July 2022 - openSUSE Factory - openSUSE Mailing Lists

New Tumbleweed snapshot 20220713 released!
by Dominique Leuenberger 14 Jul '22

14 Jul '22

Please note that this mail was generated by a script. The described changes are computed based on the x86_64 DVD. The full online repo contains too many changes to be listed here. Please check the known defects of this snapshot before upgrading: https://openqa.opensuse.org/tests/overview?distri=opensuse&groupid=1&versio… Please do not reply to this email to report issues, rather file a bug on bugzilla.opensuse.org. For more information on filing bugs please see https://en.opensuse.org/openSUSE:Submitting_bug_reports Packages changed: abseil-cpp autoyast2 (4.5.1 -> 4.5.2) chromaprint (1.5.0 -> 1.5.1) dar (2.6.15 -> 2.7.6) ddclient gnome-desktop (42.2 -> 42.3) hwdata (0.360 -> 0.361) libadwaita (1.1.2 -> 1.1.3) libaio (0.3.112+29.696a5e6483ba -> 0.3.113) libhandy (1.6.2 -> 1.6.3) libnettle (3.7.3 -> 3.8) ncurses (6.3.20220618 -> 6.3.20220709) net-snmp nghttp2 (1.47.0 -> 1.48.0) patterns-fonts perl (5.34.1 -> 5.36.0) perl-MIME-tools (5.509 -> 5.510) plymouth python-argcomplete (1.12.3 -> 2.0.0) python-pycares (4.0.0 -> 4.2.1) redis (7.0.2 -> 7.0.3) selinux-policy speex (1.2 -> 1.2.1) suse-module-tools (16.0.20 -> 16.0.21) systemd (250.6 -> 251.2) texlive wavpack (5.4.0 -> 5.5.0) xorg-x11-server xwayland (22.1.2 -> 22.1.3) yast2-perl-bindings (4.5.0 -> 4.5.1) === Details === ==== abseil-cpp ==== - Add Fix-maes-msse41-leaking-into-pkgconfig.patch * Do not make programs compiled with abseil require new-ish CPUs. ==== autoyast2 ==== Version update (4.5.1 -> 4.5.2) Subpackages: autoyast2-installation - Run the registration step early only on the Online installation medium which does not provide any packages. On the other media run the registration step later. Fixes crash in the SLE Micro when the AutoYaST profile enables the registration step. (bsc#1200803) - 4.5.2 ==== chromaprint ==== Version update (1.5.0 -> 1.5.1) - update to 1.5.1: * No functional source code changes. * Fixed some compiler warnings. * Rebuilt the fpcalc binaries with FFmpeg 4.4.1. * Added support for fat binaries on macOS, with support for both Intel and Apple silicon. ==== dar ==== Version update (2.6.15 -> 2.7.6) Subpackages: dar-doc dar-lang libdar64-6000 - Update to version 2.7.6 * adding -f option to dar_cp * adding static version of dar_cp (dar_cp_static) as compilation outcome * added FAQ for tape usage with dar * fixing error in libdar header file installation * fixed bug met when interrupting the creation of a block compressed backup (always used by lzo compression and by other algorithm only when performing multi-threaded compression) * typo fixes in documentation * fixed message in lax mode used to obtain from the user the archive format when this information is corrupted in the archive. * fixing lax mode condition that popped up without being requested * fixing bug met when reading slice an special block device by mean of a symlink * adapting sanity checks to the case of a backup read from a special device in sequential-read mode. * fixed bug that lead dar to report CRC error while reading a backup from a pipe with the help of an isolated catalogue * adding -V option to dar_split (was using -v) for homogeneity with other commands ==== ddclient ==== - fix for boo#1191885 add SupplementaryGroups=maildrop to service file - rename ddclient-3.8.1-config.patch to ddclient-config.patch - rebase patch * ddclient-delay-main-process-for-systemd.patch (p0) - update ddclient-config.patch * fix PID file path - merge ddclient-replace-varrun-with-run.patch into ddclient-config.patc - merge changes file with SLES Maintained pkg - update Source to %{name}-%{version} ==== gnome-desktop ==== Version update (42.2 -> 42.3) Subpackages: gnome-desktop-lang gnome-version libgnome-desktop-3-19 libgnome-desktop-3_0-common libgnome-desktop-4-1 typelib-1_0-GnomeDesktop-3_0 typelib-1_0-GnomeDesktop-4_0 - Update to version 42.3: + No changes, version bump only. ==== hwdata ==== Version update (0.360 -> 0.361) - update to 0.361: + Updated pci, usb and vendor ids. ==== libadwaita ==== Version update (1.1.2 -> 1.1.3) Subpackages: libadwaita-1-0 libadwaita-lang typelib-1_0-Adw-1 - Update to version 1.1.3: + AdwLeaflet: Fix a broken link in docs. + AdwPreferencesGroup: Fix accessibility labels. + AdwToast: Fix the example in docs. + Stylesheet: Add missing borders in high contrast version. ==== libaio ==== Version update (0.3.112+29.696a5e6483ba -> 0.3.113) - add fix-splice-signature.patch to fix build on 32bit - update to 0.3.113: * cases/16.t: loongarch only supports eventfd2 * Add loongarch to supported architectures in libaio.spec * Add endian detection and bit width detection for loongarch * Use generic syscall number schema for loongarch * Fix struct io_iocb_vector padding for 32bit architectures * struct io_iocb_sockaddr padding for 32bit architectures * Verify structure padding is correct at build time * harness: add test for aio poll missed events ==== libhandy ==== Version update (1.6.2 -> 1.6.3) Subpackages: libhandy-1-0 libhandy-lang typelib-1_0-Handy-1_0 - Update to version 1.6.3: + Remove a non-existent property mention from HdyDeck docs. + Remove a leftover debug message from HdySettings. + Updated translations. ==== libnettle ==== Version update (3.7.3 -> 3.8) Subpackages: libhogweed6 libhogweed6-32bit libnettle8 libnettle8-32bit - update to 3.8: This release includes a couple of new features, and many performance improvements. It adds assembly code for two more architectures: ARM64 and S390x. The new version is intended to be fully source and binary compatible with Nettle-3.6. The shared library names are libnettle.so.8.5 and libhogweed.so.6.5, with sonames libnettle.so.8 and libhogweed.so.6. New features: * AES keywrap (RFC 3394), contributed by Nicolas Mora. * SM3 hash function, contributed by Tianjia Zhang. * New functions cbc_aes128_encrypt, cbc_aes192_encrypt, cbc_aes256_encrypt. On processors where AES is fast enough, e.g., x86_64 with aesni instructions, the overhead of using Nettle's general cbc_encrypt can be significant. The new functions can be implemented in assembly, to do multiple blocks with reduced per-block overhead. Note that there's no corresponding new decrypt functions, since the general cbc_decrypt doesn't suffer from the same performance problem. Bug fixes: * Fix fat builds for x86_64 windows, these appear to never have worked. Optimizations: * New ARM64 implementation of AES, GCM, Chacha, SHA1 and SHA256, for processors supporting crypto extensions. Great speedups, and fat builds are supported. Contributed by Mamone Tarsha. * New s390x implementation of AES, GCM, Chacha, memxor, SHA1, SHA256, SHA512 and SHA3. Great speedups, and fat builds are supported. Contributed by Mamone Tarsha. * New PPC64 assembly for ecc modulo/redc operations, contributed by Amitay Isaacs, Martin Schwenke and Alastair D´Silva. * The x86_64 AES implementation using aesni instructions has been reorganized with one separate function per key size, each interleaving the processing of two blocks at a time (when the caller processes multiple blocks with each call). This gives a modest performance improvement on some processors. * Rewritten and faster x86_64 poly1305 assembly. - drop libnettle-s390x-CPACF-SHA-AES-support.patch (included in 3.8) ==== ncurses ==== Version update (6.3.20220618 -> 6.3.20220709) Subpackages: libncurses6 ncurses-utils terminfo terminfo-base terminfo-iterm terminfo-screen - Add ncurses patch 20220709 + lock the prescreen data consistently in newterm, etc., for the pthreads configuration (report by Tom de Vries). - Add ncurses patch 20220703 + add consistency check in tic for u6/u7/u8/u9 and NQ capabilities. + use NQ to flag entries where the terminal does not support query and response -TD + use ansi+enq and decid+cpr in cases where the terminal probably supported the u6-u9 extension -TD + add/use apollo+vt132, xterm+alt47 -TD - Correct offsets of patches * ncurses-5.9-ibm327x.dif * ncurses-6.3.dif - Add ncurses patch 20220625 + improve man/curs_bkgd.3x, explaining that bkgdset can affect results for bkgd (report by Anton Vidovic). + correct dsl in dec+sl (report by Rajeev Pillai) -TD + add/use ansi+cpr, decid+cpr -TD - Correct offsets of patches * ncurses-5.9-ibm327x.dif * ncurses-6.3.dif ==== net-snmp ==== Subpackages: perl-SNMP snmp-mibs - Rename libsnmp40 subpackage to libsnmp39: the libraries are all having soversion 39. ==== nghttp2 ==== Version update (1.47.0 -> 1.48.0) - update to 1.48.0: * lib: Allow server to override RFC 9218 stream priority * lib: Add a server option to fallback to RFC 7540 priorities * lib: Add PRIORITY_UPDATE frame support * lib: Implement RFC 9218 extensible prioritization scheme * lib: Do not verify host field specific characters for response field * lib: No rfc7540 priorities * lib: Fix stream stall when initial window size is decreased * doc: Document how to change stream prioritization scheme * build: Compile with libressl 3.5 * build: EXTRA_DIST: List mruby files explicitly * build: Bump ngtcp2 and nghttp3 * build: Do not check application libraries if --enable-lib-only is given * src: Update default TLS cipher suites * nghttpx, h2load: Better pack UDP packets in one GSO write * nghttpx, h2load: Quic error handling * nghttpx, h2load: Fix QUIC performance regression * nghttp, nghttpd, nghttpx: Add ktls support * h2load: Send more packets without GSO per event loop * h2load: Add ktls support * nghttpd: Fix TLS read stall * nghttpx: Disable RFC 7540 priorities * nghttpx: Client always uses simpler TLS handshake * nghttpx: Add affinity-cookie-stickiness backend parameter * nghttpx: Fix broken session affinity * nghttpx: Limit CONNECTION_CLOSE and Retry under server amplification limit * integration: Go update * integration: Add go.mod * third-party: Bump llhttp to 75b45129db961e1fb3c56044e1b8f7721bfaee5d * third-party: Bump libbpf to v0.8.0 * third-party: Bump mruby to 3.1.0 * third-party: Bump neverbleed based on the latest head (GH-1708) ==== patterns-fonts ==== Subpackages: patterns-fonts-fonts patterns-fonts-fonts_opt - Revert to recommend noto-sans-fonts only, noto-fonts and noto-sans-cjk-fonts are too much. ==== perl ==== Version update (5.34.1 -> 5.36.0) Subpackages: perl-base perl-doc - Update to 5.36.0 * the signatures and isa features are no longer experimental and part of the v5.36 feature bundle * the v5.36 bundle also enables warnings * new '-g' command line flag (alias for -0777) * support for unicode 14.0 * regex sets are no longer considered experimental * experimental iterating over multiple values at a time * experimental new builtin module * experimental defer blocks * try/catch can now have a finally block * experimental non-ASCII delimiters for quote-like operators * a physically empty sort is now a compile-time error - Rebase perl-5.34.0.dif to perl-5.36.0.diff - Refresh perl-5.18.2-overflow.diff ==== perl-MIME-tools ==== Version update (5.509 -> 5.510) - updated to 5.510 see /usr/share/doc/packages/perl-MIME-tools/ChangeLog ==== plymouth ==== Subpackages: libply-splash-core5 libply-splash-graphics5 libply5 plymouth-dracut plymouth-lang plymouth-plugin-label plymouth-plugin-two-step plymouth-scripts plymouth-theme-bgrt plymouth-theme-spinner - Add 0004-label-ft-fix-alignment.patch: to fix alignment with label-ft in some cases (boo#959986). ==== python-argcomplete ==== Version update (1.12.3 -> 2.0.0) - update to 2.0.0: * Truncate input after cursor. * Support of path completion in fish * Drop support for Python 2.7 and 3.5 * Add support for Python 3.10 * Test, documentation, and release infrastructure improvements ==== python-pycares ==== Version update (4.0.0 -> 4.2.1) - specfile: * update copyright year - update to version 4.2.1: * core: add flexible member to ares_addrinfo to fix c-ares 1.18 - changes from version 4.2.0: * tests: ignoring 2 tests due to dead servers * tests: updated google TXT records for an outdated test * deps: update bundled c-ares - changes from version 4.1.2: * build: add PYPIREADME.rst to manifest - changes from version 4.1.1: * misc: add PyPI README without images * misc: update c-ares URL - changes from version 4.1.0: * ci: use stable 3.10, rather than dev * core: add support for Python 3.10 * doc: fix typo in README * doc: updatee changelog * doc: uppdate README * core: add support for CAA queries * core: add support for getaddrinfo() * doc: update README * core: add ability to use the system installed c-ares ==== redis ==== Version update (7.0.2 -> 7.0.3) - Update to version 7.0.3 * Performance and resource utilization improvements - Optimize zset conversion on large ZRANGESTORE (#10789) - Optimize the performance of sending PING on large clusters (#10624) - Allow for faster restart of Redis in cluster mode (#10912) * INFO fields and introspection changes - Add missing sharded pubsub keychannel count to CLIENT LIST (#10895) - Add missing pubsubshard_channels field in INFO STATS (#10929) * Module API changes - Add RM_StringToULongLong and RM_CreateStringFromULongLong (#10889) - Add RM_SetClientNameById and RM_GetClientNameById (#10839) * Changes in CLI tools - Add missing cluster-port support to redis-cli --cluster (#10344) * Other General Improvements - Account sharded pubsub channels memory consumption (#10925) - Allow ECHO in loading and stale modes (#10853) - Cluster: Throw -TRYAGAIN instead of -ASK on migrating nodes for multi-key - commands when the node only has some of the keys (#9526) * Bug Fixes - TLS: Notify clients on connection shutdown (#10931) - Fsync directory while persisting AOF manifest, RDB file, and config file (#10737) - Script that made modification will not break with unexpected NOREPLICAS error (#10855) - Cluster: Fix a bug where nodes may not acknowledge a CLUSTER FAILOVER TAKEOVER - after a replica reboots (#10798) - Cluster: Fix crash during handshake and cluster shards call (#10942) * Fixes for issues in previous releases of Redis 7.0 - TLS: Fix issues with large replies (#10909) - Correctly report the startup warning for vm.overcommit_memory (#10841) - redis-server command line allow passing config name and value in the same argument (#10866) - Support --save command line argument with no value for backwards compatibility (#10866) - Fix CLUSTER RESET command regression requiring an argument (#10898) ==== selinux-policy ==== Subpackages: selinux-policy-targeted - Update fix_systemd.patch to add cap sys_admin and kernel_dgram_send for systemd_gpt_generator_t (bsc#1200911) - postfix: Label PID files and some helpers correctly (bsc#1197242) ==== speex ==== Version update (1.2 -> 1.2.1) - update to 1.2.1: * Check for _WIN32 instead of WIN32 in preprocessor checks * wav_io: check for EOF when seeking in wav (fixes hang discovered by fuzzing) * CI: add gitlab CI integration * fixed-point: make left shift macros use unsigned to avoid undefined behaviour * math_approx: use unsigned int for LCG pseudorandom generator (avoids integer overflow) * oss-fuzz: add integration and fuzzing target * speexenc: guard against invalid channel numbers) * speexdec: make left shift macros use unsigned to avoid undefined behaviour * autotools: do not use deprecated macros - drop speex-CVE-2020-23903.patch (upstream) ==== suse-module-tools ==== Version update (16.0.20 -> 16.0.21) - Update to version 16.0.21: * kernel-scriptlets: don't pass flags to weak-modules2 (bsc#1195391) ==== systemd ==== Version update (250.6 -> 251.2) Subpackages: libsystemd0 libsystemd0-32bit libudev1 libudev1-32bit systemd-32bit systemd-container systemd-devel systemd-lang udev - pstore is no more considered as an experimental feature: move it to udev package (bsc#1197802) - Adjust rpmlintrc for shlib-policy-name-error/multibuild case so that it's not only for x86_64. - spec: %suse_version rpm macro is already reserved and has a special meaning in openSUSE distros so rename it to %archive_version instead. - Import commit e9fc337d97539fcab23078ab3e06f6b2ce3a3c8d ca0b29521f sha256: fix compilation on efi-ia32 1bbbac6a7e test: enable virtio-rng device for QEMU guests - Upgrade to v251.2 (commit 949d6bb7201dd48167ee9716ed6278764d1f4c0f) See https://github.com/openSUSE/systemd/blob/SUSE/v251/NEWS for details. This includes the following bug fixes: - upstream commit e6b169418369abbc88c8f622e02e1d704a23d4ef (bsc#1137373 bsc#1181658 bsc#1194708 bsc#1195157 bsc#1197570) * Rebased 0001-conf-parser-introduce-early-drop-ins.patch * systemd-testsuite now requires python3-pexpect due to TEST-69-SHUTDOWN relying on this module. * sysusers.d/systemd-network.conf has been moved to systemd-network sub-package since the tmpfiles configuration snippets for networkd has also been moved to this sub-package. ==== texlive ==== Subpackages: libkpathsea6 libsynctex2 - Require correct perl(Biber) version - add biber-missing-semicolon.patch * supports perl 5.36 - use biber version number - use https for urls - reference full download url for biber for source verification - use negative listing approach for luajit determination ==== wavpack ==== Version update (5.4.0 -> 5.5.0) - update to 5.5.0: * fixed: CVE-2021-44269 (encoding crafted DSD file triggers OOB read crash) * fixed: very long filenames cause stack-overflow crash in all CLI programs * fixed: the length stored in WAV headers not always corrected when using -i * fixed: attempting to encode raw DSD audio from stdin sometimes causes crash * fixed: DSD to PCM decimation: small clicks between tracks and tiny DC offset * fixed: length update in library-generated WAV headers on big-endian machines * fixed: sanitize custom extensions read from WavPack files to be alphanumeric * added: accepting brace-delimited options in the wavpack executable filename * added: "--drop" option to Windows executables for multi-file "drag-and-drop" * added" "--raw-pcm" option to wvunpack executable (does DSD --> 24-bit PCM) * added: "--no-overwrite" option to wavpack executable (to resume sessions) * improved: build system clean-up including switch to non-recursive "make" - drop wavpack-CVE-2021-44269.patch (upstream) - add doc subpackage ==== xorg-x11-server ==== Subpackages: xorg-x11-server-Xvfb xorg-x11-server-extra xorg-x11-server-sdk - U_boo1194181-001-xkb-swap-XkbSetDeviceInfo-and-XkbSetDeviceInfoCheck.patch * Out-Of-Bounds Access in CheckSetDeviceIndicators() (CVE-2022-2320, ZDI-CAN-16070, bsc#1194181) - U_boo1194179-001-xkb-rename-xkb_h-to-xkb-procs_h.patch, U_boo1194179-002-xkb-add-request-length-validation-for-XkbSetGeometry.patch * Out-Of-Bounds Access in _CheckSetSections() (CVE-2022-2319, ZDI-CAN-16062, bsc#1194179) ==== xwayland ==== Version update (22.1.2 -> 22.1.3) - Update to version 22.1.3 * os: print <signal handler called> if unw_is_signal_frame() * os: print registers in the libunwind version of xorg_backtrace() * xwayland/present: Do not send two idle notify events for flip pixmaps * xwayland: Fix check logic in sprite_check_lost_focus() * xwayland: Change randr_output status when call xwl_output_remove() * xkb: switch to array index loops to moving pointers * xkb: swap XkbSetDeviceInfo and XkbSetDeviceInfoCheck * xkb: add request length validation for XkbSetGeometry ==== yast2-perl-bindings ==== Version update (4.5.0 -> 4.5.1) - Adapted to new Perl-5.36.0 API (bsc#1200990) Details: https://github.com/yast/yast-perl-bindings/pull/30 - 4.5.1

1 0

Moving Some Noto Font Weights to an Extra Package
by Gordon Leung 14 Jul '22

14 Jul '22

Hi, With the renewed interest in reducing the file size of Noto Fonts packages, I have discovered that other distributions such as Arch and NixOS have an -extra package for extra Noto font weights. So I think Noto Fonts should move extra weights in Noto packages to noto-*-extra-fonts packages. So each script (that has extra fonts), will now have a new noto-example-extra-fonts package, while the original will no longer have the extra weights. The extra weights will be: Black*, Condensed*, Extra* ,*Medium, Semi*, and Thin* (the * is to indicate that each of these keywords can be chained together to make a new font file) as indicated here: https://github.com/NixOS/nixpkgs/pull/170355/files#diff-287418ecc15d2ca3598…. As the link says, Light*/LightItalic will be kept due to KDE requiring them. If there are no problems with this, or there are but they are quickly addressed, then I will start working on this change on July 16 over the weekend. Let me know what yall think, Gordon Leung

2 1

The Hinting Problem with Switching Noto Fonts to OTF
by Gordon Leung 14 Jul '22

14 Jul '22

I am back, On Thursday, July 14th, 2022 at 12:16 PM, Simon Lees <sflees(a)suse.de> wrote: > On 7/14/22 18:05, Wolfgang Rosenauer wrote: > > > Am 14.07.22 um 10:24 schrieb Mathias Homann: > > > > > Am Donnerstag, 14. Juli 2022, 09:56:50 CEST schrieb Wolfgang Rosenauer: > > > > > > > My monitor is not totally recent but certainly not old and quite decent. > > > > Still the font was so bad that it was already impacting my work. I don't > > > > have screenshots from that time but it was really really horrible. Most > > > > noticeable for me in Thunderbird when reading and writing mails which is > > > > what I do half of the day. > > > > > > I'm looking at FullHD on a 14" Laptop screen (it's actually 4k but who > > > would > > > WANT that resolution on a screen that small) - so I'm at 159DPI, and > > > fonts > > > look awesome. I have enabled full hinting, tho. > > > > > > I also looked over the noto fonts in kfontsel (I don't use them that > > > much), > > > and they look fine at 8x magnification. > > > > > > Oh, and I'm on TW20220712. > > > > Are you talking about the current state? > > Yes, everything is fine right now. I was talking about the few days in > > TW where the new fonts were installed. That has been rolled back a while > > ago. > > > I believe the proposal would be to fix that issue by modifying the Noto > fonts so they are smaller, instead I was going to look at fixing the > issue by creating a secondary font pattern that isn't installed by > default so that its easier to provide a curated set of fonts for users > who don't need anything special and a second set for users who have a > use case where something extra would benefit them, such as this case > where someone may have need or a desire to have a good set of fonts > covering all languages or maybe providing a larger selection that people > doing some form of design work can have out of the box. > > I also need to double check that on languages the Adobe fonts don't > cover the noto fonts get pulled in via supplements and used by default. > But I may not get to that for a couple of weeks. Oh, then I should hold off on doing the -extra package split (as detailed in another email I sent to this list which is not in this thread). I just wanted to make sure that there wasn't a problem with the Noto Fonts packages themselves. The only benefit is split would have had is that the individual packages would be smaller.(Because a bunch of them would have the "extra" weights moved into new packages) But the shell script used to generate the specfile would be further complicated. I however still think that Unhinted OTFs (and later Variable fonts once the upstream issues are fixed) would be beneficial as the formats encode the same amount of information, but in much less space. This especially goes for Variable fonts where not only are they even smaller than the Unhinted OTF files they replace, they also can support way more font weights and other neat features while being very compact.(for any font nerds like me, look up what variable fonts can do if you haven't already, its pretty amazing) If as Alois said: > On openSUSE, the autohinter is disabled for CFF fonts to prevent > interfering with native hinting. Maybe Fedora doesn't do this, so the > autohinter kicks in? It should be technically feasible to re-enable the> autohinter using fontconfig for Noto fonts if this is indeed the difference. Then I should start looking into enabling the autohinter as that probably explains why the Unhinted OTFs looked ugly to some people. I was not aware that it was disabled on openSUSE since I had read elsewhere that it is "enabled by default for Linux". Thanks, Gordon Leung

1 0

Update on python3 and ruby in Leap 15.5
by Lubos Kocman 14 Jul '22

14 Jul '22

Hello openSUSE! Sticking with old/unsupported python and ruby was a pain as it became virtually impossible to update some community packages. I have some good news! We can't change the default "system python" for code-stream 15, so we've agreed to a plan to offer a parallel python 3.10+ full-stack in 15 SP5. There is a request to add 3.11 interpret to SP5, so perhaps we should consider that version instead. Essentially we have the same scenario for ruby 2.7, where we plan to re-build all rubygems also for 2.7. Both requests are blocked on packaging resources. To solve this problem, we have requested an additional maintainer for ruby and python for FY23 (Which for SUSE starts on October 22). DimStar came up with excellent mitigation of the risk of being late. If we could not land features on time during the Beta phase of SLE 15 P5, we would make them available only in the buildroot and would not ship them to SLE customers due to quality concerns. openSUSE Leap 15.5 would still inherit these packages, and SLE customers would have them available via the Package HUB module as a tech preview. This would then change in 15 SP6, where packages would be released in SLES. [0] https://code.opensuse.org/leap/features/issue/74 (jsc#SLE-23096) Provide Ruby 2.7+ to support Rails 7 (preferably 3.1) [1] https://code.opensuse.org/leap/features/issue/17 (jsc#PED-68) [3] https://code.opensuse.org/leap/features/issue/85 (jsc#PED-189)

9 20

The Hinting Problem with Switching Noto Fonts to OTF
by Gordon Leung 14 Jul '22

14 Jul '22

On Thursday, July 14th, 2022 at 7:05 AM, Wolfgang Rosenauer <wolfgang(a)rosenauer.org> wrote: > Hi, > > Am 14.07.22 um 07:27 schrieb Gordon Leung: > > > Hi, > > > > Due to concern over the file sizes of the Noto packages, I am > > considering retrying a switch to Unhinted OTF as it takes up > > significantly less space when compared to Hinted TTFs.[1] I am not > > considering a switch to Variable fonts with Unhinted OTFs filling in the > > gaps (which would save even more space) as there are a number of > > upstream issues with the variable fonts as of now.[2] The only problem > > with this is that when I had previously attempted to switch to Unhinted > > OTFs, I had gotten bug reports from a number of (KDE Plasma) users that > > the hinting wasn't good on their system. (Note that Unhinted OTFs > > actually do have hinting applied to them on openSUSE DEs including KDE > > Plasma) So it would be great if someone could look more into this. I > > closed it for now as I have reverted the fonts to Hinted TTF, but it > > will probably reappear if the fonts switch to Unhinted OTFs again... > > https://bugzilla.opensuse.org/show_bug.cgi?id=1199938 > > https://bugzilla.opensuse.org/show_bug.cgi?id=1199938 > > > > I guess an alternative is to... ignore them. But that doesn't feel good > > to me. > > > I'm not at all a font expert but I was complaining about the heavily > ugly fonts back then when it was first introduced. > > But honestly I don't fully understand your request. > > > So it would be great if someone could look more into this. > > > You mean someone should check why the fonts are rendered badly on > openSUSE while they are supposed to work? > In that case I'm out technically, sorry. No worries, just checking to see if anyone here does. I think its a problem specific to openSUSE as Fedora had already switched their Noto package to variable (which would also mean using Unhinted OTFs to fill in fonts that do not have a variable version yet) but there doesn't seem to be a similar problem there. > > I guess an alternative is to... ignore them. > > > Who is "them"? The fonts? Or the bugs? Or anything else? The "them" refers to the many people who reported how the switch had made the hinting on these fonts quite bad on KDE Plasma. The whole point of a font is that its supposed to be able to display characters while also looking good, so it doesn't sit right with me to introduce a change that makes them look ugly to some people. (Personally I couldn't tell a difference, but apparently the difference is more noticeable on older monitors) Unfortunately this is the only message I can respond to, I gotta do something urgent. Will be back in a couple of hours. Thanks, Gordon Leung

5 5

Unannounced RUNPATH policy change in TW
by Stefan Brüns 14 Jul '22

14 Jul '22

Hi, apparently, the policy for RUNPATH in TW has been changed silently, causing lots of build failures. As it is holiday season currently and many package maintainers are on leave, the remaining time span for fixing any build failures is significantly reduced. I think both doing it silently, and doing it now, is quite bad. Can this please be reverted for now, and then repeated in a coordinated manner later? Regards, Stefan -- Stefan Brüns / Bergstraße 21 / 52062 Aachen home: +49 241 53809034 mobile: +49 151 50412019

1 0

The Hinting Problem with Switching Noto Fonts to OTF
by Gordon Leung 14 Jul '22

14 Jul '22

Hi, Due to concern over the file sizes of the Noto packages, I am considering retrying a switch to Unhinted OTF as it takes up significantly less space when compared to Hinted TTFs.[1] I am not considering a switch to Variable fonts with Unhinted OTFs filling in the gaps (which would save even more space) as there are a number of upstream issues with the variable fonts as of now.[2] The only problem with this is that when I had previously attempted to switch to Unhinted OTFs, I had gotten bug reports from a number of (KDE Plasma) users that the hinting wasn't good on their system. (Note that Unhinted OTFs actually do have hinting applied to them on openSUSE DEs including KDE Plasma) So it would be great if someone could look more into this. I closed it for now as I have reverted the fonts to Hinted TTF, but it will probably reappear if the fonts switch to Unhinted OTFs again... https://bugzilla.opensuse.org/show_bug.cgi?id=1199938 I guess an alternative is to... ignore them. But that doesn't feel good to me. Let me know what yall think, Gordon Leung [1] https://github.com/NixOS/nixpkgs/issues/166953#issuecomment-1100272951 [2] https://github.com/NixOS/nixpkgs/issues/166953#issuecomment-1100812824 (note that most of the linked issues have yet to be resolved)

2 1

ALP WG Meeting minutes - 12 July 2022
by Axel Braun 13 Jul '22

13 Jul '22

Hey Geekos, the last minutes of the ALP working group has caused a huge discussion on the list. I feel this is good, as it shows the interest, as well as concerns, of the community. Work goes on, and in yesterdays meeting we came up with the idea to use current MicroOS Desktop as proof of concept for the next-gen workstation. MicroOS Desktop current has a GNOME and KDE options. KDE is temporarily removed for now, this is expected to be changed in next two weeks. ALP POC doesn't have Desktop in requirements, but we can expect Desktop to land perhaps in a snapshot or two after the POC. Dan kindly volunteeed to create a Video + Text on how to install MicroOS Audience: powerusers, regular users, everyone Not many people might be willing to install this on a primary laptop or workstation. So the idea is to come up with a "a bribe" T-shirt for people who are willing to do that during a defined timeslot (e.g. spend a week with MicroOS desktop in next two months). And IMO, testing installation in a VM should count as well We would be giving users a set of tasks, and provide a support and feedback channel (that might require less than 2 months window, as that's quite a lot of effort). We could make something like QA Leap sheet https://docs.google.com/ spreadsheets/d/1AGKijKpKiJCB616-bHVoNQuhWHpQLHPWCb3m1p6gXPc/edit#gid=455031777 or a lucid spark workshop. Step 1: Prepare a space for FAQ for the effort with MicroOS Desktop (we'd polish it for ALP POC on Asia summit). Step 2: Probably a pre-recorded video of installation (so people know where to take it, how to install it etc). Step 3: Work on the list of actions/tasks that users should complete (e.g. a gdocs sheet, or a wiki page). Step 4: Prepare the feedback look (survey-o-o vs lucid spark). Step 5: Doug comes with cool t-shirts as a price for all participants Proposed timeline for the annoucement would be 2nd August, we'd collect feedback week after for first round, and two weeks after for the second round. Most likely Lubos will hold a rowkshop on openSUSE Asia summit on how to install MicroOS - to be confirmed. Beside this we takled about issues with hplip in flatpak world, printing subsystem cups vs driverless. To be raised with Network configuration and management Workgroup (lead Stefan Fent) https://jira.suse.com/browse/PED-270 / https://lwn.net/Articles/857502/ Seems like there is a common topic of accessing devices from flatpak app (usb redirection, boxes, 3d printer fw-u pdates, printing itself). Some stones on the road, feel free to join to get these removed! Cheers Axel

5 4

New Tumbleweed snapshot 20220712 released!
by Dominique Leuenberger 13 Jul '22

13 Jul '22

Please note that this mail was generated by a script. The described changes are computed based on the x86_64 DVD. The full online repo contains too many changes to be listed here. Please check the known defects of this snapshot before upgrading: https://openqa.opensuse.org/tests/overview?distri=opensuse&groupid=1&versio… Please do not reply to this email to report issues, rather file a bug on bugzilla.opensuse.org. For more information on filing bugs please see https://en.opensuse.org/openSUSE:Submitting_bug_reports Packages changed: ell (0.50 -> 0.51) git hidapi (0.11.0 -> 0.12.0) libnbd (1.12.2 -> 1.12.4) nbdkit (1.29.4 -> 1.30.7) nfs-utils python-SQLAlchemy (1.4.37 -> 1.4.39) python-chardet (4.0.0 -> 5.0.0) python-lxml (4.9.0 -> 4.9.1) python-numexpr (2.8.1 -> 2.8.3) python-pandas (1.4.2 -> 1.4.3) python-tabulate (0.8.9 -> 0.8.10) salt xconsole (1.0.7 -> 1.0.8) xmodmap (1.0.10 -> 1.0.11) yast2-installation (4.5.2 -> 4.5.3) yast2-trans (84.87.20220704.8f5a2d0352 -> 84.87.20220709.5ead98f887) === Details === ==== ell ==== Version update (0.50 -> 0.51) - update to 0.51: * Fix issue with handling bad prefixes per RFC4862 * Fix issue with handling DHCP retransmission timers * Add support for DHCP override and static IP API ==== git ==== Subpackages: git-core git-email git-gui git-svn git-web gitk perl-Git - Add /etc/bash_completion.d/git-prompt: checks for git-prompt.sh and source it if available. Some users rely on the __git_ps1 function becoming available when bash-completion is loaded. Continue to load this library at bash-completion startup for now, to ease the transition to a world order where the prompt function is requested separately. Inspired by Debian. ==== hidapi ==== Version update (0.11.0 -> 0.12.0) - update to 0.12.0: * libusb: improved CMake dependency on Iconv (#405) - as a result, better support for NetBSD; * general: documentation improvements; * general: small code cleanups/improvements; * many windows specific fixes - spec-cleaner cleanups ==== libnbd ==== Version update (1.12.2 -> 1.12.4) - Update to version 1.12.4: * python: Correctly use PyGILState * olang: aio_buffer.go: Explicit panic() on invalid usage * python: Enhance tests of nbd.Buffer * python: Plug uninit leak in nbd.Buffer.to_bytearray * python: Avoid memleak on (unlikely) module failure * python: Accept buffers in nbd.Buffer.from_bytearray() - Enable building python module and utilities ==== nbdkit ==== Version update (1.29.4 -> 1.30.7) Subpackages: nbdkit-basic-filters nbdkit-basic-plugins nbdkit-curl-plugin nbdkit-nbd-plugin nbdkit-python-plugin nbdkit-server nbdkit-ssh-plugin nbdkit-vddk-plugin - Update to version 1.30.7: * Version 1.30.7. * configure: Use printf to print $ISOPROG * configure: Print the version of curl, libnbd and libssh * configure: Make the 'feature' function print the dots * tests/test-protect.c: Fix typo in previous commit * tests/test-protect.c: Skip test if "disk" was not created * common/include: Rename BUILD_BUG_ON_ZERO to something more meaningful * common/include/test-array-size.c: Avoid Clang warning * vddk: Demote another "phone home" error message to debug * Use ARRAY_SIZE macro in various places in nbdkit * common/include: Add ARRAY_SIZE macro * tests/test-parallel-sh.sh: Skip test under valgrind + debuginfod * tests/test-parallel-sh.sh: Small cleanups * server: Work around incorrect include in gnutls/socket.h * server: Display kTLS setting in debug output * exitwhen: Check nbdkit doesn't exit before the pipe is closed * valgrind: Update comment about valgrind bug affecting OCaml * ocaml: Add further valgrind suppression * ocaml: Add valgrind suppression for OCaml 4.14 bug * perl: Move GCC diagnostic ignored earlier * docs/nbdkit-protocol.pod: block size support was added in 1.30 - Enable linuxdisk plugin - Update to version 1.30.6: * Version 1.30.6. * exitwhen: Link to example of how to use exit-when-pipe-closed * rate: Miscellaneous enhancements to comments and debug messages * ext2: Improve comments * eval: Provide default for .can_cache * eval: Always end scripts in newline * tests: Add regression test for NBDKIT_EMULATE_CACHE fix * server: Fix NBDKIT_CACHE_EMULATE * server: Fix NBDKIT_ZERO_EMULATE from filters * tests: Give sensible name to test sockets * linuxdisk: Add new subdir/.gitignore to EXTRA_DIST * tests: Fix test-disk2data.sh use of $PYTHON * blocksize: Test of aligned writes vs. RMW race * blocksize: Avoid losing aligned writes to RMW race * linuxdisk: Reduce size of test * multi-conn: Fix test with newer libnbd * data: Small clarification to the manual * vddk: advise user on obscure thumbprint mismatch error condition * configure: Fix incorrect comment about supported VDDK version * Version 1.30.5. * .gitignore: Ignore __pycache__ directory wherever it occurs * vddk: Demote another useless phone-home error message to debug * nbd: Hide some state machine debugging behind a debug flag * Remove pointless 'global' declarations. * Use RuntimeError instead of Exception * Drop pointless error check. * tests: Print a warning if nbdkit does not start after 30 seconds * file: Fix leaked file descriptor on error path * eval: Fix memory leak on error path * multi-conn: Fix memory leak on error path * Version 1.30.4. * pause: Delete control socket on server shutdown * rust: Use a local filename for the test socket * Update to nix 0.24.1 * rust: Use correct char type for characters * readahead: Simplify the test replacing use of sh plugin * readahead: Simplify the test by inlining some files * torrent: Don't attempt to rm -rf (null) on unload * readahead: Minor whitespace changes * ssh: Move and document thread model declaration * docs/nbdkit-filter.pod: Fix invalid use of E<> in verbatim section * Version 1.30.3. * tests: ssh: Don't run unused sshd * ocaml: Declare that we use caml_-namespaced symbols, for OCaml 4.14 * tests/test-cache-max-size.sh: Fix tests in locales with different date format * common: Add hexdigit.h containing common functions for handling hex * data: Fix typo in comment * tests: web-server: Avoid rare race in shutdown * server, tests: Prefix both uses of perror ("accept") * data: Check failure from some <(...) scripts * curl: scripts: Multiple fixes * python: Also plug leak on error case * tests/test-protect-ranges.sh: Don't try to use port 10809 * python: Avoid memory leak in .list_exports and .extents * fuzzing: Recommend combining fuzzing with ASAN * configure: Add --disable-linker-script flag * configure: Disable linker script for server when using libfuzzer * configure: Reorganise USE_LINKER_SCRIPT_FOR_SERVER conditional * configure: Report --enable-libfuzzer flag in summary output * Version 1.30.2. * server: Improve error message for --dump-plugin when plugin name is missing * docs: Use I<> to quote command line flag * tests: Skip vsock tests unless the vsock_loopback module is loaded * build: Fix GCC diagnostics for GCC 4.4 in RHEL 6 * podwrapper: Don't add the =encoding header for RHEL 6-era POD * podwrapper: Verify that all man pages are valid UTF-8 * podwrapper: Add suggestion for debugging Perl scripts * docs: Document how to write plugins/filters in C++ * vddk: Don't use uninitialized values when computing preferred block size * Update Nix to 0.23.0 * Update Mockall to 0.11.0 * fuzzing: Document alternate method for AFL++ using clang-LTO * server: Export __afl_* symbols for American Fuzzy Lop * tests: Fix some failing tests in RHEL / CentOS / AlmaLinux 8 * tests: Generally prefer nbdinfo over qemu-img info * tests: Remove unnecessary 'requires qemu-img --version' * tests: Add regression test that nbdkit isn't linked to OpenSSL * Version 1.30.1. * ci: Add perl-dev back to Alpine * ci: Regenerate CI files using latest libvirt-ci/lcitool * ci/manifest.yml: Add PERL skip variables * Use new string-vector and const-string-vector types * common/utils: Add new string vector types * common/utils/nbdkit-string.h: Rename header guards * cc: Rename misnamed string_vector to params_vector * ci: Rename centos-8 to almalinux-8 * ci: Remove fedora-34 tests * plugins/sh: Add ENOEXEC fallback for inline scripts * tests/functions.sh.in: Add comment about enabling vsock in tests * ci: Skip perl on clang targets * ci/build.sh: Combine mingw cases * ci: alpine: Include perl-dev package to build perl plugin * data: Remove bash-ism from <(...) script * docs, tests: Remove bash-ism from --run parameter * ci: Temporarily remove check-valgrind, dist tests * ci: Install 'dnf-command(config-manager)' first * ci: Use almalinux-8 instead of centos-8 * Add CI for GitLab * tests: Check that /dev/loop-control is accessible when needed * libssh: Check also for SSH_OPTIONS_DELAY * test-sh-errors: Allow for more variation of error messages * Fix libraries for mingw32 builds * test-delay-shutdown: Do not assume 124 as timeout's exit code and rather probe for it * test-exportname: Only check exportname.out contents if it exists * tests: Explicitly run $(plugin) under sh in expect scripts * test-swap: Only run with proper capability * plugins/rust: Run more tests * golang: Make sure used golang version supports modules * Only use vsock if VMADDR_CID_ANY is defined * Include sys/vsock.h only if it exists * plugins/file: Include linux/fs.h only if it exists * ocaml: Minor simplification of set_field function * todo: Suggest new thread model numbering in API V3 * Version 1.30.0. * docs: Final release notes for 1.30 * sh: Use new string type instead of reallocing return buffers * common/utils: Add extensible string, based on vector * todo: Drop item about minimum request size * todo: Document possibility of extending .block_size in API v3 * blocksize: Default internal block size based on plugin * blocksize: Defer per-handle initialization to .prepare * blocksize: Refactor to per-handle constraints * vddk: Use nested MAX macros * common/include: Fix MIN and MAX macros so they can be nested * common/include: Change unique-name macros to use __COUNTER__ * blocksize: Minor doc improvements * windows: tests/test-foreground.sh: Skip this test * windows: Skip some nbdkit-cow-filter tests * windows: tests: Add requires_run for some tests * windows: server/debug.c: Call close_memstream instead of fclose * windows: server/debug.c: Don't crash if open_memstream fails * windows: Add replacement posix_memalign function * windows: Add replacement for sysconf() POSIX function * swab: Implement .block_size callback * windows: blocksize-policy: Don't use reserved word ERROR * todo: Remove retry filter suggestion which was already implemented * blocksize: Don't open-code is_power_of_2 * log: Implement .block_size callback * docs: Fix typo & line too long in nbdkit-release-notes-1.30.pod * docs: Update release notes for nbdkit 1.30 * blocksize: Export block size constraints * cache, cow: Export block size constraints * cc: Implement .block_size callback * vddk: Export block size information for this plugin * Version 1.29.16. * blocksize-policy: Check the correct policy * server/plugins.c: Simplify error handling of .block_size callback * python: Implement .block_size callback * ocaml: Implement .block_size callback * nbd: In error messages, just display nbd_get_error() * nbd: Implement .block_size callback * New filter: nbdkit-blocksize-policy-filter * nbd: Use lowercase in error message * tests: Add a simple test for block size constraints * eval, sh: Implement block_size method * server: Add new plugin/filter .block_size callback * tls-fallback: Enhance comments about required callbacks * Version 1.29.15. * vddk: Document our current thinking on VMware out of memory errors * nbd: Opt in to libnbd pread_initialize speedup * error: Tweak filter name for easier debugging * zero: Fix doc typo * plugins: python: Make threading model explicit * plugins: python: Add error example * common/allocators: Always align mlock buffer * common/utils: Add vector_reserve_page_aligned * tests: Use a larger buffer for mlock test * common/allocators/malloc.c: Avoid calling munlock (NULL, ...) * Rename top level directory bash -> bash-completion * docs: Add preliminary outline release notes for 1.30 * docs: Add link to nbdkit-release-notes-1.28 to main manual page * Version 1.29.14. * server/sockets: get rid of AI_ADDRCONFIG * server/sockets: supply trailing newline for gai_strerror() format string * tests: Fix VDDK tests * vddk: Allow create-hwversion to be specified as a number * vddk: Fix documentation of new create flag * vddk: Implement VMDK creation * Version 1.29.13. * ocaml: Pass count parameter of pread as OCaml int * ocaml: Use int64 as the OCaml type for counts * ocaml: Convert offset values correctly in several callbacks * common/replacements/poll.h: Fix redefinition of poll * common/replacements: Use _get_osfhandle correctly * info: Add info_address=yes if mode=address is supported * valgrind: Fix suppressions for dlopen * info, ip: Make this plugin and filter work without inet_ntop/inet_pton * gzip: Don't try calling set_cloexec on Windows * exportname: Use replacement open_memstream * exitwhen: Remove some Windows warnings * cow: Include assert.h and call abort instead of assert (false) * split: Remove some unused variables when compiling on Windows * server/sockets.c: Remove variable-length array in Windows-only code * server, common, filters: Include "strndup.h" in places which use strndup * server: Remove some dead code when compiling on Windows * server: Use replacement open_memstream function * common/include/unique-name.h: Rename UNIQUE_NAME macro * server: Use ATTRIBUTE_FORMAT_PRINTF instead of __attribute__((printf)) * include: Disable printf format warnings on platforms that don't support %m * Version 1.29.12. * ocaml: NBDKit.set_field should not be @@noalloc * ocaml: Use magic_config_key in the example * python: More precise Python parameter passing * tests/test-python-plugin.py: Allow test to use large disks * tests: python: Use Python unittest function descriptions consistently * ocaml: Document the version string in the example * plugins/python: Fix extents() count format string * Version 1.29.11. * tests: Add configure --disable-libguestfs-tests flag * tests: Use mke2fs -d to create ext4 test image * Version 1.29.10. * windows: Link plugins/filters that use utils + pread/pwrite with libcompat * windows: Use replacement pread and pwrite functions * windows: Replace vector .size with .len in a few places * file: Fix implementation of cache=none for writes ==== nfs-utils ==== Subpackages: libnfsidmap1 nfs-client nfs-kernel-server - 0004-modprobe-protect-against-sysctl-errors.patch 0005-modprobe-avoid-error-messages-if-sbin-sysctl-fail.patch Suppress any errors from /sbin/sysctl, if for example, it isn't installed (bsc#1200710) ==== python-SQLAlchemy ==== Version update (1.4.37 -> 1.4.39) - update to version 1.4.39: * orm + [orm] [bug] [regression] Fixed regression caused by #8133 where the pickle format for mutable attributes was changed, without a fallback to recognize the old format, causing in-place upgrades of SQLAlchemy to no longer be able to read pickled data from previous versions. A check plus a fallback for the old format is now in place. References: #8133 - changes from version 1.4.38: * orm + [orm] [bug] [regression] Fixed regression caused by #8064 where a particular check for column correspondence was made too liberal, resulting in incorrect rendering for some ORM subqueries such as those using PropComparator.has() or PropComparator.any() in conjunction with joined-inheritance queries that also use legacy aliasing features. References: [#8162] + [orm] [bug] [sql] Fixed an issue where GenerativeSelect.fetch() would not be applied when executing a statement using the ORM. References: #8091 + [orm] [bug] Fixed issue where a with_loader_criteria() option could not be pickled, as is necessary when it is carried along for propagation to lazy loaders in conjunction with a caching scheme. Currently, the only form that is supported as picklable is to pass the ?where criteria? as a fixed module-level callable function that produces a SQL expression. An ad-hoc ?lambda? can?t be pickled, and a SQL expression object is usually not fully picklable directly. References: #8109 * engine + [engine] [bug] Repaired a deprecation warning class decorator that was preventing key objects such as Connection from having a proper __weakref__ attribute, causing operations like Python standard library inspect.getmembers() to fail. References: [#8115] * sql + [sql] [bug] Fixed multiple observed race conditions related to lambda_stmt(), including an initial ?dogpile? issue when a new Python code object is initially analyzed among multiple simultaneous threads which created both a performance issue as well as some internal corruption of state. Additionally repaired observed race condition which could occur when ?cloning? an expression construct that is also in the process of being compiled or otherwise accessed in a different thread due to memoized attributes altering the __dict__ while iterated, for Python versions prior to 3.10; in particular the lambda SQL construct is sensitive to this as it holds onto a single statement object persistently. The iteration has been refined to use dict.copy() with or without an additional iteration instead. References: #8098 + [sql] [bug] Enhanced the mechanism of Cast and other ?wrapping? column constructs to more fully preserve a wrapped Label construct, including that the label name will be preserved in the .c collection of a Subquery. The label was already able to render in the SQL correctly on the outside of the construct which it was wrapped inside. References: #8084 + [sql] [bug] Adjusted the fix made for #8056 which adjusted the escaping of bound parameter names with special characters such that the escaped names were translated after the SQL compilation step, which broke a published recipe on the FAQ illustrating how to merge parameter names into the string output of a compiled SQL string. The change restores the escaped names that come from compiled.params and adds a conditional parameter to SQLCompiler.construct_params() named escape_names that defaults to True, restoring the old behavior by default. References: [#8113] * schema + [schema] [bug] Fixed bugs involving the Table.include_columns and the Table.resolve_fks parameters on Table; these little-used parameters were apparently not working for columns that refer to foreign key constraints. In the first case, not-included columns that refer to foreign keys would still attempt to create a ForeignKey object, producing errors when attempting to resolve the columns for the foreign key constraint within reflection; foreign key constraints that refer to skipped columns are now omitted from the table reflection process in the same way as occurs for Index and UniqueConstraint objects with the same conditions. No warning is produced however, as we likely want to remove the include_columns warnings for all constraints in 2.0. In the latter case, the production of table aliases or subqueries would fail on an FK related table not found despite the presence of resolve_fks=False; the logic has been repaired so that if a related table is not found, the ForeignKey object is still proxied to the aliased table or subquery (these ForeignKey objects are normally used in the production of join conditions), but it is sent with a flag that it?s not resolvable. The aliased table / subquery will then work normally, with the exception that it cannot be used to generate a join condition automatically, as the foreign key information is missing. This was already the behavior for such foreign key constraints produced using non-reflection methods, such as joining Table objects from different MetaData collections. References: #8100, #8101 + [schema] [bug] [mssql] Fixed issue where Table objects that made use of IDENTITY columns with a Numeric datatype would produce errors when attempting to reconcile the ?autoincrement? column, preventing construction of the Column from using the Column.autoincrement parameter as well as emitting errors when attempting to invoke an Insert construct. References: #8111 * extensions + [extensions] [bug] Fixed bug in Mutable where pickling and unpickling of an ORM mapped instance would not correctly restore state for mappings that contained multiple Mutable-enabled attributes. References: #8133 ==== python-chardet ==== Version update (4.0.0 -> 5.0.0) - Update to 5.0.0 * This release is the first release of chardet that no longer supports Python < 3.6 * Added a prober for Johab Korean (#207, @grizlupo) * Added a prober for UTF-16/32 BE/LE (#109, #206, @jpz) * Added test data for Croatian, Czech, Hungarian, Polish, Slovak, Slovene, Greek, and Turkish, which should help prevent future errors with those languages * Improved XML tag filtering, which should improve accuracy for XML files (#208) * Tweaked SingleByteCharSetProber confidence to match latest uchardet (#209) * Made detect_all return child prober confidences (#210) * Updated examples in docs (#223, @domdfcoding) * Documentation fixes (#212, #224, #225, #226, #220, #221, #244 from too many to mention) * Minor performance improvements (#252, @deedy5) * Add support for Python 3.10 when testing (#232, @jdufresne) * Lots of little development cycle improvements, mostly thanks to @jdufresne - Canonicalize alternatives creation ==== python-lxml ==== Version update (4.9.0 -> 4.9.1) - update to version 4.9.1: * Bugs fixed + A crash was resolved when using iterwalk() (or canonicalize()) after parsing certain incorrect input. Note that iterwalk() can crash on valid input parsed with the same parser after failing to parse the incorrect input. ==== python-numexpr ==== Version update (2.8.1 -> 2.8.3) - specfile: * require python >= 3.7 - update to version 2.8.3: * Support for Python 3.6 has been dropped due to the need to substitute the flag NPY_ARRAY_WRITEBACKIFCOPY for NPY_ARRAY_UPDATEIFCOPY. This flag change was initiated in NumPy 1.14 and finalized in 1.23. The only changes were made to cases where an unaligned constant was passed in with a pre-allocated output variable: ``` x = np.empty(5, dtype=np.uint8)[1:].view(np.int32) ne.evaluate('3', out=x) ``` We think the risk of issues is very low, but if you are using NumExpr as a expression evaluation tool you may want to write a test for this edge case. * Thanks to Matt Einhorn (@matham) for improvements to the GitHub Actions build process to add support for Apple Silicon and aarch64. * Thanks to Biswapriyo Nath (@biswa96) for a fix to allow mingw builds on Windows. * There have been some changes made to not import platform.machine() on sparc but it is highly advised to upgrade to Python 3.9+ to avoid this issue with the Python core package platform. - changes from version 2.8.2: * skipped due to an error in uploading to PyPi. ==== python-pandas ==== Version update (1.4.2 -> 1.4.3) - update to version 1.4.3: * Behavior of concat with empty or all-NA DataFrame columns The behavior change in version 1.4.0 to stop ignoring the data type of empty or all-NA columns with float or object dtype in concat() (Ignoring dtypes in concat with empty or all-NA columns) has been reverted (GH45637). * Fixed regressions + Fixed regression in DataFrame.replace() when the replacement value was explicitly None when passed in a dictionary to to_replace also casting other columns to object dtype even when there were no values to replace (GH46634) + Fixed regression in DataFrame.to_csv() raising error when DataFrame contains extension dtype categorical column (GH46297, GH46812) + Fixed regression in representation of dtypes attribute of MultiIndex (GH46900) + Fixed regression when setting values with DataFrame.loc() updating RangeIndex when index was set as new column and column was updated afterwards (GH47128) + Fixed regression in DataFrame.fillna() and DataFrame.update() creating a copy when updating inplace (GH47188) + Fixed regression in DataFrame.nsmallest() led to wrong results when the sorting column has np.nan values (GH46589) + Fixed regression in read_fwf() raising ValueError when widths was specified with usecols (GH46580) + Fixed regression in concat() not sorting columns for mixed column names (GH47127) + Fixed regression in Groupby.transform() and Groupby.agg() failing with engine="numba" when the index was a MultiIndex (GH46867) + Fixed regression in NaN comparison for Index operations where the same object was compared (GH47105) + Fixed regression is Styler.to_latex() and Styler.to_html() where buf failed in combination with encoding (GH47053) + Fixed regression in read_csv() with index_col=False identifying first row as index names when header=None (GH46955) + Fixed regression in DataFrameGroupBy.agg() when used with list-likes or dict-likes and axis=1 that would give incorrect results; now raises NotImplementedError (GH46995) + Fixed regression in DataFrame.resample() and DataFrame.rolling() when used with list-likes or dict-likes and axis=1 that would raise an unintuitive error message; now raises NotImplementedError (GH46904) + Fixed regression in testing.assert_index_equal() when check_order=False and Index has extension or object dtype (GH47207) + Fixed regression in read_excel() returning ints as floats on certain input sheets (GH46988) + Fixed regression in DataFrame.shift() when axis is columns and fill_value is absent, freq is ignored (GH47039) + Fixed regression in DataFrame.to_json() causing a segmentation violation when DataFrame is created with an index parameter of the type PeriodIndex (GH46683) * Bug fixes + Bug in pandas.eval(), DataFrame.eval() and DataFrame.query() where passing empty local_dict or global_dict was treated as passing None (GH47084) + Most I/O methods no longer suppress OSError and ValueError when closing file handles (GH47136) * Other + The minimum version of Cython needed to compile pandas is now 0.29.30 (GH41935) ==== python-tabulate ==== Version update (0.8.9 -> 0.8.10) - specfile: * update copyright year - update to version 0.8.10: * Python 3.10 support. * Bug fixes. * Column width parameter. ==== salt ==== Subpackages: python3-salt salt-master salt-minion salt-transactional-update - Add support for gpgautoimport in zypperpkg module - Update Salt to work with Jinja >= and <= 3.1.0 (bsc#1198744) - Fix salt.states.file.managed() for follow_symlinks=True and test=True (bsc#1199372) - Make Salt 3004 compatible with pyzmq >= 23.0.0 (bsc#1201082) - Added: * fix-jinja2-contextfuntion-base-on-version-bsc-119874.patch * add-support-for-gpgautoimport-539.patch * fix-62092-catch-zmq.error.zmqerror-to-set-hwm-for-zm.patch * fix-salt.states.file.managed-for-follow_symlinks-tru.patch - Add support for name, pkgs and diff_attr parameters to upgrade function for zypper and yum (bsc#1198489) - Added: * add-support-for-name-pkgs-and-diff_attr-parameters-t.patch - Fix ownership of salt thin directory when using the Salt Bundle - Set default target for pip from VENV_PIP_TARGET environment variable - Normalize package names once with pkg.installed/removed using yum (bsc#1195895) - Save log to logfile with docker.build - Use Salt Bundle in dockermod - Ignore erros on reading license files with dpkg_lowpkg (bsc#1197288) - Added: * normalize-package-names-once-with-pkg.installed-remo.patch * use-salt-bundle-in-dockermod.patch * fix-ownership-of-salt-thin-directory-when-using-the-.patch * ignore-erros-on-reading-license-files-with-dpkg_lowp.patch * set-default-target-for-pip-from-venv_pip_target-envi.patch * save-log-to-logfile-with-docker.build.patch ==== xconsole ==== Version update (1.0.7 -> 1.0.8) - Update to version 1.0.8: * Update README for gitlab migration * Update configure.ac bug URL for gitlab migration * gitlab CI: add a basic build test * Build xz tarballs instead of bzip2 * Use _CONST_X_STRING to make libXt declare String as const char * * unifdef sgi * Reduce scope of ttydev & ptydev * gitlab CI: stop requiring Signed-off-by in commits ==== xmodmap ==== Version update (1.0.10 -> 1.0.11) - Update to version 1.0.11 * gitlab CI: add a basic build test * Build xz tarballs instead of bzip2 * handle.c: avoid leaks when realloc() fails * PrintModifierMapping: stop leaking the map returned by XGetKeyboardMapping Fix warning about number of mouse buttons ==== yast2-installation ==== Version update (4.5.2 -> 4.5.3) - Do not restart services when updating the package (bsc#1199480, bsc#1200274) -4.5.3 ==== yast2-trans ==== Version update (84.87.20220704.8f5a2d0352 -> 84.87.20220709.5ead98f887) Subpackages: yast2-trans-af yast2-trans-ar yast2-trans-bg yast2-trans-bn yast2-trans-bs yast2-trans-ca yast2-trans-cs yast2-trans-cy yast2-trans-da yast2-trans-de yast2-trans-el yast2-trans-en_GB yast2-trans-es yast2-trans-et yast2-trans-fa yast2-trans-fi yast2-trans-fr yast2-trans-gl yast2-trans-gu yast2-trans-hi yast2-trans-hr yast2-trans-hu yast2-trans-id yast2-trans-it yast2-trans-ja yast2-trans-jv yast2-trans-ka yast2-trans-km yast2-trans-ko yast2-trans-lo yast2-trans-lt yast2-trans-mk yast2-trans-mr yast2-trans-nb yast2-trans-nl yast2-trans-pa yast2-trans-pl yast2-trans-pt yast2-trans-pt_BR yast2-trans-ro yast2-trans-ru yast2-trans-si yast2-trans-sk yast2-trans-sl yast2-trans-sr yast2-trans-sv yast2-trans-ta yast2-trans-th yast2-trans-tr yast2-trans-uk yast2-trans-vi yast2-trans-wa yast2-trans-xh yast2-trans-zh_CN yast2-trans-zh_TW yast2-trans-zu - Update to version 84.87.20220709.5ead98f887: * Translated using Weblate (Macedonian) * Translated using Weblate (Russian)

1 0

Re: ALP WG Meeting minutes - July 5th 2022
by Dan Čermák 13 Jul '22

13 Jul '22

Hi Lars, Lars Vogdt <lars(a)linux-schulserver.de> writes: > Am 5. Juli 2022 15:46:29 UTC schrieb "Dan Čermák" <dcermak(a)suse.de>: >>tl;dr; Fear not, you will not be forced to download random flatpaks or >>containers from dockerhub/flathub. We (probably) just change the deliver >>method, but the packaging workflow will stay the same. > > 1) What are the advantages of flatpak/container vs RPM? Flatpaks support sandboxing when configured properly giving you greater security benefits in comparison to traditional rpms. The main benefit however is that we would be able to build certain desktop applications (e.g. Firefox, Thunderbird, LibreOffice, etc.) only on top of Tumbleweed, put them into a flatpak and distribute them to Tumbleweed, Leap and ALP users. This is something that is impossible to achieve with RPMs unless you bundle *everything* into a single RPM and even then it will probably not work. So this will save our packagers a tremendous amount of work. > 2) Will ALP still allow one 'zypper up' call to update all software (container, flatpak, rpm) on a system? > > 3) Will OBS auto-convert current rpms to flatpaks? Theoretically that could be done, but I think that's at the moment out of scope. > 4) Will there still be some kind of repo (incl. package informations like version-release, changelog, etc.) for flatpaks? > > 5) Will Yast-Software manage all these different kind of formats (container, flatpak, rpm) transparently for the enduser? I don't know what the plans of the Yast team wrt flatpaks are, but GNOME Software center supports flatpaks and rpms seamlessly. > 6) Can configuration still be expected at the usual places ($HOME/.config/ : /etc : /usr/etc/)? That depends on the flatpak, but generally flatpaks put their configuration into ~/.var/app/ > 7) Are current backups (rsync, Bareos) and configuration management (ansible, Salt) still be supported? > > 8) Will the whole delivery chain stay secure (signing of flatpaks, SSL, DNSSec, ...)? Yes. This is one of the strict requirements to ALP. > 9) Will autoyast still allow automated deployment? > > Regards, > Lars -- Dan Čermák <dcermak(a)suse.com> Software Engineer Development tools SUSE Software Solutions Germany GmbH Frankenstrasse 146 90461 Nürnberg Germany (HRB 36809, AG Nürnberg) Managing Director/Geschäftsführer: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman

16 44