[Evolution-hackers] [Fwd: How to scroll one full page in message list view]
Found the following in evolution/mail/message-list.c. Seems like the vertical scrolling adjustment is left to default values. static void message_list_init (MessageList *message_list) { struct _MessageListPrivate *p; GtkAdjustment *adjustment; GdkAtom matom; adjustment = (GtkAdjustment *) gtk_adjustment_new (0.0, 0.0, G_MAXDOUBLE, 0.0, 0.0, 0.0); gtk_scrolled_window_set_vadjustment ((GtkScrolledWindow *) message_list, adjustment); gtk_scrolled_window_set_policy (GTK_SCROLLED_WINDOW (message_list), GTK_POLICY_NEVER, GTK_POLICY_ALWAYS); Weitergeleitete Nachricht Von: Thomas Mittelstaedt tmsta...@t-mittelstaedt.de Reply-to: tmsta...@t-mittelstaedt.de An: evolution-l...@gnome.org Betreff: How to scroll one full page in message list view Datum: Tue, 24 Feb 2009 01:43:08 +0100 Hallo, When clicking an area outside the scroll bar in the message list view, the display only moves by half a page or so, making it difficult at times to follow, if the list is very big, like a large newsgroup. How can I configure evolution or gtk or gnome to scroll one full page? Thank you in advance for hints. thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Evolution branched for GNOME 2.28
git branch -r does not list gnome-2-28 for the evolution project, and a merge failed, of course. For evolution-data-server and gtkhtml, it worked. Am Dienstag, den 11.08.2009, 20:20 +0530 schrieb Chenthill: Hi, We have made a early branching for evolution for GNOME 2.28. We have been discussing on this in the past and to have the context, please go through the following threads, http://mail.gnome.org/archives/evolution-hackers/2009-August/msg8.html and http://mail.gnome.org/archives/release-team/2009-June/msg00018.html The following modules have been branched, gtkhtml, evolution-data-server, evolution and evolution-exchange. The branch is gnome-2-28. Only critical fixes will be made for evolution-2.28. By critical fixes, I mean blockers and important bugs identified as must-fix by module maintainers. All freezes would apply for gnome-2-28 branch. Please note that these patches needs to be committed to master and gnome-2-28 branch. The kill-bonobo-branch and eds-dbus-port will be merged into master. So it means master will not have any freezes. Most of the active work will be going on master branch to get it usable for evolution 2.29.1. Once the branches are merged to trunk, we would be replying back to this thread mentioning the same. We have made a rough road-map for Evolution 3.0 - http://www.go-evolution.org/Evo3.0 which will be discussed and modified further. We will also be discussing over some pending tasks from Evolution 2.28 and include the ones which are possible for http://www.go-evolution.org/Evo2.28 . Thanks, Chenthill. ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] [timezone problem] Evolution calendar blew transition from CEST to CET
Well, as you might know, today there was the switch from savings time to CET here in Europe. At 3 am, clocks are reset to 2 am. Well, at about 1:40 am, I entered a new event into calendar, setting the alarm notifier to 1 minute before the event. But no alarm notifier occurred. I had forgotten that the time switch was about to occur. Then I added a new event, tried to set the time using the now button, but it gave the wrong time, one hour back to the actual time. I reconfigured tzdata, checked the timezone preferences in evolution, which was set to system time. Well, finally, when it was 3 am and my system clock would make the switch as expected, evolution would behave correctly again. Seems, as if evolution doesn't make the switch at the correct time! -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] Can't reach bugzilla.gnome.org for second straight day
Hallo, What's wrong with bugzilla.gnome.org and git.gnome.org. Can't reach them with a browser or ping them. Here is the traceroute output: traceroute to bugzilla.gnome.org (209.132.180.171), 30 hops max, 60 byte packets 1 fritz.box (xxx.xxx.xxx.xxx) 0.319 ms 0.742 ms 0.881 ms 2 rdsl-pass-de01.nw.mediaways.net (213.20.57.6) 21.220 ms 21.832 ms * 3 * * * 4 * * * 5 * * * 6 * * * 7 * * * 8 xmws-frnk-de16-chan-2.nw.mediaways.net (62.52.50.162) 38.633 ms 39.519 ms 40.743 ms 9 cr1.fra003.internap.net (80.81.194.45) 42.917 ms 43.979 ms * 10 * * * 11 * * * 12 lon005-wdc005-1069-cr1.wdc005.pnap.internap.net (66.79.147.37) 144.828 ms * * 13 * * acs007-wdc005-844-cr1.acs007.internap.net (66.79.147.197) 146.078 ms 14 * * * 15 * acs007-dal005-841-cr2.dal005.internap.net (66.79.147.190) 166.999 ms 167.033 ms 16 cr1-cr2.dal005.internap.net (66.79.147.185) 166.754 ms 167.463 ms 166.993 ms 17 dal005-phx007-833-cr2.phx007.internap.net (66.79.147.178) 190.942 ms 191.099 ms 191.173 ms 18 cr1-cr2.phx007.internap.net (66.79.147.173) 191.677 ms 191.048 ms 190.488 ms 19 phx007-phx003-847-core3.phx003.internap.net.144.79.66.in-addr.arpa (66.79.144.226) 190.939 ms 191.147 ms 191.106 ms 20 border2.po2-bbnet2.phx004.pnap.net (69.25.168.70) 191.685 ms 192.062 ms 192.416 ms 21 * * * 22 * * * 23 * * * 24 * * * 25 * * * 26 * * * 27 * * * 28 * * * 29 * * * 30 * * * -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] stacktraces, debug: How to link evolution with debug version of glib, gtk on ubuntu karmic
I try to get glib debug information for the glib libraries in backtraces on ubuntu karmic. Even though I installed the debug packages, gdb does not pick up the non-stripped library: lrwxrwxrwx 1 root root 23 2009-12-24 20:00 /lib/libglib-2.0.so.0 - libglib-2.0.so.0.2200.3 -rw-r--r-- 1 root root 743912 2009-12-03 15:00 /lib/libglib-2.0.so.0.2200.3 -rw-r--r-- 1 root root 1159711 2009-12-03 15:00 /usr/lib/debug/lib/libglib-2.0.so.0.2200.3 show debug-file-directory The directory where separate debug symbols are searched for is /usr/lib/debug/. I tried to set the symlink to point to the one in /usr/lib/debug, but that did not work. Do I have to build glib myself and install it in the same prefix as evolution? Thanks for help! -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] stacktraces, debug: How to link evolution with debug version of glib, gtk on ubuntu karmic
Well, In the case of libglib, I helped myself with the following: (gdb) inf shared libglib FromTo Syms Read Shared Object Library 0x055c8040 0x05628d78 Yes (*) /lib/libglib-2.0.so.0 (*): Shared library is missing debugging information. (gdb) add-symbol-file /usr/lib/debug/lib/libglib-2.0.so.0.2200.3 0x055c8040 -readnow To do this for all libraries loaded, is a little cumbersome, though. I am using gdb 7. Am Donnerstag, den 24.12.2009, 20:43 +0100 schrieb Thomas Mittelstaedt: I try to get glib debug information for the glib libraries in backtraces on ubuntu karmic. Even though I installed the debug packages, gdb does not pick up the non-stripped library: ... I tried to set the symlink to point to the one in /usr/lib/debug, but that did not work. Do I have to build glib myself and install it in the same prefix as evolution? Thanks for help! -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] stacktraces, debug: How to link evolution with debug version of glib, gtk on ubuntu karmic
Am Freitag, den 25.12.2009, 01:49 +0100 schrieb Tobias Mueller: Heya :) On 24.12.2009 20:43, Thomas Mittelstaedt wrote: I try to get glib debug information for the glib libraries in backtraces on ubuntu karmic. Even though I installed the debug packages, gdb does not pick up the non-stripped library: Sounds like a bug to me. How do you know it doesn't pick the debug symbols up though? Do I have to build glib myself and install it in the same prefix as evolution? Doesn't really matter where you install it, as long as the non-stripped binaries get pulled. Thus, you need to tell the linker to use the new library, i.e. with LD_LIBRARY_PATH or LD_PRELOAD. Well, do you mean compile-time or run-time. I just tried runtime, i.e. I did a set env LD_LIBRARY_PATH=/usr/lib/debug/lib:... in my .gdbinit. Now, gdb tells me FromTo Syms Read Shared Object Library 0x055c8040 0x05628d78 No /lib/libglib-2.0.so.0 without the star which would indicate that no debugging symbols were found. So, gdb found something, but when I did a bt, there were still those question marks for libglib. Additionally the program segfaulted shortly after. -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] Problems with Migration from 2.28.4 to 2.30 (master): calendar, task, memo not there
Hallo, Just tried the latest git version (master). Building and compiling went relatively smooth. After following advice at http://www.mail-archive.com/evolution-hackers@gnome.org/msg03339.html the app came up okay. Unfortunately, when I opened the calendar, the different calendars would show up in the list to the left, but no data is displayed, even though the ics files clearly exist and contain data. When I tried to import one of those ics files the app crashed. What is the correct way to migrate from 2.28.x to the latest version? -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Problems with Migration from 2.28.4 to 2.30 (master): calendar, task, memo not there
Am Samstag, den 13.03.2010, 14:35 -0500 schrieb Matthew Barnes: On Fri, 2010-03-12 at 20:13 +0100, Thomas Mittelstaedt wrote: Just tried the latest git version (master). Building and compiling went relatively smooth. After following advice at http://www.mail-archive.com/evolution-hackers@gnome.org/msg03339.html the app came up okay. Unfortunately, when I opened the calendar, the different calendars would show up in the list to the left, but no data is displayed, even though the ics files clearly exist and contain data. When I tried to import one of those ics files the app crashed. What is the correct way to migrate from 2.28.x to the latest version? The address book and calendar backend processes are now D-Bus services instead of Bonobo servers in 2.30. If you've installed your builds into a non-standard prefix (i.e. something other than /usr), then D-Bus probably doesn't know how to start the services and thus can't provide any data to Evolution. And Evolution isn't equipped to handle that gracefully at the moment, which would likely explain the crashes. If that's the case then you'll need to configure D-Bus so it can find the .service files you installed. Create a /etc/dbus-1/session-local.conf file and populate it with: !DOCTYPE busconfig PUBLIC -//freedesktop//DTD D-BUS Bus Configuration 1.0//EN http://www.freedesktop.org/standards/dbus/1.0/busconfig.dtd; busconfig servicedir/path/to/your/dbus-1/services/servicedir /busconfig Using an appropriate servicedir path for your particular setup. You may need to log out and log in again for the new settings to take effect. (There's probably a way to do it without having to log out, but I'm not sure how.) Hope this helps, Matthew Barnes Thank you, that worked!! -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] evolution master does not build
Hallo, In latest git master, I had to update the Makefile.am in evolution/capplet/settings to make it build: latest changeset: commit 04305b8fa35d87d97e48b11f93d24604cef5155d Author: Matthew Barnes mbar...@redhat.com Date: Mon Mar 15 12:11:26 2010 -0400 diff --git a/capplet/settings/Makefile.am b/capplet/settings/Makefile.am index c3df948..b7d7a34 100644 --- a/capplet/settings/Makefile.am +++ b/capplet/settings/Makefile.am @@ -11,6 +11,7 @@ AM_CPPFLAGS = \ -I$(top_srcdir)/shell \ -I$(top_srcdir)/smime/lib \ -I$(top_srcdir)/smime/gui \ + -I$(top_srcdir)/e-util \ $(GNOME_PLATFORM_CFLAGS)\ $(EVOLUTION_MAIL_CFLAGS)\ $(CERT_UI_CFLAGS) \ @@ -29,7 +30,7 @@ AM_CPPFLAGS = \ -DG_LOG_DOMAIN=\evolution-mail\ \ -DPACKAGE_LOCALE_DIR=\$(prefix)/$(DATADIRNAME)/locale\ \ -DPACKAGE_DATA_DIR=\$(datadir)\ - + AM_CFLAGS =\ -Wall\ -g -O0 @@ -64,6 +65,8 @@ libevolution_mail_settings_la_LIBADD = \ $(GNOME_PLATFORM_LIBS) \ $(top_builddir)/widgets/misc/libemiscwidgets.la \ $(top_builddir)/filter/libfilter.la \ - $(top_builddir)/mail/libevolution-mail.la + $(top_builddir)/mail/libevolution-mail.la \ + $(top_builddir)/e-util/libeutil.la \ + $(top_builddir)/shell/libeshell.la ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] evolution master does not build in evolution/modules/network-manager
Got the error: .libs/libevolution_module_network_manager_la-evolution-network-manager.o: In function `e_network_manager_register_type': /home/tom/src/evolution/git/obj/evolution/modules/network-manager/../../../../evolution/modules/network-manager/evolution-network-manager.c:54: undefined reference to `e_extension_get_type' The following is needed, i think: diff --git a/modules/network-manager/Makefile.am b/modules/network-manager/Makefile.am index 24cd80a..f7a01a6 100644 --- a/modules/network-manager/Makefile.am +++ b/modules/network-manager/Makefile.am @@ -13,6 +13,7 @@ libevolution_module_network_manager_la_SOURCES = \ libevolution_module_network_manager_la_LIBADD = \ $(top_builddir)/shell/libeshell.la \ + $(top_builddir)/e-util/libeutil.la \ $(GNOME_PLATFORM_LIBS) \ $(DBUS_GLIB_LIBS) ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] evolution 2.28 does not load calendars anymore
Milan, I just blew a couple of those G_DEFINE_TYPE's. Got to concentrate better. Weitergeleitete Nachricht Von: Evolution-Data-Server bugzi...@gnome.org An: tmsta...@t-mittelstaedt.de Betreff: [Bug 603854] evolution segfaults in e_book_new, when right-clicking in task list in calendar view Datum: Tue, 30 Mar 2010 18:08:14 + (UTC) https://bugzilla.gnome.org/show_bug.cgi?id=603854 Evolution-Data-Server | Contacts | 2.28.x Thomas tmstaedt changed: What|Removed |Added Attachment #157202|0 |1 is obsolete|| --- Comment #15 from Thomas tmsta...@t-mittelstaedt.de 2010-03-30 18:08:08 UTC --- Created an attachment (id=157521) View: https://bugzilla.gnome.org/attachment.cgi?id=157521 Review: https://bugzilla.gnome.org/review?bug=603854attachment=157521 Correction for the last patch In last patch I falsely did cast to G_TYPE_OBJECT for a couple of classes. The ugly consequence was, that no calendars came up anymore!! Dooohh. Bummer! Am Dienstag, den 30.03.2010, 09:28 +0200 schrieb Milan Crha: On Tue, 2010-03-30 at 05:03 +0200, Thomas Mittelstaedt wrote: There also is an error message: (evolution:6613): libecal-WARNING **: ../../../../evolution-data-server/calendar/libecal/e-cal.c:1046: Could not activate calendar factory (OAFIID:GNOME_Evolution_DataServer_CalFactory:1.2) I wiped out my whole /opt/evo directory and installed everything afresh. No change. Hi, there is definitely something with your bonobo activation server. Check if yours $PREFIX/lib/bonobo/servers/GNOME_Evolution_DataServer_1.2.server contains proper path to an evolution-data-server-2.28 within the CalFactory section, and whether you can run it on a console and whether your bonobo-activation-server knows about it (it doesn't, most likely). Try to search http://live.gnome.org/Evolution/FAQ for bonobo questions. Hope that helps, Milan ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] build Evolution with jhbuild - unsuccessfull result
Ivan, Am Donnerstag, den 01.04.2010, 19:03 +0300 schrieb Ivan A. Kostyuk: Hello Thomas. My environment - Ubuntu 8.04 64-bit Evolution 2.28 So, you are on ubuntu, have a 64-bit system and wish to build evolution 2.28, right? On http://live.gnome.org/Evolution/#Developer_Resources, there is a nice little link to http://mad-scientist.us/evolution.html, which uses a Makefile file rather than jhbuild to check out the sources from git and build it. When I started to build evolution a little more than a year ago, I did just that and it worked very nicely for me. Under How to use it, it says: By default, the makefile checks out and builds the latest version on the Git trunk for all packages. If you want to build the latest on a specific branch, set the branch make variable to the version number; e.g. branch := 2.28. Hope that helps. On Sun, 2010-03-28 at 21:30 +0200, Thomas Mittelstaedt wrote: Which version of evolution are you trying to build? On which operating system / distribution are you? Output below hints that you are on a 64-bit system, right? Am Sonntag, den 28.03.2010, 21:41 +0300 schrieb Ivan A. Kostyuk: Hello dear colleagues. I did the build with help of jhbuild I got following errors: Running aclocal-1.11... /home/WORK/opt/gnome2/share/aclocal/audiofile.m4:12: warning: underquoted definition of AM_PATH_AUDIOFILE /home/WORK/opt/gnome2/share/aclocal/audiofile.m4:12: run info '(automake)Extending aclocal' /home/WORK/opt/gnome2/share/aclocal/audiofile.m4:12: or see http://sources.redhat.com/automake/automake.html#Extending-aclocal configure.in:44: warning: AM_NLS is m4_require'd but not m4_defun'd m4/intltool.m4:27: IT_PROG_INTLTOOL is expanded from... configure.in:44: the top level Running autoconf... configure.in:44: warning: AM_NLS is m4_require'd but not m4_defun'd m4/intltool.m4:27: IT_PROG_INTLTOOL is expanded from... configure.in:44: the top level configure:3087: error: possibly undefined macro: AM_NLS If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation. *** Error during phase configure of gnome-doc-utils: ## Error running ./autogen.sh --prefix /home/WORK/opt/gnome2 --libdir '/home/WORK/opt/gnome2/lib64' --disable-static --disable-gtk-doc *** [8/68] *** Configuring icon-naming-utils *** [48/68] ./configure --prefix /home/WORK/opt/gnome2 --libdir '/home/WORK/opt/gnome2/lib64' --disable-static --disable-gtk-doc checking for a BSD-compatible install... /home/eventissimus/.local/bin/install-check checking whether build environment is sane... yes checking for a thread-safe mkdir -p... /bin/mkdir -p checking for gawk... no checking for mawk... mawk checking whether make sets $(MAKE)... yes checking whether to enable maintainer-specific portions of Makefiles... no checking for perl... /usr/bin/perl checking for XML::Simple... configure: error: XML::Simple perl module is required for icon-naming-utils *** Error during phase configure of icon-naming-utils: ## Error running ./configure --prefix /home/WORK/opt/gnome2 --libdir '/home/WORK/opt/gnome2/lib64' --disable-static --disable-gtk-doc *** [48/68] I gave up the modules and in the final I got that the following modules were not built: gnome-doc-utils gtk-doc glib libIDL ORBit2 cairo gobject-introspection pango atk shared-mime-info gtk+ dbus dbus-glib libbonobo libglade eggdbus polkit gconf gstreamer enchant libproxy libsoup pygobject libgweather PolicyKit hal gnome-keyring evolution-data-server icon-naming-utils gnome-icon-theme gtkhtml desktop-file-utils gnome-mime-data dbus-python pycairo pygtk avahi gamin gnome-vfs libgnome libgnomecanvas libbonoboui libgnomeui libnotify evolution What should I do in order to fix that modules? Thank you in advance, Ivan ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers -- thomas ___ Evolution-hackers mailing list Evolution-hackers@gnome.org http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Hey, experts, Just installed a fresh ubuntu maverick and try to build the master tree with Paul Smith's makefile. Unfortunately, it won't link. What do I miss? tuxdis...@ubuntu-desktop:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ ls e-book-marshal.c libebook_1_2_la-e-book-marshal.lo libebook_1_2_la-e-name-western.lo e-book-marshal.h libebook_1_2_la-e-book-query.lo libebook_1_2_la-e-vcard.lo libebook-1.2.la libebook_1_2_la-e-book-view.lo libebook-1.2.pc libebook_1_2_la-e-address-western.lo libebook_1_2_la-e-contact.lo libebook.pc libebook_1_2_la-e-book.lo libebook_1_2_la-e-destination.lo Makefile tuxdis...@ubuntu-desktop:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ /bin/bash ../../libtool --mode=install /usr/bin/install -c libebook-1.2.la '/opt/evo/lib' libtool: install: warning: relinking `libebook-1.2.la' libtool: install: (cd /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook; /bin/bash /home/tuxdistro/src/evolution/obj/evolution-data-server/libtool --silent --tag CC --mode=relink ccache gcc -g -version-info 13:1:3 -Wl,--no-undefined -o libebook-1.2.la -rpath /opt/evo/lib libebook_1_2_la-e-book-marshal.lo libebook_1_2_la-e-address-western.lo libebook_1_2_la-e-book-query.lo libebook_1_2_la-e-book-view.lo libebook_1_2_la-e-book.lo libebook_1_2_la-e-contact.lo libebook_1_2_la-e-destination.lo libebook_1_2_la-e-name-western.lo libebook_1_2_la-e-vcard.lo ../../addressbook/libegdbus/libegdbus-book.la ../../camel/libcamel-1.2.la ../../libedataserver/libedataserver-1.2.la -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lxml2 -lgconf-2 -lglib-2.0 -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lglib-2.0 -lnss3 -lnssutil3 -lsmime3 -lssl3 -lplds4 -lplc4 -lnspr4 -lpthread -ldl -lsqlite3 -lz -L/usr/lib -lkrb5 -lk5crypto -lcom_err -lgssapi_krb 5 ) .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' collect2: ld returned 1 exit status libtool: install: error: relink `libebook-1.2.la' with the above command before installing it -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
More errors trying to compile gtkhtml: Making all in gtkhtml make[2]: Entering directory `/home/tuxdistro/src/evolution/obj/gtkhtml/gtkhtml' CC htmlengine-edit-cursor.lo ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c: In function ‘draw_cursor_rectangle’: ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: error: implicit declaration of function ‘gdk_gc_new’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: warning: nested extern declaration of ‘gdk_gc_new’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: warning: assignment makes pointer from integer without a cast ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:131: error: implicit declaration of function ‘gdk_rgb_find_color’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:131: warning: nested extern declaration of ‘gdk_rgb_find_color’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:132: error: implicit declaration of function ‘gdk_gc_set_foreground’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:132: warning: nested extern declaration of ‘gdk_gc_set_foreground’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:135: error: implicit declaration of function ‘gdk_gc_set_background’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:135: warning: nested extern declaration of ‘gdk_gc_set_background’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:136: error: implicit declaration of function ‘gdk_gc_set_line_attributes’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:136: warning: nested extern declaration of ‘gdk_gc_set_line_attributes’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:137: error: implicit declaration of function ‘gdk_gc_set_dashes’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:137: warning: nested extern declaration of ‘gdk_gc_set_dashes’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:138: error: implicit declaration of function ‘gdk_draw_rectangle’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:138: warning: nested extern declaration of ‘gdk_draw_rectangle’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c: In function ‘html_engine_draw_cursor_in_area’: ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:344: error: implicit declaration of function ‘gdk_draw_line’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:344: warning: nested extern declaration of ‘gdk_draw_line’ make[2]: *** [htmlengine-edit-cursor.lo] Error 1 make[2]: Leaving directory `/home/tuxdistro/src/evolution/obj/gtkhtml/gtkhtml' Am Samstag, den 04.09.2010, 05:30 +0200 schrieb Thomas Mittelstaedt: Hey, experts, Just installed a fresh ubuntu maverick and try to build the master tree with Paul Smith's makefile. Unfortunately, it won't link. What do I miss? tuxdis...@ubuntu-desktop:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ ls e-book-marshal.c libebook_1_2_la-e-book-marshal.lo libebook_1_2_la-e-name-western.lo e-book-marshal.h libebook_1_2_la-e-book-query.lo libebook_1_2_la-e-vcard.lo libebook-1.2.la libebook_1_2_la-e-book-view.lo libebook-1.2.pc libebook_1_2_la-e-address-western.lo libebook_1_2_la-e-contact.lo libebook.pc libebook_1_2_la-e-book.lo libebook_1_2_la-e-destination.lo Makefile tuxdis...@ubuntu-desktop:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ /bin/bash ../../libtool --mode=install /usr/bin/install -c libebook-1.2.la '/opt/evo/lib' libtool: install: warning: relinking `libebook-1.2.la' libtool: install: (cd /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook; /bin/bash /home/tuxdistro/src/evolution/obj/evolution-data-server/libtool --silent --tag CC --mode=relink ccache gcc -g -version-info 13:1:3 -Wl,--no-undefined -o libebook-1.2.la -rpath /opt/evo/lib libebook_1_2_la-e-book-marshal.lo libebook_1_2_la-e-address-western.lo libebook_1_2_la-e-book-query.lo libebook_1_2_la-e-book-view.lo libebook_1_2_la-e-book.lo libebook_1_2_la-e-contact.lo libebook_1_2_la-e-destination.lo libebook_1_2_la-e-name-western.lo libebook_1_2_la-e-vcard.lo ../../addressbook/libegdbus/libegdbus-book.la ../../camel/libcamel-1.2.la ../../libedataserver/libedataserver-1.2.la -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lxml2 -lgconf-2 -lglib-2.0 -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lglib-2.0 -lnss3 -lnssutil3 -lsmime3 -lssl3 -lplds4 -lplc4 -lnspr4 -lpthread -ldl -lsqlite3 -lz -L/usr/lib -lkrb5 -lk5crypto -lcom_err -lgssapi_krb 5 ) .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' collect2: ld returned 1 exit status libtool: install: error: relink `libebook-1.2.la' with the above command before installing
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Montag, den 06.09.2010, 08:23 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 09:12 +0200, Thomas Mittelstaedt wrote: More errors trying to compile gtkhtml: Making all in gtkhtml make[2]: Entering directory `/home/tuxdistro/src/evolution/obj/gtkhtml/gtkhtml' CC htmlengine-edit-cursor.lo ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c: In function ‘draw_cursor_rectangle’: ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: error: implicit declaration of function ‘gdk_gc_new’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: warning: nested extern declaration of ‘gdk_gc_new’ ... Hi, is it with gtk3 or gtk2? The gtk3 isn't supported fully. I downloaded and built gtk 2.9, the latest from the gtk website, and installed it into /opt/evo, where I try to build evolution. I used Paul Smith's Makefile, which at some point in the configuration demanded gtk3. I just pulled the latest sources from master, the gtkhtml configuration script tells me, that it's using gtk2: Configuration: GTK+ package: gtk+-2.0 Still the same compilation error. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Montag, den 06.09.2010, 22:17 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 08:23 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 09:12 +0200, Thomas Mittelstaedt wrote: More errors trying to compile gtkhtml: Making all in gtkhtml make[2]: Entering directory `/home/tuxdistro/src/evolution/obj/gtkhtml/gtkhtml' CC htmlengine-edit-cursor.lo ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c: In function ‘draw_cursor_rectangle’: ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: error: implicit declaration of function ‘gdk_gc_new’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: warning: nested extern declaration of ‘gdk_gc_new’ ... Hi, is it with gtk3 or gtk2? The gtk3 isn't supported fully. I downloaded and built gtk 2.9, the latest from the gtk website, and installed it into /opt/evo, where I try to build evolution. I used Paul Smith's Makefile, which at some point in the configuration demanded gtk3. I just pulled the latest sources from master, the gtkhtml configuration script tells me, that it's using gtk2: Configuration: GTK+ package: gtk+-2.0 Still the same compilation error. Can you point me to some URL with instructions on how to build the latest evolution source apart from Paul Smith's page? What dependencies need to be installed? Should I rather not build the latest master source, but some tagged version? -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Montag, den 06.09.2010, 22:32 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 22:17 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 08:23 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 09:12 +0200, Thomas Mittelstaedt wrote: More errors trying to compile gtkhtml: Making all in gtkhtml make[2]: Entering directory `/home/tuxdistro/src/evolution/obj/gtkhtml/gtkhtml' CC htmlengine-edit-cursor.lo ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c: In function ‘draw_cursor_rectangle’: ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: error: implicit declaration of function ‘gdk_gc_new’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: warning: nested extern declaration of ‘gdk_gc_new’ ... Hi, is it with gtk3 or gtk2? The gtk3 isn't supported fully. I downloaded and built gtk 2.9, the latest from the gtk website, and installed it into /opt/evo, where I try to build evolution. I used Paul Smith's Makefile, which at some point in the configuration demanded gtk3. I just pulled the latest sources from master, the gtkhtml configuration script tells me, that it's using gtk2: Configuration: GTK+ package: gtk+-2.0 Still the same compilation error. Can you point me to some URL with instructions on how to build the latest evolution source apart from Paul Smith's page? What dependencies need to be installed? Should I rather not build the latest master source, but some tagged version? Found the following: http://ubuntuforums.org/showthread.php?t=1557826 Re: error: implicit declaration of function ‘gdk_gc_new’ I found that the file was already using gdk/gdk.h. I sovled the problem by removing: -GDK_DISABLE_DEPRECATED from configure.ac Re: error: implicit declaration of function ‘gdk_gc_new’ are you using the development or stable version of gtk? gdk_gc_new has not been deprecated in stable, so your change should not have had effect. Don't know about unstable. And even if, instead of enabling deprecated functions you should fix the code to not use them ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Montag, den 06.09.2010, 23:20 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 22:32 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 22:17 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 08:23 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 09:12 +0200, Thomas Mittelstaedt wrote: More errors trying to compile gtkhtml: Making all in gtkhtml make[2]: Entering directory `/home/tuxdistro/src/evolution/obj/gtkhtml/gtkhtml' CC htmlengine-edit-cursor.lo ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c: In function ‘draw_cursor_rectangle’: ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: error: implicit declaration of function ‘gdk_gc_new’ ../../../gtkhtml/gtkhtml/htmlengine-edit-cursor.c:129: warning: nested extern declaration of ‘gdk_gc_new’ ... Hi, is it with gtk3 or gtk2? The gtk3 isn't supported fully. I downloaded and built gtk 2.9, the latest from the gtk website, and installed it into /opt/evo, where I try to build evolution. I used Paul Smith's Makefile, which at some point in the configuration demanded gtk3. I just pulled the latest sources from master, the gtkhtml configuration script tells me, that it's using gtk2: Configuration: GTK+ package: gtk+-2.0 Still the same compilation error. Can you point me to some URL with instructions on how to build the latest evolution source apart from Paul Smith's page? What dependencies need to be installed? Should I rather not build the latest master source, but some tagged version? Found the following: http://ubuntuforums.org/showthread.php?t=1557826 Re: error: implicit declaration of function ‘gdk_gc_new’ I found that the file was already using gdk/gdk.h. I sovled the problem by removing: -GDK_DISABLE_DEPRECATED from configure.ac Re: error: implicit declaration of function ‘gdk_gc_new’ are you using the development or stable version of gtk? gdk_gc_new has not been deprecated in stable, so your change should not have had effect. Don't know about unstable. And even if, instead of enabling deprecated functions you should fix the code to not use them Also this one: https://bugzilla.gnome.org/show_bug.cgi?id=626605 ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Dienstag, den 07.09.2010, 07:06 -0400 schrieb Matthew Barnes: On Sat, 2010-09-04 at 09:12 +0200, Thomas Mittelstaedt wrote: More errors trying to compile gtkhtml: Use ./configure --disable-deprecated-warning-flags to work around the GdkGC errors. GTK+ deprecated GdkGC very late in the GNOME development cycle and we're not going to deal with it this close to a stable release. Passing this option won't be necessary once the version becomes 2.32. It's enabled by default for unstable versions, disabled by default for stable versions. ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers Thank you, that worked to build gtkhtml. Some oddity about autogen.sh, though: I usually executed autogen.sh from base_dir/obj/gtkhtml via PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g bash ./autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags But autogen.sh would not do its job, i.e. regenerating the configure script. So, it called the existing ../../gtkhtml/configure which was old. I had to go to the source directory, execute autogen.sh from there and interrupt it after it called configure, since this is supposed be done from the object directory. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Montag, den 06.09.2010, 08:18 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 05:30 +0200, Thomas Mittelstaedt wrote: .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' Hi, I suppose you do not have latest master sources of the evolution-data-server. The missing function is part of libedataserver/e-source-list.c and it links to libedataserver-1.2.la, which I see it used in the Makefile.am in addressbook/libebook. The function was added just recently. Bye, Milan ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers Indeed I did not have the latest sources even though I switched branches via git branch master followed by a git pull origing master I probably misread the message saying something like ... your tree is ahead by xxx commits. A git reset --hard origin/master fixed it. But still, evolution-data-server/addressbook/libebook does not install: make[1]: Verlasse Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' t...@ubuntu:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ make install make install-am make[1]: Betrete Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' make[2]: Betrete Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' test -z /opt/evo/lib || /bin/mkdir -p /opt/evo/lib /bin/bash ../../libtool --mode=install /usr/bin/install -c libebook-1.2.la '/opt/evo/lib' libtool: install: warning: relinking `libebook-1.2.la' libtool: install: (cd /home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook; /bin/bash /home/tom/src/evolution/obj/evolution-data-server/libtool --silent --tag CC --mode=relink ccache gcc -g -version-info 13:1:3 -Wl,--no-undefined -o libebook-1.2.la -rpath /opt/evo/lib libebook_1_2_la-e-book-marshal.lo libebook_1_2_la-e-address-western.lo libebook_1_2_la-e-book-query.lo libebook_1_2_la-e-book-view.lo libebook_1_2_la-e-book.lo libebook_1_2_la-e-contact.lo libebook_1_2_la-e-destination.lo libebook_1_2_la-e-name-western.lo libebook_1_2_la-e-vcard.lo ../../addressbook/libegdbus/libegdbus-book.la ../../camel/libcamel-1.2.la ../../libedataserver/libedataserver-1.2.la -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lxml2 -lgconf-2 -lglib-2.0 -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lglib-2.0 -lnss3 -lnssutil3 -lsmime3 -lssl3 -lplds4 -lplc4 -lnspr4 -lpthread -ldl -lsqlite3 -lz ) .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' collect2: ld returned 1 exit status libtool: install: error: relink `libebook-1.2.la' with the above command before installing it make[2]: *** [install-libLTLIBRARIES] Fehler 1 make[2]: Verlasse Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' make[1]: *** [install-am] Fehler 2 make[1]: Verlasse Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' Looking at the installed libraries, nm tells me that the function indeed is undefined: t...@ubuntu:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ nm /opt/evo/lib/libebook-1.2.so /opt/evo/lib/libedataserver-1.2.so |grep peek_def U e_source_list_peek_default_source 00015eed T e_source_list_peek_default_source But there are more functions like e_source_list... which are undefined according to nm, but about which the linker does not complain. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Freitag, den 10.09.2010, 06:30 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 08:18 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 05:30 +0200, Thomas Mittelstaedt wrote: .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' Hi, I suppose you do not have latest master sources of the evolution-data-server. The missing function is part of libedataserver/e-source-list.c and it links to libedataserver-1.2.la, which I see it used in the Makefile.am in addressbook/libebook. The function was added just recently. Bye, Milan ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers Indeed I did not have the latest sources even though I switched branches via git branch master followed by a git pull origing master I probably misread the message saying something like ... your tree is ahead by xxx commits. A git reset --hard origin/master fixed it. But still, evolution-data-server/addressbook/libebook does not install: make[1]: Verlasse Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' t...@ubuntu:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ make install make install-am make[1]: Betrete Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' make[2]: Betrete Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' test -z /opt/evo/lib || /bin/mkdir -p /opt/evo/lib /bin/bash ../../libtool --mode=install /usr/bin/install -c libebook-1.2.la '/opt/evo/lib' libtool: install: warning: relinking `libebook-1.2.la' libtool: install: (cd /home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook; /bin/bash /home/tom/src/evolution/obj/evolution-data-server/libtool --silent --tag CC --mode=relink ccache gcc -g -version-info 13:1:3 -Wl,--no-undefined -o libebook-1.2.la -rpath /opt/evo/lib libebook_1_2_la-e-book-marshal.lo libebook_1_2_la-e-address-western.lo libebook_1_2_la-e-book-query.lo libebook_1_2_la-e-book-view.lo libebook_1_2_la-e-book.lo libebook_1_2_la-e-contact.lo libebook_1_2_la-e-destination.lo libebook_1_2_la-e-name-western.lo libebook_1_2_la-e-vcard.lo ../../addressbook/libegdbus/libegdbus-book.la ../../camel/libcamel-1.2.la ../../libedataserver/libedataserver-1.2.la -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lxml2 -lgconf-2 -lglib-2.0 -pthread -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lglib-2.0 -lnss3 -lnssutil3 -lsmime3 -lssl3 -lplds4 -lplc4 -lnspr4 -lpthread -ldl -lsqlite3 -lz ) .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' collect2: ld returned 1 exit status libtool: install: error: relink `libebook-1.2.la' with the above command before installing it make[2]: *** [install-libLTLIBRARIES] Fehler 1 make[2]: Verlasse Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' make[1]: *** [install-am] Fehler 2 make[1]: Verlasse Verzeichnis '/home/tom/src/evolution/obj/evolution-data-server/addressbook/libebook' Looking at the installed libraries, nm tells me that the function indeed is undefined: t...@ubuntu:~/src/evolution/obj/evolution-data-server/addressbook/libebook$ nm /opt/evo/lib/libebook-1.2.so /opt/evo/lib/libedataserver-1.2.so |grep peek_def U e_source_list_peek_default_source 00015eed T e_source_list_peek_default_source But there are more functions like e_source_list... which are undefined according to nm, but about which the linker does not complain. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers I think, it's the -no-undefined linker flag that's causing problems on my ubuntu maverick box: diff --git a/configure.ac b/configure.ac index 0df5bdc..f4fd0ce 100644 --- a/configure.ac +++ b/configure.ac @@ -199,7 +199,8 @@ dnl -Wstrict-aliasing=2 AM_CPPFLAGS=$WARNING_FLAGS -fno-strict-aliasing AC_SUBST(AM_CPPFLAGS) -LDFLAGS=$LDFLAGS -Wl,--no-undefined +dnl LDFLAGS=$LDFLAGS -Wl,--no-undefined +LDFLAGS=$LDFLAGS dnl ** dnl Initialize maintainer mode
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Freitag, den 10.09.2010, 09:36 +0200 schrieb Thomas Mittelstaedt: Am Freitag, den 10.09.2010, 06:30 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 08:18 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 05:30 +0200, Thomas Mittelstaedt wrote: .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' Hi, I suppose you do not have latest master sources of the evolution-data-server. The missing function is part of libedataserver/e-source-list.c and it links to libedataserver-1.2.la, which I see it used in the Makefile.am in addressbook/libebook. The function was added just recently. Well, resolved it finally. The problem was the existing 2.30 installation on maverick. There was a libedataserverso in /usr/lib. Using -L/opt/evo/lib for autogen.sh solved that problem: PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib bash ../../evolution/autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags Then some test...-files failed to compile. I had to do the following: diff --git a/calendar/backends/file/Makefile.am b/calendar/backends/file/Makefile.am index c672157..b2e2fdf 100644 --- a/calendar/backends/file/Makefile.am +++ b/calendar/backends/file/Makefile.am @@ -44,6 +44,9 @@ test_interval_searches_LDADD = \ test_interval_searches_CPPFLAGS = \ $(AM_CPPFLAGS) \ + -I$(top_srcdir) \ + -I$(top_builddir) \ + -I$(top_srcdir)/calendar\ -I$(top_builddir)/calendar \ $(EVOLUTION_CALENDAR_CFLAGS)\ -DTEST_QUERY_RESULT=1 diff --git a/calendar/libedata-cal/Makefile.am b/calendar/libedata-cal/Makefile.am index f88e24e..182d68d 100644 --- a/calendar/libedata-cal/Makefile.am +++ b/calendar/libedata-cal/Makefile.am @@ -121,12 +121,15 @@ e_calendar_factory_LDADD = \ test_e_sexp_SOURCES = e-cal-backend-sexp.c e-cal-backend-sexp.h test_e_sexp_CPPFLAGS = \ $(AM_CPPFLAGS) \ + -I$(top_srcdir) \ -I$(top_srcdir)/calendar\ -I$(top_builddir)/calendar \ $(EVOLUTION_CALENDAR_CFLAGS)\ -DTESTER=1 -test_e_sexp_LDADD = libedata-cal-1.2.la $(E_DATA_SERVER_LIBS) +test_e_sexp_LDADD = libedata-cal-1.2.la \ + $(top_builddir)/libedataserver/libedataserver-1.2.la +###$(E_DATA_SERVER_LIBS) test_intervaltree_SOURCES = test-intervaltree.c e-cal-backend-intervaltree.c === Finally, I got evolution 2.31 to run, but the restoration of ics files from my 2.28 version is a problem. The memo (notes) ics-files are somehow not recognized, even if they are set correctly in the gconf keys. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Samstag, den 11.09.2010, 02:30 +0200 schrieb Thomas Mittelstaedt: Am Freitag, den 10.09.2010, 09:36 +0200 schrieb Thomas Mittelstaedt: Am Freitag, den 10.09.2010, 06:30 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 06.09.2010, 08:18 +0200 schrieb Milan Crha: On Sat, 2010-09-04 at 05:30 +0200, Thomas Mittelstaedt wrote: .libs/libebook_1_2_la-e-book.o: In function `e_book_new_default_addressbook': /home/tuxdistro/src/evolution/obj/evolution-data-server/addressbook/libebook/../../../../evolution-data-server/addressbook/libebook/e-book.c:3329: undefined reference to `e_source_list_peek_default_source' ... Finally, I got evolution 2.31 to run, but the restoration of ics files from my 2.28 version is a problem. The memo (notes) ics-files are somehow not recognized, even if they are set correctly in the gconf keys. I tried the restoration using the installed version 2.30, having made sure, that services in /usr/lib are called instead of the ones of 2.31 in /opt/evo, and the app would restore memo ics files correctly, plus tasks and calendars. Then I tried to use this new .evolution directory with 2.31, having switched to the right services again and 2.31 doesn't show data for those memos, even though the list items in the left pane show up, but no entries. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Can't build evolution-data-server master on ubuntu maverick
Am Montag, den 13.09.2010, 08:36 +0200 schrieb Milan Crha: On Sat, 2010-09-11 at 03:45 +0200, Thomas Mittelstaedt wrote: Finally, I got evolution 2.31 to run, but the restoration of ics files from my 2.28 version is a problem. The memo (notes) ics-files are somehow not recognized, even if they are set correctly in the gconf keys. I tried the restoration using the installed version 2.30, having made sure, that services in /usr/lib are called instead of the ones of 2.31 in /opt/evo, and the app would restore memo ics files correctly, plus tasks and calendars. Then I tried to use this new .evolution directory with 2.31, having switched to the right services again and 2.31 doesn't show data for those memos, even though the list items in the left pane show up, but no entries. Hi, actual master, 2.31.92 very soon, doesn't use ~/.evolution, but XDG folders for its data. Everything is migrated on the first run, when ~/.evolution exists, but those XDG folders/files not. (It tries it every start, and complains on console). For your file backends were changed protocol from file:// to local:, thus the gconf key differs. Because you run 2.30 after 2.31, then I suppose you've duplicated On This Computer/Personal sources in calendar/task/memo/addressbook. Also see release notes at: http://mail.gnome.org/archives/ftp-release-list/2010-August/msg00023.html Bye, Milan Thank you, Milan. That was the problem. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Evolution 2.91 installation error
Am Donnerstag, den 04.11.2010, 12:19 +0100 schrieb Stefano Facchini: Hi, thanks for a message. I tried to fix this in commit fd75cdf in evolution Maybe can this depend on the fact that I have an Evolution 2.30 installed from ubuntu repository? Could well be. I had the same problem and I did a LDFLAGS=-L/opt/evo/lib when I called autogen: cd evo-src/evolution PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib NOCONFIGURE=y bash ./autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags cd ../obj/evolution PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib bash ../../evolution/autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags Just built the whole thing and it ran through except for some help files at the end. Do a make -k to get to the end. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Evolution 2.91 installation error
Am Freitag, den 05.11.2010, 10:45 +0100 schrieb Stefano Facchini: Maybe can this depend on the fact that I have an Evolution 2.30 installed from ubuntu repository? Could well be. I had the same problem and I did a LDFLAGS=-L/opt/evo/lib when I called autogen: cd evo-src/evolution PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib NOCONFIGURE=y bash ./autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags cd ../obj/evolution PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib bash ../../evolution/autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags Just built the whole thing and it ran through except for some help files at the end. Do a make -k to get to the end. -- thomas Hi, are you suggesting to install evolution in /opt/evo or should I change it with /usr/local (where I put my compiled evolution-data-server)? I think the latter, but it didn't work. Here's what I did: /usr/local should be okay, too. I just chose /opt/evo in order to have a completely separate tree which I can blow away with rm -rf. You can't do that with /usr/local, since there usually is other software installed. $ make clean $ PKG_CONFIG_PATH=/usr/local/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/usr/local/lib NOCONFIGURE=y bash ./autogen.sh --prefix='/usr/local' --disable-deprecated-warning-flags $ make $ sudo make install and then the same error complaining about camel. What are the exact error messages? Do you compile in separate object directories as I outlined above. Remove all existing evo-src/obj/gtkhml, evo-src/obj/evolution-data-server and evo-src/obj/evolution directories, recreate them, and do the autogen, configure and build. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Evolution 2.91 installation error
Am Freitag, den 05.11.2010, 12:28 +0100 schrieb Thomas Mittelstaedt: Am Freitag, den 05.11.2010, 10:45 +0100 schrieb Stefano Facchini: Maybe can this depend on the fact that I have an Evolution 2.30 installed from ubuntu repository? Could well be. I had the same problem and I did a LDFLAGS=-L/opt/evo/lib when I called autogen: cd evo-src/evolution PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib NOCONFIGURE=y bash ./autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags cd ../obj/evolution PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib bash ../../evolution/autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags Just built the whole thing and it ran through except for some help files at the end. Do a make -k to get to the end. -- thomas Hi, are you suggesting to install evolution in /opt/evo or should I change it with /usr/local (where I put my compiled evolution-data-server)? I think the latter, but it didn't work. Here's what I did: /usr/local should be okay, too. I just chose /opt/evo in order to have a completely separate tree which I can blow away with rm -rf. You can't do that with /usr/local, since there usually is other software installed. $ make clean $ PKG_CONFIG_PATH=/usr/local/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/usr/local/lib NOCONFIGURE=y bash ./autogen.sh --prefix='/usr/local' --disable-deprecated-warning-flags $ make $ sudo make install and then the same error complaining about camel. What are the exact error messages? Do you compile in separate object directories as I outlined above. Remove all existing evo-src/obj/gtkhml, evo-src/obj/evolution-data-server and evo-src/obj/evolution directories, recreate them, and do the autogen, configure and build. Someone else asked for help compiling evolution 2.32 and above on the evolution list and I answered the following. Maybe, this is of help for some people here, too: Weitergeleitete Nachricht Von: Thomas Mittelstaedt tmsta...@t-mittelstaedt.de Reply-to: tmsta...@t-mittelstaedt.de An: Victoria Spagnolo v.spagn...@pobox.com Kopie: evolution-l...@gnome.org Betreff: Re: [Evolution] learning to compile, need help with finding dependecies in Ubuntu 10.04 Datum: Mon, 01 Nov 2010 09:26:34 +0100 Victoria, Am Montag, den 01.11.2010, 19:09 +1300 schrieb Victoria Spagnolo: Reid and Han, Thanks for the msg suggesting Have you tried Paul's Makefile? http://mad-scientist.us/evolution.html Last time I tried this under Ubuntu it was reasonably clear which packages needed to be installed. Haven't tried it in a while though... Yes, I got that straightaway. But it is not for lucid. DISTROS :=feisty gutsy hardy intrepid jaunty karmic \ etch I've asked at #ubuntu and searched and didn't come up with any thing to try. anyone? I would suggest to you to try to build the gnome-2-32 branch, but using a more direct approach than the above Makefile. I just tried it on my ubuntu maverick vmware and it gave me strange errors like a missing gtk+-3. A few weeks ago, I built the master and gnome-2-32 branches on this box and I'll give you an outline of what I did: I am only using gtkhtml, evolution-data-server and evolution projects. If you need others, like libgweather or evolution-exchange or evolution-mapi you need to check them out and build them as well. See http://git.gnome.org/browse/. So I created a base directory like ~/evo-src cd evo-src git clone git://git.gnome.org/gtkhtml -b gnome-2-32 git clone git://git.gnome.org/evolution-data-server -b gnome-2-32 git clone git://git.gnome.org/evolution -b gnome-2-32 mkdir obj/gtkhtml cd gtkhtml # The following does not execute the configure yet PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib NOCONFIGURE=y bash ./autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags # Now actuall execute configure cd ../obj/gtkhtml PKG_CONFIG_PATH=/opt/evo/lib/pkgconfig/ CC='ccache gcc' CFLAGS=-g LDFLAGS=-L/opt/evo/lib bash ../../gtkhtml/autogen.sh --prefix='/opt/evo' --disable-deprecated-warning-flags make make install Do the same thing for evolution-data-server and evolution and the other modules. In the case of evolution I had to do a make -k to let it run through, the build would complain about some missing help files. You may have to install some missing packages: A sudo apt-get build-dep evolution should install most of them. I had to do a sudo apt-get install gobject-introspection as well. Hope that helps. -- thomas
Re: [Evolution-hackers] evolution-2.32.1 build error
Am Dienstag, den 07.12.2010, 22:46 +0100 schrieb Sasa Ostrouska: On Tue, Dec 7, 2010 at 10:16 PM, Sasa Ostrouska cas...@gmail.com wrote: On Tue, Dec 7, 2010 at 4:20 PM, Reid Thompson reid.thomp...@ateb.com wrote: On 12/07/2010 10:11 AM, Sasa Ostrouska wrote: Hi, again me. I'm getting this time the following error during building up evolution 2.32.1: cc1: warning: /usr/include/drm: No such file or directory i had to create a link on my system lrwxrwxrwx 1 root root 19 Nov 18 09:44 drm - /usr/include/libdrm Ok, but this wont fix th error, because this is just an warning. Any idea about the error ? It can't find libeutil.so.0 because it is still compilling it , if I understand it correctly. Rgds Saxa Running make -n gives me that. make[3]: Entering directory `/home/sasa/evolution-2.32.1/libgnomecanvas' make[3]: Nothing to be done for `all-am'. make[3]: Leaving directory `/home/sasa/evolution-2.32.1/libgnomecanvas' make[2]: Leaving directory `/home/sasa/evolution-2.32.1/libgnomecanvas' Making all in e-util make[2]: Entering directory `/home/sasa/evolution-2.32.1/e-util' make all-am make[3]: Entering directory `/home/sasa/evolution-2.32.1/e-util' echo CCLD libeutil.la;/bin/sh ../libtool --silent --tag=CC --mode=link gcc -g -O2 -fno-strict-aliasing -no-undefined -Wl,--no-undefined -o libeutil.la -rpath /usr/lib/evolution/2.32 libeutil_la-e-account-utils.lo libeutil_la-e-activity.lo libeutil_la-e-alert.lo libeutil_la-e-alert-activity.lo libeutil_la-e-alert-dialog.lo libeutil_la-e-binding.lo libeutil_la-e-bit-array.lo libeutil_la-e-categories-config.lo libeutil_la-e-charset.lo libeutil_la-e-config.lo libeutil_la-e-datetime-format.lo libeutil_la-e-dialog-utils.lo libeutil_la-e-dialog-widgets.lo libeutil_la-e-event.lo libeutil_la-e-extensible.lo libeutil_la-e-extension.lo libeutil_la-e-file-utils.lo libeutil_la-e-html-utils.lo libeutil_la-e-icon-factory.lo libeutil_la-e-import.lo libeutil_la-e-io-activity.lo libeutil_la-e-logger.lo libeutil_la-e-marshal.lo libeutil_la-e-mktemp.lo libeutil_la-e-module.lo libeutil_la-e-non-intrusive-error-dialog.lo libeutil_la-e-poolv.lo libeutil_la-e-plugin.lo libeutil_la-e-plugin-ui.lo libeutil_la-e-plugin-util.lo libeutil_la-e-print.lo libeutil_la-e-profile-event.lo libeutil_la-e-selection.lo libeutil_la-e-signature.lo libeutil_la-e-signature-list.lo libeutil_la-e-signature-utils.lo libeutil_la-e-sorter.lo libeutil_la-e-sorter-array.lo libeutil_la-e-text-event-processor-emacs-like.lo libeutil_la-e-text-event-processor.lo libeutil_la-e-timeout-activity.lo libeutil_la-e-ui-manager.lo libeutil_la-e-util.lo libeutil_la-e-unicode.lo libeutil_la-e-xml-utils.lo libeutil_la-gconf-bridge.lo -pthread -Wl,-rpath, -L/usr/lib64/nspr -ledataserverui-1.2 -lgtk-x11-2.0 -lebook-1.2 -lgdk-x11-2.0 -latk-1.0 -lpangoft2-1.0 -lpangocairo-1.0 -lgdk_pixbuf-2.0 -lm -lcairo -lpng14 -lpango-1.0 -lfreetype -lfontconfig -lcamel-1.2 -lssl3 -lsmime3 -lnss3 -ledataserver-1.2 -lsqlite3 -lxml2 -lgconf-2 -lsoup-2.4 -lgio-2.0 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lglib-2.0 -lplds4 -lplc4 -lnspr4 -lpthread -ldl-lm -pthread -lgailutil -lgconf-2 -lxml2 -lgnome-desktop-2 -lstartup-notification-1 -lunique-1.0 -lgtk-x11-2.0 -lgdk-x11-2.0 -latk-1.0 -lgio-2.0 -lpangoft2-1.0 -lpangocairo-1.0 -lgdk_pixbuf-2.0 -lm -lcairo -lpng14 -lpango-1.0 -lfreetype -lfontconfig -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lglib-2.0 make[3]: Leaving directory `/home/sasa/evolution-2.32.1/e-util' make[2]: Leaving directory `/home/sasa/evolution-2.32.1/e-util' Making all in a11y make[2]: Entering directory `/home/sasa/evolution-2.32.1/a11y' echo CC libevolution_a11y_la-gal-a11y-util.lo;/bin/sh ../libtool --silent --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I.. -DG_DISABLE_DEPRECATED -DPANGO_DISABLE_DEPRECATED -DGDK_PIXBUF_DISABLE_DEPRECATED -DGTK_DISABLE_DEPRECATED -DEDS_DISABLE_DEPRECATED -DG_DISABLE_SINGLE_INCLUDES -DGTK_DISABLE_SINGLE_INCLUDES -Wall -Wextra -Wno-missing-field-initializers -Wno-sign-compare -Wno-unused-parameter -Wdeclaration-after-statement -Werror-implicit-function-declaration -Wformat-security -Winit-self -Wmissing-declarations -Wmissing-include-dirs -Wmissing-noreturn -Wnested-externs -Wpointer-arith -Wredundant-decls -Wundef -Wwrite-strings -fno-strict-aliasing -pthread -DQT_SHARED -DORBIT2=1 -I/usr/include/glib-2.0 -I/usr/lib64/glib-2.0/include -I/usr/include/gtk-2.0 -I/usr/lib64/gtk-2.0/include -I/usr/include/atk-1.0 -I/usr/include/cairo -I/usr/include/gdk-pixbuf-2.0 -I/usr/include/pango-1.0 -I/usr/include/pixman-1 -I/usr/include/freetype2 -I/usr/include/libpng14 -I/usr/lib64/qt/include/QtGui -I/usr/include/drm -I/usr/include/libdrm -I/usr/lib64/qt/include/QtCore -I/usr/include/gail-1.0 -I/usr/include/gconf/2 -I/usr/include/orbit-2.0 -I/usr/include/libxml2 -I/usr/include/gnome-desktop-2.0 -I/usr/include/startup-notification-1.0
Re: [Evolution-hackers] evolution-2.32.1 build error
Am Mittwoch, den 08.12.2010, 14:37 +0100 schrieb Sasa Ostrouska: On Wed, Dec 8, 2010 at 2:10 PM, Thomas Mittelstaedt tmsta...@t-mittelstaedt.de wrote: libevolution_a11y_la-gal-a11y-util.lo `test -f 'gal-a11y-util.c' || echo './'`gal-a11y-util.c mv -f .deps/libevolution_a11y_la-gal-a11y-util.Tpo .deps/libevolution_a11y_la-gal-a11y-util.Plo make[2]: *** No rule to make target `../e-util/libeutil.la', needed by `libevolution-a11y.la'. Stop. make[2]: Leaving directory `/home/sasa/evolution-2.32.1/a11y' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/sasa/evolution-2.32.1' make: *** [all] Error 2 s...@quadser:~/evolution-2.32.1$ Any ideas ? Try run autoconf again. See also: http://mail.gnome.org/archives/evolution-list/2010-November/msg4.html. Hope that helps, -- thomas thomas, thanks, I looked at the message and it doesnt suggest me anything. I have also tried to do an autoreconf -ivf but it doesn't help. It suggests to you to try a clean setup from scratch. See if that works. Create a fresh src directory, check out the sources, create that object dir as outlined in the above message, install the software into a clean directory like /opt/evo and then see if that still gives you errors. See also the follow-up message, http://mail.gnome.org/archives/evolution-list/2010-November/msg5.html. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Evolution 2.28 obsolete?
Am Donnerstag, den 27.01.2011, 16:04 + schrieb Matt Davey: I'm not sure this is the right list, as this is more of a policy question than a dev question. I recently logged a bug (#639970) that crops up every now and again for me, and it was closed as OBSOLETE because I raised it against Evolution 2.28. This surprised me, because this 2.28 is the current shipping version for Ubuntu 10.04 LTS (Long Term Support). Now, you could argue that: a) the gnome/evo team can't be blamed for Ubuntu shipping an obsolete version, b) the gnome/evo team simply doesn't have the resources to maintain older versions, so it's reasonable to mark older trees unsupported pretty quickly, and c) Ubuntu should be responsible for fixing the problem, or upgrade to v2.30 or whatever I think these are good arguments from the point of view of the developer, but it's not good news for the user, and it's hard for the average user to distinguish Ubuntu from Gnome in this situation. I'm not sure what the solution is, but if 2.28 cannot be maintained it would still have felt less frustrating and more reasonable if: a) I'd been told whether the bug was a known bug and has been fixed in a more recent version of evo. This would potentially give a route for downstream to port a bug fix, or to contribute a 2.28 patch. b) If it's not a known bug, it might have made sense to recommend I report this to Ubuntu rather than Gnome, and why. I really like the long term release model for my work machine, and Evolution is critical. It's not a great feeling to know that it's an obsolete version. Just my two cents. Matt I'd suggest to you to build version 2.32 from git source on your ubuntu LTS. See http://mail.gnome.org/archives/evolution-list/2010-November/msg4.html and http://mail.gnome.org/archives/evolution-list/2010-November/msg5.html. If you are not familiar with building software from source, I or maybe a few others on this list could assist you. If you would like to continue to use version 2.28 as I do currently, you may check out the patches I use at http://www.t-mittelstaedt.de/evo-patches-2-28.tar.gz and http://www.t-mittelstaedt.de/evo-eds-patches-2-28.tar.gz for evolution and evolution-data-server, respectively. You need a software called stacked git, stg, to apply them. It is available with ubuntu. Hope, that helps, -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Building Evolution
Am Sonntag, den 06.03.2011, 10:05 +0530 schrieb Manu Gupta: I am sorry for extending this thread, I am attaching the jhbuildrc which I made and I get this error checking for SCANNER... yes checking for FFI... no checking for ffi.h... configure: error: ffi.h not found In the configure phase of gobject-introspection. What should I do? Please see if my .jhbuildrc is correct / am I missing something. On Sun, 2011-03-06 at 07:57 +0530, Manu Gupta wrote: Hi All I need to get bleeding edge evolution from the git repo. I am trying to use jhbuild but the docs at the website are quite outdated http://projects.gnome.org/evolution/build.shtml Can I know the proper moduleset for building evolution ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers The following might be of help: http://mail.gnome.org/archives/evolution-list/2010-November/msg4.html http://mail.gnome.org/archives/evolution-list/2010-November/msg5.html -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] Need assistance building evolution master (3.1.xx) on ubuntu oneiric alpha
Just did a git pull of the latest master sources for evolution and evolution-data-server, built the whole thing and installed it into /opt/evo. Then moved the existing dbus service out of the way and symlinked them to my install directory. Unfortunately, evolution never comes up, with the output below. The existing 3.1.4 version on this oneiric alpha comes up with the new services, though. What do I need to do to build and install it correctly? tom@ubuntu:/opt/evo/libexec$ file /usr/lib/evolution/e-calendar-factory /usr/lib/evolution/e-calendar-factory: symbolic link to `/opt/evo/libexec/e-calendar-factory' (evolution:28458): evolution-shell-CRITICAL **: shell_settings_pspec_for_key: assertion `schema_name != NULL' failed (evolution:28458): GLib-GObject-WARNING **: cannot register existing type `CompEditor' (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed (evolution:28458): GLib-GObject-WARNING **: cannot register existing type `EMeetingStore' (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed (evolution:28458): GLib-GObject-WARNING **: cannot register existing type `EMeetingTimeSelector' (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed (evolution:28458): GLib-GObject-WARNING **: cannot register existing type `ECalModel' (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed (evolution:28458): GLib-GObject-WARNING **: cannot register existing type `ECalendarView' (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-GObject-CRITICAL **: g_type_add_interface_static: assertion `G_TYPE_IS_INSTANTIATABLE (instance_type)' failed (evolution:28458): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed [New Thread 0xb61fbb70 (LWP 28471)] [Thread 0xb61fbb70 (LWP 28471) exited] -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Need assistance building evolution master (3.1.xx) on ubuntu oneiric alpha
Am Donnerstag, den 04.08.2011, 07:37 +0200 schrieb Milan Crha: On Wed, 2011-08-03 at 23:17 +0200, Thomas Mittelstaedt wrote: Just did a git pull of the latest master sources for evolution and evolution-data-server, built the whole thing and installed it into /opt/evo. Then moved the existing dbus service out of the way and symlinked them to my install directory. Unfortunately, evolution never comes up, with the output below. The existing 3.1.4 version on this oneiric alpha comes up with the new services, though. What do I need to do to build and install it correctly? tom@ubuntu:/opt/evo/libexec$ file /usr/lib/evolution/e-calendar-factory /usr/lib/evolution/e-calendar-factory: symbolic link to `/opt/evo/libexec/e-calendar-factory' (evolution:28458): evolution-shell-CRITICAL **: shell_settings_pspec_for_key: assertion `schema_name != NULL' failed (evolution:28458): GLib-GObject-WARNING **: cannot register existing type `CompEditor' Hi, checkout [1]. Two possible fixes, one might help for sure. Please update the bug report with your findings. Thanks in advance. Bye, Milan [1] https://bugzilla.gnome.org/show_bug.cgi?id=655507 ___ That did it. Thanks. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] Need assistance installing latest evolution 3.1.5 on Fedora 15
Just installed a Fedora 15 system on a free partition and thought about testing the latest evolution a bit. The one installed is 3.0.2. Found out that there is a brand-new package for the 3.1.5 version available at https://admin.fedoraproject.org/pkgdb/builds/show/F-devel-i386/evolution/0/3.1.5/1.fc17/i686?_csrf_token=92652e49c54c39c10ffc2d7bb15abbd6031561e2. Unfortunately, it has quite a substantial dependency list. Do I need to upgrade to a F16 release? Since I am not so familiar with the fedora rpm/yum package management, I would welcome somebody on this list to point me in the right direction. Thanks in advance, -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Need assistance installing latest evolution 3.1.5 on Fedora 15
Am Mittwoch, den 17.08.2011, 08:39 +0200 schrieb Thomas Mittelstaedt: Just installed a Fedora 15 system on a free partition and thought about testing the latest evolution a bit. The one installed is 3.0.2. Found out that there is a brand-new package for the 3.1.5 version available at https://admin.fedoraproject.org/pkgdb/builds/show/F-devel-i386/evolution/0/3.1.5/1.fc17/i686?_csrf_token=92652e49c54c39c10ffc2d7bb15abbd6031561e2. Unfortunately, it has quite a substantial dependency list. Do I need to upgrade to a F16 release? Since I am not so familiar with the fedora rpm/yum package management, I would welcome somebody on this list to point me in the right direction. Thanks in advance, Hmm. There is a preupgrade (http://fedoraproject.org/wiki/PreUpgrade) to upgrade to the latest release, which is F16, right? -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] List of packages for debuging
Am Sonntag, den 28.08.2011, 15:22 -0400 schrieb François Michonneau: Hello, I am facing regular (apparently random) crashes with Evolution 3.0.2 and I would like to report these bugs. However, I seem to be missing some debugging symbols in the gdb trace. I installed evolution and bug-buddy through the Gnome 3 PPA for Ubuntu Natty. Does anybody know which additional packages I would need to install to obtain the full debugging symbols? So, you are on ubuntu 11.04 natty, right? The evolution version on this distribution version is 2.32. As far as your bread-and-butter use of evolution is concerned, why don't you stick with this for a while until all the gnome 3 stuff has consolidated a bit more. I recently installed an ubuntu 11.10 oneiric alpha version in a virtual machine, which has evolution 3.1.5 on it. It ran reasonably well, even though I did not test extensively. One other person who regularly responds to postings on this list and is part of the maintainers group of evolution, recently warned that 3.1.xx is still unstable. Your pick. I don't know if it's related but there are these warnings that appear after typing 'run' in gdb: warning: the debug information found in /usr/lib/debug//usr/lib/libxml2.so.2.7.8 does not match /usr/lib/libxml2.so.2 (CRC mismatch). warning: the debug information found in /usr/lib/debug/usr/lib/libxml2.so.2.7.8 does not match /usr/lib/libxml2.so.2 (CRC mismatch). So you grabbed the stuff from https://launchpad.net/~gnome3-team, right? They have a mailing list and IRC. Why don't you ask there? -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Updated Windows Port
Am Freitag, den 07.10.2011, 04:42 -0700 schrieb J. Kauffman: Evolution seems like a great email client, but my company lives in the windows world. I have tried several of the old evolution installs that have been ported to windows and none of them work. Several will not even start and the ones that do crash constantly. Does anyone have any plans to port the latest stable version to windows? I think there are a lot of people that would love to have this. Thanks- Jason K. ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers You could install virtual machines, e.g. from vmware, on your Windows boxes, install a current Linux on it, and get the latest and best evolution. Hope that helps, -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] [patch] segfault in gal_a11y_e_cell_popup_new
Just had a segfault in gal_a11y_e_cell_popup_new. Turned out that the cast popupcell= E_CELL_POPUP (cell_view-ecell); would turn up a broken pointer, crashing afterward. I inserted the following on my side: ECellPopup *popupcell = NULL; ECellView* child_view = NULL; if (E_IS_CELL_POPUP(cell_view-ecell)) { popupcell = E_CELL_POPUP(cell_view-ecell); } Looked up the master code in the git browser and saw that this code is still the same. No check. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] [patch] segfault in gal_a11y_e_cell_popup_new
Am Mittwoch, den 02.11.2011, 09:10 -0400 schrieb Matthew Barnes: On Wed, 2011-11-02 at 12:29 +0100, Thomas Mittelstaedt wrote: You are right. I just had another crash with the above code changes. gdb told me that popupcell-popup_cell_view-cell_view.ecell was a broken pointer and popupcell-popup_cell_view-cell_view.e_table_model was 0. So, I inserted another sanity check. Let's see if it crashes again. I favor the most foolproof solution of all to this, which involves: git rm widgets/table/gal-a11y-e-cell-popup.* Our accessibility code is bitrotted and unmaintained. :-) I am not sure if I still had accessibility enabled in gconf, /desktop/gnome/interface/accessibility. Well, it's turned off now. So, this should not creep up anymore, I would assume. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... http://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] alarm-notify patch: Proposed changes to notify dialog
I made a couple of changes to my ancient (2.28.4) version of evolution. First, I encountered a nuisance in the alarm notify dialog, that when the dialog pops up and a new alarm is coming in, the selection changes to this new item, while the user potentially ist just clicking the dismiss or snooze button. I made the attached change to simply not preselect anything, having checked that the same is still running in the current git version at https://git.gnome.org/browse/evolution. I also had a crash in the alarm-notify app and I inserted the attached safety check. -- thomas commit 20c497b8a7fa130eae9c639c191aa0821c1f8d47 Author: Thomas Mittelstaedt tmsta...@t-mittelstaedt.de Date: Thu Aug 15 06:55:43 2013 +0200 Don't set selection to new items coming in while dialog is open diff --git a/calendar/gui/alarm-notify/alarm-queue.c b/calendar/gui/alarm-notify/alarm-queue.c index 5d9177d..9a3f80b 100644 --- a/calendar/gui/alarm-notify/alarm-queue.c +++ b/calendar/gui/alarm-notify/alarm-queue.c @@ -1306,7 +1306,7 @@ open_alarm_dialog (TrayIconData *tray_data) tray_data-location, notify_dialog_cb, tray_data); - gtk_tree_selection_select_iter (selection, tray_data-iter); + /* gtk_tree_selection_select_iter (selection, tray_data-iter); */ } commit 2d7f700f3ef7de943549f4780252c9c6bcf1fc68 Author: Thomas Mittelstaedt tmsta...@t-mittelstaedt.de Date: Thu Aug 15 07:27:34 2013 +0200 Check for empty list in tray_list_remove_async diff --git a/calendar/gui/alarm-notify/alarm-queue.c b/calendar/gui/alarm-notify/alarm-queue.c index 9a3f80b..59a0884 100644 --- a/calendar/gui/alarm-notify/alarm-queue.c +++ b/calendar/gui/alarm-notify/alarm-queue.c @@ -1103,6 +1103,7 @@ static void tray_list_remove_async (Message *msg) { GList *list = tray_icons_list; + g_return_if_fail (g_list_first(list) != NULL); d(printf(%s:%d (tray_list_remove_async) - Removing %d alarms\n,__FILE__, __LINE__, g_list_length(list))); while (list != NULL) { ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] alarm-notify patch: Proposed changes to notify dialog
Am Donnerstag, den 15.08.2013, 07:47 +0200 schrieb Thomas Mittelstaedt: I made a couple of changes to my ancient (2.28.4) version of evolution. First, I encountered a nuisance in the alarm notify dialog, that when the dialog pops up and a new alarm is coming in, the selection changes to this new item, while the user potentially ist just clicking the dismiss or snooze button. I made the attached change to simply not preselect anything, having checked that the same is still running in the current git version at https://git.gnome.org/browse/evolution. I also had a crash in the alarm-notify app and I inserted the attached safety check. ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers Also make another change to show a confirmation dialog when the user presses the dismiss all button. -- thomas commit 8db79a84f68f32f07be4c139cdb4f9721d43ef0c Author: Thomas Mittelstaedt tmsta...@t-mittelstaedt.de Date: Fri Aug 16 05:07:11 2013 +0200 Show a confirmation dialog, when the user presses the dismiss all button. diff --git a/calendar/gui/alarm-notify/alarm-queue.c b/calendar/gui/alarm-notify/alarm-queue.c index 59a0884..b29b8ac 100644 --- a/calendar/gui/alarm-notify/alarm-queue.c +++ b/calendar/gui/alarm-notify/alarm-queue.c @@ -1177,6 +1177,9 @@ tray_list_remove_data (TrayIconData *data) message_push ((Message *) msg); } +static gboolean +procedure_notification_dialog (const gchar *cmd, const gchar *url); + static void notify_dialog_cb (AlarmNotifyResult result, gint snooze_mins, gpointer data) { @@ -1232,7 +1235,7 @@ notify_dialog_cb (AlarmNotifyResult result, gint snooze_mins, gpointer data) case ALARM_NOTIFY_CLOSE: d(printf(%s:%d (notify_dialog_cb) - Dialog close\n,__FILE__, __LINE__)); - if (alarm_notifications_dialog) { + if (alarm_notifications_dialog procedure_notification_dialog (Sure close?, )) { GtkTreeIter iter; GtkTreeModel *model = gtk_tree_view_get_model ( ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] alarm-notify patch: Proposed changes to notify dialog
Am Freitag, den 16.08.2013, 05:09 +0200 schrieb Thomas Mittelstaedt: Am Donnerstag, den 15.08.2013, 07:47 +0200 schrieb Thomas Mittelstaedt: I made a couple of changes to my ancient (2.28.4) version of evolution. First, I encountered a nuisance in the alarm notify dialog, that when the dialog pops up and a new alarm is coming in, the selection changes to this new item, while the user potentially ist just clicking the dismiss or snooze button. I made the attached change to simply not preselect anything, having checked that the same is still running in the current git version at https://git.gnome.org/browse/evolution. I also had a crash in the alarm-notify app and I inserted the attached safety check. Also make another change to show a confirmation dialog when the user presses the dismiss all button. ___ Forget the last attached diff. Doesn't work properly. Just as a general idea: Ask for confirmation before dismissing all alarms or even a single one. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] [Old Version 2.28] Tasks: Prevent Ctrl-X in search box to accidentally delete tasks.
I don't know if this works cleanly in the newest versions. But in my ancient version, doing a search by entering a term in the search box, selecting a piece of text of that search string and hitting Ctrl-x would result in the currently selected task to be deleted. And Ctrl-v would not bring it back. I inserted a little crutch in my old version to simply disable the cut command and that worked. See attached diff. -- thomas commit cf5444ce2def9c51d40ee6bf4091ad8c516f1a8b Author: Thomas Mittelstaedt tmsta...@t-mittelstaedt.de Date: Fri Aug 16 22:21:47 2013 +0200 Just a quick fix: Disable the TasksCut command in order not accidentally delete tasks while hitting Ctrl-x in the search text box. diff --git a/calendar/gui/tasks-control.c b/calendar/gui/tasks-control.c index a788931..b24b9a6 100644 --- a/calendar/gui/tasks-control.c +++ b/calendar/gui/tasks-control.c @@ -269,7 +269,8 @@ tasks_control_focus_changed (GtkWidget *widget, GdkEventFocus *event, struct foc static BonoboUIVerb verbs [] = { BONOBO_UI_VERB (TasksOpenTask, tasks_control_open_task_cmd), BONOBO_UI_VERB (TasksNewTask, tasks_control_new_task_cmd), - BONOBO_UI_VERB (TasksCut, tasks_control_cut_cmd), + /* BONOBO_UI_VERB (TasksCut, tasks_control_cut_cmd), */ + BONOBO_UI_VERB (TasksCut, NULL), BONOBO_UI_VERB (TasksCopy, tasks_control_copy_cmd), BONOBO_UI_VERB (TasksPaste, tasks_control_paste_cmd), BONOBO_UI_VERB (TasksDelete, tasks_control_delete_cmd), ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] [Old Version 2.28] Tasks: Prevent Ctrl-X in search box to accidentally delete tasks, For what it's worth: My comprehensive set of patches for ancient 2.28
Am Montag, den 19.08.2013, 11:05 +0200 schrieb Thomas Mittelstaedt: Am Montag, den 19.08.2013, 07:51 +0200 schrieb Milan Crha: On Fri, 2013-08-16 at 22:50 +0200, Thomas Mittelstaedt wrote: I don't know if this works cleanly in the newest versions. But in my ancient version, doing a search by entering a term in the search box, selecting a piece of text of that search string and hitting Ctrl-x would result in the currently selected task to be deleted. And Ctrl-v would not bring it back. I inserted a little crutch in my old version to simply disable the cut command and that worked. See attached diff. Hi, as I replied to one of your previous emails, please use bugzilla for issues and patches, the mailing list is not a place where these should be added, at least not for evolution-hackers. All right, Milan. I created an enhancement bug report at https://bugzilla.gnome.org/show_bug.cgi?id=706287 And to quote from that report: For what it's worth: My comprehensive set of patches for ancient 2.28 version. (edit) Product: Evolution Component: Tasks Version: 2.28.x Status: UNCONFIRMED Priority: Normal Severity: enhancement Collapse All Comments - Expand All Comments [reply] [-] Description Thomas [reporter] 2013-08-19 09:02:07 UTC Created an attachment (id=252183) [details] Set of patches Just for completeness: I have reached a very nice level of stability and user satisfaction with this old, ancient version of evolution, which I use on an ubuntu 10.04 (Lucid Lynx) box. Over time I applied several patches, fixes of simple safety checks in response to crashes to this version and I thought, I upload it as an enhancement bug report, if somebody may be interested. I use a Software called stacked git (stg) to apply these series of patches. Naming convention of these patches is issue_number_... . This hints at a concrete bug report in bugzilla and should be reachable with the following url: https://bugzilla.gnome.org/show_bug.cgi?id=number. I also uploaded my build script and the script I use to run the application. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] [evolution-alarm-notify] Crash in remove_queued_alarm() and possible remedy
Am Dienstag, den 08.07.2014, 09:18 +0200 schrieb Milan Crha: On Tue, 2014-07-08 at 07:42 +0200, Thomas Mittelstaedt wrote: These crashes recently happened more often, but not always and usually when I made intense use of alarms, snoozing some, rescheduling others, a torture test, if you will, spanning several hours. Hi, when talking about snoozing, do you have applied also this one? https://bugzilla.gnome.org/show_bug.cgi?id=709162#c3 No. Just did, thank you. Finally I tried the following remedy in tray_list_remove_icons(), calling tray_list_remove_async() directly instead of putting it in the message queue. Does that make sense to you, experts? Not much. What was it supposed to help with? I know the current git master (and 3.12) calls many things synchronously on the main thread, which can block the UI, thus maybe it is also related to that, a change being done along those lines. Bye, Milan Well, what does it make sense to call a cleanup function asynchronously by puting a message into a message queue, which happens to be executed in a separate thread, as far as I know. That could be the cause of the destroyed pointer, different threads interfering with these lists. -- thomas ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers