Bug 467016
Summary: | Dead keys are not rcognized as such | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Matěj Cepl <mcepl> | ||||||||||
Component: | scim | Assignee: | Peter Hutterer <peter.hutterer> | ||||||||||
Status: | CLOSED CURRENTRELEASE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||||||
Severity: | medium | Docs Contact: | |||||||||||
Priority: | medium | ||||||||||||
Version: | 10 | CC: | i18n-bugs, mcepl, mclasen, misek, oliva, peter.hutterer, petersen, phuang, xgl-maint | ||||||||||
Target Milestone: | --- | Keywords: | Reopened | ||||||||||
Target Release: | --- | ||||||||||||
Hardware: | All | ||||||||||||
OS: | Linux | ||||||||||||
Whiteboard: | |||||||||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||||||||
Doc Text: | Story Points: | --- | |||||||||||
Clone Of: | Environment: | ||||||||||||
Last Closed: | 2008-12-16 00:59:06 UTC | Type: | --- | ||||||||||
Regression: | --- | Mount Type: | --- | ||||||||||
Documentation: | --- | CRM: | |||||||||||
Verified Versions: | Category: | --- | |||||||||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||||
Cloudforms Team: | --- | Target Upstream Version: | |||||||||||
Embargoed: | |||||||||||||
Attachments: |
|
Created attachment 320402 [details]
/var/log/Xorg.0.log
... and of course I have no /etc/X11/xorg.conf
Created attachment 320516 [details]
xev log
Another xev log that Matej gave me yesterday. Attaching it so it doesn't get lost.
I need some more info please: how am I supposed to create ňóť? Hitting Shift and the +/= key, then n works fine to produce an ň. Likewise the others, AFAICT. Also, the xev output looks correct. What are you gnome options again? Have you tried xterm or konsole as well? Of course, silly me, should try xterm myself. Yes, there is works like a charm. Will reassign to Gnome folks. Please try gtk2-2.14.4-1.fc10. It has some fixes in dead key handling. Unfortunately, no change. Still getting ˇn ˇt ´o instead of ňťó even with gtk2-2.14.4-1.fc10.i386. What keyboard layout is this with ? And what applications are you trying it in ? Using the Czech layout, typing S-=n works fine to produce a ň, in gnome-terminal, gedit and firefox. firefox, gedit, pidgin, gnome-terminal. [matej@viklef ~]$ setxkbmap -print xkb_keymap { xkb_keycodes { include "evdev+aliases(qwertz)" }; xkb_types { include "complete" }; xkb_compat { include "complete" }; xkb_symbols { include "pc+cz(qwerty)+us:2+inet(evdev)+group(shifts_toggle)+level3(ralt_switch_multikey)+compose(rctrl)+nbsp(level3)" }; xkb_geometry { include "pc(pc104)" }; }; [matej@viklef ~]$ ň - still works, with those exact settings. Are you using any input methods ? Still reproducible with *scim* packages gone, and this in gconf: [matej@viklef ~]$ gconftool-2 -R /desktop/gnome/interface |grep gtk-im- gtk-im-status-style = callback gtk-im-module = gtk-im-context-simple gtk-im-preedit-style = callback [matej@viklef ~]$ Actually, I cannot fully reproduce your xkb configuration. I only ever get xkb_keycodes { include "evdev+aliases(qwerty)" }; And my symbols only have "cz" instead of "cz(qwerty)" Not convinced that makes any difference. Peter, looking at the matejs xev log: KeyPress event, serial 27, synthetic NO, window 0x4200001, root 0x7b, subw 0x0, time 16186204, (164,-13), root:(173,58), state 0x0, keycode 50 (keysym 0xffe1, Shift_L), same_screen YES, XLookupString gives 0 bytes: XmbLookupString gives 0 bytes: XFilterEvent returns: False KeyPress event, serial 30, synthetic NO, window 0x4200001, root 0x7b, subw 0x0, time 16186425, (164,-13), root:(173,58), state 0x1, keycode 21 (keysym 0xfe5a, dead_caron), same_screen YES, XLookupString gives 2 bytes: (cb 87) "ˇ" XmbLookupString gives 0 bytes: XFilterEvent returns: True KeyRelease event, serial 30, synthetic NO, window 0x4200001, root 0x7b, subw 0x0, time 16186485, (164,-13), root:(173,58), state 0x1, keycode 21 (keysym 0xfe5a, dead_caron), same_screen YES, XLookupString gives 2 bytes: (cb 87) "ˇ" XFilterEvent returns: False KeyRelease event, serial 30, synthetic NO, window 0x4200001, root 0x7b, subw 0x0, time 16186535, (164,-13), root:(173,58), state 0x1, keycode 50 (keysym 0xfe0a, ISO_Prev_Group), same_screen YES, XLookupString gives 0 bytes: XFilterEvent returns: False Notice how the shift press doesn't match up with the release. Why is that ? I don't see that in my own xev log, which shows a matching shift release. Matej, what is the sequence of key presses and releases that you are using to type a ň ? ie something like press shift press = release = release shift press n release n Exactly like that. Created attachment 321462 [details]
output of xkbcomp :0 -
When creating a new user, everything works smoothly. Any ideas, what goo I accquired where?
no idea, sorry. Lets close this for now, ok ? (In reply to comment #19) > Lets close this for now, ok ? Please no, I don't insist to be fixed immediately (or even soon), but it is still broken, and it is gnome* problem (xterm still works like a charm). Removed fedora-x-target from blocking bugs, so that your conscience is soothened. OK, somewhere between removing scim (again) and doing zillion changes to gconf and couple of others things, it justed start to work. No idea, what fixed it, but I am not in the mood to play with it too much. Blaming scim, but closing as INSUFFICIENT_DATA, because I am not able to say much about it. Given comment 12 I don't know why you are blaming scim which hasn't really changed for f10: xorg-x11 or possibly imsettings sound more plausible suspects, but anyway... (In reply to comment #22) > Given comment 12 I don't know why you are blaming scim which hasn't really > changed for f10: xorg-x11 or possibly imsettings sound more plausible suspects, > but anyway... I have my reasons to believe that it is not xorg-x11* (with possible exception of XIM) but it might be imsettings -- difference between *im* tools is slightly blurry for me. Too bad I'm hit by the same error, it appeared two days ago. Probably after gnome update????? The system is Fedora 10 with the last updates+updates-testing. I'm adding list what was updated since the last known working state (I'm sure it was working 2008-12-03 at 19:32:10). Dec 03 18:36:42 Updated: libtdb-1.1.1-23.fc10.x86_64 Dec 03 18:36:43 Updated: libtalloc-1.2.0-23.fc10.x86_64 Dec 03 18:36:47 Updated: 1:qt-4.4.3-6.fc10.x86_64 Dec 03 18:37:03 Updated: 1:qt-x11-4.4.3-6.fc10.x86_64 Dec 03 18:37:29 Updated: 1:qt-devel-4.4.3-6.fc10.x86_64 Dec 03 18:37:29 Updated: 1:qt-mysql-4.4.3-6.fc10.x86_64 Dec 03 18:37:29 Updated: 1:xorg-x11-drv-nouveau-0.0.11-1.20081119git65b956f.fc10.x86_64 Dec 03 18:37:59 Updated: virt-manager-0.6.0-4.fc10.x86_64 Dec 03 18:38:01 Updated: xorg-x11-drv-ati-6.9.0-59.fc10.x86_64 Dec 03 18:38:02 Updated: docbook-utils-0.6.14-15.fc10.noarch Dec 03 18:38:04 Installed: spin-kickstarts-0.10.3-1.fc10.noarch Dec 03 18:38:04 Updated: docbook-utils-pdf-0.6.14-15.fc10.noarch Dec 03 18:38:14 Updated: OmegaT-1.7.3_04-2.fc10.noarch Dec 03 18:38:27 Updated: wine-core-1.1.9-2.fc10.i386 Dec 03 18:38:28 Updated: wine-cms-1.1.9-2.fc10.i386 Dec 03 18:38:29 Updated: wine-capi-1.1.9-2.fc10.i386 Dec 03 18:38:29 Updated: wine-ldap-1.1.9-2.fc10.i386 Dec 03 18:38:30 Updated: wine-twain-1.1.9-2.fc10.i386 Dec 03 18:38:31 Updated: wine-nas-1.1.9-2.fc10.i386 Dec 03 18:38:32 Updated: wine-tools-1.1.9-2.fc10.i386 Dec 03 18:38:33 Updated: wine-jack-1.1.9-2.fc10.i386 Dec 03 18:38:34 Updated: wine-esd-1.1.9-2.fc10.i386 Dec 03 18:38:35 Updated: wine-desktop-1.1.9-2.fc10.i386 Dec 03 18:38:35 Updated: wine-1.1.9-2.fc10.i386 Dec 03 18:38:38 Updated: samba-winbind-3.2.5-0.23.fc10.x86_64 Dec 03 18:38:43 Updated: samba-common-3.2.5-0.23.fc10.x86_64 Dec 03 18:38:45 Updated: revisor-cli-2.1.3-1.fc10.noarch Dec 03 18:38:47 Updated: revisor-gui-2.1.3-1.fc10.noarch Dec 03 18:38:47 Updated: revisor-2.1.3-1.fc10.noarch Dec 03 18:38:49 Updated: libsmbclient-3.2.5-0.23.fc10.x86_64 Dec 03 18:38:54 Updated: revisor-comps-2.1.3-1.fc10.noarch Dec 03 18:38:57 Updated: samba-3.2.5-0.23.fc10.x86_64 Dec 03 23:33:10 Updated: gtk2-2.14.5-2.fc10.x86_64 Dec 03 23:33:11 Updated: 2:gimp-libs-2.6.3-2.fc10.x86_64 Dec 03 23:33:41 Updated: libgnome-2.24.1-8.fc10.x86_64 Dec 03 23:33:45 Updated: shared-mime-info-0.51-5.fc10.x86_64 Dec 03 23:33:47 Updated: check-0.9.5-3.fc10.x86_64 Dec 03 23:33:49 Updated: 1:xmms-libs-1.2.11-3.20071117cvs.fc10.x86_64 Dec 03 23:34:30 Updated: 2:gimp-2.6.3-2.fc10.x86_64 Dec 03 23:34:33 Updated: sectool-0.9.2-2.x86_64 Dec 03 23:34:34 Updated: loudmouth-1.4.3-1.fc10.x86_64 Dec 03 23:34:37 Updated: system-config-printer-libs-1.0.12-1.fc10.x86_64 Dec 03 23:34:39 Updated: libdiscid-0.2.2-1.fc10.x86_64 Dec 03 23:34:42 Updated: gtk2-2.14.5-2.fc10.i386 Dec 03 23:35:01 Updated: libgnome-2.24.1-8.fc10.i386 Dec 03 23:35:02 Updated: 2:gimp-help-browser-2.6.3-2.fc10.x86_64 Dec 03 23:35:05 Updated: 1:xmms-1.2.11-3.20071117cvs.fc10.x86_64 Dec 03 23:35:12 Updated: 1:grip-3.2.0-25.fc10.x86_64 Dec 03 23:35:32 Updated: gtk2-devel-2.14.5-2.fc10.x86_64 Dec 03 23:35:44 Updated: freeciv-2.1.7-1.fc10.x86_64 Dec 03 23:35:46 Updated: dcraw-8.89-1.fc10.x86_64 Dec 03 23:35:46 Updated: 2:vim-minimal-7.2.060-1.fc10.x86_64 Dec 03 23:35:48 Updated: openvpn-2.1-0.29.rc15.fc10.x86_64 Dec 03 23:35:55 Updated: 1:nfs-utils-1.1.4-2.fc10.x86_64 Dec 03 23:36:01 Updated: netbeans-ide9-6.1-10.fc10.noarch Dec 03 23:36:05 Updated: yum-3.2.20-4.fc10.noarch Dec 03 23:36:08 Updated: netbeans-java2-6.1-10.fc10.noarch Dec 03 23:36:09 Updated: netbeans-apisupport1-6.1-10.fc10.noarch Dec 03 23:36:10 Updated: kernel-firmware-2.6.27.7-130.fc10.noarch Dec 03 23:36:11 Updated: gnome-doc-utils-stylesheets-0.14.0-2.fc10.noarch Dec 03 23:36:16 Updated: system-config-network-tui-1.5.94-2.fc10.noarch Dec 03 23:36:18 Updated: system-config-network-1.5.94-2.fc10.noarch Dec 03 23:36:20 Updated: gnome-doc-utils-0.14.0-2.fc10.noarch Dec 03 23:36:21 Updated: netbeans-6.1-10.fc10.noarch Dec 03 23:36:24 Updated: system-config-printer-1.0.12-1.fc10.x86_64 Dec 03 23:36:25 Updated: sectool-gui-0.9.2-2.x86_64 Dec 03 23:36:26 Updated: check-devel-0.9.5-3.fc10.x86_64 Dec 03 23:36:27 Updated: libgnome-devel-2.24.1-8.fc10.x86_64 Dec 03 23:36:27 Updated: cdlabelgen-4.1.0-1.fc10.noarch Dec 03 23:36:28 Updated: etherboot-roms-kvm-5.4.4-5.fc10.x86_64 Dec 03 23:36:37 Updated: kernel-headers-2.6.27.7-130.fc10.x86_64 Dec 03 23:36:38 Updated: crontabs-1.10-25.fc10.noarch Dec 03 23:39:15 Installed: kernel-devel-2.6.27.7-130.fc10.x86_64 Dec 03 23:40:07 Installed: kernel-2.6.27.7-130.fc10.x86_64 Dec 03 23:40:07 Updated: kmod-nvidia-177.82-1.fc10.5.x86_64 Dec 03 23:40:08 Installed: kmod-nvidia-2.6.27.7-130.fc10.x86_64-177.82-1.fc10.5.x86_64 Dec 03 23:40:10 Updated: PackageKit-yum-plugin-0.3.11-3.fc10.x86_64 Dec 03 23:40:11 Updated: PackageKit-glib-0.3.11-3.fc10.x86_64 Dec 03 23:40:11 Updated: PackageKit-udev-helper-0.3.11-3.fc10.x86_64 Dec 03 23:40:16 Updated: PackageKit-0.3.11-3.fc10.x86_64 Dec 03 23:40:47 Updated: gnome-packagekit-0.3.11-3.fc10.x86_64 Dec 03 23:40:47 Updated: PackageKit-yum-0.3.11-3.fc10.x86_64 Dec 03 23:41:10 Erased: kmod-nvidia-2.6.27.5-113.fc10.x86_64 Dec 03 23:42:12 Installed: kernel-2.6.27.7-130.fc10.x86_64 Is anyone here using scim? - I don't think this is a scim but happen to be proven otherwise... Peter, do you have any light on this? No, and there are no scim packages installed on my machine. Downgrading gtk2 and gtk2-devel to 2.14.4-3.fc10 restored dead keys to me. I'm going to clone this bug for gtk2. This doesn't make sense to clone this bug -- if you blame Gtk, then let's go there, but don't make us unnecessary bugs. Closing this as duplicate of bug 475008. *** This bug has been marked as a duplicate of bug 475008 *** It doesn't make sense to assume that a bug filed two months ago is in any way related with the new bug that appeared 72 hours ago, that happens to display similar symptoms under completely different circumstances. That's why I created a separate bug, rather than reassign or close this one. I know very little about the original bug reported here; all I know is that it can't be the same bug as the one I and Vaclav ran into. So I'm reopening this, unless you, as the original reporter of this bug, can verify that it somehow is the same bug, or that it is fixed. I don't know what the issue is, but Comment #17 indicates that it's neither driver nor server (mind you, dead keys are processed in the client anyway). So one of you who sees the bug needs to be able to narrow down what causes the issue. If you experience the bug create a new user and test this user to ensure there's no stale settings that may trigger it. Matej, you still see this post F10 GA? Given comment 21 and the new gtk2 bug 475008, I reckon this can be closed now, no? (In reply to comment #31) > Matej, you still see this post F10 GA? No, I don't. |
Created attachment 320401 [details] output of xev Description of problem: Having cs_CZ locale, I need dead-key working, because some of the letters in the Czech keyboard layout are not on the keyboard directly (e.g., ňóť). Unfortunately when entering these keys in the current Fedora Rawhide I get ˇn´oˇt instead. Version-Release number of selected component (if applicable): xorg-x11-drv-keyboard-1.3.0-3.fc9.i386 xorg-x11-drv-evdev-2.0.6-2.fc10.i386 How reproducible: 100% Steps to Reproduce: 1.see above 2. 3. Actual results: see above Expected results: see above (used gnucharmap to get these) Additional info: