[Bug 2093313] New: Too many ibus warnings, "no capability of
surrounding-text feature"
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=2093313
Bug ID: 2093313
Summary: Too many ibus warnings, "no capability of
surrounding-text feature"
Product: Fedora
Version: rawhide
Status: NEW
Component: ibus
Assignee: tfujiwar(a)redhat.com
Reporter: tagoh(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org,
shawn.p.huang(a)gmail.com, tfujiwar(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
I know this is an application bug, flooding same warnings looks not good to me.
this affects UX. applications and desktop is often freezing.
$ journalctl -S 2022-06-03 | grep IBUS-WARNING | wc -l
15974
That is quite bad.
Too many warnings should be omitted.
Version-Release number of selected component (if applicable):
ibus-1.5.26-4.fc36.x86_64
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugzilla.redhat.com/show_bug.cgi?id=2093313
1 week, 5 days
[Fedora-i18n-bugs] [Bug 1790554] New: Keyboard layout of Kana Kanji wont show up
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1790554
Bug ID: 1790554
Summary: Keyboard layout of Kana Kanji wont show up
Product: Fedora
Version: rawhide
Hardware: x86_64
Status: NEW
Component: ibus
Assignee: tfujiwar(a)redhat.com
Reporter: sumukher(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org,
shawn.p.huang(a)gmail.com, tfujiwar(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
The keyboard layout for Kana Kanji Japanese doesnt show up
Version-Release number of selected component (if applicable):
Fedora-Rawhide-20200112.n.0
How reproducible:
Everytime
Steps to Reproduce:
1. Install Fedora-Rawhide-20200112.n.0 WS from live boot
2. Open settings and navigate to Region and Language
3. Add Japanese Kana Kanji in the input source
4. Click the icon looking like an eye button which should open the keyboard
layout
Actual results:
Doesn't show up the keyboard layout
Expected results:
The keyboard layout should show up like it shows up for english
Additional info:
--
You are receiving this mail because:
You are on the CC list for the bug.
1 week, 5 days
[Bug 2095164] New: Conversion region of ibus-anthy is invisible in
konsole, kwrite, kate
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=2095164
Bug ID: 2095164
Summary: Conversion region of ibus-anthy is invisible in
konsole, kwrite, kate
Product: Fedora
Version: 36
Status: NEW
Component: ibus
Assignee: tfujiwar(a)redhat.com
Reporter: mfabian(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org,
shawn.p.huang(a)gmail.com, tfujiwar(a)redhat.com
Target Milestone: ---
Classification: Fedora
Created attachment 1888253
--> https://bugzilla.redhat.com/attachment.cgi?id=1888253&action=edit
Video showing the the conversion region is visible in some places but
invisible in konsole, kwrite, kate
Using Fedora-Workstation-Live-x86_64-36-1.2.iso installed in qemu-kvm with all
current updates.
kwrite, kate, and konsole, do not show the conversion region when using
ibus-anthy
konsole5-22.04.1-1.fc36.x86_64
kwrite-22.04.1-1.fc36.x86_64
kate-21.12.2-1.fc36.x86_64
It does not matter whether these programs are used in
Plasma(X11), Plasma(Wayland), Gnome(Xorg), Gnome(Wayland), the behaviour is
always the same, the conversion region is never visible.
In konsole the behaviour is even worse because not even the cursor is shown at
the start of the conversion region.
When testing in Plasma(Wayland) or Plasma(X11) and typing with ibus-anthy into
the search field of the KDE control center, the conversion region is coloured
and thus visible. It is also visible when typing into the entry field for a
command which opens with Alt+F2. So there are only some programs in KDE where
it does not work like konsole, kwrite, kate, ... whereas it works in other
places.
See the attached video.
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugzilla.redhat.com/show_bug.cgi?id=2095164
1 week, 5 days
[Bug 2090204] New: [abrt] ibus: __strndup(): ibus-extension-gtk3
killed by SIGSEGV
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=2090204
Bug ID: 2090204
Summary: [abrt] ibus: __strndup(): ibus-extension-gtk3 killed
by SIGSEGV
Product: Fedora
Version: 36
Hardware: x86_64
Status: NEW
Whiteboard: abrt_hash:8c2b2d109d6c6600cfdf1e41632f3398d6d21e15;VAR
IANT_ID=workstation;
Component: ibus
Assignee: tfujiwar(a)redhat.com
Reporter: mfabian(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org,
shawn.p.huang(a)gmail.com, tfujiwar(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
Version-Release number of selected component:
ibus-1.5.26-6.fc36
Additional info:
reporter: libreport-2.17.1
backtrace_rating: 4
cgroup:
0::/user.slice/user-1000.slice/user@1000.service/session.slice/org.freedesktop.IBus.session.GNOME.service
cmdline: /usr/libexec/ibus-extension-gtk3
crash_function: __strndup
dso_list: /usr/libexec/ibus-extension-gtk3 ibus-1.5.26-6.fc36.x86_64
(Fedora Project) 1653401127
executable: /usr/libexec/ibus-extension-gtk3
journald_cursor:
s=3da4198b50ed40bb8e39aac19c44f062;i=15b44;b=4f87f45459b64f7299ca0a57d8a15df6;m=51ec68d6;t=5dfc2daddfca9;x=e27b7bc9d78e7740
kernel: 5.17.9-300.fc36.x86_64
rootdir: /
runlevel: N 5
type: CCpp
uid: 1000
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugzilla.redhat.com/show_bug.cgi?id=2090204
1 week, 5 days
[Bug 2070187] New: please branch ibus-kkc for epel9
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=2070187
Bug ID: 2070187
Summary: please branch ibus-kkc for epel9
Product: Fedora
Version: rawhide
Status: NEW
Component: ibus-kkc
Assignee: dueno(a)redhat.com
Reporter: petersen(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: dueno(a)redhat.com, i18n-bugs(a)lists.fedoraproject.org
Target Milestone: ---
Classification: Fedora
Description of problem:
I would like to have ibus-kkc with libkkc in epel9
and would be happy to help build it if you like.
Could you please request an epel9 branch?
If you don't mind to make me comaintainer that would be fine too.
(libkkc got orphaned during the F35 cycle, I think
so I was able to branch that already.)
Thanks
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugzilla.redhat.com/show_bug.cgi?id=2070187
1 week, 5 days
[Fedora-i18n-bugs] [Bug 1897782] New: ibus pinyin input with special characters like '/' will repeat itself non-stop and cause strange behaviours with background applications
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1897782
Bug ID: 1897782
Summary: ibus pinyin input with special characters like '/'
will repeat itself non-stop and cause strange
behaviours with background applications
Product: Fedora
Version: 33
Status: NEW
Component: ibus-libpinyin
Assignee: pwu(a)redhat.com
Reporter: yemoran-2020(a)outlook.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org,
petersen(a)redhat.com, pwu(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
In Fedora 33 Workstation, with a Chinese system language, when I try to input
with the ibus input (intelligent pinyin), and I pressed the slash key'/', then
the slash character '/' will automatically keep repeating itself, even if I
only pressed the slash key '/' once.
Not only this caused unwanted '/' to spam, but:
1. Also prevents me from pressing alphabetical characters to use pinyin input
normally, unless I delete all my remaining characters to exit pinyin prompt and
start over;
2. Can possibly delete my already-saved text, in combination with
<Ctrl-Backspace>.
It seems that I have completely lost control within the pinyin prompt. I have
met with other surprising conditions, though cannot be immediately reproduced
now, but I believe more combinations of inputs triggering different bugs will
be confirmed later.
Version-Release number of selected component (if applicable):
Any kind of version. I also tested with Fedora 32 Workstation, still have this
bug.
How reproducible:
Always
Steps to Reproduce:
1. Fetch any Fedora Workstation version, be it 32 or 33 or any updates applied.
2. Add Chinese input method 'Intelligent Pinyin' and switch to it
3. In GNOME(Wayland), open any GTK-based application, be it Firefox, GNOME
Terminal, gedit, Libreoffice, ... (take gedit as an example)
4. Input 你好，今天天气 (keypress: nihao<1>，jintian<1>) as a start up example, confirm
input use number key '1'. or keys like <Enter> or <Space>.
5. Input 怎么样/ (keypress: zenmy/）. Do not confirm immediately after 'zenmy', but
press '/' exactly once.
6. The '/' character is repeating itself within several hundred milliseconds.
7. Press <Ctrl-Backspace>. gedit will highlight "你好，今天天气" (with a lot of
trailing '/').
8. Press <Backspace> to delete these Chinese characters.
Actual results:
Not only '/' is repeating itself, but also I lost control to gedit and gedit
thinks you want to press '/' forever or even wants to delete my
already-confirmed characters before this input.
Expected results:
1. '/' Should not trigger anything, unless I have pressed this key and did not
release this key press (but I always release keys)
2. '/' Should not make me lost control over pinyin prompt to gedit
3. <Backspace> Should only delete pinyin alphabets in pinyin prompt panel, and
should not do anything to the background gedit texts, unless I have exited the
pinyin prompt panel (because I confirmed my input or have deleted every pinyin
alphabet in it)
Additional info:
1. I did not found this bug on Debian buster or Arch with latest updates.
Additionally, '/' did not even input a single character '/' in these
distributions, but I cannot confirm if this is the expected behaviour. In
Fedora 33 KDE, '/' did input a single character '/', but immediately stopped.
2. I didn't find this behaviour in qt-based applications (like Fedora Image
Writer or Octave) in GNOME, or any application in other desktop environments
(including KDE, or even GTK-based Cinnamon Desktop). This bug seems to only
occur on GTK-based applications on GNOME.
3. I can be very sure that this is not caused by something wrong with my
specific keyboard, because (a) Other distributions worked very well (b) Other
Fedora spins worked very well (c) Qt-based applications in GNOME worked very
well (d) GNOME X11 session worked very well (e) I replicated this bug on a
fresh install of Fedora 32 Workstation in another laptop of mine.
4. I guess other special characters (or keys) like '[', ']', '\', '<Esc>',
'<Backspace>', etc. may trigger similar bugs, but I cannot confirm. '/' will
surely trigger this. ',' and '.' are used in pinyin input method to flip
candidate characters pages, so they doesn't trigger this bug. ' does not
trigger this bug (used in xi'an for 西安)
--
You are receiving this mail because:
You are on the CC list for the bug.
1 week, 5 days
[Fedora-i18n-bugs] [Bug 1847347] New: gnome-terminal and xfce4-terminal get surrounding text from previous application
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1847347
Bug ID: 1847347
Summary: gnome-terminal and xfce4-terminal get surrounding text
from previous application
Product: Fedora
Version: 32
Status: NEW
Component: ibus
Assignee: tfujiwar(a)redhat.com
Reporter: mfabian(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org,
shawn.p.huang(a)gmail.com, tfujiwar(a)redhat.com
Target Milestone: ---
Classification: Fedora
I made the following screenshots and video by testing on Gnome-Xorg, but the
problem occurs on Gnome-Wayland as well.
How to reproduce (see also attached screenshots and video):
- To show what happens with the surrounding text, I first opened the
setup tool of ibus-typing-booster and set the debug level in the
options tab to 3.
When the debug level is high, the auxiliary text above the candidate
list shows what context has been obtained by using surrounding text
(if surrounding text is available, if it is not available like when
using xterm, the context is just remembered from what was typed
last).
- I opened 3 windows: gedit, firefox, and gnome-terminal.
- I focus on gedit and type "a b c d e"
On top of the candidate list one can see: "Context: b c d"
This is because when the letter "e" was typed, ibus-typing-booster
got the surrounding text at the cursor position and parsed the last
3 tokens left of the cursor out of the result. And these
last 3 tokens left of the cursor are "b c d".
- Now I focus on firefox and type "f g h i j"
On top of the candidate list one can see: "Context: g h i"
I.e. the correct context "g h i" to the left of the just typed "j"
(which is still in preedit) has been found using surrounding text.
- Now I focus on gnome-terminal and type "k l m"
On top of the candidate list one can see: "Context h i j"
This comes from the surrounding text left of the cursor in
*firefox*, *not* from gnome-terminal.
Although gnome-terminal reports that surrounding text is supported, i.e.
self.client_capabilities & IBus.Capabilite.SURROUNDING_TEXT
is True (see the code to get the context at:
https://github.com/mike-fabian/ibus-typing-booster/blob/master/engine/hun...
)
gnome-terminal does not really seem to support surrounding text.
The surrounding text comes from the previous client which really supported
surrounding text.
If the previous client was firefox, one still gets the surrounding text from
firefox while typing in gnome-terminal.
Same with gedit, if the previous client before focussing on gnome-terminal
was gedit, one still gets the surrounding text from gedit while typing in
gnome-terminal.
- The same problem occurs when using xfce4-terminal instead of gnome-terminal.
- When using xterm, the problem does *not* occur because when using xterm
self.client_capabilities & IBus.Capabilite.SURROUNDING_TEXT
is False and then the get_context() immediately returns and the context
remembered from the last text typed is used as a fallback.
--
You are receiving this mail because:
You are on the CC list for the bug.
1 week, 5 days
[Bug 1974983] New: After upgrade xkeyboard-config to version
2.33-1.fc35 layout indicator stop react to layout switch by Ctrl-Shift key
combination.
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1974983
Bug ID: 1974983
Summary: After upgrade xkeyboard-config to version 2.33-1.fc35
layout indicator stop react to layout switch by
Ctrl-Shift key combination.
Product: Fedora
Version: rawhide
Status: NEW
Component: xkeyboard-config
Assignee: peter.hutterer(a)redhat.com
Reporter: mikhail.v.gavrilov(a)gmail.com
QA Contact: extras-qa(a)fedoraproject.org
CC: ajax(a)redhat.com, caillon+fedoraproject(a)gmail.com,
i18n-bugs(a)lists.fedoraproject.org,
negativo17(a)gmail.com, peter.hutterer(a)redhat.com,
rhughes(a)redhat.com, rstrode(a)redhat.com,
sandmann(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
After upgrade xkeyboard-config to version 2.33-1.fc35 layout indicator stop
react to layout switch by Ctrl-Shift key combination.
The bug is affected only Wayland session and only non standard layout switch
key combination.
Last good version is 2.32-3.fc35
--
You are receiving this mail because:
You are on the CC list for the bug.
2 months
[Bug 2088665] New: Noto Sans is chosen to display symbol characters
it doesn't contain
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=2088665
Bug ID: 2088665
Summary: Noto Sans is chosen to display symbol characters it
doesn't contain
Product: Fedora
Version: 36
Status: NEW
Component: google-noto-fonts
Assignee: tagoh(a)redhat.com
Reporter: talk(a)danielflaum.net
QA Contact: extras-qa(a)fedoraproject.org
CC: fonts-bugs(a)lists.fedoraproject.org,
i18n-bugs(a)lists.fedoraproject.org,
petersen(a)redhat.com, psatpute(a)redhat.com,
pwu(a)redhat.com, tagoh(a)redhat.com
Target Milestone: ---
Classification: Fedora
Created attachment 1881507
--> https://bugzilla.redhat.com/attachment.cgi?id=1881507&action=edit
A zipped sample PDF and image of relevant portion of PDF when affected by the
issue
Description of problem:
Given a PDF lacking embedded fonts which use certain characters (including →
and ≥), GNOME's Evince on Fedora 36 chooses to substitute the Noto Sans font,
which does not include these characters.
Version-Release number of selected component (if applicable):
How reproducible:
Successfully reproduced by two people independently.
Steps to Reproduce:
1. Boot a fresh copy of Fedora 36 (the Live version in a VM will do).
2. Open the attached sample PDF in GNOME Evince (aka Document Viewer).
3. Observe the missing characters in the second paragraph from the top of the
page.
Actual results:
See attached image.
Expected results:
The missing characters should be displayed properly as → (that is,
https://unicode-table.com/en/2192/).
Additional info:
The filer initially sought help at
https://ask.fedoraproject.org/t/missing-characters-in-pdfs-since-upgrade-...,
which may be informative in reproducing the issue.
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugzilla.redhat.com/show_bug.cgi?id=2088665
2 months
[Bug 2093080] New: Default fonts for Arabic do not match the font
packages list
by bugzilla＠redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=2093080
Bug ID: 2093080
Summary: Default fonts for Arabic do not match the font
packages list
Product: Fedora
Version: 36
Hardware: All
OS: Linux
Status: NEW
Component: fontconfig
Severity: medium
Assignee: tagoh(a)redhat.com
Reporter: awilliam(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: ajax(a)redhat.com, caillon+fedoraproject(a)gmail.com,
fonts-bugs(a)lists.fedoraproject.org,
gnome-sig(a)lists.fedoraproject.org,
i18n-bugs(a)lists.fedoraproject.org, mclasen(a)redhat.com,
pnemade(a)redhat.com, rhughes(a)redhat.com,
rstrode(a)redhat.com, sandmann(a)redhat.com,
tagoh(a)redhat.com
Target Milestone: ---
Classification: Fedora
There's a test case:
https://fedoraproject.org/wiki/QA:Testcase_i18n_default_fonts
which requires checking the default fonts for various languages against a list,
http://tagoh.fedorapeople.org/fonts/fc-test.sh .
The current default fonts for Arabic installs do not match the list. The list
states sans should be DejaVu Sans, serif should be FreeSerif or MPH 2B Damase,
and mono should be DejaVu Sans Mono. These may have been changed recently, as
our openQA reference text file expects them to be Noto Naskh Arabic (for both
sans and serif?) and PakType Naskh Basic for mono.
In any case, what we actually see doesn't match either the list or the openQA
reference file. We see "Noto Sans Arabic" and "PakType Naqsh" in the output
from the test, I think for serif (yes really) and monospace respectively.
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugzilla.redhat.com/show_bug.cgi?id=2093080
2 months