https://bugs.documentfoundation.org/show_bug.cgi?id=152969
Pierre Fortin <p...@pfortin.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |INVALID
Status|NEEDINFO |RESOLVED
--- Comment #6 from Pierre Fortin <p...@pfortin.com> ---
(In reply to Michael Weghorn from comment #5)
> Was there any recent update of ICU on your system as well, or just Qt?
Looks like ICU also:
$ rpm -qa --last | grep icu
lib64icu-devel-72.1-1.mga9.x86_64 Sun 08 Jan 2023 08:08:18 AM EST
lib64icu72-72.1-1.mga9.x86_64 Sun 08 Jan 2023 08:05:05 AM EST
icu72-data-72.1-1.mga9.noarch Sun 08 Jan 2023 08:05:05 AM EST
### The following are still in the system
lib64icu71-71.1-2.mga9.x86_64 Mon 20 Jun 2022 09:15:09 AM EDT
icu71-data-71.1-2.mga9.noarch Mon 20 Jun 2022 09:15:09 AM EDT
lib64basicusageenvironment1-2021.06.25-3.mga9.x86_64 Thu 24 Mar 2022 11:54:19
AM EDT
lib64icu69-69.1-2.mga9.x86_64 Thu 24 Mar 2022 11:48:03 AM EDT
icu69-data-69.1-2.mga9.noarch Thu 24 Mar 2022 11:48:03 AM EDT
> Some commands you might check: What do these commands output for you?
> (You'll have to adapt paths to where the libraries are on Mageia). Following
> what I see on Debian testing):
>
>
> $ nm -gD /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 | grep u_str
> U u_strToLower_72
> U u_strToUpper_72
$ nm -gD /lib64/libQt5Core.so.5 | grep u_str
U u_strToLower
U u_strToUpper
> $ ldd /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 | grep icu
> libicui18n.so.72 => /lib/x86_64-linux-gnu/libicui18n.so.72
> (0x00007f6ef1e00000)
> libicuuc.so.72 => /lib/x86_64-linux-gnu/libicuuc.so.72
> (0x00007f6ef1c02000)
> libicudata.so.72 => /lib/x86_64-linux-gnu/libicudata.so.72
> (0x00007f6eef600000)
$ ldd /lib64/libQt5Core.so.5 | grep icu
libicui18n.so.72 => /lib64/libicui18n.so.72 (0x00007fae2be00000)
libicuuc.so.72 => /lib64/libicuuc.so.72 (0x00007fae2bc0a000)
libicudata.so.72 => /lib64/libicudata.so.72 (0x00007fae2c133000)
> $ nm -gD /lib/x86_64-linux-gnu/libicuuc.so.72 | grep -i u_strToLower
> 0000000000143330 T u_strToLower_72
$ nm -gD /lib64/libicuuc.so.72 | grep -i u_strToLower
000000000013b0c0 T u_strToLower
> (In reply to Pierre Fortin from comment #3)
> > Operating System: Mageia 9
>
> Is that a stable Mageia release or (as a quick look on the Mageia website
> suggests) some alpha/pre-release?
Other than LOTS of updates; I've been using it as my primary OS since Feb/22
when I acquired the machine. On the whole, I have no problem with alpha/beta
software; support tends to be better than with released software... :)
> > > You can try this:
> > > - launch a terminal/console
> > > - type:
> > > export SAL_USE_VCLPLUGIN=gen
> > > - scalc &
> >
> > Splash and quits; no error message with this var. Just installed today's
> > build; same...
For completeness:
$ scalc --safe-mode
warn:sal.osl:377911:377911:sal/osl/unx/module.cxx:152:
dlopen(/home/pfortin/libreoffice/opt/libreofficedev7.6/program/libvclplug_kf5lo.so,
257): /lib64/libQt5Core.so.5: undefined symbol: u_strToLower
warn:sal.osl:377911:377911:sal/osl/unx/module.cxx:152:
dlopen(/home/pfortin/libreoffice/opt/libreofficedev7.6/program/libvclplug_gtk3_kde5lo.so,
257): /lib64/libQt5Core.so.5: undefined symbol: u_strToLower
warn:desktop.app:377911:377911:unotools/source/ucbhelper/localfilehelper.cxx:87:
error removing directory
file:///home/pfortin/libreoffice/opt/libreofficedev7.6/program/../../UserInstallation/user/extensions
The last line is new.
$ ll libreoffice/opt/UserInstallation/user/
total 16
drwxr-xr-x 2 pfortin pfortin 4096 Jan 11 14:58 config/
drwxr-xr-x 2 pfortin pfortin 4096 Jan 11 14:58 extensions/
-rw------- 1 pfortin pfortin 329 Jan 11 14:58 registrymodifications.xcu
drwxr-xr-x 13 pfortin pfortin 4096 Jan 11 14:57 SafeMode/
$ ll libreoffice/opt/UserInstallation/user/extensions/
total 4
-rw-r--r-- 1 pfortin pfortin 40 Jan 11 14:58 buildid
> Does this still happen after deleting your LO profile or when you start in
> safe mode?
Safe mode: see above.
$ mv libreoffice/opt/UserInstallation/user
libreoffice/opt/UserInstallation/user.off
No change.
> Does it also happen when you start 'soffice --writer' or just 'soffice'
> instead of 'scalc'?
Interesting: 'soffice --writer' & 'soffice --calc' both start up.
'scalc' and 'swriter' fail as originally reported...
> Does the LibreOffice installed via Mageia packages work, in particular the
> kf5 VCL plugin (i.e. the one that says "VCL: kf5 (...)" in "Help" -> "About
> LibreOffice"? (Don't know whether Mageia provides that as a package.)
Never saw VCL (whatever that is)... Here's the version info:
Version: 7.4.3.2
Build ID: 40(Build:2)
CPU threads: 20; OS: Linux 6.1; UI render: default; VCL: gtk3
Locale: en-US (en_US.UTF-8); UI: en-US
Calc: threaded Jumbo
And for 7.6:
Version: 7.6.0.0.alpha0+ (X86_64) / LibreOffice Community
Build ID: 28ff4647e9dac8eebe3a169e828bacc8dc78e363
CPU threads: 20; OS: Linux 6.1; UI render: default; VCL: gtk3
Locale: en-US (en_US.UTF-8); UI: en-US
Calc: threaded
Argh... this may be an invalid bug report... When I thought both 7.4.3.2 and
7.6.0.0 were shutdown, I got this:
$ ps aux | grep office
pfortin 1814092 0.0 0.0 164560 5684 pts/16 Sl Jan10 0:00
/home/pfortin/libreoffice/opt/libreofficedev7.6/program/oosplash --calc
pfortin 1814127 0.1 1.2 38046616 1708112 pts/16 Sl Jan10 3:01
/home/pfortin/libreoffice/opt/libreofficedev7.6/program/soffice.bin --calc
--splash-pipe=5
Somehow, these windows were nowhere to be found. I did find the konsole from
where they were started and fg + Ctrl+C got rid of them. Now scalc and swriter
start just fine.
Hmm... I can do 'scalc &' multiple times and get a new instance each time.
Somehow, I thought all previous instances were closed (no calc windows or
taskbar icons); these appear to have been causing this issue. Now that I can
start multiple instances; I don't get how the instances without windows caused
this... Closing as INVALID vs UNCONFIRMED -- unless there's something here you
feel needs a second look... Sorry for the noise that still leaves me
puzzled... no idea how 2 windows got disowned or ?
--
You are receiving this mail because:
You are the assignee for the bug.
