
+cleanup_notification (notification, user_data)
+ NotifyNotification *notification;
+ gpointer user_data;
+{
+ Lisp_Object callback_arg = (Lisp_Object) user_data;
+ handle_notification_event (notification, -1, callback_arg);
+}
+
+DEFUN ("show-status-icon-message", Fshow_status_icon_message,
+ Sshow_status_icon_message,
+ 2, MANY, 0,
+ doc: /* Post a notification message attached to the icon ICON.
+ICON is a status icon created by `make-status-icon', or nil for a
+standalone notification.
+SUMMARY is the message to display, a string.
+
+The remaining arguments, if any, are a property list specifying
+additional parameters of the notification:
+
+ :body STRING Set the body text of the notification.
+ :timeout VALUE Set the timeout value.
+ nil means use the default.
+ t means never time out.
+ An integer specifies the timeout in milliseconds.
+ :urgency VALUE Set the urgency level. VALUE is a symbol, one of
+ 'low', 'normal', 'critical'. The default is 'normal'.
+ :icon VALUE Set the notification icon. VALUE is a symbol, currently
+ one of 'warning', 'info', 'question', 'error'. The
+ default is 'info'.
+ :action (LABEL . FUNCTION)
+ Add an action to the notification, with label LABEL.
+ FUNCTION is called with no arguments if the action
+ is chosen.
+ :closed-callback FUNCTION
+ Call FUNCTION with no arguments when the notification
+ is closed. This is not called when the user clicks
+ an action button.
+
+usage: (show-status-icon-message ICON SUMMARY &rest ARGS) */)
+ (nargs, args)
+ int nargs;
+ Lisp_Object *args;
+{
+ struct status_icon *sicon;
+ struct gcpro gcpro1, gcpro2, gcpro3, gcpro4;
+ Lisp_Object gc_temp = Qnil;
+ Lisp_Object callback_list = Qnil;
+ Lisp_Object closed_callback = Qnil;
+ Lisp_Object callback_arg = Qnil;
+ NotifyNotification *notification;
+
+ int i;
+ int action_count = 0;
+ Lisp_Object icon = args[0];
+ Lisp_Object summary = args[1];
+
+ check_x ();
+ if ((nargs - 2) % 2 != 0)
+ error ("Invalid number of arguments");
+
+ if (! NILP (icon))
+ CHECK_LIVE_STATUS_ICON (icon);
+ CHECK_STRING (summary);
+
+ BLOCK_INPUT;
+
+ if (! notify_is_initted () && ! notify_init ("Emacs"))
+ {
+ UNBLOCK_INPUT;
+ error ("Couldn't connect to notification server");
+ }
+
+ GCPRO4 (gc_temp, callback_list, closed_callback, callback_arg);
+ gc_temp = ENCODE_UTF_8 (summary);
+
+ /* Pre-allocate the cons that will be passed to all callbacks. We
+ will fill it in later. */
+ callback_arg = Fcons (Qnil, Qnil);
+
+ if (! NILP (icon))
+ {
+ sicon = XSTATUS_ICON (icon);
+ notification
+ = notify_notification_new_with_status_icon (SDATA (gc_temp), NULL,
+ GTK_STOCK_DIALOG_INFO,
+ GET_ICON (XSTATUS_ICON (icon)));
+ }
+ else
+ {
+ icon = Vstandalone_icon;
+ sicon = XSTATUS_ICON (icon);
+ notification
+ = notify_notification_new (SDATA (gc_temp), NULL, GTK_STOCK_DIALOG_INFO,
+ NULL);
+ }
+
+ for (i = 2; i < nargs; i += 2)
+ {
+ Lisp_Object key = args[i];
+ Lisp_Object value = args[i + 1];
+ /* FIXME: still no access to: setting the icon from a pixbuf,
+ maybe hints (what are they good for?), closing the
+ notification by caller. */
+ if (EQ (key, QCbody))
+ {
+ /* Ignore things we don't understand. */
+ if (STRINGP (value))
+ {
+ gc_temp = ENCODE_UTF_8 (value);
+ g_object_set (G_OBJECT (notification), "body", SDATA (gc_temp),
+ NULL);
+ }
+ }
+ else if (EQ (key, QCtimeout))
+ {
+ gint timeout = NOTIFY_EXPIRES_DEFAULT;
+ if (EQ (value, Qt))
+ timeout = NOTIFY_EXPIRES_NEVER;
+ else if (INTEGERP (value))
+ {
+ timeout = XINT (value);
+ if (timeout <= 0)
+ timeout = NOTIFY_EXPIRES_DEFAULT;
+ }
+ notify_notification_set_timeout (notification, timeout);
+ }
+ else if (EQ (key, QCurgency))
+ {
+ NotifyUrgency urgency = NOTIFY_URGENCY_NORMAL;
+ if (EQ (value, Qlow))
+ urgency = NOTIFY_URGENCY_LOW;
+ else if (EQ (value, Qcritical))
+ urgency = NOTIFY_URGENCY_CRITICAL;
+ notify_notification_set_urgency (notification, urgency);
+ }
+ else if (EQ (key, QCicon))
+ {
+ char *icon_name = GTK_STOCK_DIALOG_INFO;
+ if (EQ (value, Qwarning))
+ icon_name = GTK_STOCK_DIALOG_WARNING;
+ else if (EQ (value, Qquestion))
+ icon_name = GTK_STOCK_DIALOG_QUESTION;
+ else if (EQ (value, Qerror))
+ icon_name = GTK_STOCK_DIALOG_ERROR;
+ g_object_set (G_OBJECT (notification), "icon-name", icon_name, NULL);
+ }
+ else if (EQ (key, QCaction))
+ {
+ /* Ignore if we don't understand the value. */
+ if (CONSP (value) && STRINGP (XCAR (value)))
+ {
+ char actionid[20];
+ Lisp_Object label = XCAR (value);
+ gc_temp = ENCODE_UTF_8 (label);
+ /* Use a plain integer for the action id, because we
+ will use it later to look up the action. */
+ sprintf (actionid, "%d", action_count);
+ ++action_count;
+ notify_notification_add_action (notification, actionid,
+ SDATA (gc_temp),
+ action_click,
+ (gpointer) callback_arg, NULL);
+
+ /* We build the list from the back, the reverse it
+ later. */
+ callback_list = Fcons (XCDR (value), callback_list);
+ }
+ }
+ else if (EQ (key, QCclosed_callback))
+ closed_callback = value;
+ }
+
+ /* Put the callback list into the right order, then put the 'closed'
+ callback at the start of the list. */
+ callback_list = Fcons (closed_callback, Fnreverse (callback_list));
+
+ XSETCAR (callback_arg, icon);
+ XSETCDR (callback_arg, callback_list);
+ sicon->notification_alist = Fcons (callback_arg, sicon->notification_alist);
+
+ /* Arrange to clean up when this notification goes away. */
+ g_signal_connect (G_OBJECT (notification), "closed",
+ G_CALLBACK (cleanup_notification),
+ (gpointer) callback_arg);
+
+ /* Ignore errors here for now. */
+ notify_notification_show (notification, NULL);
+
+ UNBLOCK_INPUT;
+
+ RETURN_UNGCPRO (Qnil);
+}
+
+#endif /* HAVE_LIBNOTIFY */
+
+DEFUN ("status-iconp", Fstatus_iconp, Sstatus_iconp,
+ 1, 1, 0, doc: /* Return non-nil if OBJECT is a status icon. */)
+ (object)
+ Lisp_Object object;
+{
+ return STATUS_ICONP (object) ? Qt : Qnil;
+}
+
+DEFUN ("status-icon-live-p", Fstatus_icon_live_p, Sstatus_icon_live_p,
+ 1, 1, 0,
+ doc: /* Return non-nil if OBJECT is a status icon which has not been deleted.*/)
+ (object)
+ Lisp_Object object;
+{
+ return ((STATUS_ICONP (object) && STATUS_ICON_LIVE_P (XSTATUS_ICON (object)))
+ ? Qt : Qnil);
+}
+
+DEFUN ("status-icon-handle-click-event", Fstatus_icon_handle_click_event,
+ Sstatus_icon_handle_click_event,
+ 1, 1, "e",
+ doc: /* Internal handler for status icon click events.
+A status icon click event is generated in response to the user clicking
+on a status icon. This handler calls the specified callback function,
+if any.*/)
+ (event)
+ Lisp_Object event;
+{
+ Lisp_Object fun;
+ struct gcpro gcpro1;
+
+ /* Just ignore garbage. */
+ if (! CONSP (event) || ! EQ (XCAR (event), Qstatus_icon_click_event)
+ || ! FUNCTIONP (XCDR (event)))
+ return Qnil;
+
+ GCPRO1 (fun);
+ fun = XCDR (event);
+ Ffuncall (1, &fun);
+ RETURN_UNGCPRO (Qnil);
+}
+
+void
+syms_of_systray ()
+{
+ struct status_icon *sicon;
+
+ Qstatus_icon_live_p = intern ("status-icon-live-p");
+ staticpro (&Qstatus_icon_live_p);
+ Qblinking = intern ("blinking");
+ staticpro (&Qblinking);
+ Qclick_callback = intern ("click-callback");
+ staticpro (&Qclick_callback);
+ QCbody = intern (":body");
+ staticpro (&QCbody);
+ QCtimeout = intern (":timeout");
+ staticpro (&QCtimeout);
+ QCurgency = intern (":urgency");
+ staticpro (&QCurgency);
+ QCicon = intern (":icon");
+ staticpro (&QCicon);
+ QCaction = intern (":action");
+ staticpro (&QCaction);
+ QCclosed_callback = intern (":closed-callback");
+ staticpro (&QCclosed_callback);
+ Qlow = intern ("low");
+ staticpro (&Qlow);
+ Qcritical = intern ("critical");
+ staticpro (&Qcritical);
+ Qwarning = intern ("warning");
+ staticpro (&Qwarning);
+ Qquestion = intern ("question");
+ staticpro (&Qquestion);
+ Qerror = intern ("error");
+ staticpro (&Qerror);
+ Qpopup_menu = intern ("popup-menu");
+ staticpro (&Qpopup_menu);
+
+ staticpro (&Vstatus_icon_list);
+ Vstatus_icon_list = Qnil;
+
+ defsubr (&Smake_status_icon);
+ defsubr (&Smodify_status_icon_parameters);
+ defsubr (&Sdelete_status_icon);
+ defsubr (&Sstatus_icon_parameters);
+ defsubr (&Sstatus_icon_list);
+#ifdef HAVE_LIBNOTIFY
+ defsubr (&Sshow_status_icon_message);
+#endif /* HAVE_LIBNOTIFY */
+ defsubr (&Sstatus_iconp);
+ defsubr (&Sstatus_icon_live_p);
+ defsubr (&Sstatus_icon_handle_click_event);
+
+ /* We make a dummy status icon so that standalone notifications have
+ a place to store their user-data. This icon never has a GUI or
+ associated native state. */
+ staticpro (&Vstandalone_icon);
+ sicon = allocate_status_icon ();
+ sicon->param_alist = Qnil;
+ sicon->notification_alist = Qnil;
+ XSETSTATUS_ICON (Vstandalone_icon, sicon);
+
+ Fprovide (intern ("systray"), Qnil);
+}
+
+#else /* HAVE_GTK_STATUS_ICON */
+
+void
+syms_of_systray ()
+{
+ /* Nothing. */
+}
+
+#endif
Index: src/systray.h
===================================================================
RCS file: src/systray.h
diff -N src/systray.h
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ src/systray.h 11 Jan 2008 23:47:14 -0000
@@ -0,0 +1,56 @@
+/* Define systray object for GNU Emacs.
+ Copyright (C) 2007, 2008 Free Software Foundation, Inc.
+
+This file is part of GNU Emacs.
+
+GNU Emacs is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GNU Emacs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Emacs; see the file COPYING. If not, write to
+the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA. */
+
+#ifndef EMACS_SYSTRAY_H
+#define EMACS_SYSTRAY_H
+
+extern Lisp_Object Qstatus_icon_live_p;
+
+struct status_icon
+{
+ EMACS_UINT size;
+ struct Lisp_Vector *next;
+
+ /* Parameter alist of this status icon. */
+ Lisp_Object param_alist;
+
+ /* List holding objects we want to keep live while a notification is
+ up. */
+ Lisp_Object notification_alist;
+
+ /* Beyond here, there should be no more Lisp_Object components. */
+
+ /* The icon widget, or NULL if it has been destroyed. Really a
+ 'GtkStatusIcon *'. */
+ void *icon;
+};
+
+#define XSTATUS_ICON(p) \
+ (eassert (GC_STATUS_ICONP (p)), (struct status_icon *) XPNTR (p))
+
+#define XSETSTATUS_ICON(a, b) (XSETPSEUDOVECTOR (a, b, PVEC_STATUS_ICON))
+
+#define STATUS_ICON_LIVE_P(s) (((s)->icon) != 0)
+
+#define CHECK_LIVE_STATUS_ICON(s) \
+ CHECK_TYPE (STATUS_ICONP (s) && STATUS_ICON_LIVE_P (XSTATUS_ICON (s)), \
+ Qstatus_icon_live_p, s)
+
+#endif /* EMACS_SYSTRAY_H */
Index: src/termhooks.h
===================================================================
RCS file: /sources/emacs/emacs/src/termhooks.h,v
retrieving revision 1.90
diff -u -r1.90 termhooks.h
--- src/termhooks.h 2 Dec 2007 16:23:40 -0000 1.90
+++ src/termhooks.h 11 Jan 2008 23:47:14 -0000
@@ -201,6 +201,8 @@
, DBUS_EVENT
#endif
+ , STATUS_ICON_CLICK_EVENT
+
#ifdef WINDOWSNT
/* Generated when an APPCOMMAND event is received, in response to
Multimedia or Internet buttons on some keyboards.
Index: src/xmenu.c
===================================================================
RCS file: /sources/emacs/emacs/src/xmenu.c,v
retrieving revision 1.324
diff -u -r1.324 xmenu.c
--- src/xmenu.c 13 Oct 2007 12:10:07 -0000 1.324
+++ src/xmenu.c 11 Jan 2008 23:47:14 -0000
@@ -1,6 +1,6 @@
/* X Communication module for terminals which understand the X protocol.
Copyright (C) 1986, 1988, 1993, 1994, 1996, 1999, 2000, 2001, 2002, 2003,
- 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+ 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Emacs.
@@ -51,6 +51,7 @@
#include "charset.h"
#include "coding.h"
#include "sysselect.h"
+#include "systray.h"
#ifdef MSDOS
#include "msdos.h"
@@ -156,7 +157,7 @@
Lisp_Object, Lisp_Object));
static int update_frame_menubar P_ ((struct frame *));
static Lisp_Object xmenu_show P_ ((struct frame *, int, int, int, int,
- Lisp_Object, char **));
+ Lisp_Object, char **, void *, int));
static void keymap_panes P_ ((Lisp_Object *, int, int));
static void single_keymap_panes P_ ((Lisp_Object, Lisp_Object, Lisp_Object,
int, int));
@@ -780,6 +781,9 @@
corner of WINDOW. (WINDOW may be a window or a frame object.)
This controls the position of the top left of the menu as a whole.
If POSITION is t, it means to use the current mouse position.
+POSITION can also be a list (STATUS-ICON BUTTON-TIME),
+where STATUS-ICON is a status icon and BUTTON is an integer.
+In this case the menu is displayed over the indicated status icon.
MENU is a specifier for a menu. For the simplest case, MENU is a keymap.
The menu items come from key bindings that have a menu string as well as
@@ -829,6 +833,8 @@
int for_click = 0;
int specpdl_count = SPECPDL_INDEX ();
struct gcpro gcpro1;
+ void *status_icon = NULL;
+ int status_button = 0;
#ifdef HAVE_MENUS
if (! NILP (position))
@@ -843,6 +849,19 @@
{
get_current_pos_p = 1;
}
+ else if (CONSP (position) && STATUS_ICONP (XCAR (position)))
+ {
+ Lisp_Object icon = XCAR (position);
+
+ CHECK_LIVE_STATUS_ICON (icon);
+ status_icon = XSTATUS_ICON (icon)->icon;
+ status_button = XINT (Fcar (Fcdr (position)));
+ for_click = 1;
+ /* Placate later code. */
+ XSETINT (x, 0);
+ XSETINT (y, 0);
+ window = selected_window;
+ }
else
{
tem = Fcar (position);
@@ -1029,7 +1048,8 @@
BLOCK_INPUT;
selection = xmenu_show (f, xpos, ypos, for_click,
- keymaps, title, &error_name);
+ keymaps, title, &error_name,
+ status_icon, status_button);
UNBLOCK_INPUT;
discard_menu_items ();
@@ -2664,18 +2684,22 @@
menu pops down.
menu_item_selection will be set to the selection. */
static void
-create_and_show_popup_menu (f, first_wv, x, y, for_click)
+create_and_show_popup_menu (f, first_wv, x, y, for_click,
+ status_icon, status_button)
FRAME_PTR f;
widget_value *first_wv;
int x;
int y;
int for_click;
+ void *status_icon;
+ int status_button;
{
int i;
GtkWidget *menu;
GtkMenuPositionFunc pos_func = 0; /* Pop up at pointer. */
struct next_popup_x_y popup_x_y;
int specpdl_count = SPECPDL_INDEX ();
+ gpointer user_data = &popup_x_y;
if (! FRAME_X_P (f))
abort ();
@@ -2702,6 +2726,14 @@
i = 0; /* gtk_menu_popup needs this to be 0 for a non-button popup. */
}
+#ifdef HAVE_GTK_STATUS_ICON
+ else if (status_icon)
+ {
+ user_data = status_icon;
+ i = status_button;
+ pos_func = gtk_status_icon_position_menu;
+ }
+#endif /* HAVE_GTK_STATUS_ICON */
else
{
for (i = 0; i < 5; i++)
@@ -2711,7 +2743,8 @@
/* Display the menu. */
gtk_widget_show_all (menu);
- gtk_menu_popup (GTK_MENU (menu), 0, 0, pos_func, &popup_x_y, i, 0);
+ gtk_menu_popup (GTK_MENU (menu), 0, 0, pos_func, user_data, i,
+ gtk_get_current_event_time ());
record_unwind_protect (pop_down_menu, make_save_value (menu, 0));
@@ -2772,14 +2805,18 @@
/* Pop up the menu for frame F defined by FIRST_WV at X/Y and loop until the
menu pops down.
- menu_item_selection will be set to the selection. */
+ menu_item_selection will be set to the selection.
+ The status-icon-related arguments are ignore in this implementation. */
static void
-create_and_show_popup_menu (f, first_wv, x, y, for_click)
+create_and_show_popup_menu (f, first_wv, x, y, for_click,
+ status_icon, status_button)
FRAME_PTR f;
widget_value *first_wv;
int x;
int y;
int for_click;
+ void *status_icon;
+ int status_button;
{
int i;
Arg av[2];
@@ -2848,7 +2885,8 @@
#endif /* not USE_GTK */
static Lisp_Object
-xmenu_show (f, x, y, for_click, keymaps, title, error)
+xmenu_show (f, x, y, for_click, keymaps, title, error,
+ status_icon, status_button)
FRAME_PTR f;
int x;
int y;
@@ -2856,6 +2894,8 @@
int keymaps;
Lisp_Object title;
char **error;
+ void *status_icon;
+ int status_button;
{
int i;
widget_value *wv, *save_wv = 0, *first_wv = 0, *prev_wv = 0;
@@ -3062,7 +3102,8 @@
menu_item_selection = 0;
/* Actually create and show the menu until popped down. */
- create_and_show_popup_menu (f, first_wv, x, y, for_click);
+ create_and_show_popup_menu (f, first_wv, x, y, for_click,
+ status_icon, status_button);
/* Free the widget_value objects we used to specify the contents. */
free_menubar_widget_value_tree (first_wv);
@@ -3523,6 +3564,10 @@
int keymaps;
Lisp_Object title;
char **error;
+ /* Note that status-icon-related arguments are ignored in this
+ implementation */
+ void *status_icon;
+ int status_button;
{
Window root;
XMenu *menu;
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-11 23:28 RFC: status icon support Tom Tromey
@ 2008-01-12 1:57 ` Dan Nicolaescu
2008-01-12 1:28 ` Tom Tromey
` (2 more replies)
2008-01-14 3:47 ` YAMAMOTO Mitsuharu
1 sibling, 3 replies; 55+ messages in thread
From: Dan Nicolaescu @ 2008-01-12 1:57 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs-devel
Tom Tromey <tromey@redhat.com> writes:
> +/* Callback which is called when the user clicks on a status icon. */
> +static void
> +activate_icon (widget, data)
> + GtkWidget *widget;
> + gpointer data;
> +{
Please no K&R in new code.
I am not sure if we actually have a policy about this. If we don't we
should make it now required to have K&R in new code. It's already not
possible to compile emacs with a K&R compiler, so please let's not add
to this.
The last thing we want is emacs to regarded as being a backwards project
because it insists on using a language that was standardized in a
different way 19 years ago.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 1:57 ` Dan Nicolaescu
@ 2008-01-12 1:28 ` Tom Tromey
2008-01-12 1:38 ` Tom Tromey
2008-01-12 11:11 ` Richard Stallman
2 siblings, 0 replies; 55+ messages in thread
From: Tom Tromey @ 2008-01-12 1:28 UTC (permalink / raw)
To: Dan Nicolaescu; +Cc: emacs-devel
>>>>> "Dan" == Dan Nicolaescu <dann@ics.uci.edu> writes:
>> +/* Callback which is called when the user clicks on a status icon. */
>> +static void
>> +activate_icon (widget, data)
>> + GtkWidget *widget;
>> + gpointer data;
>> +{
Dan> Please no K&R in new code.
Hahaha -- I actually went and changed everything to K&R, thinking that
was still the official Emacs style.
I will change it back.
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 1:57 ` Dan Nicolaescu
2008-01-12 1:28 ` Tom Tromey
@ 2008-01-12 1:38 ` Tom Tromey
2008-01-12 8:45 ` Ulrich Mueller
2008-01-14 1:03 ` Michael Olson
2008-01-12 11:11 ` Richard Stallman
2 siblings, 2 replies; 55+ messages in thread
From: Tom Tromey @ 2008-01-12 1:38 UTC (permalink / raw)
To: Dan Nicolaescu; +Cc: emacs-devel
>>>>> "Dan" == Dan Nicolaescu <dann@ics.uci.edu> writes:
Dan> Please no K&R in new code.
Here is the updated patch.
Tom
ChangeLog:
2008-01-11 Tom Tromey <tromey@redhat.com>
* configure: Rebuild.
* configure.in: Check for GtkStatusIcon, libnotify.
doc/lispref/ChangeLog:
2008-01-11 Tom Tromey <tromey@redhat.com>
* frames.texi (Status Icons): New node.
(Frames): Update.
src/ChangeLog:
2008-01-11 Tom Tromey <tromey@redhat.com>
* xmenu.c: Include systray.h.
(Fx_popup_menu): Update documentation. Handle status-icon
position argument.
(create_and_show_popup_menu): Add status_icon and status_button
arguments. Update. Use gtk_get_current_event_time.
(xmenu_show): Add status_icon and status_button arguments.
Update.
* termhooks.h (enum event_kind) <STATUS_ICON_CLICK_EVENT>: New
constant.
* print.c (print_object): Handle status icons.
* lisp.h (enum pvec_type) <PVEC_STATUS_ICON> New constant.
<PVEC_TYPE_MASK>: Update.
(STATUS_ICONP): New macro.
(GC_STATUS_ICONP): Likewise.
(allocate_status_icon, syms_of_systray): Declare.
* keyboard.c (Qstatus_icon_click_event): New global.
(kbd_buffer_get_event): Handle STATUS_ICON_CLICK_EVENT.
(make_lispy_event): Likewise.
(syms_of_keyboard): Initialize Qstatus_icon_click_event. Define
"status-icon-click-event" event.
* emacs.c (main): Call syms_of_systray.
* alloc.c: Include systray.h.
(allocate_status_icon): New function.
* config.in: Rebuild.
* Makefile.in (LIBNOTIFY_CFLAGS): New variable.
(LIBNOTIFY_LIBS): Likewise.
(ALL_CFLAGS): Add LIBNOTIFY_CFLAGS.
(GTK_OBJ): Add systray.o
(LIBES): Add LIBNOTIFY_LIBS.
(systray.o): New target.
* systray.c: New file.
* systray.h: New file.
Index: configure.in
===================================================================
RCS file: /sources/emacs/emacs/configure.in,v
retrieving revision 1.490
diff -u -r1.490 configure.in
--- configure.in 6 Jan 2008 11:49:25 -0000 1.490
+++ configure.in 12 Jan 2008 02:09:06 -0000
@@ -4,7 +4,7 @@
dnl in the directory containing this script.
dnl
dnl Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2003,
-dnl 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+dnl 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
dnl
dnl This file is part of GNU Emacs.
dnl
@@ -1777,6 +1777,15 @@
[Define to 1 if GTK has both file selection and chooser dialog.])
fi
+ dnl Check if we have GtkStatusIcon.
+ HAVE_GTK_STATUS_ICON=no
+ AC_CHECK_FUNCS(gtk_status_icon_new, HAVE_GTK_STATUS_ICON=yes)
+
+ if test "$HAVE_GTK_STATUS_ICON" = yes; then
+ AC_DEFINE(HAVE_GTK_STATUS_ICON, 1,
+ [Define to 1 if GTK has GtkStatusIcon.])
+ fi
+
dnl Check if pthreads are available. Emacs only needs this when using
dnl gtk_file_chooser under Gnome.
if test "$HAVE_GTK_FILE_CHOOSER" = yes; then
@@ -1807,6 +1816,14 @@
fi
fi
+dnl Check if we have libnotify.
+PKG_CHECK_MODULES(LIBNOTIFY, libnotify >= 0.4.1, HAVE_LIBNOTIFY=yes,
+ HAVE_LIBNOTIFY=no)
+if test "$HAVE_LIBNOTIFY" = yes; then
+ AC_DEFINE(HAVE_LIBNOTIFY, 1,
+ [Define to 1 if the system has libnotify.])
+fi
+
### Link with -lXft if available to work around a bug.
HAVE_XFT=maybe
if test "${HAVE_GTK}" = "yes"; then
Index: doc/lispref/frames.texi
===================================================================
RCS file: /sources/emacs/emacs/doc/lispref/frames.texi,v
retrieving revision 1.5
diff -u -r1.5 frames.texi
--- doc/lispref/frames.texi 28 Dec 2007 22:26:13 -0000 1.5
+++ doc/lispref/frames.texi 12 Jan 2008 02:09:07 -0000
@@ -1,7 +1,7 @@
@c -*-texinfo-*-
@c This is part of the GNU Emacs Lisp Reference Manual.
@c Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1998, 1999, 2001,
-@c 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+@c 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
@c See the file elisp.texi for copying conditions.
@setfilename ../../info/frames
@node Frames, Positions, Windows, Top
@@ -75,6 +75,7 @@
* Text Terminal Colors:: Defining colors for text-only terminals.
* Resources:: Getting resource values from the server.
* Display Feature Testing:: Determining the features of a terminal.
+* Status Icons:: Status icons and notifications.
@end menu
@xref{Display}, for information about the related topic of
@@ -2218,6 +2219,128 @@
width and height of an X Window frame, measured in pixels.
@end ignore
+@node Status Icons
+@section Status Icons and Notifications
+@cindex status icons
+@cindex system tray
+
+Some platforms provide a ``system tray'', an area holding icons tied
+to particular applications. Applications use these icons to
+unobtrusively communicate information to the user. Status icons also
+support ``notifications'', which are pop-up windows displaying some
+text and perhaps having some buttons. A notification is less
+intrusive than a dialog because it does not steal the focus and need
+not be manually dismissed. Emacs can create and manipulate status
+icons, and send notifications.
+
+If status icons are available, the feature @code{systray} is provided.
+
+@defun make-status-icon &optional alist
+This function creates and returns a new status icon. The optional
+argument @var{alist} provides initial properties for the icon. Some
+properties are specially recognized; unrecognized properties are
+retained (for use by @code{status-icon-parameters}) but otherwise
+ignored.
+
+The recognized properties are:
+@table @code
+@item icon-name
+The value is the file name of the image to display on the icon.
+@code{nil} means that Emacs should display a default image.
+
+@item blinking
+If the value is non-@code{nil}, the icon will blink. The default is
+@code{nil}.
+
+@item visible
+If the value is @code{nil}, the icon will not be shown. The default
+is @code{t}.
+
+@item help-echo
+The value is a string which will displayed as a tooltip for the icon.
+
+@item click-callback
+The value is a no-argument function which will be called when the user
+clicks on the icon.
+
+@item menu
+The value is a menu (of any kind recognized by @code{popup-menu})
+which will be displayed when the user right-clicks on the icon.
+@end table
+@end defun
+
+@defun delete-status-icon icon
+Destroy @var{icon}.
+@end defun
+
+@defun modify-status-icon-parameters icon alist
+Modify the parameters of @var{icon} according. @var{alist} is an
+alist of parameters of the same kind accepted by
+@code{make-status-icon}.
+@end defun
+
+@defun status-iconp object
+Return non-@code{nil} if @var{object} is a status icon.
+@end defun
+
+@defun status-icon-live-p object
+Return non-@code{nil} if @var{object} is an undestroyed status icon.
+@end defun
+
+@defun status-icon-list
+Return a list of all undestroyed status icons.
+@end defun
+
+@defun show-status-icon-message icon summary &rest args
+This function creates and displays a new notification. Note that this
+function may not be available on all platforms, and can be missing
+even if the basic status icon support is available.
+
+@var{icon} is the status icon to which the notification should attach.
+If @var{icon} is @code{nil}, then the notification will not be
+attached to an icon, but instead will display in some default
+location.
+
+@var{summary} is the summary text of the notification, a string.
+
+@var{args}, if given, is a list of key-value pairs. The recognized
+keys and their meanings are:
+@table @code
+@item :body @var{string}
+Set the body text of the notification. This is ordinarily displayed
+in a different font than the summary text.
+
+@item :timeout @var{value}
+By default a notification will time out. The default timeout interval
+is system-specific. If @var{value} is @code{nil}, the default is
+used; if it is @code{t}, the notification will never time out.
+Otherwise, if @var{value} is an integer, then it specifies the amount
+of time in milliseconds.
+
+@item :urgency @var{value}
+Set the urgency level. @var{value} is a symbol, one of @samp{low},
+@samp{normal}, @samp{critical}. The default is @samp{normal}. This
+setting may affect how the notification is displayed; for instance a
+@samp{critical} notification may have a red background.
+
+@item :icon @var{value}
+Set the notification's icon. @samp{value} is a symbol, currently one
+of @samp{warning}, @samp{info}, @samp{question}, @samp{error}. The
+default is @samp{info}.
+
+@item :action (@var{label} . @var{function})
+Add an action to the notification, with label @var{label}. An action
+typically appears on the notification as a small button. Multiple
+actions can be added, to give the user a choice of how to react to a
+given event. @var{function} is called with no arguments if the action
+is chosen.
+
+@item :closed-callback @var{function}
+Call @var{function} with no arguments when the notification is closed.
+This is not called when the user clicks an action button.
+@end table
+@end defun
+
@ignore
arch-tag: 94977df6-3dca-4730-b57b-c6329e9282ba
@end ignore
Index: src/Makefile.in
===================================================================
RCS file: /sources/emacs/emacs/src/Makefile.in,v
retrieving revision 1.361
diff -u -r1.361 Makefile.in
--- src/Makefile.in 2 Dec 2007 16:23:40 -0000 1.361
+++ src/Makefile.in 12 Jan 2008 02:09:08 -0000
@@ -1,6 +1,6 @@
# Makefile for GNU Emacs.
# Copyright (C) 1985, 1987, 1988, 1993, 1994, 1995, 1999, 2000, 2001, 2002,
-# 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+# 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
# This file is part of GNU Emacs.
@@ -268,6 +268,11 @@
DBUS_OBJ = dbusbind.o
#endif
+#ifdef HAVE_LIBNOTIFY
+LIBNOTIFY_CFLAGS = @LIBNOTIFY_CFLAGS@
+LIBNOTIFY_LIBS = @LIBNOTIFY_LIBS@
+#endif
+
/* DO NOT use -R. There is a special hack described in lastfile.c
which is used instead. Some initialized data areas are modified
at initial startup, then labeled as part of the text area when
@@ -281,7 +286,7 @@
/* C_SWITCH_X_SITE must come before C_SWITCH_X_MACHINE and C_SWITCH_X_SYSTEM
since it may have -I options that should override those two. */
-ALL_CFLAGS=-Demacs -DHAVE_CONFIG_H $(TOOLKIT_DEFINES) $(MYCPPFLAGS) -I. -I${srcdir} C_SWITCH_MACHINE C_SWITCH_SYSTEM C_SWITCH_SITE C_SWITCH_X_SITE C_SWITCH_X_MACHINE C_SWITCH_X_SYSTEM C_SWITCH_SYSTEM_TEMACS ${CFLAGS_SOUND} ${RSVG_CFLAGS} ${DBUS_CFLAGS} ${CFLAGS}
+ALL_CFLAGS=-Demacs -DHAVE_CONFIG_H $(TOOLKIT_DEFINES) $(MYCPPFLAGS) -I. -I${srcdir} C_SWITCH_MACHINE C_SWITCH_SYSTEM C_SWITCH_SITE C_SWITCH_X_SITE C_SWITCH_X_MACHINE C_SWITCH_X_SYSTEM C_SWITCH_SYSTEM_TEMACS ${CFLAGS_SOUND} ${RSVG_CFLAGS} ${DBUS_CFLAGS} ${LIBNOTIFY_CFLAGS} ${CFLAGS}
.c.o:
$(CC) -c $(CPPFLAGS) $(ALL_CFLAGS) $<
@@ -309,7 +314,7 @@
#ifdef HAVE_MENUS
#ifdef USE_GTK
-GTK_OBJ= gtkutil.o
+GTK_OBJ= gtkutil.o systray.o
#endif
/* The X Menu stuff is present in the X10 distribution, but missing
@@ -945,6 +950,7 @@
duplicated symbols. If the standard libraries were compiled
with GCC, we might need gnulib again after them. */
LIBES = $(LOADLIBES) $(LIBS) $(LIBX) $(LIBSOUND) $(RSVG_LIBS) $(DBUS_LIBS) \
+ $(LIBNOTIFY_LIBS) \
LIBGPM LIBRESOLV LIBS_SYSTEM LIBS_MACHINE LIBS_TERMCAP \
LIBS_DEBUG $(GETLOADAVG_LIBS) $(GNULIB_VAR) LIB_MATH LIB_STANDARD \
$(GNULIB_VAR)
@@ -1190,6 +1196,7 @@
sysdep.o: sysdep.c syssignal.h systty.h systime.h syswait.h blockinput.h \
process.h dispextern.h termhooks.h termchar.h termopts.h \
frame.h atimer.h window.h msdos.h dosfns.h keyboard.h cm.h $(config_h)
+systray.o: systray.c $(config_h) lisp.h frame.h
term.o: term.c termchar.h termhooks.h termopts.h $(config_h) cm.h frame.h \
disptab.h dispextern.h keyboard.h charset.h coding.h ccl.h msdos.h \
window.h keymap.h blockinput.h atimer.h systime.h
Index: src/alloc.c
===================================================================
RCS file: /sources/emacs/emacs/src/alloc.c,v
retrieving revision 1.432
diff -u -r1.432 alloc.c
--- src/alloc.c 16 Nov 2007 21:24:59 -0000 1.432
+++ src/alloc.c 12 Jan 2008 02:09:08 -0000
@@ -1,6 +1,6 @@
/* Storage allocation and gc for GNU Emacs Lisp interpreter.
Copyright (C) 1985, 1986, 1988, 1993, 1994, 1995, 1997, 1998, 1999,
- 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+ 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Emacs.
@@ -56,6 +56,7 @@
#include "charset.h"
#include "syssignal.h"
#include "termhooks.h" /* For struct terminal. */
+#include "systray.h"
#include <setjmp.h>
/* GC_MALLOC_CHECK defined means perform validity checks of malloc'd
@@ -3021,6 +3022,17 @@
return f;
}
+struct status_icon *
+allocate_status_icon ()
+{
+ struct status_icon *s = ALLOCATE_PSEUDOVECTOR (struct status_icon,
+ notification_alist,
+ PVEC_STATUS_ICON);
+ /* Zero out the non-GC'd fields. FIXME: This should be made unnecessary. */
+ bzero (&(s->param_alist),
+ ((char*)(s+1)) - ((char*)&(s->notification_alist)));
+ return s;
+}
struct Lisp_Process *
allocate_process ()
Index: src/emacs.c
===================================================================
RCS file: /sources/emacs/emacs/src/emacs.c,v
retrieving revision 1.413
diff -u -r1.413 emacs.c
--- src/emacs.c 2 Dec 2007 16:23:40 -0000 1.413
+++ src/emacs.c 12 Jan 2008 02:09:08 -0000
@@ -1,6 +1,6 @@
/* Fully extensible Emacs, running on Unix, intended for GNU.
Copyright (C) 1985, 1986, 1987, 1993, 1994, 1995, 1997, 1998, 1999,
- 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+ 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Emacs.
@@ -1643,6 +1643,10 @@
syms_of_dbusbind ();
#endif /* HAVE_DBUS */
+#ifdef HAVE_GTK
+ syms_of_systray ();
+#endif
+
#ifdef SYMS_SYSTEM
SYMS_SYSTEM;
#endif
Index: src/keyboard.c
===================================================================
RCS file: /sources/emacs/emacs/src/keyboard.c,v
retrieving revision 1.937
diff -u -r1.937 keyboard.c
--- src/keyboard.c 6 Jan 2008 21:34:57 -0000 1.937
+++ src/keyboard.c 12 Jan 2008 02:09:09 -0000
@@ -500,6 +500,8 @@
#ifdef HAVE_DBUS
Lisp_Object Qdbus_event;
#endif
+Lisp_Object Qstatus_icon_click_event;
+
/* Lisp_Object Qmouse_movement; - also an event header */
/* Properties of event headers. */
@@ -4298,6 +4300,11 @@
kbd_fetch_ptr = event + 1;
}
#endif
+ else if (event->kind == STATUS_ICON_CLICK_EVENT)
+ {
+ obj = make_lispy_event (event);
+ kbd_fetch_ptr = event + 1;
+ }
else
{
/* If this event is on a different frame, return a switch-frame this
@@ -6174,6 +6181,9 @@
}
#endif /* HAVE_DBUS */
+ case STATUS_ICON_CLICK_EVENT:
+ return Fcons (Qstatus_icon_click_event, event->arg);
+
#ifdef HAVE_GPM
case GPM_CLICK_EVENT:
{
@@ -11781,6 +11791,9 @@
staticpro (&Qdbus_event);
#endif
+ Qstatus_icon_click_event = intern ("status-icon-click-event");
+ staticpro (&Qstatus_icon_click_event);
+
Qmenu_enable = intern ("menu-enable");
staticpro (&Qmenu_enable);
Qmenu_alias = intern ("menu-alias");
@@ -12517,6 +12530,9 @@
initial_define_lispy_key (Vspecial_event_map, "dbus-event",
"dbus-handle-event");
#endif
+
+ initial_define_lispy_key (Vspecial_event_map, "status-icon-click-event",
+ "status-icon-handle-click-event");
}
/* Mark the pointers in the kboard objects.
Index: src/lisp.h
===================================================================
RCS file: /sources/emacs/emacs/src/lisp.h,v
retrieving revision 1.603
diff -u -r1.603 lisp.h
--- src/lisp.h 22 Nov 2007 01:01:26 -0000 1.603
+++ src/lisp.h 12 Jan 2008 02:09:10 -0000
@@ -349,7 +349,8 @@
PVEC_HASH_TABLE = 0x40000,
PVEC_TERMINAL = 0x80000,
PVEC_OTHER = 0x100000,
- PVEC_TYPE_MASK = 0x1ffe00
+ PVEC_STATUS_ICON = 0x200000,
+ PVEC_TYPE_MASK = 0x2ffe00
#if 0 /* This is used to make the value of PSEUDOVECTOR_FLAG available to
GDB. It doesn't work on OS Alpha. Moved to a variable in
@@ -1552,6 +1553,8 @@
#define GC_BOOL_VECTOR_P(x) GC_PSEUDOVECTORP (x, PVEC_BOOL_VECTOR)
#define FRAMEP(x) PSEUDOVECTORP (x, PVEC_FRAME)
#define GC_FRAMEP(x) GC_PSEUDOVECTORP (x, PVEC_FRAME)
+#define STATUS_ICONP(x) PSEUDOVECTORP (x, PVEC_STATUS_ICON)
+#define GC_STATUS_ICONP(x) GC_PSEUDOVECTORP (x, PVEC_STATUS_ICON)
#define SUB_CHAR_TABLE_P(x) (CHAR_TABLE_P (x) && NILP (XCHAR_TABLE (x)->top))
@@ -2616,6 +2620,7 @@
extern struct Lisp_Hash_Table *allocate_hash_table P_ ((void));
extern struct window *allocate_window P_ ((void));
extern struct frame *allocate_frame P_ ((void));
+extern struct status_icon *allocate_status_icon P_ ((void));
extern struct Lisp_Process *allocate_process P_ ((void));
extern struct terminal *allocate_terminal P_ ((void));
extern int gc_in_progress;
@@ -3248,6 +3253,9 @@
extern void syms_of_sound P_ ((void));
extern void init_sound P_ ((void));
+/* Defined in systray.c */
+extern void syms_of_systray P_ ((void));
+
/* Defined in category.c */
extern void init_category_once P_ ((void));
extern void syms_of_category P_ ((void));
Index: src/print.c
===================================================================
RCS file: /sources/emacs/emacs/src/print.c,v
retrieving revision 1.243
diff -u -r1.243 print.c
--- src/print.c 22 Nov 2007 16:16:02 -0000 1.243
+++ src/print.c 12 Jan 2008 02:09:10 -0000
@@ -2031,6 +2031,8 @@
strout (buf, -1, -1, printcharfun, 0);
PRINTCHAR ('>');
}
+ else if (STATUS_ICONP (obj))
+ strout ("#<status-icon>", -1, -1, printcharfun, 0);
else
{
EMACS_INT size = XVECTOR (obj)->size;
Index: src/systray.c
===================================================================
RCS file: src/systray.c
diff -N src/systray.c
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ src/systray.c 12 Jan 2008 02:09:10 -0000
@@ -0,0 +1,667 @@
+/* Elisp bindings system tray icons
+ Copyright (C) 2007, 2008 Free Software Foundation, Inc.
+
+This file is part of GNU Emacs.
+
+GNU Emacs is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GNU Emacs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Emacs; see the file COPYING. If not, write to
+the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA. */
+
+#include "config.h"
+
+#include "lisp.h"
+#include "systray.h"
+#include "frame.h"
+#include "charset.h"
+#include "coding.h"
+#include "blockinput.h"
+#include "termhooks.h"
+#include "xterm.h"
+
+#include <gtk/gtk.h>
+
+#ifdef HAVE_LIBNOTIFY
+#include <libnotify/notify.h>
+#endif
+
+#ifdef HAVE_GTK_STATUS_ICON
+
+/* FIXME: should be in a header. */
+extern Lisp_Object Qhelp_echo;
+extern Lisp_Object Qmenu;
+extern Lisp_Object Qstatus_icon_click_event;
+
+Lisp_Object Vstatus_icon_list;
+Lisp_Object Vstandalone_icon;
+
+Lisp_Object Qblinking;
+Lisp_Object Qstatus_icon_live_p;
+Lisp_Object Qclick_callback;
+Lisp_Object QCbody, QCtimeout, QCurgency, QCicon, QCaction, QCclosed_callback;
+Lisp_Object Qlow, Qcritical, Qwarning, Qquestion, Qerror;
+Lisp_Object Qpopup_menu;
+
+#define GET_ICON(s) ((GtkStatusIcon *) ((s)->icon))
+
+/* Callback which is called when the user clicks on a status icon. */
+static void
+activate_icon (GtkWidget *widget, gpointer data)
+{
+ Lisp_Object icon = (Lisp_Object) data;
+
+ /* Just ignore garbage. */
+ if (STATUS_ICON_LIVE_P (XSTATUS_ICON (icon)))
+ {
+ struct status_icon *sicon = XSTATUS_ICON (icon);
+ Lisp_Object elt = Fassq (Qmenu, sicon->param_alist);
+ if (CONSP (elt) && FUNCTIONP (XCDR (elt)))
+ {
+ struct input_event event;
+ EVENT_INIT (event);
+ event.kind = STATUS_ICON_CLICK_EVENT;
+ event.frame_or_window = Qnil;
+ event.arg = XCDR (elt);
+ kbd_buffer_store_event (&event);
+ }
+ }
+}
+
+static void
+pop_up_status_menu (GtkStatusIcon *gicon, guint button,
+ guint activate_time, gpointer user_data)
+{
+ Lisp_Object icon = (Lisp_Object) user_data;
+
+ /* Just ignore garbage. */
+ if (STATUS_ICON_LIVE_P (XSTATUS_ICON (icon)))
+ {
+ struct status_icon *sicon = XSTATUS_ICON (icon);
+ Lisp_Object elt = Fassq (Qmenu, sicon->param_alist);
+ if (CONSP (elt) && ! NILP (XCDR (elt)))
+ {
+ struct gcpro gcpro1;
+ struct input_event event;
+ Lisp_Object function;
+
+ GCPRO1 (function);
+ /* Construct the 'position' list. */
+ function = Fcons (icon,
+ Fcons (make_number (button), Qnil));
+ function = Fcons (Qquote,
+ Fcons (function, Qnil));
+ /* Construct the call to `popup-menu'. */
+ function = Fcons (Qpopup_menu,
+ Fcons (Fcons (Qquote,
+ Fcons (XCDR (elt), Qnil)),
+ Fcons (function, Qnil)));
+ /* Construct the function. */
+ function = Fcons (Qlambda,
+ Fcons (Qnil,
+ Fcons (function, Qnil)));
+
+ EVENT_INIT (event);
+ event.kind = STATUS_ICON_CLICK_EVENT;
+ event.frame_or_window = Qnil;
+ event.arg = function;
+ kbd_buffer_store_event (&event);
+ }
+ }
+}
+
+/* Apply property KEY with value VALUE to icon SICON. */
+static void
+apply_one_icon_property (struct status_icon *sicon, Lisp_Object key,
+ Lisp_Object value)
+{
+ if (EQ (key, Qicon_name))
+ {
+ struct gcpro gcpro1;
+ Lisp_Object filename = Qnil;
+ GCPRO1 (filename);
+ if (NILP (value))
+ {
+ /* Default to the GNU. */
+ filename = build_string ("emacs_48.png");
+ }
+ else
+ {
+ CHECK_STRING (value);
+ filename = value;
+ }
+ /* Not clear if x_find_image_file returns a string with the
+ correct encoding for Gtk. However, it is already used this
+ way in gtkutil.c. */
+ filename = x_find_image_file (filename);
+ if (NILP (filename))
+ {
+ /* No good choice, so choose a semi-appropriate stock
+ icon. */
+ gtk_status_icon_set_from_stock (GET_ICON (sicon),
+ GTK_STOCK_DIALOG_QUESTION);
+ }
+ else
+ gtk_status_icon_set_from_file (GET_ICON (sicon),
+ (char *) SDATA (filename));
+ UNGCPRO;
+ }
+ else if (EQ (key, Qblinking))
+ gtk_status_icon_set_blinking (GET_ICON (sicon), ! NILP (value));
+ else if (EQ (key, Qvisible))
+ gtk_status_icon_set_visible (GET_ICON (sicon), ! NILP (value));
+ else if (EQ (key, Qhelp_echo))
+ {
+ struct gcpro gcpro1;
+ Lisp_Object tooltip = Qnil;
+ GCPRO1 (tooltip);
+ if (NILP (value))
+ gtk_status_icon_set_tooltip (GET_ICON (sicon), NULL);
+ else
+ {
+ /* FIXME: should accept a sexp to eval here. */
+ CHECK_STRING (value);
+ tooltip = ENCODE_UTF_8 (value);
+ gtk_status_icon_set_tooltip (GET_ICON (sicon),
+ (char *) SDATA (tooltip));
+ }
+ UNGCPRO;
+ }
+ else
+ {
+ /* Ignore things we don't understand. */
+ }
+}
+
+/* Apply the ALIST of properties to status icon ICON. */
+static void
+apply_status_icon_alist (Lisp_Object icon, Lisp_Object alist)
+{
+ Lisp_Object tail;
+ struct status_icon *sicon;
+
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+
+ for (tail = alist; CONSP (tail); tail = XCDR (tail))
+ {
+ Lisp_Object key, value;
+ Lisp_Object elt = XCAR (tail);
+ if (! CONSP (elt))
+ continue;
+ key = XCAR (elt);
+ value = XCDR (elt);
+
+ apply_one_icon_property (sicon, key, value);
+
+ /* Update frame parameters. */
+ elt = Fassq (key, sicon->param_alist);
+ if (NILP (elt))
+ sicon->param_alist = Fcons (Fcons (key, value), sicon->param_alist);
+ else
+ Fsetcdr (elt, value);
+ }
+}
+
+DEFUN ("make-status-icon", Fmake_status_icon, Smake_status_icon,
+ 0, 1, 0,
+ doc: /* Return a newly created status icon.
+Optional argument ALIST is an alist of parameters for the new frame.
+See `modify-status-icon-parameters' for a list of recognized parameters.*/)
+ (alist)
+ Lisp_Object alist;
+{
+ struct status_icon *sicon;
+ Lisp_Object icon, deflist;
+ struct gcpro gcpro1;
+
+ check_x ();
+
+ BLOCK_INPUT;
+
+ sicon = allocate_status_icon ();
+ XSETSTATUS_ICON (icon, sicon);
+ sicon->param_alist = Qnil;
+ sicon->notification_alist = Qnil;
+ sicon->icon = gtk_status_icon_new ();
+
+ g_signal_connect (sicon->icon, "activate", G_CALLBACK (activate_icon),
+ (gpointer) icon);
+ g_signal_connect (sicon->icon, "popup-menu", G_CALLBACK (pop_up_status_menu),
+ (gpointer) icon);
+
+ Vstatus_icon_list = Fcons (icon, Vstatus_icon_list);
+
+ GCPRO1 (deflist);
+
+ /* Make some defaults. */
+ deflist = Fcons (Fcons (Qclick_callback, Qnil),
+ Fcons (Fcons (Qicon_name, Qnil),
+ Fcons (Fcons (Qblinking, Qnil),
+ Fcons (Fcons (Qvisible, Qt),
+ Fcons (Fcons (Qhelp_echo, Qnil),
+ alist)))));
+ apply_status_icon_alist (icon, deflist);
+
+ UNBLOCK_INPUT;
+
+ RETURN_UNGCPRO (icon);
+}
+
+DEFUN ("modify-status-icon-parameters", Fmodify_status_icon_parameters,
+ Smodify_status_icon_parameters,
+ 2, 2, 0,
+ doc: /* Modify the parameters of status icon ICON according to ALIST.
+Each element of alist has the form (PARM . VALUE), where PARM is a symbol.
+Undefined PARMs are ignored, but stored in the frame's parameter list
+so that `status-icon-parameters' will return them.
+
+Currently defined parameters and their values are:
+
+ icon-name The file name of an icon to display.
+ blinking If non-nil, the status icon will blink.
+ visible If nil, status icon is invisible.
+ help-echo A string which will be displayed as a tooltip for the
+ status icon.
+ click-callback A no-argument function which will be called when the
+ user clicks on the icon.
+ menu A menu which will pop up when the user clicks on the
+ icon. May not be available on all platforms.*/)
+ (icon, alist)
+ Lisp_Object icon, alist;
+{
+ BLOCK_INPUT;
+ apply_status_icon_alist (icon, alist);
+ UNBLOCK_INPUT;
+ return Qnil;
+}
+
+DEFUN ("status-icon-parameters", Fstatus_icon_parameters,
+ Sstatus_icon_parameters,
+ 1, 1, 0,
+ doc: /* Return the parameters-alist of the status icon ICON.
+It is a list of elements of the form (PARM . VALUE), where PARM is a symbol.*/)
+ (icon)
+ Lisp_Object icon;
+{
+ struct status_icon *sicon;
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+ return Fcopy_alist (sicon->param_alist);
+}
+
+DEFUN ("status-icon-list", Fstatus_icon_list, Sstatus_icon_list,
+ 0, 0, 0,
+ doc: /* Return a list of all status icons. */)
+ ()
+{
+ return Fcopy_sequence (Vstatus_icon_list);
+}
+
+DEFUN ("delete-status-icon", Fdelete_status_icon, Sdelete_status_icon,
+ 1, 1, 0,
+ doc: /* Delete the status icon ICON. */)
+ (icon)
+ Lisp_Object icon;
+{
+ struct status_icon *sicon;
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+ BLOCK_INPUT;
+ g_object_unref (sicon->icon);
+ UNBLOCK_INPUT;
+ sicon->icon = NULL;
+ Vstatus_icon_list = Fdelq (icon, Vstatus_icon_list);
+ return Qnil;
+}
+
+#ifdef HAVE_LIBNOTIFY
+
+/* Called in response to notification action clicks. */
+static void
+handle_notification_event (NotifyNotification *notification,
+ int action_number, Lisp_Object callback_arg)
+{
+ Lisp_Object icon = XCAR (callback_arg);
+ struct status_icon *sicon = XSTATUS_ICON (icon);
+ Lisp_Object callback = Fnth (make_number (action_number + 2), callback_arg);
+ if (FUNCTIONP (callback))
+ {
+ struct input_event event;
+ EVENT_INIT (event);
+ event.kind = STATUS_ICON_CLICK_EVENT;
+ event.frame_or_window = Qnil;
+ event.arg = callback;
+ kbd_buffer_store_event (&event);
+ }
+ sicon->notification_alist = Fdelq (callback_arg,
+ sicon->notification_alist);
+ g_object_unref (notification);
+}
+
+static void
+action_click (NotifyNotification *notification, gchar *actionid,
+ gpointer user_data)
+{
+ int id = atoi (actionid);
+ Lisp_Object callback_arg = (Lisp_Object) user_data;
+ handle_notification_event (notification, id, callback_arg);
+}
+
+static void
+cleanup_notification (NotifyNotification *notification, gpointer user_data)
+{
+ Lisp_Object callback_arg = (Lisp_Object) user_data;
+ handle_notification_event (notification, -1, callback_arg);
+}
+
+DEFUN ("show-status-icon-message", Fshow_status_icon_message,
+ Sshow_status_icon_message,
+ 2, MANY, 0,
+ doc: /* Post a notification message attached to the icon ICON.
+ICON is a status icon created by `make-status-icon', or nil for a
+standalone notification.
+SUMMARY is the message to display, a string.
+
+The remaining arguments, if any, are a property list specifying
+additional parameters of the notification:
+
+ :body STRING Set the body text of the notification.
+ :timeout VALUE Set the timeout value.
+ nil means use the default.
+ t means never time out.
+ An integer specifies the timeout in milliseconds.
+ :urgency VALUE Set the urgency level. VALUE is a symbol, one of
+ 'low', 'normal', 'critical'. The default is 'normal'.
+ :icon VALUE Set the notification icon. VALUE is a symbol, currently
+ one of 'warning', 'info', 'question', 'error'. The
+ default is 'info'.
+ :action (LABEL . FUNCTION)
+ Add an action to the notification, with label LABEL.
+ FUNCTION is called with no arguments if the action
+ is chosen.
+ :closed-callback FUNCTION
+ Call FUNCTION with no arguments when the notification
+ is closed. This is not called when the user clicks
+ an action button.
+
+usage: (show-status-icon-message ICON SUMMARY &rest ARGS) */)
+ (nargs, args)
+ int nargs;
+ Lisp_Object *args;
+{
+ struct status_icon *sicon;
+ struct gcpro gcpro1, gcpro2, gcpro3, gcpro4;
+ Lisp_Object gc_temp = Qnil;
+ Lisp_Object callback_list = Qnil;
+ Lisp_Object closed_callback = Qnil;
+ Lisp_Object callback_arg = Qnil;
+ NotifyNotification *notification;
+
+ int i;
+ int action_count = 0;
+ Lisp_Object icon = args[0];
+ Lisp_Object summary = args[1];
+
+ check_x ();
+ if ((nargs - 2) % 2 != 0)
+ error ("Invalid number of arguments");
+
+ if (! NILP (icon))
+ CHECK_LIVE_STATUS_ICON (icon);
+ CHECK_STRING (summary);
+
+ BLOCK_INPUT;
+
+ if (! notify_is_initted () && ! notify_init ("Emacs"))
+ {
+ UNBLOCK_INPUT;
+ error ("Couldn't connect to notification server");
+ }
+
+ GCPRO4 (gc_temp, callback_list, closed_callback, callback_arg);
+ gc_temp = ENCODE_UTF_8 (summary);
+
+ /* Pre-allocate the cons that will be passed to all callbacks. We
+ will fill it in later. */
+ callback_arg = Fcons (Qnil, Qnil);
+
+ if (! NILP (icon))
+ {
+ sicon = XSTATUS_ICON (icon);
+ notification
+ = notify_notification_new_with_status_icon (SDATA (gc_temp), NULL,
+ GTK_STOCK_DIALOG_INFO,
+ GET_ICON (XSTATUS_ICON (icon)));
+ }
+ else
+ {
+ icon = Vstandalone_icon;
+ sicon = XSTATUS_ICON (icon);
+ notification
+ = notify_notification_new (SDATA (gc_temp), NULL, GTK_STOCK_DIALOG_INFO,
+ NULL);
+ }
+
+ for (i = 2; i < nargs; i += 2)
+ {
+ Lisp_Object key = args[i];
+ Lisp_Object value = args[i + 1];
+ /* FIXME: still no access to: setting the icon from a pixbuf,
+ maybe hints (what are they good for?), closing the
+ notification by caller. */
+ if (EQ (key, QCbody))
+ {
+ /* Ignore things we don't understand. */
+ if (STRINGP (value))
+ {
+ gc_temp = ENCODE_UTF_8 (value);
+ g_object_set (G_OBJECT (notification), "body", SDATA (gc_temp),
+ NULL);
+ }
+ }
+ else if (EQ (key, QCtimeout))
+ {
+ gint timeout = NOTIFY_EXPIRES_DEFAULT;
+ if (EQ (value, Qt))
+ timeout = NOTIFY_EXPIRES_NEVER;
+ else if (INTEGERP (value))
+ {
+ timeout = XINT (value);
+ if (timeout <= 0)
+ timeout = NOTIFY_EXPIRES_DEFAULT;
+ }
+ notify_notification_set_timeout (notification, timeout);
+ }
+ else if (EQ (key, QCurgency))
+ {
+ NotifyUrgency urgency = NOTIFY_URGENCY_NORMAL;
+ if (EQ (value, Qlow))
+ urgency = NOTIFY_URGENCY_LOW;
+ else if (EQ (value, Qcritical))
+ urgency = NOTIFY_URGENCY_CRITICAL;
+ notify_notification_set_urgency (notification, urgency);
+ }
+ else if (EQ (key, QCicon))
+ {
+ char *icon_name = GTK_STOCK_DIALOG_INFO;
+ if (EQ (value, Qwarning))
+ icon_name = GTK_STOCK_DIALOG_WARNING;
+ else if (EQ (value, Qquestion))
+ icon_name = GTK_STOCK_DIALOG_QUESTION;
+ else if (EQ (value, Qerror))
+ icon_name = GTK_STOCK_DIALOG_ERROR;
+ g_object_set (G_OBJECT (notification), "icon-name", icon_name, NULL);
+ }
+ else if (EQ (key, QCaction))
+ {
+ /* Ignore if we don't understand the value. */
+ if (CONSP (value) && STRINGP (XCAR (value)))
+ {
+ char actionid[20];
+ Lisp_Object label = XCAR (value);
+ gc_temp = ENCODE_UTF_8 (label);
+ /* Use a plain integer for the action id, because we
+ will use it later to look up the action. */
+ sprintf (actionid, "%d", action_count);
+ ++action_count;
+ notify_notification_add_action (notification, actionid,
+ SDATA (gc_temp),
+ action_click,
+ (gpointer) callback_arg, NULL);
+
+ /* We build the list from the back, the reverse it
+ later. */
+ callback_list = Fcons (XCDR (value), callback_list);
+ }
+ }
+ else if (EQ (key, QCclosed_callback))
+ closed_callback = value;
+ }
+
+ /* Put the callback list into the right order, then put the 'closed'
+ callback at the start of the list. */
+ callback_list = Fcons (closed_callback, Fnreverse (callback_list));
+
+ XSETCAR (callback_arg, icon);
+ XSETCDR (callback_arg, callback_list);
+ sicon->notification_alist = Fcons (callback_arg, sicon->notification_alist);
+
+ /* Arrange to clean up when this notification goes away. */
+ g_signal_connect (G_OBJECT (notification), "closed",
+ G_CALLBACK (cleanup_notification),
+ (gpointer) callback_arg);
+
+ /* Ignore errors here for now. */
+ notify_notification_show (notification, NULL);
+
+ UNBLOCK_INPUT;
+
+ RETURN_UNGCPRO (Qnil);
+}
+
+#endif /* HAVE_LIBNOTIFY */
+
+DEFUN ("status-iconp", Fstatus_iconp, Sstatus_iconp,
+ 1, 1, 0, doc: /* Return non-nil if OBJECT is a status icon. */)
+ (object)
+ Lisp_Object object;
+{
+ return STATUS_ICONP (object) ? Qt : Qnil;
+}
+
+DEFUN ("status-icon-live-p", Fstatus_icon_live_p, Sstatus_icon_live_p,
+ 1, 1, 0,
+ doc: /* Return non-nil if OBJECT is a status icon which has not been deleted.*/)
+ (object)
+ Lisp_Object object;
+{
+ return ((STATUS_ICONP (object) && STATUS_ICON_LIVE_P (XSTATUS_ICON (object)))
+ ? Qt : Qnil);
+}
+
+DEFUN ("status-icon-handle-click-event", Fstatus_icon_handle_click_event,
+ Sstatus_icon_handle_click_event,
+ 1, 1, "e",
+ doc: /* Internal handler for status icon click events.
+A status icon click event is generated in response to the user clicking
+on a status icon. This handler calls the specified callback function,
+if any.*/)
+ (event)
+ Lisp_Object event;
+{
+ Lisp_Object fun;
+ struct gcpro gcpro1;
+
+ /* Just ignore garbage. */
+ if (! CONSP (event) || ! EQ (XCAR (event), Qstatus_icon_click_event)
+ || ! FUNCTIONP (XCDR (event)))
+ return Qnil;
+
+ GCPRO1 (fun);
+ fun = XCDR (event);
+ Ffuncall (1, &fun);
+ RETURN_UNGCPRO (Qnil);
+}
+
+void
+syms_of_systray ()
+{
+ struct status_icon *sicon;
+
+ Qstatus_icon_live_p = intern ("status-icon-live-p");
+ staticpro (&Qstatus_icon_live_p);
+ Qblinking = intern ("blinking");
+ staticpro (&Qblinking);
+ Qclick_callback = intern ("click-callback");
+ staticpro (&Qclick_callback);
+ QCbody = intern (":body");
+ staticpro (&QCbody);
+ QCtimeout = intern (":timeout");
+ staticpro (&QCtimeout);
+ QCurgency = intern (":urgency");
+ staticpro (&QCurgency);
+ QCicon = intern (":icon");
+ staticpro (&QCicon);
+ QCaction = intern (":action");
+ staticpro (&QCaction);
+ QCclosed_callback = intern (":closed-callback");
+ staticpro (&QCclosed_callback);
+ Qlow = intern ("low");
+ staticpro (&Qlow);
+ Qcritical = intern ("critical");
+ staticpro (&Qcritical);
+ Qwarning = intern ("warning");
+ staticpro (&Qwarning);
+ Qquestion = intern ("question");
+ staticpro (&Qquestion);
+ Qerror = intern ("error");
+ staticpro (&Qerror);
+ Qpopup_menu = intern ("popup-menu");
+ staticpro (&Qpopup_menu);
+
+ staticpro (&Vstatus_icon_list);
+ Vstatus_icon_list = Qnil;
+
+ defsubr (&Smake_status_icon);
+ defsubr (&Smodify_status_icon_parameters);
+ defsubr (&Sdelete_status_icon);
+ defsubr (&Sstatus_icon_parameters);
+ defsubr (&Sstatus_icon_list);
+#ifdef HAVE_LIBNOTIFY
+ defsubr (&Sshow_status_icon_message);
+#endif /* HAVE_LIBNOTIFY */
+ defsubr (&Sstatus_iconp);
+ defsubr (&Sstatus_icon_live_p);
+ defsubr (&Sstatus_icon_handle_click_event);
+
+ /* We make a dummy status icon so that standalone notifications have
+ a place to store their user-data. This icon never has a GUI or
+ associated native state. */
+ staticpro (&Vstandalone_icon);
+ sicon = allocate_status_icon ();
+ sicon->param_alist = Qnil;
+ sicon->notification_alist = Qnil;
+ XSETSTATUS_ICON (Vstandalone_icon, sicon);
+
+ Fprovide (intern ("systray"), Qnil);
+}
+
+#else /* HAVE_GTK_STATUS_ICON */
+
+void
+syms_of_systray ()
+{
+ /* Nothing. */
+}
+
+#endif
Index: src/systray.h
===================================================================
RCS file: src/systray.h
diff -N src/systray.h
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ src/systray.h 12 Jan 2008 02:09:10 -0000
@@ -0,0 +1,56 @@
+/* Define systray object for GNU Emacs.
+ Copyright (C) 2007, 2008 Free Software Foundation, Inc.
+
+This file is part of GNU Emacs.
+
+GNU Emacs is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GNU Emacs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Emacs; see the file COPYING. If not, write to
+the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA. */
+
+#ifndef EMACS_SYSTRAY_H
+#define EMACS_SYSTRAY_H
+
+extern Lisp_Object Qstatus_icon_live_p;
+
+struct status_icon
+{
+ EMACS_UINT size;
+ struct Lisp_Vector *next;
+
+ /* Parameter alist of this status icon. */
+ Lisp_Object param_alist;
+
+ /* List holding objects we want to keep live while a notification is
+ up. */
+ Lisp_Object notification_alist;
+
+ /* Beyond here, there should be no more Lisp_Object components. */
+
+ /* The icon widget, or NULL if it has been destroyed. Really a
+ 'GtkStatusIcon *'. */
+ void *icon;
+};
+
+#define XSTATUS_ICON(p) \
+ (eassert (GC_STATUS_ICONP (p)), (struct status_icon *) XPNTR (p))
+
+#define XSETSTATUS_ICON(a, b) (XSETPSEUDOVECTOR (a, b, PVEC_STATUS_ICON))
+
+#define STATUS_ICON_LIVE_P(s) (((s)->icon) != 0)
+
+#define CHECK_LIVE_STATUS_ICON(s) \
+ CHECK_TYPE (STATUS_ICONP (s) && STATUS_ICON_LIVE_P (XSTATUS_ICON (s)), \
+ Qstatus_icon_live_p, s)
+
+#endif /* EMACS_SYSTRAY_H */
Index: src/termhooks.h
===================================================================
RCS file: /sources/emacs/emacs/src/termhooks.h,v
retrieving revision 1.90
diff -u -r1.90 termhooks.h
--- src/termhooks.h 2 Dec 2007 16:23:40 -0000 1.90
+++ src/termhooks.h 12 Jan 2008 02:09:10 -0000
@@ -201,6 +201,8 @@
, DBUS_EVENT
#endif
+ , STATUS_ICON_CLICK_EVENT
+
#ifdef WINDOWSNT
/* Generated when an APPCOMMAND event is received, in response to
Multimedia or Internet buttons on some keyboards.
Index: src/xmenu.c
===================================================================
RCS file: /sources/emacs/emacs/src/xmenu.c,v
retrieving revision 1.324
diff -u -r1.324 xmenu.c
--- src/xmenu.c 13 Oct 2007 12:10:07 -0000 1.324
+++ src/xmenu.c 12 Jan 2008 02:09:10 -0000
@@ -1,6 +1,6 @@
/* X Communication module for terminals which understand the X protocol.
Copyright (C) 1986, 1988, 1993, 1994, 1996, 1999, 2000, 2001, 2002, 2003,
- 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
+ 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Emacs.
@@ -51,6 +51,7 @@
#include "charset.h"
#include "coding.h"
#include "sysselect.h"
+#include "systray.h"
#ifdef MSDOS
#include "msdos.h"
@@ -156,7 +157,7 @@
Lisp_Object, Lisp_Object));
static int update_frame_menubar P_ ((struct frame *));
static Lisp_Object xmenu_show P_ ((struct frame *, int, int, int, int,
- Lisp_Object, char **));
+ Lisp_Object, char **, void *, int));
static void keymap_panes P_ ((Lisp_Object *, int, int));
static void single_keymap_panes P_ ((Lisp_Object, Lisp_Object, Lisp_Object,
int, int));
@@ -780,6 +781,9 @@
corner of WINDOW. (WINDOW may be a window or a frame object.)
This controls the position of the top left of the menu as a whole.
If POSITION is t, it means to use the current mouse position.
+POSITION can also be a list (STATUS-ICON BUTTON-TIME),
+where STATUS-ICON is a status icon and BUTTON is an integer.
+In this case the menu is displayed over the indicated status icon.
MENU is a specifier for a menu. For the simplest case, MENU is a keymap.
The menu items come from key bindings that have a menu string as well as
@@ -829,6 +833,8 @@
int for_click = 0;
int specpdl_count = SPECPDL_INDEX ();
struct gcpro gcpro1;
+ void *status_icon = NULL;
+ int status_button = 0;
#ifdef HAVE_MENUS
if (! NILP (position))
@@ -843,6 +849,19 @@
{
get_current_pos_p = 1;
}
+ else if (CONSP (position) && STATUS_ICONP (XCAR (position)))
+ {
+ Lisp_Object icon = XCAR (position);
+
+ CHECK_LIVE_STATUS_ICON (icon);
+ status_icon = XSTATUS_ICON (icon)->icon;
+ status_button = XINT (Fcar (Fcdr (position)));
+ for_click = 1;
+ /* Placate later code. */
+ XSETINT (x, 0);
+ XSETINT (y, 0);
+ window = selected_window;
+ }
else
{
tem = Fcar (position);
@@ -1029,7 +1048,8 @@
BLOCK_INPUT;
selection = xmenu_show (f, xpos, ypos, for_click,
- keymaps, title, &error_name);
+ keymaps, title, &error_name,
+ status_icon, status_button);
UNBLOCK_INPUT;
discard_menu_items ();
@@ -2664,18 +2684,22 @@
menu pops down.
menu_item_selection will be set to the selection. */
static void
-create_and_show_popup_menu (f, first_wv, x, y, for_click)
+create_and_show_popup_menu (f, first_wv, x, y, for_click,
+ status_icon, status_button)
FRAME_PTR f;
widget_value *first_wv;
int x;
int y;
int for_click;
+ void *status_icon;
+ int status_button;
{
int i;
GtkWidget *menu;
GtkMenuPositionFunc pos_func = 0; /* Pop up at pointer. */
struct next_popup_x_y popup_x_y;
int specpdl_count = SPECPDL_INDEX ();
+ gpointer user_data = &popup_x_y;
if (! FRAME_X_P (f))
abort ();
@@ -2702,6 +2726,14 @@
i = 0; /* gtk_menu_popup needs this to be 0 for a non-button popup. */
}
+#ifdef HAVE_GTK_STATUS_ICON
+ else if (status_icon)
+ {
+ user_data = status_icon;
+ i = status_button;
+ pos_func = gtk_status_icon_position_menu;
+ }
+#endif /* HAVE_GTK_STATUS_ICON */
else
{
for (i = 0; i < 5; i++)
@@ -2711,7 +2743,8 @@
/* Display the menu. */
gtk_widget_show_all (menu);
- gtk_menu_popup (GTK_MENU (menu), 0, 0, pos_func, &popup_x_y, i, 0);
+ gtk_menu_popup (GTK_MENU (menu), 0, 0, pos_func, user_data, i,
+ gtk_get_current_event_time ());
record_unwind_protect (pop_down_menu, make_save_value (menu, 0));
@@ -2772,14 +2805,18 @@
/* Pop up the menu for frame F defined by FIRST_WV at X/Y and loop until the
menu pops down.
- menu_item_selection will be set to the selection. */
+ menu_item_selection will be set to the selection.
+ The status-icon-related arguments are ignore in this implementation. */
static void
-create_and_show_popup_menu (f, first_wv, x, y, for_click)
+create_and_show_popup_menu (f, first_wv, x, y, for_click,
+ status_icon, status_button)
FRAME_PTR f;
widget_value *first_wv;
int x;
int y;
int for_click;
+ void *status_icon;
+ int status_button;
{
int i;
Arg av[2];
@@ -2848,7 +2885,8 @@
#endif /* not USE_GTK */
static Lisp_Object
-xmenu_show (f, x, y, for_click, keymaps, title, error)
+xmenu_show (f, x, y, for_click, keymaps, title, error,
+ status_icon, status_button)
FRAME_PTR f;
int x;
int y;
@@ -2856,6 +2894,8 @@
int keymaps;
Lisp_Object title;
char **error;
+ void *status_icon;
+ int status_button;
{
int i;
widget_value *wv, *save_wv = 0, *first_wv = 0, *prev_wv = 0;
@@ -3062,7 +3102,8 @@
menu_item_selection = 0;
/* Actually create and show the menu until popped down. */
- create_and_show_popup_menu (f, first_wv, x, y, for_click);
+ create_and_show_popup_menu (f, first_wv, x, y, for_click,
+ status_icon, status_button);
/* Free the widget_value objects we used to specify the contents. */
free_menubar_widget_value_tree (first_wv);
@@ -3523,6 +3564,10 @@
int keymaps;
Lisp_Object title;
char **error;
+ /* Note that status-icon-related arguments are ignored in this
+ implementation */
+ void *status_icon;
+ int status_button;
{
Window root;
XMenu *menu;
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 1:38 ` Tom Tromey
@ 2008-01-12 8:45 ` Ulrich Mueller
2008-01-12 17:45 ` Tom Tromey
2008-01-14 1:03 ` Michael Olson
1 sibling, 1 reply; 55+ messages in thread
From: Ulrich Mueller @ 2008-01-12 8:45 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs, emacs-devel
>>>>> On Fri, 11 Jan 2008, Tom Tromey wrote:
> Here is the updated patch.
Would it be possible to add an explicit option --with{,out}-libnotify
to configure, so that this feature can be switched on or off at
compile time? I think the same arguments as for D-Bus support [1]
apply here.
Ulrich
[1] See thread "D-Bus integration into Emacs" from beginning of
December, e.g. the following message:
<http://lists.gnu.org/archive/html/emacs-devel/2007-12/msg00189.html>
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 8:45 ` Ulrich Mueller
@ 2008-01-12 17:45 ` Tom Tromey
2008-01-14 2:01 ` Richard Stallman
0 siblings, 1 reply; 55+ messages in thread
From: Tom Tromey @ 2008-01-12 17:45 UTC (permalink / raw)
To: Ulrich Mueller; +Cc: emacs, emacs-devel
>>>>> "Ulrich" == Ulrich Mueller <ulm@gentoo.org> writes:
Ulrich> Would it be possible to add an explicit option
Ulrich> --with{,out}-libnotify to configure, so that this feature can
Ulrich> be switched on or off at compile time? I think the same
Ulrich> arguments as for D-Bus support [1] apply here.
Sure. I wrote the needed change, but I think I will wait for other
feedback on the patch rather than posting an updated version
immediately.
Ulrich> [1] See thread "D-Bus integration into Emacs" from beginning of
Ulrich> December, e.g. the following message:
Ulrich> <http://lists.gnu.org/archive/html/emacs-devel/2007-12/msg00189.html>
I read this thread. RMS says that the default should be to have dbus
enabled, here:
http://lists.gnu.org/archive/html/emacs-devel/2007-12/msg00308.html
But in configure.in, the default is to have dbus disabled:
EMACS_ARG_Y([dbus],[use D-Bus])
Could someone change that to EMACS_ARG_N?
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 17:45 ` Tom Tromey
@ 2008-01-14 2:01 ` Richard Stallman
2008-01-14 1:35 ` Tom Tromey
2008-01-14 1:41 ` Tom Tromey
0 siblings, 2 replies; 55+ messages in thread
From: Richard Stallman @ 2008-01-14 2:01 UTC (permalink / raw)
To: Tom Tromey; +Cc: ulm, emacs-devel, emacs
But in configure.in, the default is to have dbus disabled:
EMACS_ARG_Y([dbus],[use D-Bus])
Could someone change that to EMACS_ARG_N?
Would someone please add comments explaining the meaning of
EMACS_ARG_Y and EMACS_ARG_N? I cannot understand their definitions,
and I am not sure what they do.
Perhaps their names should be changed; it would be good to use
names whose meanings are self-evident.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 2:01 ` Richard Stallman
@ 2008-01-14 1:35 ` Tom Tromey
2008-01-14 17:26 ` Richard Stallman
2008-01-14 1:41 ` Tom Tromey
1 sibling, 1 reply; 55+ messages in thread
From: Tom Tromey @ 2008-01-14 1:35 UTC (permalink / raw)
To: rms; +Cc: ulm, emacs-devel, emacs
rms> Would someone please add comments explaining the meaning of
rms> EMACS_ARG_Y and EMACS_ARG_N? I cannot understand their definitions,
rms> and I am not sure what they do.
Here's my understanding of them:
dnl EMACS_ARG_Y(NAME, HELP-STRING, [IF-GIVEN], [IF-NOT-GIVEN])
dnl Define a new --with option which is disabled by default.
dnl NAME is the base name of the option.
dnl HELP-STRING is the help text for the option. It should not include
dnl the option name; this macro provides that to AS_HELP_STRING.
dnl IF-GIVEN and IF-NOT-GIVEN are the same as for AC_ARG_WITH.
AC_DEFUN([EMACS_ARG_Y],[dnl
AC_ARG_WITH([$1],[AS_HELP_STRING([--with-$1],[$2])],[$3],[$4])dnl
])dnl
dnl EMACS_ARG_N(NAME, HELP-STRING, [IF-YES], [IF-NO])
dnl Define a new --with option which is enabled by default.
dnl Arguments are the same as EMACS_ARG_Y.
AC_DEFUN([EMACS_ARG_N],[dnl
AC_ARG_WITH([$1],[AS_HELP_STRING([--without-$1],[$2])],[$3],[$4])dnl
])dnl
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 1:35 ` Tom Tromey
@ 2008-01-14 17:26 ` Richard Stallman
2008-01-19 5:18 ` Tom Tromey
0 siblings, 1 reply; 55+ messages in thread
From: Richard Stallman @ 2008-01-14 17:26 UTC (permalink / raw)
To: Tom Tromey; +Cc: ulm, emacs-devel, emacs
dnl EMACS_ARG_Y(NAME, HELP-STRING, [IF-GIVEN], [IF-NOT-GIVEN])
dnl Define a new --with option which is disabled by default.
If that is the meaning, the name is counterintuitive.
Let's rename EMACS_ARG_Y to OPTION_DEFAULT_OFF,
and rename EMACS_ARG_N to OPTION_DEFAULT_ON.
dnl NAME is the base name of the option.
dnl HELP-STRING is the help text for the option. It should not include
dnl the option name; this macro provides that to AS_HELP_STRING.
dnl IF-GIVEN and IF-NOT-GIVEN are the same as for AC_ARG_WITH.
That text is clear.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 17:26 ` Richard Stallman
@ 2008-01-19 5:18 ` Tom Tromey
2008-01-20 6:14 ` Richard Stallman
0 siblings, 1 reply; 55+ messages in thread
From: Tom Tromey @ 2008-01-19 5:18 UTC (permalink / raw)
To: rms; +Cc: emacs-devel
> dnl EMACS_ARG_Y(NAME, HELP-STRING, [IF-GIVEN], [IF-NOT-GIVEN])
> dnl Define a new --with option which is disabled by default.
rms> If that is the meaning, the name is counterintuitive.
rms> Let's rename EMACS_ARG_Y to OPTION_DEFAULT_OFF,
rms> and rename EMACS_ARG_N to OPTION_DEFAULT_ON.
Here's a patch that renames these and cleans up the uses. With the
new macros there are no 3rd/4th arguments, just the result variable
which can be tested separately.
This patch also changes --with-pkg-config-prog not to use
OPTION_DEFAULT_ON, since it is not really a binary option, but rather
a path. And, while I was at it, I cleaned up the code implementing
this option so that the work is done in a single place.
Tom
ChangeLog:
2008-01-19 Tom Tromey <tromey@redhat.com>
* configure.in (OPTION_DEFAULT_ON, OPTION_DEFAULT_OFF): New
macros.
(EMACS_ARG_Y, EMACS_ARG_N): Remove.
Update all users.
Index: configure.in
===================================================================
RCS file: /sources/emacs/emacs/configure.in,v
retrieving revision 1.495
diff -u -r1.495 configure.in
--- configure.in 17 Jan 2008 11:41:10 -0000 1.495
+++ configure.in 19 Jan 2008 05:50:48 -0000
@@ -43,47 +43,63 @@
gameuser=games
-dnl Autoconf is so much less fun under VMS, maybe
-dnl because everything is less fun under VMS. --ttn
-AC_DEFUN([EMACS_ARG_Y],[dnl
-AC_ARG_WITH([$1],[AS_HELP_STRING([--with-$1],[$2])],[$3],[$4])dnl
+dnl OPTION_DEFAULT_OFF(NAME, HELP-STRING)
+dnl Create a new --with option that defaults to being disabled.
+dnl NAME is the base name of the option. The shell variable with_NAME
+dnl will be set to either the user's value (if the option is
+dnl specified; 'yes' for a plain --with-NAME) or to 'no' (if the
+dnl option is not specified). Note that the shell variable name is
+dnl constructed as autoconf does, by replacing non-alphanumeric
+dnl characters with "_".
+dnl HELP-STRING is the help text for the option.
+AC_DEFUN([OPTION_DEFAULT_OFF], [dnl
+ AC_ARG_WITH([$1],[AS_HELP_STRING([--with-$1],[$2])],[],[dnl
+ m4_bpatsubst([with_$1], [[^0-9a-z]], [_])=no])dnl
])dnl
-AC_DEFUN([EMACS_ARG_N],[dnl
-AC_ARG_WITH([$1],[AS_HELP_STRING([--without-$1],[$2])],[$3],[$4])dnl
+
+dnl OPTION_DEFAULT_ON(NAME, HELP-STRING)
+dnl Create a new --with option that defaults to being enabled. NAME
+dnl is the base name of the option. The shell variable with_NAME
+dnl will be set either to 'no' (for a plain --without-NAME) or to
+dnl 'yes' (if the option is not specified). Note that the shell
+dnl variable name is constructed as autoconf does, by replacing
+dnl non-alphanumeric characters with "_".
+dnl HELP-STRING is the help text for the option.
+AC_DEFUN([OPTION_DEFAULT_ON], [dnl
+ AC_ARG_WITH([$1],[AS_HELP_STRING([--without-$1],[$2])],[],[dnl
+ m4_bpatsubst([with_$1], [[^0-9a-z]], [_])=yes])dnl
])dnl
-EMACS_ARG_N([gcc],[don't use GCC to compile Emacs if GCC is found])
+OPTION_DEFAULT_ON([gcc],[don't use GCC to compile Emacs if GCC is found])
-EMACS_ARG_N([pop],[don't support POP mail retrieval with movemail],
-[if test "$withval" = yes; then
+OPTION_DEFAULT_ON([pop],[don't support POP mail retrieval with movemail])
+if test "$with_pop" = yes; then
AC_DEFINE(MAIL_USE_POP)
-else :
-fi],
-AC_DEFINE(MAIL_USE_POP))
+fi
AH_TEMPLATE(MAIL_USE_POP, [Define to support POP mail retrieval.])dnl
-EMACS_ARG_Y([kerberos],[support Kerberos-authenticated POP],
-[if test "$withval" = yes; then
+OPTION_DEFAULT_OFF([kerberos],[support Kerberos-authenticated POP])
+if test "$with_kerberos" = yes; then
AC_DEFINE(KERBEROS)
-fi])
+fi
AH_TEMPLATE(KERBEROS,
[Define to support Kerberos-authenticated POP mail retrieval.])dnl
-EMACS_ARG_Y([kerberos5],[support Kerberos version 5 authenticated POP],
-[if test "${with_kerberos5+set}" = set; then
- if test "${with_kerberos+set}" != set; then
+OPTION_DEFAULT_OFF([kerberos5],[support Kerberos version 5 authenticated POP])
+if test "${with_kerberos5}" = yes; then
+ if test "${with_kerberos}" != yes; then
with_kerberos=yes
AC_DEFINE(KERBEROS)
fi
AC_DEFINE(KERBEROS5, 1, [Define to use Kerberos 5 instead of Kerberos 4.])
-fi])
+fi
-EMACS_ARG_Y([hesiod],[support Hesiod to get the POP server host],
-[if test "$withval" = yes; then
+OPTION_DEFAULT_OFF([hesiod],[support Hesiod to get the POP server host])
+if test "$with_hesiod" = yes; then
AC_DEFINE(HESIOD, 1, [Define to support using a Hesiod database to find the POP server.])
-fi])
+fi
-EMACS_ARG_N([sound],[don't compile with sound support])
+OPTION_DEFAULT_ON([sound],[don't compile with sound support])
dnl This should be the last --with option, because --with-x is
dnl added later on when we find the path of X, and it's best to
@@ -106,20 +122,28 @@
with_x_toolkit=$val
])
-EMACS_ARG_Y([xpm],[use -lXpm for displaying XPM images])
-EMACS_ARG_Y([jpeg],[use -ljpeg for displaying JPEG images])
-EMACS_ARG_Y([tiff],[use -ltiff for displaying TIFF images])
-EMACS_ARG_Y([gif],[use -lgif (or -lungif) for displaying GIF images])
-EMACS_ARG_Y([png],[use -lpng for displaying PNG images])
-EMACS_ARG_Y([gpm],[use -lgpm for mouse support on a GNU/Linux console])
-EMACS_ARG_Y([rsvg],[use -lrsvg-2 for displaying SVG images])
-EMACS_ARG_Y([gtk],[use GTK (same as --with-x-toolkit=gtk)])
-EMACS_ARG_Y([pkg-config-prog],[Path to pkg-config for finding GTK and librsvg])
-EMACS_ARG_N([toolkit-scroll-bars],[don't use Motif or Xaw3d scroll bars])
-EMACS_ARG_N([xaw3d],[don't use Xaw3d])
-EMACS_ARG_N([xim],[don't use X11 XIM])
-EMACS_ARG_N([carbon],[don't use Carbon GUI on Mac OS X])
-EMACS_ARG_Y([dbus],[use D-Bus])
+OPTION_DEFAULT_OFF([xpm],[use -lXpm for displaying XPM images])
+OPTION_DEFAULT_OFF([jpeg],[use -ljpeg for displaying JPEG images])
+OPTION_DEFAULT_OFF([tiff],[use -ltiff for displaying TIFF images])
+OPTION_DEFAULT_OFF([gif],[use -lgif (or -lungif) for displaying GIF images])
+OPTION_DEFAULT_OFF([png],[use -lpng for displaying PNG images])
+OPTION_DEFAULT_OFF([gpm],[use -lgpm for mouse support on a GNU/Linux console])
+OPTION_DEFAULT_OFF([rsvg],[use -lrsvg-2 for displaying SVG images])
+OPTION_DEFAULT_OFF([gtk],[use GTK (same as --with-x-toolkit=gtk)])
+OPTION_DEFAULT_ON([toolkit-scroll-bars],[don't use Motif or Xaw3d scroll bars])
+OPTION_DEFAULT_ON([xaw3d],[don't use Xaw3d])
+OPTION_DEFAULT_ON([xim],[don't use X11 XIM])
+OPTION_DEFAULT_ON([carbon],[don't use Carbon GUI on Mac OS X])
+OPTION_DEFAULT_ON([dbus],[use D-Bus])
+
+AC_ARG_WITH([pkg-config-prog],dnl
+ [AS_HELP_STRING([--with-pkg-config-prog=PATH],
+ [Path to pkg-config for finding GTK and librsvg])])
+if test "X${with_pkg_config_prog}" != X; then
+ if test "${with_pkg_config_prog}" != yes; then
+ PKG_CONFIG="${with_pkg_config_prog}"
+ fi
+fi
AC_ARG_ENABLE(carbon-app,
[AS_HELP_STRING([--enable-carbon-app@<:@=DIR@:>@],
@@ -1089,9 +1113,7 @@
AC_DEFUN([PKG_CHECK_MODULES], [
succeeded=no
- if test -z "$PKG_CONFIG"; then
- AC_PATH_PROG(PKG_CONFIG, pkg-config, no)
- fi
+ AC_PATH_PROG(PKG_CONFIG, pkg-config, no)
if test "$PKG_CONFIG" = "no" ; then
ifelse([$4], , [AC_MSG_ERROR([
@@ -1147,10 +1169,6 @@
ALSA_REQUIRED=1.0.0
ALSA_MODULES="alsa >= $ALSA_REQUIRED"
- dnl Check if --with-pkg-config-prog has been given.
- if test "X${with_pkg_config_prog}" != X; then
- PKG_CONFIG="${with_pkg_config_prog}"
- fi
PKG_CHECK_MODULES(ALSA, $ALSA_MODULES, HAVE_ALSA=yes, HAVE_ALSA=no)
if test $HAVE_ALSA = yes; then
SAVE_CFLAGS="$CFLAGS"
@@ -1639,11 +1657,6 @@
HAVE_RSVG=no
if test "${HAVE_X11}" = "yes" || test "${HAVE_CARBON}" = "yes"; then
if test "${with_rsvg}" != "no"; then
- dnl Check if `--with-pkg-config-prog' has been given.
- if test "X${with_pkg_config_prog}" != X; then
- PKG_CONFIG="${with_pkg_config_prog}"
- fi
-
RSVG_REQUIRED=2.0.0
RSVG_MODULE="librsvg-2.0 >= $RSVG_REQUIRED"
@@ -1674,10 +1687,6 @@
GTK_REQUIRED=2.6
GTK_MODULES="gtk+-2.0 >= $GTK_REQUIRED glib-2.0 >= $GLIB_REQUIRED"
- dnl Check if --with-pkg-config-prog has been given.
- if test "X${with_pkg_config_prog}" != X; then
- PKG_CONFIG="${with_pkg_config_prog}"
- fi
dnl Checks for libraries.
PKG_CHECK_MODULES(GTK, $GTK_MODULES, pkg_check_gtk=yes, pkg_check_gtk=no)
if test "$pkg_check_gtk" = "no" && test "$USE_X_TOOLKIT" != "maybe"; then
@@ -1765,21 +1774,20 @@
dnl other platforms. Support for higher D-Bus versions but 1.0 is
dnl also not configured.
HAVE_DBUS=no
+AC_MSG_CHECKING([maybe dbus])
if test "${with_dbus}" = "yes"; then
PKG_CHECK_MODULES(DBUS, dbus-1 >= 1.0, HAVE_DBUS=yes, HAVE_DBUS=no)
if test "$HAVE_DBUS" = yes; then
AC_DEFINE(HAVE_DBUS, 1, [Define to 1 if using D-Bus.])
fi
+ AC_MSG_RESULT($HAVE_DBUS)
+else
+ AC_MSG_RESULT(no bub)
fi
### Link with -lXft if available to work around a bug.
HAVE_XFT=maybe
if test "${HAVE_GTK}" = "yes"; then
- dnl Check if --with-pkg-config-prog has been given.
- if test "X${with_pkg_config_prog}" != X; then
- PKG_CONFIG="${with_pkg_config_prog}"
- fi
-
PKG_CHECK_MODULES(XFT, xft >= 0.13.0, , HAVE_XFT=no)
if test "$HAVE_XFT" != no; then
OLD_CFLAGS="$CPPFLAGS"
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-19 5:18 ` Tom Tromey
@ 2008-01-20 6:14 ` Richard Stallman
2008-01-23 4:00 ` Michael Olson
0 siblings, 1 reply; 55+ messages in thread
From: Richard Stallman @ 2008-01-20 6:14 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs-devel
Thanks. Would someone please install this?
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-20 6:14 ` Richard Stallman
@ 2008-01-23 4:00 ` Michael Olson
0 siblings, 0 replies; 55+ messages in thread
From: Michael Olson @ 2008-01-23 4:00 UTC (permalink / raw)
To: emacs-devel
[-- Attachment #1.1: Type: text/plain, Size: 409 bytes --]
Richard Stallman <rms@gnu.org> writes:
> Thanks. Would someone please install this?
Installed. I've also regenerated the configure script.
--
Michael Olson -- FSF Associate Member #652 |
http://mwolson.org/ -- Jabber: mwolson_at_hcoop.net | /` |\ | | |
Programmer -- Hobbies: Lisp, HCoop | |_] | \| |_|
Projects: Emacs, Muse, ERC, EMMS, ErBot, DVC, Planner |
[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]
[-- Attachment #2: Type: text/plain, Size: 142 bytes --]
_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 2:01 ` Richard Stallman
2008-01-14 1:35 ` Tom Tromey
@ 2008-01-14 1:41 ` Tom Tromey
1 sibling, 0 replies; 55+ messages in thread
From: Tom Tromey @ 2008-01-14 1:41 UTC (permalink / raw)
To: rms; +Cc: ulm, emacs-devel, emacs
Tom> But in configure.in, the default is to have dbus disabled:
Tom> EMACS_ARG_Y([dbus],[use D-Bus])
Tom> Could someone change that to EMACS_ARG_N?
... and of course, changing this is not sufficient. The code also has
to change to set the default properly.
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 1:38 ` Tom Tromey
2008-01-12 8:45 ` Ulrich Mueller
@ 2008-01-14 1:03 ` Michael Olson
2008-01-14 1:01 ` Tom Tromey
1 sibling, 1 reply; 55+ messages in thread
From: Michael Olson @ 2008-01-14 1:03 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs-devel
[-- Attachment #1.1: Type: text/plain, Size: 3169 bytes --]
Tom Tromey <tromey@redhat.com> writes:
>>>>>> "Dan" == Dan Nicolaescu <dann@ics.uci.edu> writes:
>
> Dan> Please no K&R in new code.
>
> Here is the updated patch.
I tried it out today. I can see from config.log and src/config.h that
libnotify support is being detected correctly, and that
gtk_status_icon_new is present. The problem is that the
`show-status-icon-message' acts somewhat erratically if the icon is
created and then an immediate attempt is made to display a popup
message. I suspect a race condition.
Once this issue is fixed, I'd love to see this feature hit Emacs CVS.
It will be particularly nice for ERC once your erc-status.el is adapted
to work with the new code.
Here's my test case, with instructions in comments. Let me know if you
need any additional info.
;; Save this to test.el.
;;
;; Recipe:
;;
;; emacs -Q -l test.el
;; M-x my-icon-all-in-one RET
;; => Creates icon in status area
;; => "Couldn't connect to notification server"
;; => libnotify-Message: Unable to get session bus: Failed to
;; execute dbus-launch to autolaunch D-Bus session
;; => Does not show "foo" popup
;; M-x my-icon-remove-all RET
;;
;; C-x 5 2
;; C-x 5 o
;; M-x my-icon-all-in-one RET
;; => Creates icon and shows "foo" popup, but off the screen, or
;; sometimes neat the position of the cursor, or (once) crashes
;; Emacs.
;; M-x my-icon-remove-all RET
;;
;; M-x my-icon-create RET
;; => Creates icon in status area.
;; M-x my-icon-display RET
;; => Always displays popup successfully.
;; C-x C-c
;;
;; emacs -Q -l test.el
;; M-x my-icon-two-separate RET
;; => Creates icon and shows "foo" popup, but off the screen, or
;; near the position of the cursor
;; M-x my-icon-remove RET
;; M-x my-icon-create RET
;; => Creates icon in status area.
;; M-x my-icon-display RET
;; => Always displays popup successfully.
(defvar my-status-icon nil)
(defun my-icon-create ()
(interactive)
(setq my-status-icon
(make-status-icon '((help-echo . "bar")
(icon-name . "/usr/local/share/emacs/23.0.50/etc/images/icons/hicolor/48x48/apps/emacs.png")))))
(defun my-icon-test ()
(interactive)
(show-status-icon-message my-status-icon "foo"))
(defun my-icon-remove ()
(interactive)
(delete-status-icon my-status-icon))
(defun my-icon-remove-all ()
(interactive)
(dolist (icon (status-icon-list))
(delete-status-icon icon)))
(defun my-icon-all-in-one ()
(interactive)
(show-status-icon-message
(setq my-status-icon
(make-status-icon '((help-echo . "bar")
(icon-name . "/usr/local/share/emacs/23.0.50/etc/images/icons/hicolor/48x48/apps/emacs.png"))))
"foo"))
(defun my-icon-two-separate ()
(interactive)
(my-icon-create)
(my-icon-test))
--
Michael Olson -- FSF Associate Member #652 |
http://mwolson.org/ -- Jabber: mwolson_at_hcoop.net | /` |\ | | |
Programmer -- Hobbies: Lisp, HCoop | |_] | \| |_|
Projects: Emacs, Muse, ERC, EMMS, ErBot, DVC, Planner |
[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]
[-- Attachment #2: Type: text/plain, Size: 142 bytes --]
_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 1:03 ` Michael Olson
@ 2008-01-14 1:01 ` Tom Tromey
2008-01-14 7:03 ` Jan Djärv
2008-01-15 6:01 ` Michael Olson
0 siblings, 2 replies; 55+ messages in thread
From: Tom Tromey @ 2008-01-14 1:01 UTC (permalink / raw)
To: Michael Olson; +Cc: emacs-devel
>>>>> "Michael" == Michael Olson <mwolson@gnu.org> writes:
Michael> [ ... excellent bug report ]
Michael> I suspect a race condition.
Yes, I think you are right. I am not sure but maybe the answer is to
call gtk_main_iteration to flush the gtk event queue. I am testing
that. If that doesn't work, I guess I'll have to learn more about the
Emacs event stuff... scary. I may need help.
Michael> Once this issue is fixed, I'd love to see this feature hit
Michael> Emacs CVS. It will be particularly nice for ERC once your
Michael> erc-status.el is adapted to work with the new code.
I've appended my updated erc-status.el. I haven't tried it at all, I
just rewrote it to the new API.
Michael> ;; => "Couldn't connect to notification server"
Michael> ;; => libnotify-Message: Unable to get session bus: Failed to
Michael> ;; execute dbus-launch to autolaunch D-Bus session
Michael> ;; => Does not show "foo" popup
Are you running dbus? And do you have a system tray in your panel?
I'm not an expert here, either, but I would only expect these errors
if one of those answers is "no".
FWIW I do see the racy behavior you pointed out, but I don't see this
error. I'm running a rather standard Gnome desktop -- whatever is in
Fedora Core 6. My panel does have a system tray.
The notification stuff is cross-desktop, so it should also work if you
are a KDE user. I have not tested that with the Emacs-based code, but
I have tried out other system-tray-using Gnome applications on KDE.
Thanks for taking the time to try this out.
Tom
;;; erc-status.el --- notification area support for ERC
;; Copyright (C) 2007, 2008 Tom Tromey <tromey@redhat.com>
;; Author: Tom Tromey <tromey@redhat.com>
;; Version: 0.2
;; Keywords: comm
;; This file is not (yet) part of GNU Emacs.
;; However, it is distributed under the same license.
;; GNU Emacs is free software; you can redistribute it and/or modify
;; it under the terms of the GNU General Public License as published by
;; the Free Software Foundation; either version 3, or (at your option)
;; any later version.
;; GNU Emacs is distributed in the hope that it will be useful,
;; but WITHOUT ANY WARRANTY; without even the implied warranty of
;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
;; GNU General Public License for more details.
;; You should have received a copy of the GNU General Public License
;; along with GNU Emacs; see the file COPYING. If not, write to the
;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
;; Boston, MA 02110-1301, USA.
;;; Commentary:
;; This provides nice support for the notification area to ERC. In
;; particular it:
;; * Will blink the icon when you get a private message or are paged
;; in a channel.
;; * Left-click on the blinking icon will show the appropriate channel
;; buffer in some frame (which is then raised). If there are
;; multiple pages at once, it will show one and you can click again
;; to go to the next one.
;; * Will pop up notification bubbles when you connect to or
;; disconnect from a server.
;; This is regular erc module named 'status'; you can enable it as you
;; would any other module.
;; TO DO:
;; - make tool tip show some kind of real status ...?
;; - use a nicer icon
;; - add a menu
;; - integrate with auto-query a bit better
;; - let left click use specified frame or make a new frame?
;; - when last server connection is closed, remove icon
(require 'systray)
(require 'erc)
;; The status icon object.
(defvar erc-status-status-icon nil)
;; List of ERC buffers that caused the status icon to blink.
(defvar erc-status-buffer-list nil)
(defun erc-status-remove-buffer (buffer)
;; If the list is not empty, and removing an element makes the list
;; empty, stop blinking.
(and erc-status-buffer-list
(not (setq erc-status-buffer-list (delq buffer erc-status-buffer-list)))
(modify-status-icon erc-status-status-icon '((blinking . nil)))))
(defun erc-status-add-buffer (buffer)
(and (not (erc-buffer-visible buffer))
(progn
(modify-status-icon erc-status-status-icon '((blinking . t)))
(or (memq buffer erc-status-buffer-list)
(setq erc-status-buffer-list (cons buffer
erc-status-buffer-list))))))
(defun erc-status-match-hook (match-type nick message)
;; Look for user's nick and make the icon blink.
(if (eq match-type 'current-nick)
(erc-status-add-buffer (current-buffer))))
(defun erc-status-buffer-killed ()
;; If one of our buffers was killed, remove it.
(erc-status-remove-buffer (current-buffer)))
(defun erc-status-window-configuration-changed ()
(let ((new-list)
(iter erc-status-buffer-list))
(while iter
(or (erc-buffer-visible (car iter))
(setq new-list (cons (car iter) new-list)))
(setq iter (cdr iter)))
(or (setq erc-status-buffer-list new-list)
(modify-status-icon erc-status-status-icon '((blinking . nil))))))
(defun erc-status-disconnected (nick ip reason)
;; FIXME: should mention the server from which we were disconnected.
;; FIXME: add a :action to reconnect.
(show-status-icon-message erc-status-status-icon
(concat "Disconnected: " reason)))
(defun erc-status-after-connect (server nick)
(show-status-icon-message erc-status-status-icon
(concat "Connected to " server " as " nick)))
(defun erc-status-select-first-buffer ()
"Switch to the first ERC buffer requiring your attention.
If there is no such buffer, do nothing."
(when erc-status-buffer-list
(switch-to-buffer (car erc-status-buffer-list))
(raise-frame)))
\f
;; From: http://www.emacswiki.org/cgi-bin/wiki/ErcPageMe
;; Then modified to suit.
(defun erc-status-PRIVMSG (proc parsed)
(let* ((nick (car (erc-parse-user (erc-response.sender parsed))))
(target (car (erc-response.command-args parsed)))
(msg (erc-response.contents parsed))
(query (if (not erc-query-on-unjoined-chan-privmsg)
nick
(if (erc-current-nick-p target)
nick
target))))
(when (and (erc-current-nick-p target)
(not (erc-is-message-ctcp-and-not-action-p msg)))
;; Note: assumes you are using auto-query.
(erc-status-add-buffer (erc-get-buffer query proc))))
;; Always return nil.
nil)
\f
(define-erc-module status nil
"Notification area support for ERC."
;; Enable.
((unless erc-status-status-icon
(setq erc-status-status-icon
(make-status-icon '((help-echo . "ERC - IRC client for Emacs")
(click-callback . #'erc-status-select-first-buffer)))))
(add-hook 'erc-text-matched-hook 'erc-status-match-hook)
(add-hook 'kill-buffer-hook 'erc-status-buffer-killed)
(add-hook 'window-configuration-change-hook
'erc-status-window-configuration-changed)
(add-hook 'erc-after-connect 'erc-status-after-connect)
(add-hook 'erc-disconnected-hook 'erc-status-disconnected)
;; FIXME: Must come *after* erc-auto-query. Some sort of
;; auto-query hook or the like would be good here.
(add-hook 'erc-server-PRIVMSG-functions 'erc-status-PRIVMSG t))
;; Disable.
((when erc-status-status-icon
(delete-status-icon erc-status-status-icon)
(setq erc-status-status-icon nil))
(remove-hook 'erc-text-matched-hook 'erc-status-match-hook)
(remove-hook 'kill-buffer-hook 'erc-status-buffer-killed)
(remove-hook 'window-configuration-change-hook
'erc-status-window-configuration-changed)
(remove-hook 'erc-after-connect 'erc-status-after-connect)
(remove-hook 'erc-disconnected-hook 'erc-status-disconnected)
(remove-hook 'erc-server-PRIVMSG-functions 'erc-status-PRIVMSG)))
;;; erc-status.el ends here
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 1:01 ` Tom Tromey
@ 2008-01-14 7:03 ` Jan Djärv
2008-01-15 6:01 ` Michael Olson
1 sibling, 0 replies; 55+ messages in thread
From: Jan Djärv @ 2008-01-14 7:03 UTC (permalink / raw)
To: Tom Tromey; +Cc: Michael Olson, emacs-devel
Tom Tromey skrev:
>>>>>> "Michael" == Michael Olson <mwolson@gnu.org> writes:
>
> Michael> [ ... excellent bug report ]
> Michael> I suspect a race condition.
>
> Yes, I think you are right. I am not sure but maybe the answer is to
> call gtk_main_iteration to flush the gtk event queue. I am testing
> that. If that doesn't work, I guess I'll have to learn more about the
> Emacs event stuff... scary. I may need help.
gdk_window_process_all_updates flushes the event queue. But if there are
timers involved, you must do something more complicated as the Emacs main loop
does not do Gtk+/Glib timers. Search for
xg_process_timeouts/xg_start_timer/xg_stop_timer in gtkutil.c in that case and
something similar.
Jan D.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 1:01 ` Tom Tromey
2008-01-14 7:03 ` Jan Djärv
@ 2008-01-15 6:01 ` Michael Olson
2008-01-16 1:10 ` Tom Tromey
1 sibling, 1 reply; 55+ messages in thread
From: Michael Olson @ 2008-01-15 6:01 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs-devel
[-- Attachment #1.1: Type: text/plain, Size: 1408 bytes --]
Tom Tromey <tromey@redhat.com> writes:
> Michael> Once this issue is fixed, I'd love to see this feature hit
> Michael> Emacs CVS. It will be particularly nice for ERC once your
> Michael> erc-status.el is adapted to work with the new code.
>
> I've appended my updated erc-status.el. I haven't tried it at all, I
> just rewrote it to the new API.
I've checked this into a new branch of ERC called "erc-status" (at
http://git.sv.gnu.org/gitweb/?p=erc.git) and have done a bit of extra
hacking on it.
> Michael> ;; => "Couldn't connect to notification server"
> Michael> ;; => libnotify-Message: Unable to get session bus: Failed
> Michael> ;; to execute dbus-launch to autolaunch D-Bus session
> Michael> ;; => Does not show "foo" popup
>
> Are you running dbus?
Yep.
> And do you have a system tray in your panel?
Yep.
> FWIW I do see the racy behavior you pointed out, but I don't see this
> error. I'm running a rather standard Gnome desktop -- whatever is in
> Fedora Core 6. My panel does have a system tray.
I'm using Ubuntu hardy (the experimental not-yet-released version of
Ubuntu).
--
Michael Olson -- FSF Associate Member #652 |
http://mwolson.org/ -- Jabber: mwolson_at_hcoop.net | /` |\ | | |
Programmer -- Hobbies: Lisp, HCoop | |_] | \| |_|
Projects: Emacs, Muse, ERC, EMMS, ErBot, DVC, Planner |
[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]
[-- Attachment #2: Type: text/plain, Size: 142 bytes --]
_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-15 6:01 ` Michael Olson
@ 2008-01-16 1:10 ` Tom Tromey
2008-01-16 4:10 ` Michael Olson
0 siblings, 1 reply; 55+ messages in thread
From: Tom Tromey @ 2008-01-16 1:10 UTC (permalink / raw)
To: Michael Olson; +Cc: emacs-devel
>>>>> "Michael" == Michael Olson <mwolson@gnu.org> writes:
Michael> I've checked this into a new branch of ERC called "erc-status" (at
Michael> http://git.sv.gnu.org/gitweb/?p=erc.git) and have done a bit of extra
Michael> hacking on it.
Thanks!
Michael> ;; => "Couldn't connect to notification server"
Michael> ;; => libnotify-Message: Unable to get session bus: Failed
Michael> ;; to execute dbus-launch to autolaunch D-Bus session
Michael> ;; => Does not show "foo" popup
Michael> [ ... set-up appears normal ]
I really don't know about this one :(. Does notify-send (command line
tool) work ok?
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-16 1:10 ` Tom Tromey
@ 2008-01-16 4:10 ` Michael Olson
0 siblings, 0 replies; 55+ messages in thread
From: Michael Olson @ 2008-01-16 4:10 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs-devel
[-- Attachment #1.1: Type: text/plain, Size: 893 bytes --]
Tom Tromey <tromey@redhat.com> writes:
> Michael> ;; => "Couldn't connect to notification server"
> Michael> ;; => libnotify-Message: Unable to get session bus: Failed
> Michael> ;; to execute dbus-launch to autolaunch D-Bus session
> Michael> ;; => Does not show "foo" popup
>
> Michael> [ ... set-up appears normal ]
>
> I really don't know about this one :(. Does notify-send (command line
> tool) work ok?
Yep, notify-send works fine. The problem goes away as noted in the test
case, so I=it's probably due to trying to pop up a notification
immediately after creating the icon (namely, the race).
--
Michael Olson -- FSF Associate Member #652 |
http://mwolson.org/ -- Jabber: mwolson_at_hcoop.net | /` |\ | | |
Programmer -- Hobbies: Lisp, HCoop | |_] | \| |_|
Projects: Emacs, Muse, ERC, EMMS, ErBot, DVC, Planner |
[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]
[-- Attachment #2: Type: text/plain, Size: 142 bytes --]
_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 1:57 ` Dan Nicolaescu
2008-01-12 1:28 ` Tom Tromey
2008-01-12 1:38 ` Tom Tromey
@ 2008-01-12 11:11 ` Richard Stallman
2008-01-12 11:25 ` David Kastrup
2008-01-12 13:52 ` Dan Nicolaescu
2 siblings, 2 replies; 55+ messages in thread
From: Richard Stallman @ 2008-01-12 11:11 UTC (permalink / raw)
To: Dan Nicolaescu; +Cc: tromey, emacs-devel
Please no K&R in new code.
We have no policy against K&R style. I recently accepted non-K&R
function definitions in Emacs sources, but I still do not particularly
like it. K&R style is easier to read anyway.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 11:11 ` Richard Stallman
@ 2008-01-12 11:25 ` David Kastrup
2008-01-12 11:27 ` Andreas Schwab
2008-01-12 13:52 ` Dan Nicolaescu
1 sibling, 1 reply; 55+ messages in thread
From: David Kastrup @ 2008-01-12 11:25 UTC (permalink / raw)
To: rms; +Cc: tromey, Dan Nicolaescu, emacs-devel
Richard Stallman <rms@gnu.org> writes:
> Please no K&R in new code.
>
> We have no policy against K&R style. I recently accepted non-K&R
> function definitions in Emacs sources, but I still do not particularly
> like it. K&R style is easier to read anyway.
It stops the compiler from doing type checking and type conversions, for
one thing.
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 11:25 ` David Kastrup
@ 2008-01-12 11:27 ` Andreas Schwab
2008-01-12 11:46 ` David Kastrup
0 siblings, 1 reply; 55+ messages in thread
From: Andreas Schwab @ 2008-01-12 11:27 UTC (permalink / raw)
To: David Kastrup; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
David Kastrup <dak@gnu.org> writes:
> Richard Stallman <rms@gnu.org> writes:
>
>> Please no K&R in new code.
>>
>> We have no policy against K&R style. I recently accepted non-K&R
>> function definitions in Emacs sources, but I still do not particularly
>> like it. K&R style is easier to read anyway.
>
> It stops the compiler from doing type checking and type conversions, for
> one thing.
Unless you add prototypes.
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 11:27 ` Andreas Schwab
@ 2008-01-12 11:46 ` David Kastrup
2008-01-12 14:10 ` Andreas Schwab
0 siblings, 1 reply; 55+ messages in thread
From: David Kastrup @ 2008-01-12 11:46 UTC (permalink / raw)
To: Andreas Schwab; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
Andreas Schwab <schwab@suse.de> writes:
> David Kastrup <dak@gnu.org> writes:
>
>> Richard Stallman <rms@gnu.org> writes:
>>
>>> Please no K&R in new code.
>>>
>>> We have no policy against K&R style. I recently accepted non-K&R
>>> function definitions in Emacs sources, but I still do not particularly
>>> like it. K&R style is easier to read anyway.
>>
>> It stops the compiler from doing type checking and type conversions, for
>> one thing.
>
> Unless you add prototypes.
In which case you have
a) undefined behavior (the compiler need not check whether the prototype
matches the function declaration).
b) not K&R compatible C, so where is the point?
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 11:46 ` David Kastrup
@ 2008-01-12 14:10 ` Andreas Schwab
2008-01-12 14:19 ` David Kastrup
0 siblings, 1 reply; 55+ messages in thread
From: Andreas Schwab @ 2008-01-12 14:10 UTC (permalink / raw)
To: David Kastrup; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
David Kastrup <dak@gnu.org> writes:
> Andreas Schwab <schwab@suse.de> writes:
>
>> David Kastrup <dak@gnu.org> writes:
>>
>>> Richard Stallman <rms@gnu.org> writes:
>>>
>>>> Please no K&R in new code.
>>>>
>>>> We have no policy against K&R style. I recently accepted non-K&R
>>>> function definitions in Emacs sources, but I still do not particularly
>>>> like it. K&R style is easier to read anyway.
>>>
>>> It stops the compiler from doing type checking and type conversions, for
>>> one thing.
>>
>> Unless you add prototypes.
>
> In which case you have
> a) undefined behavior (the compiler need not check whether the prototype
> matches the function declaration).
There is nothing undefined in old-style function defintions.
> b) not K&R compatible C, so where is the point?
The Emacs sources use macros that keep prototypes from being visible
from obsolte compilers. What's your point?
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:10 ` Andreas Schwab
@ 2008-01-12 14:19 ` David Kastrup
2008-01-12 17:33 ` Andreas Schwab
2008-01-14 2:00 ` Richard Stallman
0 siblings, 2 replies; 55+ messages in thread
From: David Kastrup @ 2008-01-12 14:19 UTC (permalink / raw)
To: Andreas Schwab; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
Andreas Schwab <schwab@suse.de> writes:
> David Kastrup <dak@gnu.org> writes:
>
>> Andreas Schwab <schwab@suse.de> writes:
>>
>>> David Kastrup <dak@gnu.org> writes:
>>>
>>>> Richard Stallman <rms@gnu.org> writes:
>>>>
>>>>> Please no K&R in new code.
>>>>>
>>>>> We have no policy against K&R style. I recently accepted non-K&R
>>>>> function definitions in Emacs sources, but I still do not particularly
>>>>> like it. K&R style is easier to read anyway.
>>>>
>>>> It stops the compiler from doing type checking and type conversions, for
>>>> one thing.
>>>
>>> Unless you add prototypes.
>>
>> In which case you have
>> a) undefined behavior (the compiler need not check whether the prototype
>> matches the function declaration).
>
> There is nothing undefined in old-style function defintions.
There certainly is. If they don't match the prototype, the compiler is
free to do whatever it thinks correct. It is not required to generate
an error.
>> b) not K&R compatible C, so where is the point?
>
> The Emacs sources use macros that keep prototypes from being visible
> from obsolte compilers.
We stopped mandating this.
> What's your point?
Where is the point in combining a prototype that won't compile under K&R
compilers with a K&R definition (which won't compile under C++ or newer
standard compilers)?
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:19 ` David Kastrup
@ 2008-01-12 17:33 ` Andreas Schwab
2008-01-14 2:00 ` Richard Stallman
1 sibling, 0 replies; 55+ messages in thread
From: Andreas Schwab @ 2008-01-12 17:33 UTC (permalink / raw)
To: David Kastrup; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
David Kastrup <dak@gnu.org> writes:
> Andreas Schwab <schwab@suse.de> writes:
>
>> There is nothing undefined in old-style function defintions.
>
> There certainly is. If they don't match the prototype, the compiler is
> free to do whatever it thinks correct.
Wrong. In the presence of a prototype there is no difference between an
old-style and a new-style definition.
> It is not required to generate an error.
A compiler is never required to generate an error. The standard calls
this a diagnostic, and the compiler can do whatever it pleases after it
has issued a diagnostic.
> Where is the point in combining a prototype that won't compile under K&R
> compilers with a K&R definition (which won't compile under C++ or newer
> standard compilers)?
A standard compiler is required to accept them.
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:19 ` David Kastrup
2008-01-12 17:33 ` Andreas Schwab
@ 2008-01-14 2:00 ` Richard Stallman
2008-01-14 2:25 ` Stefan Monnier
2008-01-14 7:05 ` Jan Djärv
1 sibling, 2 replies; 55+ messages in thread
From: Richard Stallman @ 2008-01-14 2:00 UTC (permalink / raw)
To: David Kastrup; +Cc: tromey, schwab, dann, emacs-devel
Where is the point in combining a prototype that won't compile under K&R
compilers with a K&R definition (which won't compile under C++ or newer
standard compilers)?
Our prototypes compile just fine in a K&R compiler, due to _P.
The reason I prefer K&R style in function definitions is that the
argument type declarations are easier to read when not inside the
parentheses.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 2:00 ` Richard Stallman
@ 2008-01-14 2:25 ` Stefan Monnier
2008-01-14 7:05 ` Jan Djärv
1 sibling, 0 replies; 55+ messages in thread
From: Stefan Monnier @ 2008-01-14 2:25 UTC (permalink / raw)
To: rms; +Cc: tromey, schwab, dann, emacs-devel
> The reason I prefer K&R style in function definitions is that the
> argument type declarations are easier to read when not inside the
> parentheses.
That is simply not so. You may find them easier to read, but I find
them harder to read instead. I suspect it has to do with habit.
Stefan
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 2:00 ` Richard Stallman
2008-01-14 2:25 ` Stefan Monnier
@ 2008-01-14 7:05 ` Jan Djärv
1 sibling, 0 replies; 55+ messages in thread
From: Jan Djärv @ 2008-01-14 7:05 UTC (permalink / raw)
To: rms; +Cc: tromey, schwab, dann, emacs-devel
Richard Stallman skrev:
> Where is the point in combining a prototype that won't compile under K&R
> compilers with a K&R definition (which won't compile under C++ or newer
> standard compilers)?
>
> Our prototypes compile just fine in a K&R compiler, due to _P.
>
That would be P_.
Jan D.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 11:11 ` Richard Stallman
2008-01-12 11:25 ` David Kastrup
@ 2008-01-12 13:52 ` Dan Nicolaescu
2008-01-12 14:13 ` Andreas Schwab
1 sibling, 1 reply; 55+ messages in thread
From: Dan Nicolaescu @ 2008-01-12 13:52 UTC (permalink / raw)
To: rms; +Cc: tromey, emacs-devel
Richard Stallman <rms@gnu.org> writes:
> Please no K&R in new code.
>
> We have no policy against K&R style. I recently accepted non-K&R
> function definitions in Emacs sources, but I still do not particularly
> like it. K&R style is easier to read anyway.
It might be easier to read for you personally, but it is harder for
people that have never written/read any K&R code. Some emacs
contributors have started programming after C was standardized, so they
never had a chance to know any different (and its quite possible that
some were even born after the standardization).
So yes, this preference has a price, and one that is not in emacs' best
interest to pay.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 13:52 ` Dan Nicolaescu
@ 2008-01-12 14:13 ` Andreas Schwab
2008-01-12 14:26 ` Dan Nicolaescu
2008-01-12 14:33 ` David Kastrup
0 siblings, 2 replies; 55+ messages in thread
From: Andreas Schwab @ 2008-01-12 14:13 UTC (permalink / raw)
To: Dan Nicolaescu; +Cc: tromey, rms, emacs-devel
Dan Nicolaescu <dann@ics.uci.edu> writes:
> Richard Stallman <rms@gnu.org> writes:
>
> > Please no K&R in new code.
> >
> > We have no policy against K&R style. I recently accepted non-K&R
> > function definitions in Emacs sources, but I still do not particularly
> > like it. K&R style is easier to read anyway.
>
> It might be easier to read for you personally, but it is harder for
> people that have never written/read any K&R code. Some emacs
> contributors have started programming after C was standardized, so they
> never had a chance to know any different (and its quite possible that
> some were even born after the standardization).
Old-style function definitions are still part of the C standard. They
are only marked obsolescent.
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:13 ` Andreas Schwab
@ 2008-01-12 14:26 ` Dan Nicolaescu
2008-01-12 17:36 ` Andreas Schwab
2008-01-12 14:33 ` David Kastrup
1 sibling, 1 reply; 55+ messages in thread
From: Dan Nicolaescu @ 2008-01-12 14:26 UTC (permalink / raw)
To: Andreas Schwab; +Cc: tromey, rms, emacs-devel
Andreas Schwab <schwab@suse.de> writes:
> Dan Nicolaescu <dann@ics.uci.edu> writes:
>
> > Richard Stallman <rms@gnu.org> writes:
> >
> > > Please no K&R in new code.
> > >
> > > We have no policy against K&R style. I recently accepted non-K&R
> > > function definitions in Emacs sources, but I still do not particularly
> > > like it. K&R style is easier to read anyway.
> >
> > It might be easier to read for you personally, but it is harder for
> > people that have never written/read any K&R code. Some emacs
> > contributors have started programming after C was standardized, so they
> > never had a chance to know any different (and its quite possible that
> > some were even born after the standardization).
>
> Old-style function definitions are still part of the C standard. They
> are only marked obsolescent.
And because of that they are not being used, people are not being taught
about them. They might exist from the language lawyer point of view, but
for practical purposes they don't exist for large number of
programmers.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:26 ` Dan Nicolaescu
@ 2008-01-12 17:36 ` Andreas Schwab
2008-01-12 18:59 ` Dan Nicolaescu
0 siblings, 1 reply; 55+ messages in thread
From: Andreas Schwab @ 2008-01-12 17:36 UTC (permalink / raw)
To: Dan Nicolaescu; +Cc: tromey, rms, emacs-devel
Dan Nicolaescu <dann@ics.uci.edu> writes:
> Andreas Schwab <schwab@suse.de> writes:
>
> > Dan Nicolaescu <dann@ics.uci.edu> writes:
> >
> > > Richard Stallman <rms@gnu.org> writes:
> > >
> > > > Please no K&R in new code.
> > > >
> > > > We have no policy against K&R style. I recently accepted non-K&R
> > > > function definitions in Emacs sources, but I still do not particularly
> > > > like it. K&R style is easier to read anyway.
> > >
> > > It might be easier to read for you personally, but it is harder for
> > > people that have never written/read any K&R code. Some emacs
> > > contributors have started programming after C was standardized, so they
> > > never had a chance to know any different (and its quite possible that
> > > some were even born after the standardization).
> >
> > Old-style function definitions are still part of the C standard. They
> > are only marked obsolescent.
>
> And because of that they are not being used, people are not being taught
> about them.
I claim that the majority of existing C sources still uses old-style
function definitions. I cannot prove that, but given the vast amount of
existing old software I would be surprised if a programmer can learn C
without having been confronted with them.
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 17:36 ` Andreas Schwab
@ 2008-01-12 18:59 ` Dan Nicolaescu
0 siblings, 0 replies; 55+ messages in thread
From: Dan Nicolaescu @ 2008-01-12 18:59 UTC (permalink / raw)
To: Andreas Schwab; +Cc: tromey, rms, emacs-devel
Andreas Schwab <schwab@suse.de> writes:
> Dan Nicolaescu <dann@ics.uci.edu> writes:
>
> > Andreas Schwab <schwab@suse.de> writes:
> >
> > > Dan Nicolaescu <dann@ics.uci.edu> writes:
> > >
> > > > Richard Stallman <rms@gnu.org> writes:
> > > >
> > > > > Please no K&R in new code.
> > > > >
> > > > > We have no policy against K&R style. I recently accepted non-K&R
> > > > > function definitions in Emacs sources, but I still do not particularly
> > > > > like it. K&R style is easier to read anyway.
> > > >
> > > > It might be easier to read for you personally, but it is harder for
> > > > people that have never written/read any K&R code. Some emacs
> > > > contributors have started programming after C was standardized, so they
> > > > never had a chance to know any different (and its quite possible that
> > > > some were even born after the standardization).
> > >
> > > Old-style function definitions are still part of the C standard. They
> > > are only marked obsolescent.
> >
> > And because of that they are not being used, people are not being taught
> > about them.
>
> I claim that the majority of existing C sources still uses old-style
> function definitions. I cannot prove that, but given the vast amount of
> existing old software I would be surprised if a programmer can learn C
> without having been confronted with them.
And what is your point? That we should insist on using K&R? Let's not
forget the starting point of this discussion and stop discussing
strawmen just for the sake of discussion.
And I completely disagree with your point. I won't make judgements in
general, but in my personal experience, I have only seen K&R in GNU
software. I have seen a lot of software that was written in K&R style
that was converted to ISO C mainly because of better compiler error
checking. The conversion is trivial (and as you know, a lot of GNU
software was converted too).
Second, "having been confronted" is not the same as being familiar,
understanding the differences, and actually writing code in the style.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:13 ` Andreas Schwab
2008-01-12 14:26 ` Dan Nicolaescu
@ 2008-01-12 14:33 ` David Kastrup
2008-01-12 17:45 ` Andreas Schwab
2008-01-14 2:01 ` Richard Stallman
1 sibling, 2 replies; 55+ messages in thread
From: David Kastrup @ 2008-01-12 14:33 UTC (permalink / raw)
To: Andreas Schwab; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
Andreas Schwab <schwab@suse.de> writes:
> Dan Nicolaescu <dann@ics.uci.edu> writes:
>
>> Richard Stallman <rms@gnu.org> writes:
>>
>> > Please no K&R in new code.
>> >
>> > We have no policy against K&R style. I recently accepted non-K&R
>> > function definitions in Emacs sources, but I still do not particularly
>> > like it. K&R style is easier to read anyway.
>>
>> It might be easier to read for you personally, but it is harder for
>> people that have never written/read any K&R code. Some emacs
>> contributors have started programming after C was standardized, so they
>> never had a chance to know any different (and its quite possible that
>> some were even born after the standardization).
>
> Old-style function definitions are still part of the C standard. They
> are only marked obsolescent.
"only"?
Obsolescing K&R Function Declaration Style
The K&R function declaration style in the form of
func(n, m) /* return type is unspecified */
int n; double m; /* function parameters */
{
/*function body*/
}
was used in K&R. It was later replaced by the more common form:
int func(int n, double m)
{
/*function body*/
}
The K&R function declaration style is obsolescent in C99. This means
that compilers are allowed to reject such code or accept it with a
warning message. In either case, they aren't allowed to ignore it.
Compilers are not allowed to accept this style of declaration without
warning. They are free to reject it completely, however.
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:33 ` David Kastrup
@ 2008-01-12 17:45 ` Andreas Schwab
2008-01-12 18:07 ` David Kastrup
2008-01-14 2:01 ` Richard Stallman
1 sibling, 1 reply; 55+ messages in thread
From: Andreas Schwab @ 2008-01-12 17:45 UTC (permalink / raw)
To: David Kastrup; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
David Kastrup <dak@gnu.org> writes:
> Compilers are not allowed to accept this style of declaration without
> warning. They are free to reject it completely, however.
Obsolescent just means that the feature may be removed from the next
standard.
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 17:45 ` Andreas Schwab
@ 2008-01-12 18:07 ` David Kastrup
2008-01-12 18:16 ` Andreas Schwab
0 siblings, 1 reply; 55+ messages in thread
From: David Kastrup @ 2008-01-12 18:07 UTC (permalink / raw)
To: Andreas Schwab; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
Andreas Schwab <schwab@suse.de> writes:
> David Kastrup <dak@gnu.org> writes:
>
>> Compilers are not allowed to accept this style of declaration without
>> warning. They are free to reject it completely, however.
>
> Obsolescent just means that the feature may be removed from the next
> standard.
Have you even read what I quoted?
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 18:07 ` David Kastrup
@ 2008-01-12 18:16 ` Andreas Schwab
0 siblings, 0 replies; 55+ messages in thread
From: Andreas Schwab @ 2008-01-12 18:16 UTC (permalink / raw)
To: David Kastrup; +Cc: tromey, Dan Nicolaescu, rms, emacs-devel
David Kastrup <dak@gnu.org> writes:
> Have you even read what I quoted?
The C standard says nothing like that.
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-12 14:33 ` David Kastrup
2008-01-12 17:45 ` Andreas Schwab
@ 2008-01-14 2:01 ` Richard Stallman
2008-01-14 2:47 ` Dan Nicolaescu
2008-01-14 9:14 ` David Kastrup
1 sibling, 2 replies; 55+ messages in thread
From: Richard Stallman @ 2008-01-14 2:01 UTC (permalink / raw)
To: David Kastrup; +Cc: tromey, schwab, dann, emacs-devel
The K&R function declaration style is obsolescent in C99. This means
that compilers are allowed to reject such code or accept it with a
warning message. In either case, they aren't allowed to ignore it.
Compilers are not allowed to accept this style of declaration without
warning.
Not allowed? Nobody can stop us!
The point is that Emacs does NOT have a convention against K&R
function definitions. When Dan told Tom Tromey to rewrite to ANSI
style, he implied that that was our convention. That is not so.
I do not want to have a discussion about changing this, because there
are other, more important things to spend our time on. For instance,
a few bugs remain in Emacs 22.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 2:01 ` Richard Stallman
@ 2008-01-14 2:47 ` Dan Nicolaescu
2008-01-14 17:26 ` Richard Stallman
2008-01-14 9:14 ` David Kastrup
1 sibling, 1 reply; 55+ messages in thread
From: Dan Nicolaescu @ 2008-01-14 2:47 UTC (permalink / raw)
To: rms; +Cc: tromey, schwab, emacs-devel
Richard Stallman <rms@gnu.org> writes:
> The K&R function declaration style is obsolescent in C99. This means
> that compilers are allowed to reject such code or accept it with a
> warning message. In either case, they aren't allowed to ignore it.
>
>
> Compilers are not allowed to accept this style of declaration without
> warning.
>
> Not allowed? Nobody can stop us!
>
> The point is that Emacs does NOT have a convention against K&R
> function definitions. When Dan told Tom Tromey to rewrite to ANSI
> style, he implied that that was our convention. That is not so.
Sorry, but that is not accurate characterization of my message. Here it
is for the record:
Please no K&R in new code.
I am not sure if we actually have a policy about this. If we don't we
should make it now required to have K&R in new code. It's already not
possible to compile emacs with a K&R compiler, so please let's not add
to this.
The last thing we want is emacs to regarded as being a backwards project
because it insists on using a language that was standardized in a
different way 19 years ago.
> I do not want to have a discussion about changing this, because there
> are other, more important things to spend our time on. For instance,
> a few bugs remain in Emacs 22.
As you can see from the message cited above the discussion was not about
changing what we have now, but about what to do with new code.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 2:47 ` Dan Nicolaescu
@ 2008-01-14 17:26 ` Richard Stallman
0 siblings, 0 replies; 55+ messages in thread
From: Richard Stallman @ 2008-01-14 17:26 UTC (permalink / raw)
To: Dan Nicolaescu; +Cc: tromey, schwab, emacs-devel
As you can see from the message cited above the discussion was not about
changing what we have now, but about what to do with new code.
K&R definitions are ok in new code. They have always been ok.
Until a couple of years ago I insisted on K&R style function
definitions. Then I decided to accept ANSI style also, but that is as
far as we have gone.
I don't want to reconsider the question at this time,
because there are many other more important things to do
than reconsider this.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 2:01 ` Richard Stallman
2008-01-14 2:47 ` Dan Nicolaescu
@ 2008-01-14 9:14 ` David Kastrup
2008-01-14 17:26 ` Richard Stallman
1 sibling, 1 reply; 55+ messages in thread
From: David Kastrup @ 2008-01-14 9:14 UTC (permalink / raw)
To: rms; +Cc: tromey, schwab, dann, emacs-devel
Richard Stallman <rms@gnu.org> writes:
> The K&R function declaration style is obsolescent in C99. This
> means that compilers are allowed to reject such code or accept
> it with a warning message. In either case, they aren't allowed
> to ignore it.
>
>
> Compilers are not allowed to accept this style of declaration without
> warning.
>
> Not allowed? Nobody can stop us!
What about "in C99" did you not understand? In a similar vein, nothing
can stop you from moving a rook diagonally. Sure. But it is no longer
chess you are playing then.
--
David Kastrup
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-11 23:28 RFC: status icon support Tom Tromey
2008-01-12 1:57 ` Dan Nicolaescu
@ 2008-01-14 3:47 ` YAMAMOTO Mitsuharu
2008-01-14 3:49 ` Tom Tromey
2008-01-14 17:26 ` Richard Stallman
1 sibling, 2 replies; 55+ messages in thread
From: YAMAMOTO Mitsuharu @ 2008-01-14 3:47 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs-devel
>>>>> On Fri, 11 Jan 2008 16:28:18 -0700, Tom Tromey <tromey@redhat.com> said:
> --- src/lisp.h 22 Nov 2007 01:01:26 -0000 1.603
> +++ src/lisp.h 11 Jan 2008 23:47:14 -0000
> @@ -349,7 +349,8 @@
> PVEC_HASH_TABLE = 0x40000,
> PVEC_TERMINAL = 0x80000,
> PVEC_OTHER = 0x100000,
> - PVEC_TYPE_MASK = 0x1ffe00
> + PVEC_STATUS_ICON = 0x200000,
> + PVEC_TYPE_MASK = 0x2ffe00
The introduction of a new Lisp data type for this purpose looks rather
unusual compared with what other kinds of existing objects do. For
example, images, which are more fundamental notion I think, do not
have a dedicated Lisp data type but are represented by image
descriptors that are just lists of the form `(image . PROPS)'.
YAMAMOTO Mitsuharu
mituharu@math.s.chiba-u.ac.jp
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 3:47 ` YAMAMOTO Mitsuharu
@ 2008-01-14 3:49 ` Tom Tromey
2008-01-14 13:35 ` Stefan Monnier
2008-01-14 21:40 ` YAMAMOTO Mitsuharu
2008-01-14 17:26 ` Richard Stallman
1 sibling, 2 replies; 55+ messages in thread
From: Tom Tromey @ 2008-01-14 3:49 UTC (permalink / raw)
To: YAMAMOTO Mitsuharu; +Cc: emacs-devel
>>>>> ">" == YAMAMOTO Mitsuharu <mituharu@math.s.chiba-u.ac.jp> writes:
>> - PVEC_TYPE_MASK = 0x1ffe00
>> + PVEC_STATUS_ICON = 0x200000,
>> + PVEC_TYPE_MASK = 0x2ffe00
>> The introduction of a new Lisp data type for this purpose looks rather
>> unusual compared with what other kinds of existing objects do. For
>> example, images, which are more fundamental notion I think, do not
>> have a dedicated Lisp data type but are represented by image
>> descriptors that are just lists of the form `(image . PROPS)'.
Ok. I suppose I could pretty easily change the lisp-visible
representation to a cons like (status-icon . N), where N is an index
into a table in systray.c. What would you think of that?
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 3:49 ` Tom Tromey
@ 2008-01-14 13:35 ` Stefan Monnier
2008-01-14 21:40 ` YAMAMOTO Mitsuharu
1 sibling, 0 replies; 55+ messages in thread
From: Stefan Monnier @ 2008-01-14 13:35 UTC (permalink / raw)
To: Tom Tromey; +Cc: YAMAMOTO Mitsuharu, emacs-devel
>>> - PVEC_TYPE_MASK = 0x1ffe00
>>> + PVEC_STATUS_ICON = 0x200000,
>>> + PVEC_TYPE_MASK = 0x2ffe00
>>> The introduction of a new Lisp data type for this purpose looks rather
>>> unusual compared with what other kinds of existing objects do. For
>>> example, images, which are more fundamental notion I think, do not
>>> have a dedicated Lisp data type but are represented by image
>>> descriptors that are just lists of the form `(image . PROPS)'.
> Ok. I suppose I could pretty easily change the lisp-visible
> representation to a cons like (status-icon . N), where N is an index
> into a table in systray.c. What would you think of that?
If the internal data is not accessible anyway, then I think it doesn't
matter what representation you use. So, in that case, just go with the
simplest implementation.
Stefan
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 3:49 ` Tom Tromey
2008-01-14 13:35 ` Stefan Monnier
@ 2008-01-14 21:40 ` YAMAMOTO Mitsuharu
2008-01-16 1:17 ` Tom Tromey
1 sibling, 1 reply; 55+ messages in thread
From: YAMAMOTO Mitsuharu @ 2008-01-14 21:40 UTC (permalink / raw)
To: tromey; +Cc: emacs-devel
>>>>> On Sun, 13 Jan 2008 20:49:42 -0700, Tom Tromey <tromey@redhat.com> said:
>>> The introduction of a new Lisp data type for this purpose looks
>>> rather unusual compared with what other kinds of existing objects
>>> do. For example, images, which are more fundamental notion I
>>> think, do not have a dedicated Lisp data type but are represented
>>> by image descriptors that are just lists of the form `(image
>>> . PROPS)'.
> Ok. I suppose I could pretty easily change the lisp-visible
> representation to a cons like (status-icon . N), where N is an index
> into a table in systray.c. What would you think of that?
A tool bar is represented as a keymap. If status icons are similar to
tool bar icons (application-level as opposed to frame-level), the use
of the keymap data structure looks more consistent to me.
YAMAMOTO Mitsuharu
mituharu@math.s.chiba-u.ac.jp
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 21:40 ` YAMAMOTO Mitsuharu
@ 2008-01-16 1:17 ` Tom Tromey
2008-01-16 11:55 ` YAMAMOTO Mitsuharu
0 siblings, 1 reply; 55+ messages in thread
From: Tom Tromey @ 2008-01-16 1:17 UTC (permalink / raw)
To: YAMAMOTO Mitsuharu; +Cc: emacs-devel
>>>>> ">" == YAMAMOTO Mitsuharu <mituharu@math.s.chiba-u.ac.jp> writes:
>> A tool bar is represented as a keymap. If status icons are similar to
>> tool bar icons (application-level as opposed to frame-level), the use
>> of the keymap data structure looks more consistent to me.
Thanks for responding. I've been thinking about this a bit, and I
don't understand the benefits of this versus just using a simple cons.
(I do understand, I think, why you would prefer not to make a new
type.)
Also, I do not know how to implement this.
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-16 1:17 ` Tom Tromey
@ 2008-01-16 11:55 ` YAMAMOTO Mitsuharu
0 siblings, 0 replies; 55+ messages in thread
From: YAMAMOTO Mitsuharu @ 2008-01-16 11:55 UTC (permalink / raw)
To: tromey; +Cc: emacs-devel
>>>>> On Tue, 15 Jan 2008 18:17:05 -0700, Tom Tromey <tromey@redhat.com> said:
>>> A tool bar is represented as a keymap. If status icons are
>>> similar to tool bar icons (application-level as opposed to
>>> frame-level), the use of the keymap data structure looks more
>>> consistent to me.
> Thanks for responding. I've been thinking about this a bit, and I
> don't understand the benefits of this versus just using a simple
> cons. (I do understand, I think, why you would prefer not to make a
> new type.)
I think it would be obviously beneficial if the same data structure is
used for similar concepts (I'm not so familiar with status icons, so
please point out if they are not so similar to tool bar icons). Emacs
Lisp programmers who have get used to the tool bars can easily
learn/imagine how to operate on the status icons. Some implementation
codes can be shared and that reduces the chance of bugs. Also, there
would be possible to complement some missing features in tool bars by
analogy from those in status icons.
> Also, I do not know how to implement this.
You can look at the implementation of tool bars. Of course, it's not
as easy as creating some black-box Lisp objects for GTK+ objects and
direct bridges/wrappers to GTK+ functions. It might be helpful to
understand the basic strategy of Emacs redisplay:
* Lisp functions basically modify buffers and/or Lisp data
structures (i.e., models), but don't usually do immediate display
operations. (Of course, there are several exceptions.)
* When `redisplay' is triggered, it calculates the changes from the
previous redisplay in the models and do appropriate display
operations.
YAMAMOTO Mitsuharu
mituharu@math.s.chiba-u.ac.jp
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 3:47 ` YAMAMOTO Mitsuharu
2008-01-14 3:49 ` Tom Tromey
@ 2008-01-14 17:26 ` Richard Stallman
2008-01-14 17:10 ` Tom Tromey
1 sibling, 1 reply; 55+ messages in thread
From: Richard Stallman @ 2008-01-14 17:26 UTC (permalink / raw)
To: YAMAMOTO Mitsuharu; +Cc: tromey, emacs-devel
The introduction of a new Lisp data type for this purpose looks rather
unusual compared with what other kinds of existing objects do.
We should not create a new data type for status icons unless there is
a compelling reason for it -- that is to say, not making the data type
would imply serious problems or ugliness.
What is the motive for this data type? Is it so as to handle X events
for the icon?
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 17:26 ` Richard Stallman
@ 2008-01-14 17:10 ` Tom Tromey
2008-01-16 2:42 ` Richard Stallman
0 siblings, 1 reply; 55+ messages in thread
From: Tom Tromey @ 2008-01-14 17:10 UTC (permalink / raw)
To: rms; +Cc: YAMAMOTO Mitsuharu, emacs-devel
>>>>> "rms" == Richard Stallman <rms@gnu.org> writes:
rms> We should not create a new data type for status icons unless there is
rms> a compelling reason for it -- that is to say, not making the data type
rms> would imply serious problems or ugliness.
rms> What is the motive for this data type? Is it so as to handle X events
rms> for the icon?
There's no deep reason for it, it is just the approach I picked, by
analogy to frames and other programming languages, where adding new
types is unexceptional. I don't mind changing it to be a cons if that
is the preferred Emacs way.
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2008-01-14 17:10 ` Tom Tromey
@ 2008-01-16 2:42 ` Richard Stallman
0 siblings, 0 replies; 55+ messages in thread
From: Richard Stallman @ 2008-01-16 2:42 UTC (permalink / raw)
To: Tom Tromey; +Cc: mituharu, emacs-devel
There's no deep reason for it, it is just the approach I picked, by
analogy to frames and other programming languages, where adding new
types is unexceptional. I don't mind changing it to be a cons if that
is the preferred Emacs way.
It would be better to avoid adding a data type for this, just because
there's a lot of conceptual overhead for that.
^ permalink raw reply [flat|nested] 55+ messages in thread
* RFC: status icon support
@ 2007-12-30 19:56 Tom Tromey
2007-12-31 17:18 ` Dan Nicolaescu
0 siblings, 1 reply; 55+ messages in thread
From: Tom Tromey @ 2007-12-30 19:56 UTC (permalink / raw)
To: emacs-devel
This patch implements status icon support for Emacs.
A status icon is an icon that sits in a part of the panel called the
"system tray". It can be used to display notifications that are
somewhat less intrusive than a real dialog box. A status icon can
also have a menu (unimplemented here), react to button presses, and
can display notification messages.
This patch only works when using Gtk. It could in theory be made to
work with other toolkits -- there is a cross-desktop systray
specification, and both Windows and the Mac support something similar.
However, I don't know these other platforms, so someone else would
have to implement that.
This patch is still incomplete. It needs a configure change and some
#ifdefs. It also needs a ChangeLog entry and, I think, a section in
the elisp manual.
I'm looking for feedback on the general approach, the code, etc.
FWIW I've been using a version of this based on an external utility
for about a year now. The best use I've found for it is integrating
with ERC; I have an ERC module that I'll rewrite and submit once this
code goes in. I've also got hacks for EMMS and the calendar, and
other things are possible (mail notification comes to mind).
Tom
Index: alloc.c
===================================================================
RCS file: /sources/emacs/emacs/src/alloc.c,v
retrieving revision 1.432
diff -u -r1.432 alloc.c
--- alloc.c 16 Nov 2007 21:24:59 -0000 1.432
+++ alloc.c 30 Dec 2007 19:47:59 -0000
@@ -56,6 +56,7 @@
#include "charset.h"
#include "syssignal.h"
#include "termhooks.h" /* For struct terminal. */
+#include "systray.h"
#include <setjmp.h>
/* GC_MALLOC_CHECK defined means perform validity checks of malloc'd
@@ -3021,6 +3022,16 @@
return f;
}
+struct status_icon *
+allocate_status_icon ()
+{
+ struct status_icon *s = ALLOCATE_PSEUDOVECTOR (struct status_icon,
+ param_alist, PVEC_STATUS_ICON);
+ /* Zero out the non-GC'd fields. FIXME: This should be made unnecessary. */
+ bzero (&(s->param_alist),
+ ((char*)(s+1)) - ((char*)&(s->param_alist)));
+ return s;
+}
struct Lisp_Process *
allocate_process ()
Index: emacs.c
===================================================================
RCS file: /sources/emacs/emacs/src/emacs.c,v
retrieving revision 1.413
diff -u -r1.413 emacs.c
--- emacs.c 2 Dec 2007 16:23:40 -0000 1.413
+++ emacs.c 30 Dec 2007 19:48:01 -0000
@@ -1579,6 +1579,7 @@
/* Called before init_window_once for Mac OS Classic. */
syms_of_search ();
syms_of_frame ();
+ syms_of_systray ();
#endif
syms_of_syntax ();
syms_of_terminal ();
Index: keyboard.c
===================================================================
RCS file: /sources/emacs/emacs/src/keyboard.c,v
retrieving revision 1.935
diff -u -r1.935 keyboard.c
--- keyboard.c 10 Dec 2007 03:48:05 -0000 1.935
+++ keyboard.c 30 Dec 2007 19:48:03 -0000
@@ -500,6 +500,8 @@
#ifdef HAVE_DBUS
Lisp_Object Qdbus_event;
#endif
+Lisp_Object Qstatus_icon_click_event;
+
/* Lisp_Object Qmouse_movement; - also an event header */
/* Properties of event headers. */
@@ -4298,6 +4300,11 @@
kbd_fetch_ptr = event + 1;
}
#endif
+ else if (event->kind == STATUS_ICON_CLICK_EVENT)
+ {
+ obj = make_lispy_event (event);
+ kbd_fetch_ptr = event + 1;
+ }
else
{
/* If this event is on a different frame, return a switch-frame this
@@ -6174,6 +6181,9 @@
}
#endif /* HAVE_DBUS */
+ case STATUS_ICON_CLICK_EVENT:
+ return Fcons (Qstatus_icon_click_event, event->arg);
+
#ifdef HAVE_GPM
case GPM_CLICK_EVENT:
{
@@ -11780,6 +11790,9 @@
staticpro (&Qdbus_event);
#endif
+ Qstatus_icon_click_event = intern ("status-icon-click-event");
+ staticpro (&Qstatus_icon_click_event);
+
Qmenu_enable = intern ("menu-enable");
staticpro (&Qmenu_enable);
Qmenu_alias = intern ("menu-alias");
@@ -12516,6 +12529,9 @@
initial_define_lispy_key (Vspecial_event_map, "dbus-event",
"dbus-handle-event");
#endif
+
+ initial_define_lispy_key (Vspecial_event_map, "status-icon-click-event",
+ "status-icon-handle-click-event");
}
/* Mark the pointers in the kboard objects.
Index: lisp.h
===================================================================
RCS file: /sources/emacs/emacs/src/lisp.h,v
retrieving revision 1.603
diff -u -r1.603 lisp.h
--- lisp.h 22 Nov 2007 01:01:26 -0000 1.603
+++ lisp.h 30 Dec 2007 19:48:03 -0000
@@ -349,7 +349,8 @@
PVEC_HASH_TABLE = 0x40000,
PVEC_TERMINAL = 0x80000,
PVEC_OTHER = 0x100000,
- PVEC_TYPE_MASK = 0x1ffe00
+ PVEC_STATUS_ICON = 0x200000,
+ PVEC_TYPE_MASK = 0x2ffe00
#if 0 /* This is used to make the value of PSEUDOVECTOR_FLAG available to
GDB. It doesn't work on OS Alpha. Moved to a variable in
@@ -1552,6 +1553,8 @@
#define GC_BOOL_VECTOR_P(x) GC_PSEUDOVECTORP (x, PVEC_BOOL_VECTOR)
#define FRAMEP(x) PSEUDOVECTORP (x, PVEC_FRAME)
#define GC_FRAMEP(x) GC_PSEUDOVECTORP (x, PVEC_FRAME)
+#define STATUS_ICONP(x) PSEUDOVECTORP (x, PVEC_STATUS_ICON)
+#define GC_STATUS_ICONP(x) GC_PSEUDOVECTORP (x, PVEC_STATUS_ICON)
#define SUB_CHAR_TABLE_P(x) (CHAR_TABLE_P (x) && NILP (XCHAR_TABLE (x)->top))
@@ -2616,6 +2620,7 @@
extern struct Lisp_Hash_Table *allocate_hash_table P_ ((void));
extern struct window *allocate_window P_ ((void));
extern struct frame *allocate_frame P_ ((void));
+extern struct status_icon *allocate_status_icon P_ ((void));
extern struct Lisp_Process *allocate_process P_ ((void));
extern struct terminal *allocate_terminal P_ ((void));
extern int gc_in_progress;
@@ -3248,6 +3253,9 @@
extern void syms_of_sound P_ ((void));
extern void init_sound P_ ((void));
+/* Defined in systray.c */
+extern void syms_of_systray P_ ((void));
+
/* Defined in category.c */
extern void init_category_once P_ ((void));
extern void syms_of_category P_ ((void));
Index: print.c
===================================================================
RCS file: /sources/emacs/emacs/src/print.c,v
retrieving revision 1.243
diff -u -r1.243 print.c
--- print.c 22 Nov 2007 16:16:02 -0000 1.243
+++ print.c 30 Dec 2007 19:48:03 -0000
@@ -2031,6 +2031,8 @@
strout (buf, -1, -1, printcharfun, 0);
PRINTCHAR ('>');
}
+ else if (STATUS_ICONP (obj))
+ strout ("#<status-icon>", -1, -1, printcharfun, 0);
else
{
EMACS_INT size = XVECTOR (obj)->size;
Index: systray.c
===================================================================
RCS file: systray.c
diff -N systray.c
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ systray.c 30 Dec 2007 19:48:03 -0000
@@ -0,0 +1,446 @@
+/* Elisp bindings system tray icons
+ Copyright (C) 2007 Free Software Foundation, Inc.
+
+This file is part of GNU Emacs.
+
+GNU Emacs is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GNU Emacs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Emacs; see the file COPYING. If not, write to
+the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA. */
+
+#include "config.h"
+
+#include "lisp.h"
+#include "systray.h"
+#include "frame.h"
+#include "charset.h"
+#include "coding.h"
+#include "blockinput.h"
+#include "termhooks.h"
+
+#include <gtk/gtk.h>
+#include <libnotify/notify.h>
+
+/* FIXME: should be in a header. */
+extern Lisp_Object Qhelp_echo;
+extern Lisp_Object Qstatus_icon_click_event;
+
+Lisp_Object Vstatus_icon_list;
+
+Lisp_Object Qblinking;
+Lisp_Object Qstatus_icon_live_p;
+Lisp_Object Qclick_callback;
+Lisp_Object QCbody, QCtimeout, QCurgency, QCicon;
+Lisp_Object Qlow, Qcritical, Qwarning, Qquestion, Qerror;
+
+#define GET_ICON(s) ((GtkStatusIcon *) ((s)->icon))
+
+/* Callback which is called when the user clicks on a status icon. */
+static void
+activate_icon (GtkWidget *widget, gpointer data)
+{
+ Lisp_Object icon = (Lisp_Object) data;
+ struct input_event event;
+ EVENT_INIT (event);
+ event.kind = STATUS_ICON_CLICK_EVENT;
+ event.frame_or_window = Qnil;
+ event.arg = icon;
+ kbd_buffer_store_event (&event);
+}
+
+/* Apply property KEY with value VALUE to icon SICON. */
+static void
+apply_one_icon_property (sicon, key, value)
+ struct status_icon *sicon;
+ Lisp_Object key, value;
+{
+ if (EQ (key, Qicon_name))
+ {
+ struct gcpro gcpro1;
+ Lisp_Object filename = Qnil;
+ GCPRO1 (filename);
+ if (NILP (value))
+ {
+ /* Default to the GNU. */
+ filename = build_string ("emacs_48.png");
+ }
+ else
+ {
+ CHECK_STRING (value);
+ filename = value;
+ }
+ /* Not clear if x_find_image_file returns a string with the
+ correct encoding for Gtk. However, it is already used this
+ way in gtkutil.c. */
+ filename = x_find_image_file (filename);
+ if (NILP (filename))
+ {
+ /* No good choice, so choose a semi-appropriate stock
+ icon. */
+ gtk_status_icon_set_from_stock (GET_ICON (sicon),
+ GTK_STOCK_DIALOG_QUESTION);
+ }
+ else
+ gtk_status_icon_set_from_file (GET_ICON (sicon),
+ (char *) SDATA (filename));
+ UNGCPRO;
+ }
+ else if (EQ (key, Qblinking))
+ gtk_status_icon_set_blinking (GET_ICON (sicon), ! NILP (value));
+ else if (EQ (key, Qvisible))
+ gtk_status_icon_set_visible (GET_ICON (sicon), ! NILP (value));
+ else if (EQ (key, Qhelp_echo))
+ {
+ struct gcpro gcpro1;
+ Lisp_Object tooltip = Qnil;
+ GCPRO1 (tooltip);
+ if (NILP (value))
+ gtk_status_icon_set_tooltip (GET_ICON (sicon), NULL);
+ else
+ {
+ /* FIXME: should accept a sexp to eval here. */
+ CHECK_STRING (value);
+ tooltip = ENCODE_UTF_8 (value);
+ gtk_status_icon_set_tooltip (GET_ICON (sicon),
+ (char *) SDATA (tooltip));
+ }
+ UNGCPRO;
+ }
+ else
+ {
+ /* Ignore things we don't understand. */
+ }
+}
+
+/* Apply the ALIST of properties to status icon ICON. */
+static void
+apply_status_icon_alist (icon, alist)
+ Lisp_Object icon, alist;
+{
+ Lisp_Object tail;
+ struct status_icon *sicon;
+
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+
+ for (tail = alist; CONSP (tail); tail = XCDR (tail))
+ {
+ Lisp_Object key, value;
+ Lisp_Object elt = XCAR (tail);
+ if (! CONSP (elt))
+ continue;
+ key = XCAR (elt);
+ value = XCDR (elt);
+
+ apply_one_icon_property (sicon, key, value);
+
+ /* Update frame parameters. */
+ elt = Fassq (key, sicon->param_alist);
+ if (NILP (elt))
+ sicon->param_alist = Fcons (Fcons (key, value), sicon->param_alist);
+ else
+ Fsetcdr (elt, value);
+ }
+}
+
+DEFUN ("make-status-icon", Fmake_status_icon, Smake_status_icon,
+ 0, 1, 0,
+ doc: /* Return a newly created status icon.
+Optional argument ALIST is an alist of parameters for the new frame.
+See `modify-status-icon-parameters' for a list of recognized parameters.*/)
+ (alist)
+ Lisp_Object alist;
+{
+ struct status_icon *sicon;
+ Lisp_Object icon, deflist;
+ struct gcpro gcpro1;
+
+ BLOCK_INPUT;
+
+ sicon = allocate_status_icon ();
+ XSETSTATUS_ICON (icon, sicon);
+ sicon->param_alist = Qnil;
+ sicon->icon = gtk_status_icon_new ();
+
+ g_signal_connect (sicon->icon, "activate", G_CALLBACK (activate_icon),
+ (gpointer) icon);
+
+ Vstatus_icon_list = Fcons (icon, Vstatus_icon_list);
+
+ GCPRO1 (deflist);
+
+ /* Make some defaults. */
+ deflist = Fcons (Fcons (Qclick_callback, Qnil),
+ Fcons (Fcons (Qicon_name, Qnil),
+ Fcons (Fcons (Qblinking, Qnil),
+ Fcons (Fcons (Qvisible, Qt),
+ Fcons (Fcons (Qhelp_echo, Qnil),
+ alist)))));
+ apply_status_icon_alist (icon, deflist);
+
+ UNBLOCK_INPUT;
+
+ RETURN_UNGCPRO (icon);
+}
+
+DEFUN ("modify-status-icon-parameters", Fmodify_status_icon_parameters,
+ Smodify_status_icon_parameters,
+ 2, 2, 0,
+ doc: /* Modify the parameters of status icon ICON according to ALIST.
+Each element of alist has the form (PARM . VALUE), where PARM is a symbol.
+Undefined PARMs are ignored, but stored in the frame's parameter list
+so that `status-icon-parameters' will return them.
+
+Currently defined parameters and their values are:
+
+ icon-name The file name of an icon to display.
+ blinking If non-nil, the status icon will blink.
+ visible If nil, status icon is invisible.
+ help-echo A string which will be displayed as a tooltip for the
+ status icon.*/)
+ (icon, alist)
+ Lisp_Object icon, alist;
+{
+ BLOCK_INPUT;
+ apply_status_icon_alist (icon, alist);
+ UNBLOCK_INPUT;
+ return Qnil;
+}
+
+DEFUN ("status-icon-parameters", Fstatus_icon_parameters,
+ Sstatus_icon_parameters,
+ 1, 1, 0,
+ doc: /* Return the parameters-alist of the status icon ICON.
+It is a list of elements of the form (PARM . VALUE), where PARM is a symbol.*/)
+ (icon)
+ Lisp_Object icon;
+{
+ struct status_icon *sicon;
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+ return Fcopy_alist (sicon->param_alist);
+}
+
+DEFUN ("status-icon-list", Fstatus_icon_list, Sstatus_icon_list,
+ 0, 0, 0,
+ doc: /* Return a list of all status icons. */)
+ ()
+{
+ return Fcopy_sequence (Vstatus_icon_list);
+}
+
+DEFUN ("delete-status-icon", Fdelete_status_icon, Sdelete_status_icon,
+ 1, 1, 0,
+ doc: /* Delete the status icon ICON. */)
+ (icon)
+ Lisp_Object icon;
+{
+ struct status_icon *sicon;
+ CHECK_LIVE_STATUS_ICON (icon);
+ sicon = XSTATUS_ICON (icon);
+ BLOCK_INPUT;
+ g_object_unref (sicon->icon);
+ UNBLOCK_INPUT;
+ sicon->icon = NULL;
+ Vstatus_icon_list = Fdelq (icon, Vstatus_icon_list);
+ return Qnil;
+}
+
+DEFUN ("show-status-icon-message", Fshow_status_icon_message,
+ Sshow_status_icon_message,
+ 2, MANY, 0,
+ doc: /* Post a notification message attached to the icon ICON.
+ICON is a status icon created by `make-status-icon'.
+SUMMARY is the message to display, a string.
+
+The remaining arguments, if any, are a property list specifying
+additional parameters of the notification:
+
+ :body TEXT Set the body text of the notification.
+ :timeout VALUE Set the timeout value.
+ nil means use the default.
+ t means never time out.
+ An integer specifies the timeout in milliseconds.
+ :urgency VALUE Set the urgency level. VALUE is a symbol, one of
+ 'low', 'normal', 'critical'. The default is 'normal'.
+ :icon VALUE Set the notification icon. VALUE is a symbol, currently
+ one of 'warning', 'info', 'question', 'error'. The
+ default is 'info'. */)
+ (nargs, args)
+ int nargs;
+ Lisp_Object *args;
+{
+ struct status_icon *sicon;
+ struct gcpro gcpro1;
+ Lisp_Object gc_temp = Qnil;
+ GError *gerr = NULL;
+ NotifyNotification *notification;
+
+ int i;
+ Lisp_Object icon = args[0];
+ Lisp_Object summary = args[1];
+
+ if ((nargs - 2) % 2 != 0)
+ error ("Invalid number of arguments");
+
+ CHECK_LIVE_STATUS_ICON (icon);
+ CHECK_STRING (summary);
+
+ if (! notify_is_initted () && ! notify_init ("Emacs"))
+ error ("Couldn't connect to notification server");
+
+ sicon = XSTATUS_ICON (icon);
+
+ GCPRO1 (gc_temp);
+
+ gc_temp = ENCODE_UTF_8 (summary);
+
+ BLOCK_INPUT;
+
+ /* FIXME: allow actions, setting the icon, callback when the
+ notification is closed, maybe hints. */
+ notification
+ = notify_notification_new_with_status_icon (SDATA (gc_temp), NULL,
+ GTK_STOCK_DIALOG_INFO,
+ GET_ICON (XSTATUS_ICON (icon)));
+
+ for (i = 2; i < nargs; i += 2)
+ {
+ Lisp_Object key = args[i];
+ Lisp_Object value = args[i + 1];
+ if (EQ (key, QCbody))
+ {
+ /* Ignore things we don't understand. */
+ if (STRINGP (value))
+ {
+ gc_temp = ENCODE_UTF_8 (value);
+ g_object_set (G_OBJECT (notification), "body", SDATA (gc_temp),
+ NULL);
+ }
+ }
+ else if (EQ (key, QCtimeout))
+ {
+ gint timeout = NOTIFY_EXPIRES_DEFAULT;
+ if (EQ (value, Qt))
+ timeout = NOTIFY_EXPIRES_NEVER;
+ else if (INTEGERP (value))
+ {
+ timeout = XINT (value);
+ if (timeout <= 0)
+ timeout = NOTIFY_EXPIRES_DEFAULT;
+ }
+ notify_notification_set_timeout (notification, timeout);
+ }
+ else if (EQ (key, QCurgency))
+ {
+ NotifyUrgency urgency = NOTIFY_URGENCY_NORMAL;
+ if (EQ (value, Qlow))
+ urgency = NOTIFY_URGENCY_LOW;
+ else if (EQ (value, Qcritical))
+ urgency = NOTIFY_URGENCY_CRITICAL;
+ notify_notification_set_urgency (notification, urgency);
+ }
+ else if (EQ (key, QCicon))
+ {
+ char *icon_name = GTK_STOCK_DIALOG_INFO;
+ if (EQ (value, Qwarning))
+ icon_name = GTK_STOCK_DIALOG_WARNING;
+ else if (EQ (value, Qquestion))
+ icon_name = GTK_STOCK_DIALOG_QUESTION;
+ else if (EQ (value, Qerror))
+ icon_name = GTK_STOCK_DIALOG_ERROR;
+ g_object_set (G_OBJECT (notification), "icon-name", icon_name, NULL);
+ }
+ }
+
+ notify_notification_show (notification, &gerr);
+ /* FIXME: error handling? */
+ g_object_unref (notification);
+
+ UNBLOCK_INPUT;
+
+ RETURN_UNGCPRO (Qnil);
+}
+
+DEFUN ("status-icon-handle-click-event", Fstatus_icon_handle_click_event,
+ Sstatus_icon_handle_click_event,
+ 1, 1, "e",
+ doc: /* Internal handler for status icon click events.
+A status icon click event is generated in response to the user clicking
+on a status icon. This handler calls the specified callback function,
+if any.*/)
+ (event)
+ Lisp_Object event;
+{
+ struct status_icon *sicon;
+ Lisp_Object elt;
+
+ /* Just ignore garbage. */
+ if (! CONSP (event) || ! EQ (XCAR (event), Qstatus_icon_click_event)
+ || ! STATUS_ICONP (XCDR (event))
+ || ! STATUS_ICON_LIVE_P (XSTATUS_ICON (XCDR (event))))
+ return Qnil;
+
+ sicon = XSTATUS_ICON (XCDR (event));
+ elt = Fassq (Qclick_callback, sicon->param_alist);
+ if (CONSP (elt) && FUNCTIONP (XCDR (elt)))
+ {
+ struct gcpro gcpro1;
+ Lisp_Object fun = XCDR (elt);
+ GCPRO1 (fun);
+ Ffuncall (1, &fun);
+ UNGCPRO;
+ }
+
+ return Qnil;
+}
+
+void
+syms_of_systray ()
+{
+ Qstatus_icon_live_p = intern ("status-icon-live-p");
+ staticpro (&Qstatus_icon_live_p);
+ Qblinking = intern ("blinking");
+ staticpro (&Qblinking);
+ Qclick_callback = intern ("click-callback");
+ staticpro (&Qclick_callback);
+ QCbody = intern (":body");
+ staticpro (&QCbody);
+ QCtimeout = intern (":timeout");
+ staticpro (&QCtimeout);
+ QCurgency = intern (":urgency");
+ staticpro (&QCurgency);
+ QCicon = intern (":icon");
+ staticpro (&QCicon);
+ Qlow = intern ("low");
+ staticpro (&Qlow);
+ Qcritical = intern ("critical");
+ staticpro (&Qcritical);
+ Qwarning = intern ("warning");
+ staticpro (&Qwarning);
+ Qquestion = intern ("question");
+ staticpro (&Qquestion);
+ Qerror = intern ("error");
+ staticpro (&Qerror);
+
+ staticpro (&Vstatus_icon_list);
+ Vstatus_icon_list = Qnil;
+
+ defsubr (&Smake_status_icon);
+ defsubr (&Smodify_status_icon_parameters);
+ defsubr (&Sdelete_status_icon);
+ defsubr (&Sstatus_icon_parameters);
+ defsubr (&Sstatus_icon_list);
+ defsubr (&Sshow_status_icon_message);
+ defsubr (&Sstatus_icon_handle_click_event);
+}
Index: systray.h
===================================================================
RCS file: systray.h
diff -N systray.h
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ systray.h 30 Dec 2007 19:48:03 -0000
@@ -0,0 +1,52 @@
+/* Define systray object for GNU Emacs.
+ Copyright (C) 2007 Free Software Foundation, Inc.
+
+This file is part of GNU Emacs.
+
+GNU Emacs is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GNU Emacs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Emacs; see the file COPYING. If not, write to
+the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA. */
+
+#ifndef EMACS_SYSTRAY_H
+#define EMACS_SYSTRAY_H
+
+extern Lisp_Object Qstatus_icon_live_p;
+
+struct status_icon
+{
+ EMACS_UINT size;
+ struct Lisp_Vector *next;
+
+ /* Parameter alist of this status icon. */
+ Lisp_Object param_alist;
+
+ /* Beyond here, there should be no more Lisp_Object components. */
+
+ /* The icon widget, or NULL if it has been destroyed. Really a
+ 'GtkStatusIcon *'. */
+ void *icon;
+};
+
+#define XSTATUS_ICON(p) \
+ (eassert (GC_STATUS_ICONP (p)), (struct status_icon *) XPNTR (p))
+
+#define XSETSTATUS_ICON(a, b) (XSETPSEUDOVECTOR (a, b, PVEC_STATUS_ICON))
+
+#define STATUS_ICON_LIVE_P(s) (((s)->icon) != 0)
+
+#define CHECK_LIVE_STATUS_ICON(s) \
+ CHECK_TYPE (STATUS_ICONP (s) && STATUS_ICON_LIVE_P (XSTATUS_ICON (s)), \
+ Qstatus_icon_live_p, s)
+
+#endif /* EMACS_SYSTRAY_H */
Index: termhooks.h
===================================================================
RCS file: /sources/emacs/emacs/src/termhooks.h,v
retrieving revision 1.90
diff -u -r1.90 termhooks.h
--- termhooks.h 2 Dec 2007 16:23:40 -0000 1.90
+++ termhooks.h 30 Dec 2007 19:48:03 -0000
@@ -201,6 +201,8 @@
, DBUS_EVENT
#endif
+ , STATUS_ICON_CLICK_EVENT
+
#ifdef WINDOWSNT
/* Generated when an APPCOMMAND event is received, in response to
Multimedia or Internet buttons on some keyboards.
Index: Makefile.in
===================================================================
RCS file: /sources/emacs/emacs/src/Makefile.in,v
retrieving revision 1.361
diff -u -r1.361 Makefile.in
--- Makefile.in 2 Dec 2007 16:23:40 -0000 1.361
+++ Makefile.in 30 Dec 2007 19:47:59 -0000
@@ -593,7 +593,7 @@
/* lastfile must follow all files
whose initialized data areas should be dumped as pure by dump-emacs. */
obj= dispnew.o frame.o scroll.o xdisp.o $(XMENU_OBJ) window.o \
- charset.o coding.o category.o ccl.o \
+ charset.o coding.o category.o ccl.o systray.o \
cm.o term.o terminal.o xfaces.o $(XOBJ) $(GTK_OBJ) $(DBUS_OBJ) \
emacs.o keyboard.o macros.o keymap.o sysdep.o \
buffer.o filelock.o insdel.o marker.o \
@@ -944,7 +944,7 @@
Note that SunOS needs -lm to come before -lc; otherwise, you get
duplicated symbols. If the standard libraries were compiled
with GCC, we might need gnulib again after them. */
-LIBES = $(LOADLIBES) $(LIBS) $(LIBX) $(LIBSOUND) $(RSVG_LIBS) $(DBUS_LIBS) \
+LIBES = $(LOADLIBES) $(LIBS) -lnotify $(LIBX) $(LIBSOUND) $(RSVG_LIBS) $(DBUS_LIBS) \
LIBGPM LIBRESOLV LIBS_SYSTEM LIBS_MACHINE LIBS_TERMCAP \
LIBS_DEBUG $(GETLOADAVG_LIBS) $(GNULIB_VAR) LIB_MATH LIB_STANDARD \
$(GNULIB_VAR)
@@ -1190,6 +1190,7 @@
sysdep.o: sysdep.c syssignal.h systty.h systime.h syswait.h blockinput.h \
process.h dispextern.h termhooks.h termchar.h termopts.h \
frame.h atimer.h window.h msdos.h dosfns.h keyboard.h cm.h $(config_h)
+systray.o: systray.c $(config_h) lisp.h frame.h
term.o: term.c termchar.h termhooks.h termopts.h $(config_h) cm.h frame.h \
disptab.h dispextern.h keyboard.h charset.h coding.h ccl.h msdos.h \
window.h keymap.h blockinput.h atimer.h systime.h
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2007-12-30 19:56 Tom Tromey
@ 2007-12-31 17:18 ` Dan Nicolaescu
2007-12-31 18:29 ` Tom Tromey
0 siblings, 1 reply; 55+ messages in thread
From: Dan Nicolaescu @ 2007-12-31 17:18 UTC (permalink / raw)
To: Tom Tromey; +Cc: emacs-devel
Tom Tromey <tromey@redhat.com> writes:
> This patch implements status icon support for Emacs.
>
> A status icon is an icon that sits in a part of the panel called the
> "system tray". It can be used to display notifications that are
> somewhat less intrusive than a real dialog box. A status icon can
> also have a menu (unimplemented here), react to button presses, and
> can display notification messages.
>
> This patch only works when using Gtk. It could in theory be made to
> work with other toolkits -- there is a cross-desktop systray
> specification, and both Windows and the Mac support something similar.
> However, I don't know these other platforms, so someone else would
> have to implement that.
>
> This patch is still incomplete. It needs a configure change and some
> #ifdefs. It also needs a ChangeLog entry and, I think, a section in
> the elisp manual.
>
> I'm looking for feedback on the general approach, the code, etc.
Very interesting idea!
What would the lisp code for this look like?
Another possible use that comes to mind: emit some kind of notification
when a (long) compilation finishes.
^ permalink raw reply [flat|nested] 55+ messages in thread
* Re: RFC: status icon support
2007-12-31 17:18 ` Dan Nicolaescu
@ 2007-12-31 18:29 ` Tom Tromey
0 siblings, 0 replies; 55+ messages in thread
From: Tom Tromey @ 2007-12-31 18:29 UTC (permalink / raw)
To: Dan Nicolaescu; +Cc: emacs-devel
>>>>> "Dan" == Dan Nicolaescu <dann@ics.uci.edu> writes:
Dan> Very interesting idea!
Dan> What would the lisp code for this look like?
Use make-status-icon to make an icon. Then use
modify-status-icon-parameters to change it, or use show-status-icon-message
to display a notification.
E.g., for my ERC module I have the icon blink when someone mentions my
name in a channel. With the current API this would look like:
(modify-status-icon-parameters erc-status-status-icon '((blinking . t)))
It is also set up so that clicking on the status icon will switch to
the first buffer where someone has mentioned my nick:
(modify-status-icon-parameters erc-status-status-icon
'((click-callback . erc-status-select-first-buffer)))
erc-status.el isn't really ready yet though... it is still written
using the old API (using an external program). I didn't want to
rewrite it for real until something goes in.
Dan> Another possible use that comes to mind: emit some kind of notification
Dan> when a (long) compilation finishes.
Yeah. This is easy:
(defvar compilation-status-icon nil)
(defun compilation-finish-notification (buffer status)
(unless compilation-status-icon
;; No need for it to be visible; we just want to post
;; notifications for now.
(setq compilation-status-icon (make-status-icon '((visible . nil)))))
(let ((ok (string-match "finished" status)))
(show-status-icon-message compilation-status-icon
(if ok
"Compilation finished"
"Compilation failed")
:icon (if ok 'info 'warning))))
(add-to-list 'compilation-finish-functions #'compilation-finish-notification)
Currently I have code (using my old zenity-based approach) for ERC,
EMMS (a pause/play button), and calendar (clicking shows the calendar,
and in theory appointments are notifications, though this has never
really worked). Compilation mode would be useful (IMO particularly if
the compilation buffer is not visible; you could also animate the icon
during compilation). Mail notification would be easy. There are
probably other possibilities too; an emacs-based replacement for the
Gnome sticky notes applet, for instance, would be pretty simple.
The other thing I wanted to do is add menu support. This looks kind
of complicated, though. Also there's an oddity in that the current
frame at the time the menu pops up may or may not be related to the
display that the status icon appears on. Maybe this doesn't matter?
This all reminds me, though -- I need to implement status-icon-p and
status-icon-live-p.
A few more specific questions for Emacs maintainers:
* The current code is Gtk-only. Is this ok? Or do I need to get
someone to write Windows and MacOS ports first? (I don't know much
about those platforms and I have no idea whether the same
functionality works there or not.)
* Is the current API ok? modify-status-icon-parameters uses an alist,
like modify-frame-parameters. But show-status-icon-message uses
keyword arguments, since that seemed nicer there. How's that?
What about the names of the keywords and alist keys?
* I considered making a status icon a special kind of frame, but this
looked a lot more difficult. Plus, I wasn't sure what the benefit
would be. The current code is a bit strange though -- a status icon
appears on some display, currently just whatever the default is.
But if you have multiple displays you may want to have a status icon
on each. (It is unclear to me if that is expected to work on the
Gtk side -- the status icon code always seems to use the default
display, so we'd have to switch displays around in the display
manager to get this working.)
I'm not sure what to do about the display-related oddities in
general. x-close-connection should probably close the related
status icons (though we'd have to do the bookkeeping ourselves,
since AFAICT there is no way to get this info out of Gtk). For
things like erc-status or a compile-finish-function, it seems weird
to show status on an icon that may not be readily visible to the
user. (But maybe it is ok to punt on this one.)
* Or, to sum it all up: aside from the configury bits, what do you
want to see before you will check this in?
Tom
^ permalink raw reply [flat|nested] 55+ messages in thread
end of thread, other threads:[~2008-01-23 4:00 UTC | newest]
Thread overview: 55+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-01-11 23:28 RFC: status icon support Tom Tromey
2008-01-12 1:57 ` Dan Nicolaescu
2008-01-12 1:28 ` Tom Tromey
2008-01-12 1:38 ` Tom Tromey
2008-01-12 8:45 ` Ulrich Mueller
2008-01-12 17:45 ` Tom Tromey
2008-01-14 2:01 ` Richard Stallman
2008-01-14 1:35 ` Tom Tromey
2008-01-14 17:26 ` Richard Stallman
2008-01-19 5:18 ` Tom Tromey
2008-01-20 6:14 ` Richard Stallman
2008-01-23 4:00 ` Michael Olson
2008-01-14 1:41 ` Tom Tromey
2008-01-14 1:03 ` Michael Olson
2008-01-14 1:01 ` Tom Tromey
2008-01-14 7:03 ` Jan Djärv
2008-01-15 6:01 ` Michael Olson
2008-01-16 1:10 ` Tom Tromey
2008-01-16 4:10 ` Michael Olson
2008-01-12 11:11 ` Richard Stallman
2008-01-12 11:25 ` David Kastrup
2008-01-12 11:27 ` Andreas Schwab
2008-01-12 11:46 ` David Kastrup
2008-01-12 14:10 ` Andreas Schwab
2008-01-12 14:19 ` David Kastrup
2008-01-12 17:33 ` Andreas Schwab
2008-01-14 2:00 ` Richard Stallman
2008-01-14 2:25 ` Stefan Monnier
2008-01-14 7:05 ` Jan Djärv
2008-01-12 13:52 ` Dan Nicolaescu
2008-01-12 14:13 ` Andreas Schwab
2008-01-12 14:26 ` Dan Nicolaescu
2008-01-12 17:36 ` Andreas Schwab
2008-01-12 18:59 ` Dan Nicolaescu
2008-01-12 14:33 ` David Kastrup
2008-01-12 17:45 ` Andreas Schwab
2008-01-12 18:07 ` David Kastrup
2008-01-12 18:16 ` Andreas Schwab
2008-01-14 2:01 ` Richard Stallman
2008-01-14 2:47 ` Dan Nicolaescu
2008-01-14 17:26 ` Richard Stallman
2008-01-14 9:14 ` David Kastrup
2008-01-14 17:26 ` Richard Stallman
2008-01-14 3:47 ` YAMAMOTO Mitsuharu
2008-01-14 3:49 ` Tom Tromey
2008-01-14 13:35 ` Stefan Monnier
2008-01-14 21:40 ` YAMAMOTO Mitsuharu
2008-01-16 1:17 ` Tom Tromey
2008-01-16 11:55 ` YAMAMOTO Mitsuharu
2008-01-14 17:26 ` Richard Stallman
2008-01-14 17:10 ` Tom Tromey
2008-01-16 2:42 ` Richard Stallman
-- strict thread matches above, loose matches on Subject: below --
2007-12-30 19:56 Tom Tromey
2007-12-31 17:18 ` Dan Nicolaescu
2007-12-31 18:29 ` Tom Tromey
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs.git
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).