[Fedora-i18n-bugs] [Bug 1784650] New: Fontconfig is slow, causing stuttering and freezing
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1784650
Bug ID: 1784650
Summary: Fontconfig is slow, causing stuttering and freezing
Product: Fedora
Version: 31
Status: NEW
Component: fontconfig
Severity: high
Assignee: tagoh(a)redhat.com
Reporter: bepvte+bugzilla(a)gmail.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,
john.j5live(a)gmail.com, 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
Description of problem:
Fontconfig is much much slower than on other distros, and it stutters or
freezes applications that use it.
Version-Release number of selected component (if applicable):
Name : fontconfig
Version : 2.13.92
Release : 3.fc31
Architecture: x86_64
How reproducible:
I can reproduce this bug on a fresh Fedora 31 vm with the Xfce desktop and
google-noto-sans-* fonts installed.
Steps to Reproduce:
1. dnf install google-noto-sans-*
2. run gedit on the attached example file
alternatively
1. dnf install google-noto-sans-*
2. open firefox and browse to https://www.wikidata.org/wiki/Q52 (page with lots
of languages)
Actual results:
It takes around 60 seconds for gedit to become responsive to scrolling and
input. Mousepad is faster but still slow.
It takes firefox upwards of 5 minutes to get to first paint on a page with many
fonts or languages, compared to a simpler page.
Expected results:
Gedit should load files with many fonts at a similar speed as other distros.
The page should load quickly, like on Debian and others.
Additional info:
I have tried to diagnose the source of this issue in many ways.
Running `perf trace` on what sysprof indicated was the most busy function
(FcStrCmpIgnoreCaseAndDelims), shows that every name of every font family is
being compared to every other name of every other font family. I do not know if
this is a normal behaviour of fontconfig.
I have noticed the amount of calls to "FcStrCmpIgnoreCaseAndDelims" and program
startup time both drop to a similar amount as Debian's when all of the
"google-noto" configuration files in /etc/fonts/conf.d/ are deleted (These
files are not present in Debian). However, this might not be the source of the
problem:
In the Debian vm, with a copy of my computer's /etc/fonts/, including the
google-noto files, (I took care to ensure that there would be no broken
symlinks) and /usr/share/fonts, fontconfig does not stall any programs. The
amount of calls to FcStrCmpIgnoreCaseAndDelims is also much lower as well.
This led me to believe that it was a difference caused by compiler flags but
this does not seem to be the case. I tried to replace the optflags in the
package, except for the rpmbuild required debug ones, and found no difference.
I also checked to ensure that it was not caused by GCC version differences.
Debian results for mousepad:
1,845,449 calls to FcStrCmpIgnoreCaseAndDelims
Time: 5 seconds
Fedora results for mousepad:
11,658,380 calls to FcStrCmpIgnoreCaseAndDelims
Time: 23 seconds
https://perfht.ml/2tleJxN
Here is a link to a Firefox profiler result of the wikidata page, where in the
flame graph you can see that Firefox is spending most of its time in
fontconfig. You can also see "FirstNonBlankPaint" is at 50 seconds in the
marker table.
TLDR: Fontconfig matching is slow with all google-noto fonts installed, unless
you remove the noto config files. Using the same exact font directory and
config directory (including the noto config files) on Debian does not cause the
same problem.
--
You are receiving this mail because:
You are on the CC list for the bug.
1 year, 3 months
[Fedora-i18n-bugs] [Bug 1815510] New: Input hangul -> cursor becomes ahead in preedit
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1815510
Bug ID: 1815510
Summary: Input hangul -> cursor becomes ahead in preedit
Product: Fedora
Version: 32
Hardware: x86_64
OS: Linux
Status: NEW
Component: ibus-hangul
Assignee: pwu(a)redhat.com
Reporter: sangu.fedora(a)gmail.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org, pwu(a)redhat.com,
shawn.p.huang(a)gmail.com, tfujiwar(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
Input hangul -> cursor becomes ahead in preedit
Version-Release number of selected component (if applicable):
1.5.3-2.fc32.x86_64
How reproducible:
always in haungul input state
Steps to Reproduce:
1. gedit starts
2. switch hangul
3. input hangul
Actual results:
Expected results:
Additional info:
ibus-1.5.22-4.fc32.x86_64
gtk3-3.24.14-1.fc32.x86_64
--
You are receiving this mail because:
You are on the CC list for the bug.
1 year, 6 months
[Fedora-i18n-bugs] [Bug 1659748] New: Characters get deleted as you type
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1659748
Bug ID: 1659748
Summary: Characters get deleted as you type
Product: Fedora
Version: 29
Status: NEW
Component: ibus-m17n
Severity: high
Assignee: pnemade(a)redhat.com
Reporter: lohang(a)gmail.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org, pnemade(a)redhat.com,
shawn.p.huang(a)gmail.com
Target Milestone: ---
Classification: Fedora
Description of problem:
When you type Sinhala using Wijesekera keyboard layout through ibus the first
character gets deleted as soon as you start typing the second character. This
is similar to the bug previously reported for Fedora 28
https://bugzilla.redhat.com/show_bug.cgi?id=1617978
How reproducible:
(1) Steps to Reproduce:
1. Open Libre Office Writer (6.1.2.1)
2. Switch to Sinhala; Sinhalese (wijesekera (m17n))
3. Type keys vksIal kjSka
4. Hit space to separate the second word from the first word. And hit space at
the end of the first word.
Actual results: ක වීන්
Expected results: ඩනිෂ්ක නවීන්
Additional info:
(2) When you try this with emacs the results are a little different. I am
including it here assuming this is related to the same issue:
This is how to reproduce with emacs:
1. Open emacs 26.1 (build 1, x86_64-redhat-linux-gnu, GTK+ Version 3.23.2)
of 2018-08-13
2. Switch to Sinhala; Sinhalese (wijesekera (m17n))
3. Type keys vksIal kjSka
4. Hit space to separate the second word from the first word. And hit space at
the end of the first word.
Actual results: ඩනිෂ් කනවී න්
Expected results: ඩනිෂ්ක නවීන්
This adds a space before the last character of the first word, not where you
want the space to be.
--
You are receiving this mail because:
You are on the CC list for the bug.
1 year, 7 months
[Fedora-i18n-bugs] [Bug 1786596] New: Vertical alignment is off with PANGO_GRAVITY_EAST and PANGO_GRAVITY_HINT_LINE
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1786596
Bug ID: 1786596
Summary: Vertical alignment is off with PANGO_GRAVITY_EAST and
PANGO_GRAVITY_HINT_LINE
Product: Fedora
Version: 31
OS: Linux
Status: NEW
Component: pango
Assignee: pwu(a)redhat.com
Reporter: abetakehiko(a)gmail.com
QA Contact: extras-qa(a)fedoraproject.org
CC: caillon+fedoraproject(a)gmail.com,
fonts-bugs(a)lists.fedoraproject.org,
gnome-sig(a)lists.fedoraproject.org,
i18n-bugs(a)lists.fedoraproject.org,
john.j5live(a)gmail.com, mclasen(a)redhat.com,
pwu(a)redhat.com, rhughes(a)redhat.com,
rstrode(a)redhat.com, sandmann(a)redhat.com,
tagoh(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
Japanese chars and Latin chars in the same line do not vertically align when
the pango context's gravity is set to PANGO_GRAVITY_EAST and its hint set to
PANGO_GRAVITY_HINT_LINE.
Version-Release number of selected component (if applicable):
pango-1.44.7-1.fc31.x86_64
How reproducible:
Always
Steps to Reproduce:
pango-view --text="あーいうえお abcde" --gravity east --gravity-hint line
--font="NotoSerifJP 24"
Actual results:
Japanese chars and Latin chars in the same line do not vertically align.
Expected results:
Japanese chars and Latin chars in the same line vertically align as before.
Additional info:
--
You are receiving this mail because:
You are on the CC list for the bug.
1 year, 9 months
[Fedora-i18n-bugs] [Bug 1677534] New: texttopaps OOMs with 4GB text file!
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1677534
Bug ID: 1677534
Summary: texttopaps OOMs with 4GB text file!
Product: Fedora
Version: rawhide
Hardware: x86_64
OS: Linux
Status: NEW
Component: paps
Severity: high
Assignee: tagoh(a)redhat.com
Reporter: petersen(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: desktop-qa-list(a)redhat.com, eng-i18n-bugs(a)redhat.com,
i18n-bugs(a)lists.fedoraproject.org, rhel(a)vlasiu.net,
tagoh(a)redhat.com
Depends On: 1635160
Target Milestone: ---
Classification: Fedora
(Cloned from a RHEL7 report)
+++ This bug was initially created as a clone of Bug #1635160 +++
Description of problem:
Trying to print with cups a 4Gb text file fail.
texttopaps consume all the memory and is killed bye the system:
[1023082.003989] Out of memory: Kill process xxxx (texttopaps) score 955 or
sacrifice child
[1023082.005120] Killed process xxxx (texttopaps) total-vm:19958848kB,
anon-rss:7531872kB, file-rss:84kB, shmem-rss:0kB
/usr/lib/cups/filter/texttopaps 1 user1 "example" 1 "" sample-4Gb.txt
--- Additional comment from Akira TAGOH on 2018-10-03 11:54:48 SGT ---
Hmm, that log says it all. I don't think OOM is a software bug.
--- Additional comment from GV on 2018-10-03 13:00:48 SGT ---
Of course OOM is not a software bug. Where exactly did I said that?
The problem is texttopaps that allocate memory out of control.
texttopaps should allocate memory, use-it, release-it. Like any sane program.
The issue was not present on RHEL 6. I had printed the same file multiple times
with cups on RHEL6 and it worked just fine (not sure it was texttopaps cups
filter involved). Printing the file does not work on RHEL 7 and it should.
The 4Gb file is a testcase for our application we develop.
--- Additional comment from Jens Petersen on 2018-10-18 17:49:35 SGT ---
I don't think the paps in RHEL7 is significantly different from RHEL6.
As you say it could be due to changes in Cups?
The cups versions in RHEL 6, 7 and Fedora are certainly quite different.
Unless texttopaps really worked in RHEL6 it might make to reassign this to
cups?
--- Additional comment from GV on 2018-10-25 18:29:46 SGT ---
Since the real computer running RHEL 6 was already scrapped, I tried to make a
virtual machine using RHEL 6. Unfortunately, now I also get a crash on RHEL 6
when printing the 4GB file. I can't recall how much memory was in that
computer. The virtual machine had 8Gb of memory allocated.
Still, it does not matter that much.
I think texttopaps have an issue and it should be fixed or at least an
workaround should be available (other than 'don't print a file that large').
I'm afraid this is not a cups issue since I can reproduce the issue running the
texttopaps standalone.
Sincerely,
Gabriel
--- Additional comment from Akira TAGOH on 2018-11-16 17:59:04 SGT ---
paps isn't supposed to work with large files. to support it, most of code needs
to rewritten. this isn't realistic to provide an update for existing products.
for a workaround, you can remove paps package (or simply remove
/usr/share/cups/mime/paps.convs file) to fall back the text filter to texttops
which CUPS originally provides. this should works for this issue and as long as
you don't need the internationalization support for printing.
Hope that helps.
--- Additional comment from GV on 2018-12-07 14:03:33 SGT ---
It helps. Thank you!
Referenced Bugs:
https://bugzilla.redhat.com/show_bug.cgi?id=1635160
[Bug 1635160] texttopaps OOMs with 4GB text file!
--
You are receiving this mail because:
You are on the CC list for the bug.
1 year, 11 months
[Fedora-i18n-bugs] [Bug 1753295] New: Pango no longer supports bitmap fonts
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1753295
Bug ID: 1753295
Summary: Pango no longer supports bitmap fonts
Product: Fedora
Version: 31
Status: NEW
Component: pango
Assignee: tagoh(a)redhat.com
Reporter: jsbillin(a)umich.edu
QA Contact: extras-qa(a)fedoraproject.org
CC: caillon+fedoraproject(a)gmail.com,
fonts-bugs(a)lists.fedoraproject.org,
gnome-sig(a)lists.fedoraproject.org,
i18n-bugs(a)lists.fedoraproject.org,
john.j5live(a)gmail.com, mclasen(a)redhat.com,
pwu(a)redhat.com, rhughes(a)redhat.com,
rstrode(a)redhat.com, sandmann(a)redhat.com,
tagoh(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
In pango 1.44, pango dropped support for bitmap fonts, so font packages like
terminus-fonts and ucs-miscfixed-fonts no longer appear in GNOME applications
that use pango, such as Terminal. If you did a dnf system-upgrade from Fedora
30 and were using a bitmap font like Terminus, all your terminals will show the
rectangular placeholders the first time you start a terminal.
Bitmap fonts like terminus and ucs-miscfixed are much easier to read in a
terminal since they are pixel perfect. If Fedora 31 is going to disable
support for bitmap fonts, it needs to be announced and the package maintainers
of those bitmap font packages are going to need to find a way to convert their
fonts.
Version-Release number of selected component (if applicable):
pango-1.44.6-1.fc31.x86_64
How reproducible:
always
Steps to Reproduce:
1. install 'terminus-fonts'
2. Start GNOME terminal
3. Open Terminal Preferences
4. Attempt to change the font to Terminus font
Actual results:
If you upgraded from Fedora 30 and had Terminus fonts as the default font, the
first time you launch Terminal, all the text will be the rectangular
placeholders. If you search for the Terminus font in the preferences, you
won't be able to find it.
Expected results:
Use the Terminus font as usual
Additional info:
It appears that pango switched from using FreeType to HarfBuzz, which only
supports truetype. I rebuilt the Fedora 30 pango package (1.43) for Fedora 31,
downgraded it, and now my terminals are fine showing my bitmap fonts.
--
You are receiving this mail because:
You are on the CC list for the bug.
2 years
[Fedora-i18n-bugs] [Bug 1711646] New: Scriptlet error when updating through SSH
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1711646
Bug ID: 1711646
Summary: Scriptlet error when updating through SSH
Product: Fedora
Version: 30
Status: NEW
Component: ibus
Assignee: tfujiwar(a)redhat.com
Reporter: dreua(a)posteo.de
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:
When I update or reinstall ibus-1.5.20-4.fc30 over SSH, dnf outputs the error
"X11 connection rejected because of wrong authentication.".
This is caused by
https://src.fedoraproject.org/rpms/ibus/blob/master/f/ibus.spec#_338
running `sudo ibus write-cache --system` results in the same error when called
over SSH.
Version-Release number of selected component (if applicable):
ibus-1.5.20-4.fc30
How reproducible:
Always over SSH, never when using the terminal directly on the machine.
Steps to Reproduce:
1. Reinstall or update to ibus-1.5.20-4.fc30
Actual results:
Running scriptlet: ibus-1.5.20-4.fc30.x86_64
X11 connection rejected because of wrong authentication.
X11 connection rejected because of wrong authentication.
Verifying : ibus-1.5.20-4.fc30.x86_64
Expected results:
Running scriptlet: ibus-1.5.20-4.fc30.x86_64
Verifying : ibus-1.5.20-4.fc30.x86_64
--
You are receiving this mail because:
You are on the CC list for the bug.
2 years
[Fedora-i18n-bugs] [Bug 1749943] New: pango contain header with wrong hb.h include
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1749943
Bug ID: 1749943
Summary: pango contain header with wrong hb.h include
Product: Fedora
Version: rawhide
Status: NEW
Component: pango
Assignee: tagoh(a)redhat.com
Reporter: vascom2(a)gmail.com
QA Contact: extras-qa(a)fedoraproject.org
CC: caillon+fedoraproject(a)gmail.com,
fonts-bugs(a)lists.fedoraproject.org,
gnome-sig(a)lists.fedoraproject.org,
i18n-bugs(a)lists.fedoraproject.org,
john.j5live(a)gmail.com, mclasen(a)redhat.com,
pwu(a)redhat.com, rhughes(a)redhat.com,
rstrode(a)redhat.com, sandmann(a)redhat.com,
tagoh(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
I am try to build eiskaltdcpp and it fsilrd with error:
BUILDSTDERR: /usr/include/pango-1.0/pango/pango-coverage.h:28:10: fatal error:
hb.h: No such file or directory
BUILDSTDERR: 28 | #include <hb.h>
BUILDSTDERR: | ^~~~~~
Full build log
https://kojipkgs.fedoraproject.org/work/tasks/5226/37235226/build.log
Version-Release number of selected component (if applicable):
pango-1.44.6-1.fc32
How reproducible:
Always.
As I see this line was added in pango 1.44. But harfbuzz's hb.h in Fedora
placed at /usr/include/harfbuzz/
Need to solve it for build other packages.
--
You are receiving this mail because:
You are on the CC list for the bug.
2 years, 2 months
[Fedora-i18n-bugs] [Bug 1807534] New: pyzy requires Python 2 to build
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1807534
Bug ID: 1807534
Summary: pyzy requires Python 2 to build
Product: Fedora
Version: rawhide
Status: NEW
Component: pyzy
Assignee: pwu(a)redhat.com
Reporter: pviktori(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org,
petersen(a)redhat.com, pwu(a)redhat.com
Blocks: 1803205 (BRPY27)
Target Milestone: ---
Classification: Fedora
Python 2 reached upstream end-of-life in January 2020. In Fedora Rawhide, it's
now provided from the compat package `python27`.
Packages that only use Python 2 at build time, like pyzy, had a general
exception to keep using it in Fedora 31. Now, the dependency should be removed.
Let us know if you need any help investigating or removing the dependency.
(There are dozens of packages like this, so we didn't investigate this one
thoroughly. We assume you know the package best.)
If it's possible that the dependency won't be removed in Fedora 33. Please
request a FESCo exception. You can refer to the exception for mercurial as an
example: https://pagure.io/fesco/issue/2243
It's good to mention:
- What is the reason for the Python 2 build dependency?
- What are the upstream/community plans/timelines regarding Python 2?
- What is the guidance for porting the build to Python 3? (Assuming that there
is someone who generally knows how to port to Python 3, but doesn't know
anything about the particular package, what are the next steps to take?)
If you need anything from us, or something is unclear, please mention it here.
Referenced Bugs:
https://bugzilla.redhat.com/show_bug.cgi?id=1803205
[Bug 1803205] Tracking: Packages BuildRequiring python27
--
You are receiving this mail because:
You are on the CC list for the bug.
2 years, 4 months
[Fedora-i18n-bugs] [Bug 1707001] New: Ibus-pinyin ERROR
by bugzilla@redhat.com
https://bugzilla.redhat.com/show_bug.cgi?id=1707001
Bug ID: 1707001
Summary: Ibus-pinyin ERROR
Product: Fedora
Version: 30
Hardware: All
OS: Linux
Status: NEW
Component: ibus-pinyin
Severity: urgent
Assignee: pwu(a)redhat.com
Reporter: 1990konger(a)gmail.com
QA Contact: extras-qa(a)fedoraproject.org
CC: i18n-bugs(a)lists.fedoraproject.org, pwu(a)redhat.com,
shawn.p.huang(a)gmail.com, tfujiwar(a)redhat.com
Target Milestone: ---
Classification: Fedora
Description of problem:
The input method is incorrect and cannot be recognized correctly. The text to
be entered is required.
For example, 'nihao' should be right, '你好'.
Now enter 'nihao' to show 'niha o'.Unrecognized what you need
'ni' is a word.
'hao' is a word.
'ha' is a word but the display is wrong at this time and cannot be selected.
Unable to recognize 'hao'. Affect all input suggestions to withdraw to the old
version
Version-Release number of selected component (if applicable):
Name : ibus-pinyin
Version : 1.5.0
Release : 16.fc30
How reproducible:
'nihao' is meaning is 'Hello'
'niha o', it is an error
Steps to Reproduce:
1.All Chinese input is wrong
Additional info:
--
You are receiving this mail because:
You are on the CC list for the bug.
2 years, 4 months