e
update, it seems to be correct. Please note this is a workaround, a
fix for another issue and not the intended security update. I work on
three more security fixes of ICU for Trixie, it is currently in the
testing.
Regards,
Laszlo/GCS
sting
Regards,
Laszlo/GCS
diff -Nru patch-2.8/debian/changelog patch-2.8/debian/changelog
--- patch-2.8/debian/changelog 2025-04-06 21:10:31.0 +0200
+++ patch-2.8/debian/changelog 2025-06-13 20:59:38.0 +0200
@@ -1,3 +1,13 @@
+patch (2.8-2) unstable; urgency=medium
+
+ [ Samuel Thi
appreciated. Upload is in progress.
Cheers,
Laszlo/GCS
of future point-releases.
There is the above mentioned build break. Please discuss that first.
Thanks,
Laszlo/GCS
[1] https://bugs.debian.org/1107408
. May I wait for something? Trixie release is coming soon.
Regards,
Laszlo/GCS
[1]
https://github.com/unicode-org/icu/commit/2c667e31cfd0b6bb1923627a932fd3453a5bac77
[2] https://unicode-org.atlassian.net/browse/ICU-22973
ense by the way as the report is from November 2024 and 77.1 release
is made later in March, 2025. I need to dig into it, probably with Git
bisect to find the fixing commit.
I couldn't test the 74.1 version and earliter yet.
Regards,
Laszlo/GCS
On Tue, May 27, 2025 at 4:06 PM Sebastian Ramacher wrote:
> > This symbol is introduced in -5 so it should be 3.46.1~ (~ for
> > backports).
>
> Sorry, it should have been 3.46.1-5~.
My bad. I should upload a fix. Anything else I might have missed?
Thanks,
Laszlo/GCS
the d/changelog
[x] I reviewed all changes and I approve them
[x] attach debdiff against the package in testing
Regards,
Laszlo/GCS
[1] https://issues.chromium.org/issues/415397143
diff -Nru sqlite3-3.46.1/debian/changelog sqlite3-3.46.1/debian/changelog
--- sqlite3-3.46.1/debian/change
is affects the installer due to an udeb, but none of these changes
risk it in my opinion.
[ Checklist ]
[x] all changes are documented in the d/changelog
[x] I reviewed all changes and I approve them
[x] attach debdiff against the package in testing
Regards,
Laszlo/GCS
diff -Nru fuse3-3.1
xie or can I defer this upload?
Cheers,
Laszlo/GCS
On Thu, May 8, 2025 at 1:48 PM Paul Gevers wrote:
> If the upload can happen soonish (next couple of days), please go ahead.
Thanks, it is uploaded and for all release architectures it is
already built and installed to the pool.
Laszlo/GCS
On Thu, May 8, 2025 at 6:51 AM Salvatore Bonaccorso wrote:
> Thanks a lot for double-checking. So that is bit worriesome if the
> tagged version and the released version do not match.
Yes, it is. Contacted upstream directly. I got the reply that they
are investigating this issue.
Laszlo/GCS
issue. It seems the tag is wrong. If you
download that and check VERSION.txt it will state 4.8.1 which is
wrong. You need to download the tarball from the releases page. That's
correct.
Regards,
Laszlo/GCS
On Sun, May 4, 2025 at 8:23 PM Debian FTP Masters
wrote:
>* Update debian/patches/utsn.patch for hurd-any (Closes: #1104235)
Please communicate before NMUs.
Thanks,
Laszlo/GCS
n't see the patch attached.
Cheers,
Laszlo/GCS
ore fuse_session unconditionally.
+ * Backport upstream fixes:
+- fix meson function tests,
+- meson.build: make special_funcs check more reliable.
+
+ -- Laszlo Boszormenyi (GCS) Sun, 27 Apr 2025 08:10:01 +0200
+
fuse3 (3.17.1+git250416-1) unstable; urgency=medium
* New git snapshot
sion in NTLM auth in 0.34.0,
+- add docs for ne_ssl_proto_name(), ne_ssl_protovers() and
+ ne_get_request_target() functions.
+
+ -- Laszlo Boszormenyi (GCS) Sat, 26 Apr 2025 16:44:36 +0200
+
neon27 (0.34.0-1) unstable; urgency=medium
* New upstream release.
diff -Nur -x po -x doc
using Appstream information, like the isenkram
> package, will know that this package is useful on machines where the USB
> IDs are discovered.
Do you consider this forwarding to upstream? It would help all users
of liquidctl, not just Debian ones.
Regards,
Laszlo/GCS
Hi Nick,
Can you please try upgrading your system? It was #1093826 [1] I
believe, a bug in python3-pil itself.
Regards,
Laszlo/GCS
[1] https://bugs.debian.org/1093826
aphic error
>
> My guess is that this is linked to the CVE-2025-27809 fix.
Just found, it is [1]. I will do other tests if I can fix it without
upstream involvement.
Thanks for the confirmation,
Laszlo/GCS
[1]
https://github.com/Mbed-TLS/mbedtls/commit/20c7748575320ff721d99ab64456532c0205c8d3
vulnerability is
in the concat_ws() function, which can cause a memory error if the
separator string is very large (hundreds of megabytes). The fix is
already in place and a small one. I plan to upload it tomorrow
afternoon.
Hope this helps,
Laszlo/GCS
Hi Bastian,
Thanks for your pointers. Might you check the work in progress update
of socat if the copyright file is now correct and complete?
Thanks,
Laszlo/GCS
[1] dget -x https://people.debian.org/~gcs/socat_1.8.0.3-1.dsc
still broken. I'm going to retest it, but
sure something strange is going on.
Regards,
Laszlo/GCS
Control: unmerge -1
Control: reassign -1 src:loggedfs
Control: tags -1 +patch
Hi,
Your package forces C++11 standard during compilation. However it uses
ICU which is C++17 now. I attached a patch to use the C++17
standard and make your package compile again.
Regards,
Laszlo/GCS
Description: fix
Control: unmerge -1
Control: reassign -1 src:libxml++2.6
Control: tags -1 +patch
Hi,
Your package forces C++11 standard during compilation. However it uses
ICU which is C++17 now. I attached a patch not to force the C++
standard and make your package compile again.
Regards,
Laszlo/GCS
and uses that C++ standard. Then if you still would like to
force the C++ standard, compile with this: "g++ test.cpp -o test
-std=c++17".
I attach a patch for both cases, choose the one that you want and fix
_your_ source.
Regards,
Laszlo/GCS
Description: fix FTBFS with ICU 75.1+
Do not f
ith the former, might use different flags / values with FUSE v3.17.1
- need to check.
But if this is the reason, gvfs will need a binNMU with the final FUSE
v3.17.1 release.
Regards,
Laszlo/GCS
[1]
https://github.com/libfuse/libfuse/commit/3ae5ca7443348aabad9bc71b9d5b0999f8292379
;fuse: use
> fuse_(un)set_feature_flag instead of setting manually" from
> <https://gitlab.gnome.org/GNOME/gvfs/-/merge_requests/248>?
Yes, exactly commit 77581bc4 which can be viewed only at:
https://gitlab.gnome.org/GNOME/gvfs/-/commit/77581bc426025bd04bd7ccb3b61e9c20114bc44b?merge_request_iid=248
Laszlo/GCS
posed fix is already provided [4].
In very short, gvfs needs to be fixed. I can't do anything on the FUSE side.
Regards,
Laszlo/GCS
[1]
https://github.com/libfuse/libfuse/blob/fuse-3.17.x/include/fuse_common.h#L596
[2]
https://github.com/libfuse/libfuse/blob/fuse-3.17.x/include/fuse
.
Regards,
Laszlo/GCS
[1] https://github.com/libexpat/libexpat/issues/980#issue-2922101832
[2] https://github.com/libexpat/libexpat/issues/980#issuecomment-2734559744
[3] https://github.com/libexpat/libexpat/issues/980#issuecomment-2735149399
On Fri, Feb 28, 2025 at 10:09 AM Emilio Pozuelo Monfort
wrote:
> Sounds good, but let's wait for kibi's ack.
Almost two weeks have passed. Friendly ping, do you ACK or NACK this
fuse3 transition kibi?
Cheers,
Laszlo/GCS
ou could take the approach from my previous patch, and
> upgrade to a newer upstream release.
That would require a double transition, starting with protobuf first.
I need to check it with the new abseil version in Sid.
Regards,
Laszlo/GCS
On Mon, Mar 10, 2025 at 9:15 PM wrote:
> grpc was noted before, but failed for new reasons. Filed #1100044 with
> an NMU, it looks like GCS will incorporate it into a maintainer upload
> soon.
It is just accepted to the archive. Thanks for giving me some time to do it.
You can cance
On Sat, Mar 8, 2025 at 9:25 PM Mike Gabriel wrote:
> I am ready to upload qtcontacts-sqlite whenever sqlite3 +
> libsqlite3-ext-icu has landed in unstable. Please go ahead.
It's already there. You can go on with your upload.
Laszlo/GCS
patch provided [12] which just recently uploaded.
The good: I'm here and will check the transition, send patches for any
unexpected problems.
The bad: some of the packages have no active maintenance and I can't
provide NMUs.
Thanks for considering,
Laszlo/GCS
[1]
h
build
dependencies. Basically in d/control:
python3-all-dev:any (>= 3.5) ,
+ python3-setuptools ,
rename,
Please do apply this soon.
Regards,
Laszlo/GCS
Control: tags -1 +patch
Hi,
I provide a working, tested patch to fix this issue.
Regards,
Laszlo/GCS
Description: fix FTBFS with ICU 75.1+
Set correct C++ standard.
Author: Laszlo Boszormenyi (GCS)
Forwarded: no
Last-Update: 2024-11-18
---
--- aegisub-3.2.2+dfsg.orig/configure
+++ aegisub
Package: src:qmf
Version: 5.0~git20241030.1476b88-2
Severity: important
Tags: patch
User: debian-de...@lists.debian.org
Usertags: icu76
Hi,
Your package FTBFS with ICU 76.1 available from experimental.
I provide a working, tested patch to fix this issue.
Regards,
Laszlo/GCS
Description: fix
Control: tags -1 +patch
Hi,
I provide a working, tested patch to fix this issue.
Regards,
Laszlo/GCS
Description: fix FTBFS with ICU 75.1+
Set correct C++ standard.
Author: Laszlo Boszormenyi (GCS)
Forwarded: no
Last-Update: 2025-03-02
---
--- xalan-1.12.orig/CMakeLists.txt
+++ xalan-1.12
Package: src:maim
Version: 5.7.4-2
Severity: important
Tags: patch
User: debian-de...@lists.debian.org
Usertags: icu76
Hi,
Your package FTBFS with ICU 76.1 available from experimental.
I provide a working, tested patch to fix this issue.
Regards,
Laszlo/GCS
Description: fix FTBFS with ICU 75.1
Control: tags -1 +patch
Hi,
I provide a working, tested patch to fix this issue.
Regards,
Laszlo/GCS
Description: fix FTBFS with ICU 75.1+
Set correct C++ standard.
Author: Laszlo Boszormenyi (GCS)
Forwarded: no
Last-Update: 2025-03-03
---
--- dimbl-0.17.orig/src/Makefile.am
+++ dimbl-0.17
Control: tags -1 +patch
Hi,
I provide a working, tested patch to fix this issue.
Regards,
Laszlo/GCS
diff -pruN 7.6-4/debian/patches/series 7.6-4ubuntu1/debian/patches/series
--- 7.6-4/debian/patches/series 2023-05-06 19:37:16.0 +
+++ 7.6-4ubuntu1/debian/patches/series 2025-01-09 11
Control: tags -1 +patch
Hi,
I provide a working, tested patch to fix this issue.
Regards,
Laszlo/GCS
--- a/text-icu.cabal
+++ b/text-icu.cabal
@@ -125,7 +125,7 @@ library
Data.Text.ICU.Spoof.Pure
Data.Text.ICU.Text
c-sources: cbits/text_icu.c
- cc-options: -Wall -Wextra
Control: tags -1 +patch
Hi,
There are two ways to have this fixed. There's the minimal patch or
two upstream commits backported. It's your choice which way to go when
the transition starts.
Regards,
Laszlo/GCS
Description: fix FTBFS with ICU 75.1+
Set correct C++ standard.
Auth
t; deferred upload. debdiff attached, and also sent as PR:
>
> https://github.com/gcsideal/syslog-ng-debian/pull/13
Merged. Uploaded with other fixes, you can cancel the NMU.
Laszlo/GCS
i side?
Cheers,
Laszlo/GCS
libsqlite3-ext-icu, hope you don't object. Then it should depend on
libsqlite3-0 as it doesn't work on its own, right?
Laszlo/GCS
#x27;t work and
I've found different information about why it may fail. Either due to
I try to run it as root and/or that dlopen() can't handle
subdirectories itself.
Regards,
Laszlo/GCS
are explained below:\fR
.IP "Around line 234:" 4
.IX Item "Around line 234:"
Unterminated L<...> sequence
I've fixed it and you can check my iteration [1] if it meets your approval.
Thanks for your contribution,
Laszlo/GCS
[1] dget https://people.debian.org/~gcs/apt-src_0.25.4.dsc
CLI shell:
sqlite> .load libSqliteIcu.so
Error: /lib/x86_64-linux-gnu/libSqliteIcu.so: undefined symbol:
sqlite3_sqliteicu_init
But I can load the csv extension for example. Did you test the module
on your end? Maybe sqlite3 itself needs to build with ICU support [1].
Regards,
Laszlo/G
can check if non-ASCII character support works in
> lomiri-addressbook-app. Thanks!!!
Are you sure it builds on all architectures? It might be better to
upload to experimental and check it from there. Wouldn't that be more
wise?
Cheers,
Laszlo/GCS
c as fuse3 has an udeb.
Thanks for consideration,
Laszlo/GCS
[1] https://bugs.debian.org/1098735
Hi Helge,
On Wed, Feb 26, 2025 at 7:47 PM Helge Kreutzmann wrote:
> So at the weekend I'll prepare the NMU. You can fine tune it
> afterwards, but this way apt-src is in Trixie.
Please send it to me when it's ready. I'll check and sponsor the upload.
Regards,
Laszlo/GCS
]: Leaving directory '/build/vuos-0.9.1/obj-x86_64-linux-gnu'
make[3]: *** [vufuse/CMakeFiles/vufuse.dir/build.make:96:
vufuse/CMakeFiles/vufuse.dir/vufuse.c.o] Error 1
If you have a connection with your upstream, please notify them about
this issue. The transition may start in a week.
Thanks,
Laszlo/GCS
xie
> freeze)?
I do have. In the hope that at least some of the above bugs get
fixed. I do not think it is worth the risks to have those with Trixie.
But I'm open to hear your thoughts.
Regards,
Laszlo/GCS
[1] https://github.com/libfuse/libfuse/issues/945
[2] https://github.com/libfuse/libfu
ue at least. Can you retry your testing if
this issue still affects you or not?
Thanks for your work,
Laszlo/GCS
t it also help to add Provides: python3-google-api-python-client
to my package? This way it would be visible as it is available from
that binary package.
Laszlo/GCS
status.
OK, I will disable the mentioned tests in the next few days.
Regards,
Laszlo/GCS
h the
maintainers first. As several upstream releases packaging is a big
change and would be an NMU. Then he reported no success, no answer of
any kind and things stand as it is for two weeks. I do not have the
capacity now to further go with this situation.
Regards,
Laszlo/GCS
n ruby-vips. Already reported upstream and got a fix. The actual
patch is this [1]. Unfortunately it seems this package is not
maintained, at least the last upload was two years ago. Meanwhile
three upstream releases were tagged, including a somewhat big (v2.2.0)
one.
Cheers,
Laszlo/GCS
[1]
https:/
g to disappear. Making this project a
question to carry around. Sure, it will be OK for Trixie, not sure
after that. It's up to you if you can find (hope) someone to continue
with its development.
Regards,
Laszlo/GCS
[1] https://sourceforge.net/p/lsdvd/git/ci/master/tree/
[2] https://gitweb
re than ten years, has some working flaws and might not work with
Blu-Ray disks.
I was about to ask for its removal. But if the team has the power to
continue with its development as well, that would be good. I will
definitely not have time for it.
Regards,
Laszlo/GCS
d know this package. As of myself, I have
less time for a while than I want to.
Regards,
Laszlo/GCS
On Sun, Oct 27, 2024 at 11:17 PM Moritz Mühlenhoff wrote:
> Am Sun, Oct 27, 2024 at 03:58:13PM +0100 schrieb László Böszörményi (GCS):
> If we do an update anyway, we could also piggyback the straightforward
> patch for CVE-2023-52890/#1073248, what do you think?
> https://github.com
s for considering,
Laszlo/GCS
[1] https://bugs.debian.org/983359
diff -Nru ntfs-3g-2022.10.3/debian/changelog ntfs-3g-2022.10.3/debian/changelog
--- ntfs-3g-2022.10.3/debian/changelog 2024-06-23 14:34:22.0 +0200
+++ ntfs-3g-2022.10.3/debian/changelog 2024-10-27 15:16:10.0 +0100
@@
'.
> Error - Machine initialization failed.
>
> So, the vice package is unusable.
As noted by Spiro, this bug is invalid. Package documentation states
that ROM files can't be distributed due to their non-free licence. You
need to get those by yourself.
Regards,
Laszlo/GCS
On Sun, Sep 1, 2024 at 3:39 PM László Böszörményi (GCS) wrote:
> Please try the proposed package update [1]. Can you confirm that all
> Ruby parts are in place now?
Friendly ping Praveen, what's your status on this?
abandoned by its
upstream for four years. Its last real upload was more than two years
ago; the uploader is in Cc.
Regards,
Laszlo/GCS
[1] https://bugs.debian.org/1084676
back my
findings here.
Regards,
Laszlo/GCS
[2] fix. The choice is yours of course to include these or
not.
Regards,
Laszlo/GCS
[1] https://sqlite.org/src/info/074002718b2ecb9f
[2] https://sqlite.org/src/info/e6bec37ea1ca51e1
he NMU, I'm just going to
upload the update as is - crediting you of course.
Regards,
Laszlo/GCS
you
> tar up the the current state and make it available somewhere?
That's also there [1].
Regards,
Laszlo/GCS
[1] https://people.debian.org/~gcs/apt-src_0.25.4.tar.xz
On Sun, Sep 1, 2024 at 6:33 PM Helge Kreutzmann wrote:
> If anything is unclear, I'm happy to resolve this.
I started the update back then. Soon I got stuck, the current state
is online [1]. The pod can't be processed and in the next few days I
don't have time to look at it.
R
Hi Praveen,
Please try the proposed package update [1]. Can you confirm that all
Ruby parts are in place now?
Of course, I will remove protobuf_c.so from
/usr/lib/ruby/vendor_ruby/google/ , the other installed place should
be the official location.
Regards,
Laszlo/GCS
[1] dget -x https
ad the package, giving you all the credits of course.
Thanks,
Laszlo/GCS
ancel your NMU.
Thanks for the help,
Laszlo/GCS
al uploads of Graphviz also have
this fix already.
[ Checklist ]
[x] *all* changes are documents in the d/changelog
[x] I reviewed all changes and I approve them
[x] attach debdiff against the package in bullseye
[x] the issue is verified as fixed in unstable
Thanks for considering,
Las
al uploads of Graphviz also have
this fix already.
[ Checklist ]
[x] *all* changes are documents in the d/changelog
[x] I reviewed all changes and I approve them
[x] attach debdiff against the package in bullseye
[x] the issue is verified as fixed in unstable
Thanks for considering,
Las
n a stable update for
> Debian 12 and—if we are quick enough—Debian 11. This fix was included
> in an Ubuntu SRU for 22.04 LTS with a test case.
> https://launchpad.net/bugs/2015017
Thanks for your suggestion. What's up with the reported regressions
in Ubuntu [1]?
Regards,
Laszlo/GCS
[1
ormal
working in any way.
[ Checklist ]
[x] *all* changes are documents in the d/changelog
[x] I reviewed all changes and I approve them
[x] attach debdiff against the package in bullseye
[x] the issue is verified as fixed in unstable
Thanks for considering,
Laszlo/GCS
diff -Nru ntfs-3g-2022
ormal
working in any way.
[ Checklist ]
[x] *all* changes are documents in the d/changelog
[x] I reviewed all changes and I approve them
[x] attach debdiff against the package in bullseye
[x] the issue is verified as fixed in unstable
Thanks for considering,
Laszlo/GCS
diff -Nru nt
Hi Adrien,
On Thu, Jun 13, 2024 at 12:06 PM Adrien Nader wrote:
> Since it's ready and passing, I'd like to more forward with this. Do you
> have any objection?
As noted, I'm waiting for your upload. Friendly ping, how it goes?
Did you lose interest and I should take over?
Regards,
Laszlo/GCS
/debian/rocksdb ?
Please do. I will try to answer other parts of your email today.
Regards,
Laszlo/GCS
ersion will be in stable still for a while, maybe it would be
> worth
> to patch it (a patch attached).
Thanks, I'm going to upload it soon to Sid. That will make your patch
easier to be accepted to Bookworm.
Regards,
Laszlo/GCS
hing out. Please go ahead with the upload or if you
need a sponsor feel free to let me know.
Regards,
Laszlo/GCS
package in salsa.debian.org/debian namespace?
> Or at least mirror it there?
As noted, GitHub is only for the easier review for you instead of
downloading the whole repository on a slow internet connection.
Laszlo/GCS
d at:
https://github.com/gcsideal/rocksdb
Hope this helps,
Laszlo/GCS
pshot. [1]
Not yet imported the 9.2.1 changes that I have locally but for some
reason not on snapshot. Please do a quick check and let me know your
findings. Then I can correct what's needed and add the mentioned
latest changes.
Thanks,
Laszlo/GCS
[1] https://people.debian.org/~gcs/rocksdb_git.tar.gz
of it. Can you recommend to
me a way to (probably automatically) import package versions from the
snapshot archive?
Regards,
Laszlo/GCS
s are affected: balboa and sortmerna. Both build fine
with the new rocksdb release already in experimental.
Thanks for consideration,
Laszlo/GCS
(but AFAICT calling RawCompressFromIOVec() will fail on overload resolution
> like Compress()).
Nope, it was added in v1.1.10 and it was part of the archive [1],
version 1.2.0 changed the function signature as the others [2] added
CompressionOptions.
Regards,
Laszlo/GCS
[1]
https://tracker.
't
change the soname.
Then the public source tree is hosted on GitHub [1] without the issues
(report) area enabled. The AUTHORS file contains a general email
address (opensou...@google.com) [2] meaning I'm not sure if I get any
answer or I will get one soon. But I can try it if you insist.
Re
which was done due to the C++11 ABI
change with g++ 5.0 back in 2015.
Thanks for considering,
Laszlo/GCS
[1] https://bugs.debian.org/1070217
[2] https://bugs.debian.org/1070785
hon3_INCLUDE_DIR
Development) (found suitable exact version "3.10.13")
But indeed, I should do a transition of snappy with a prepared new library name.
Regards,
Laszlo/GCS
e know.
Do you have a list of people, projects that will be affected by this
change? Sure, it would help to reach them for comments.
Regards,
Laszlo/GCS
pied into the source tree missing something else. This
is what SIGABRT suggests, probably some binary or library can't be
dlopen-ed as it's (those are) not copied over.
It's the tracker autopkg testing that needs fixing.
Regards,
Laszlo/GCS
[1] https://sqlite.org/releaselog/3_45_3.h
without problems. Do you experience any odd behaviour?
Regards,
Laszlo/GCS
On Mon, Apr 22, 2024 at 8:33 PM Daniel Kahn Gillmor wrote:
> On Sun 2024-04-21 15:44:12 +0200, László Böszörményi (GCS) wrote:
> > I prefer communication first. :) Currently I'm travelling so I can
> > only check it on Tuesday.
>
> That's why i uploaded to DELAYE
logy described below).
There were some license problems in the past at least, which
prevented packaging. I will check the current situation.
Regards,
Laszlo/GCS
point of this. Can you please recheck the current
graphviz package state and report back to me?
Regards,
Laszlo/GCS
1 - 100 of 1075 matches
Mail list logo
