Bug#1061868: kodi: FTBFS: Settings.h:60:3: note: candidate expects 1 argument, 0 provided
Control: fixed -1 2:20.3+dfsg-1 Control: close -1 Dear Aurelien, I ported the patch pointed by Rechi to 20.3 and uploaded whole thing to unstable. I added that patch after I generated d/changelog so cloding it manually now. Cheers, Vasyl
Bug#1000113: marked as pending in kodi
Control: tag -1 pending Hello, Bug #1000113 in kodi reported by you has been fixed in the Git repository and is awaiting an upload. You can see the commit message below and you can check the diff of the fix at: https://salsa.debian.org/multimedia-team/kodi-media-center/kodi/-/commit/45961a01b7ad18dcd6ca76e92fe9e1f74d0f0581 Add pcre2 patch (Closes: #1000113) Thanks to Yavor Doganov! Signed-off-by: Vasyl Gello (this message was generated automatically) -- Greetings https://bugs.debian.org/1000113
Bug#1000113: kodi: depends on obsolete pcre3 library
Hi Yavor, Thanks for the patch! Greatly appreciated!!! Upstream we discussed the pcre PR and there is an old branch porting some stuff to std::regex. I have checked that branch but some problems remained. I can test your patch because I am both the user and the maintainer as you requested. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1043128: kodi: FTBFS on s390x
Hi Sebastian, Thanks for the report! I did a preliminary investigation yesterday and filed the bug against gcc-13 [1]. In the meantime I will include the workaround for it and upload the -4 (as there is another ffmpeg6-related crash fix to upload). [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1043076 -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1000113: kodi: depends on obsolete pcre3 library
Dear Nicholas, Yes I definitely see the bug. However, Kodi extensively uses pcrecpp and the only replacement I see for pcre2 is jpcre2 [1] There is an ITP bug about it since 2017 but no package. Matthew, from your experience, is jpcre2 the only C++ wrapper for pcre2 or there is something more recommended / maintainable? I did the search but found only jpcre2. [1] https://github.com/jpcre2/jpcre2 -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1028434: kodi: FTBFS in bullseye (fatal error: date/tz.h: No such file or directory)
Hi Alban, bullseye-proposed-updates needs 19.5 while bullseye-backports needs 20.0 (and corresponding set of addons!) 19.5 will be different from 19.4 currently in bullseye-backports because we can not introduce new binary packages into anything that is frozen. Doing 19.5 is on my plans for a month but I had to package 20.0 first. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1027913: kodi: FTBFS in bookworm (undeclared build-dependency on tzdata)
Control: reassign -1 howardhinnant-date 3.0.1+ds-3 Control: retitle -1 howardhinnant-date must runtime-depend on tzdata Control: affects -1 = kodi Dear Santiago, Thanks for the bug report! The "discover_tz_dir" function is directly not used by Kodi and is part of howardhinnant-date. I looked through d/control of howardhinnant-date and it appears that libhowardhinnant-tz3 must explicitly depend on tzdata just like the source package build-depends on it. I can craft the Salsa MR if Andrea wants me to do that. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1023848: kodi-pvr-hts: Packet requires kodi-api-pvr but is not available in sid
Control: close -1 Hi Leela, The new versions of addons are uploaded to sid and will be built overnight. I am closing this bug in hope the api version mismatch is now resolved. Please re-open this bug if it is not the case! -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1023848: kodi-pvr-hts: Packet requires kodi-api-pvr but is not available in sid
Hi Leela, To resolve this I will need to upload src:kodi once more to fix an overlooked issue breaking builds on non-x86 platforms and then, after it gets built for all arches, upload the addons again. Mattia might help me early next week with it, because my status (Debian Contributor) does not allow me uploading packages directly. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1018203: kodi-inputstream-adaptive: FTBFS on ppc64el (regression in endian detection)
Hi Steve! Thanks for reminding me on this! I will upstream the patch and get a new tag along with security fixes for embedded Bento4 fork. After that, we can upload new version. Then the only thing it takes is to upload new release to unstable. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1016925: kodi.log attached
Control: severity -1 important Hi Mark, Sorry for the inconvenience! I had to upload v20 with experimental ffmpeg5 support because otherwise we had no Kodi at all. This case is the second video not playing that I am aware of. I am investigating the "unsupported stream" error message and will send you a test build this week. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1004612: marked as pending in kodi
Control: tag -1 pending Hello, Bug #1004612 in kodi reported by you has been fixed in the Git repository and is awaiting an upload. You can see the commit message below and you can check the diff of the fix at: https://salsa.debian.org/multimedia-team/kodi-media-center/kodi/-/commit/0f6c33665a82bc71f5094dbd006fb936082b69ef workarounds: Add some merged and unmerged stuff * Replace 0002-floats-threshold-compare-tests.patch (merged) * Replace 0004-Allow-overriding-the-hashsums-on-local-tarball.patch (merged) * Add patch for ffmpeg 5 support (not merged) (Closes: #1004612) * Add patch for proper fix of EDL rounding (merged) * Add patch for building against libfmt 9.0.0 (merged) (Closes: #1014542) Signed-off-by: Vasyl Gello (this message was generated automatically) -- Greetings https://bugs.debian.org/1004612
Bug#1004612: marked as pending in kodi
Control: tag -1 pending Hello, Bug #1004612 in kodi reported by you has been fixed in the Git repository and is awaiting an upload. You can see the commit message below and you can check the diff of the fix at: https://salsa.debian.org/multimedia-team/kodi-media-center/kodi/-/commit/0f6c33665a82bc71f5094dbd006fb936082b69ef workarounds: Add some merged and unmerged stuff * Replace 0002-floats-threshold-compare-tests.patch (merged) * Replace 0004-Allow-overriding-the-hashsums-on-local-tarball.patch (merged) * Add patch for ffmpeg 5 support (not merged) (Closes: #1004612) * Add patch for proper fix of EDL rounding (merged) * Add patch for building against libfmt 9.0.0 (merged) (Closes: #1014542) Signed-off-by: Vasyl Gello (this message was generated automatically) -- Greetings https://bugs.debian.org/1004612
Bug#1010956: Info received (Bug#1010956: kodi: Kodi crashes on startup with "error while loading shared libraries: libwayland-client++)
block -1 by 1010921 thanks
Bug#1010956: kodi: Kodi crashes on startup with "error while loading shared libraries: libwayland-client++
Control: reassign -1 src:waylandpp 1.0.0-1 Control: affects -1 kodi Control: blocked -1 by 1010921 Dear Mark, I have already filed the patch for waylandpp 1.0.0 that fixes the issue. Hope George uploads the fixed waylandpp package soon. If you need prebuilt waylandpp binary packages to test that I made, let me know so I will upload them to fex.net. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1006812: kodi: Kodi Fails to Start
Hi Stephen, 19.4 is out and I am preparing the release. This might take some time because I left all my infrastructure in Kharkiv, Ukraine and relevant Debian RT ticket is not yet processed. Once I get access to the beefy build machine I will upload everything. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#1004214: Acknowledgement (kodi: Segfaults on menu selection)
Control: severity -1 important Hi Stephen! Thanks for the bugreport! However, the "grave" severity is too high for it given I see the following issues: * Your logfile attached in the previous message pasted to Kodi hastebin ( https://paste.kodi.tv/ovifonuziy.kodi ) shows the presence of banned addons (marked in red). You should remove them and reproduce the issue without them. * The logfile is not a debug log at all. Also there is no stacktrace in it. You need to install gdb and kodi-bin-dbgsym from Debian debug repository according to https://wiki.debian.org/HowToGetABacktrace and run Kodi under gdb, reproduce the crash and run "bt full" to get a proper stacktrace. * The kernel you run is 5.16.2-xanmod1, which is not an official Debian kernel. Did you also update Mesa from external repos by chance? If you did, this might be an issue, too. Please try running Kodi with default Debian kernel and 3D stack and capture the proper backtrace so we can fix the segfault. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#997640: pybluez: FTBFS: error in PyBluez setup command: use_2to3 is invalid.
Control: tag -1 patch And again I missed the BTS cobtrol header :p -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#997640: pybluez: FTBFS: error in PyBluez setup command: use_2to3 is invalid.
Source: pybluez Version: 0.23-3 Followup-For: Bug #997640 X-Debbugs-Cc: iwama...@debian.org, mat...@debian.org tag -1 patch Dear colleagues, Since this issue puts the whole Kodi ecosystem I maintain at risk of autoremoval from testing, I decided to modernize the package and send the patch. TO solve the build failure, I had to re-import the same source tarball 0.23 with one upstream patch applied: https://github.com/pybluez/pybluez/pull/414 Everyone expects the upstream release fixing this bug but from reading upstream Github I have a feeling it will not happen anytime soon. Apart from re-packing the source tarball, I have bumped standards, moved to debhelper-compat 13 and brushed the file structure of a Debian package. I am attaching a debdiff with all the fixes. I can also push the fix to Salsa if I am granted developer rights at https://salsa.debian.org/bluetooth-team/pybluez or send three MRs for every affected branch (master, upstream and pristine-tar). Please let me know if you need more information. Cheers, Vasyl -- System Information: Debian Release: bookworm/sid APT prefers unstable-debug APT policy: (500, 'unstable-debug'), (500, 'buildd-unstable'), (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 5.10.0-8-amd64 (SMP w/6 CPU threads) Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=C, LC_CTYPE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8), LANGUAGE not set Shell: /bin/sh linked to /bin/dash Init: unable to detect diff -Nru pybluez-0.23/debian/changelog pybluez-0.23+ds1/debian/changelog --- pybluez-0.23/debian/changelog 2021-02-09 07:17:00.0 + +++ pybluez-0.23+ds1/debian/changelog 2021-11-01 08:56:51.0 + @@ -1,3 +1,15 @@ +pybluez (0.23+ds1-0.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * New upstream version 0.23+ds1 (Closes: #997640) ++ Add https://github.com/pybluez/pybluez/pull/414 patch + to existing pristine-tar of pybluez_0.23.orig.tar.gz + and re-import it because dh_clean step throws an error + before the d/pathes are applied. + * Modernize package + + -- Vasyl Gello Mon, 01 Nov 2021 08:56:51 + + pybluez (0.23-3) unstable; urgency=medium * Add python3-gattlib to Depends. (Closes: #930990) @@ -262,4 +274,3 @@ * My first debian package. -- Albert Huang Mon, 4 Apr 2005 01:22:15 -0400 - diff -Nru pybluez-0.23/debian/compat pybluez-0.23+ds1/debian/compat --- pybluez-0.23/debian/compat 2021-02-09 07:17:00.0 + +++ pybluez-0.23+ds1/debian/compat 1970-01-01 00:00:00.0 + @@ -1 +0,0 @@ -11 diff -Nru pybluez-0.23/debian/control pybluez-0.23+ds1/debian/control --- pybluez-0.23/debian/control 2021-02-09 07:17:00.0 + +++ pybluez-0.23+ds1/debian/control 2021-11-01 08:56:51.0 + @@ -1,18 +1,19 @@ Source: pybluez -Section: python Priority: optional +Section: python Maintainer: Debian Bluetooth Maintainers Uploaders: Albert Huang -Build-Depends: debhelper (>= 11), +Build-Depends: debhelper-compat (= 13), dpkg-dev (>= 1.16.1~), - libbluetooth-dev, dh-python, + libbluetooth-dev, python3-all-dev, python3-setuptools, +Standards-Version: 4.6.0 +Rules-Requires-Root: no Vcs-Git: https://salsa.debian.org/bluetooth-team/pybluez.git Vcs-Browser: https://salsa.debian.org/bluetooth-team/pybluez Homepage: https://github.com/pybluez/pybluez -Standards-Version: 4.2.1 Package: python3-bluez Architecture: any diff -Nru pybluez-0.23/debian/copyright pybluez-0.23+ds1/debian/copyright --- pybluez-0.23/debian/copyright 2021-02-09 07:17:00.0 + +++ pybluez-0.23+ds1/debian/copyright 2021-11-01 08:56:51.0 + @@ -6,10 +6,15 @@ Files: * Copyright: 2004-2005 Massachusetts Institute of Technology License: GPL-2+ +Comment: The source tarball has been re-packaged by adding the + upstream patch https://github.com/pybluez/pybluez/pull/414 + into existing pybluez_0.23.orig.tar.gz because the dh_clean + step throws an error before the debian/patches are processed. -Files: debian/ +Files: debian/* Copyright: 2004-2005 Massachusetts Institute of Technology 2018 Diane Trout + 2021 Vasyl Gello License: GPL-2+ License: GPL-2+ diff -Nru pybluez-0.23/debian/dirs pybluez-0.23+ds1/debian/dirs --- pybluez-0.23/debian/dirs2021-02-09 07:17:00.0 + +++ pybluez-0.23+ds1/debian/dirs1970-01-01 00:00:00.0 + @@ -1,2 +0,0 @@ -usr/bin -usr/sbin diff -Nru pybluez-0.23/debian/docs pybluez-0.23+ds1/debian/docs --- pybluez-0.23/debian/docs2021-02-09 07:17:00.0 + +++ pybluez-0.23+ds1/debian/docs1970-01-01 00:00:00.0 + @@ -1 +0,0 @@ -README.md diff -Nru pybluez-0.23/debian/examples pybluez-0.23+ds1/debian/examples --- pybluez-0.23/debian/examples2021-02-09 07:17:00.0 + +++ pybluez-0.23+ds1/debian/examples1970-01-01 00:00:00.0 + @@ -1,
Bug#980846: marked as pending in kodi
Control: tag -1 pending Hello, Bug #980846 in kodi reported by you has been fixed in the Git repository and is awaiting an upload. You can see the commit message below and you can check the diff of the fix at: https://salsa.debian.org/multimedia-team/kodi-media-center/kodi/-/commit/3983d600c9896852adec4aa23bbc5a126cf52798 Declare proper Debian release splitting kodi-addons-dev (Closes: #980846) Piuparts spotted the regression that emerged from the fact that Debian release '2:19.0~beta2+dfsg1-6~' has never been accepted into Debian archive. See bug report #980846 for more information. Signed-off-by: Vasyl Gello (this message was generated automatically) -- Greetings https://bugs.debian.org/980846
Bug#980846: marked as pending in kodi
Control: tag -1 pending Hello, Bug #980846 in kodi reported by you has been fixed in the Git repository and is awaiting an upload. You can see the commit message below and you can check the diff of the fix at: https://salsa.debian.org/multimedia-team/kodi-media-center/kodi/-/commit/80d363c229b0361e5778894ecbe88904818f1c34 Declare proper Debian release splitting kodi-addons-dev (Closes: #980846) Piuparts spotted the regression that emerged from the fact that Debian release '2:19.0~beta2+dfsg1-6~' has never been accepted into Debian archive. See bug report #980846 for more information. Signed-off-by: Vasyl Gello (this message was generated automatically) -- Greetings https://bugs.debian.org/980846
Bug#980846: kodi-addons-dev-common: missing Breaks+Replaces: kodi-addons-dev (<< 2:19.0~rc1+git20210119.8c761c4+dfsg1-2)
Hi Andreas! Thanks for bringing that to my attention! 23 січня 2021 р. 01:20:27 UTC, Andreas Beckmann написав(-ла): >See policy 7.6 at >https://www.debian.org/doc/debian-policy/ch-relationships.html#overwriting-files-and-replacing-packages-replaces I think the Policy should be clarified stating that B+R must point to a version that has been accepted into Debian archive. That was unclear for me to the dare because of >There are > B+R: kodi-addons-dev (<< 2:19.0~beta2+dfsg1-6~) >but that is the wrong version. kodi-addons-dev (<< 2:19.0~beta2+dfsg1-6~) was uploaded to NEW queue but rejected by FTP Master. Mattia suggests just changing that paragraph to: kodi-addons-dev (<< 2:19.0~rc1+git20210119.8c761c4+dfsg1-2~) which I am going to upload as: 2:19.0~rc1+git20210119.8c761c4+dfsg1-3 with s390x ftbfs fix. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#980172: kodi-pvr-hts: kodi crashes when switching to TV with pvr hts plugin activated and configured
Hi Stephan! Please install the test build available at https://fex.net/s/kkea8fe It is signed with my Debian key (changes file, buildinfo & dsc). If that works, the next upload wi.l fix the issue for general public. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 15 січня 2021 р. 16:19:36 UTC, Stephan Skrodzki написав(-ла): >Package: kodi-pvr-hts >Version: 8.1.2+ds1-1 >Severity: normal > > >thanks for updating the pvr-hts plugin. After configuration, everything seems >to work fine, but choosing the TV entry on the main menu let's Kodi crash... > >Kodi is 2:19.0~beta2+dfsg1-5 > >I deleted the whole .kodi dir, so there was no old configuration there. > >Crash log is attached. > >I have no clue what to do :-) > > > >-- System Information: >Debian Release: bullseye/sid > APT prefers unstable > APT policy: (500, 'unstable') >Architecture: amd64 (x86_64) >Foreign Architectures: i386 > >Kernel: Linux 5.10.0-1-amd64 (SMP w/8 CPU threads) >Kernel taint flags: TAINT_CPU_OUT_OF_SPEC, TAINT_OOT_MODULE, >TAINT_UNSIGNED_MODULE >Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8), LANGUAGE not >set >Shell: /bin/sh linked to /usr/bin/dash >Init: systemd (via /run/systemd/system) >LSM: AppArmor: enabled > >Versions of packages kodi-pvr-hts depends on: >ii kodi [kodi-api-pvr] 2:19.0~beta2+dfsg1-5 >ii libc62.31-9 >ii libgcc-s110.2.1-6 >ii libstdc++6 10.2.1-6 > >kodi-pvr-hts recommends no packages. > >kodi-pvr-hts suggests no packages. > >-- no debconf information >
Bug#980038: marked as pending in kodi
Control: tag -1 pending Hello, Bug #980038 in kodi reported by you has been fixed in the Git repository and is awaiting an upload. You can see the commit message below and you can check the diff of the fix at: https://salsa.debian.org/multimedia-team/kodi-media-center/kodi/-/commit/e15c8192e58fbaf94b1aabcf78864d90bf8678af Refresh cdatetime-std-chrono topic (Closes: #980038) Imported as: git format-patch --stdout master..HEAD -- . \ ':(exclude)tools/depends' \ ':(exclude)xbmc/platform/darwin' \ ':(exclude)xbmc/platform/win32' \ 1>../cdatetime-std-chrono.patch Signed-off-by: Vasyl Gello (this message was generated automatically) -- Greetings https://bugs.debian.org/980038
Bug#980038: kodi-pvr-hts causing Kodi to segfault
Control: reassign -1 kodi Control: retitle -1 Kodi crashes on listing of PVR recordings Hi Mark! Thanks to the coredump you provided, I spotted a regression in Kodi that I was previously not able to catch. I have updated the cdatetime Debian patch and building the fix. After it completes, I will send it to you privately and you verify whether PVR recordings work as expected (scheduling, viewing, listing, deleting etc). -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#976069: bugs
Dear colleagues, Thanks for reporting this! Unfortunately I got the reference to the bug only from kodi tracker.d.o page. I will investigate the issue and upload the fix to Git. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#976303: kodi: FTBFS during separate binary-indep build
Dear colleagues, On Thu, 03 Dec 2020 03:34:52 +0100 Andreas Beckmann wrote: > Probably some files are being accessed that are not created in this mode: I have already fixed it and a bunch of another issues but I have not uploaded the fix yet to Salsa. Today I am going to build the latest trunk for my staging repo and push beta1 to unstable after that closing this bug. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#936805: kodi: Python2 removal in sid/bullseye
Control: fixed -1 19.0~alpha3+dfsg1-1 Control: close -1 Dear colleagues, 19.0 Alpha3 is fully compatible with Python3 and I expect Beta1 to be uploaded to unstable as tgere will be no API changes by upsrream as promised. Closing this bug manually because I forgot mentionung it in a changelog at a time of upload. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#968339: Reopening for mips/s390x failures
Control: reopen ! I am reopening this bug as mipsel test failures are caused by ld.gold usage filed separately.
Bug#968339: Accept
Dear colleagues, I had to track this issue by building Kodi 18.8 with address sanitizer (asan) and it appears that the root cause is very likely a data race introduced by libsrt_gnutls.so.1, which in turn became an ffmpeg dependency since 4.3.0: (gdb) bt #0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50 #1 0x7f5392767537 in __GI_abort () at abort.c:79 #2 0x7f53927c0828 in __libc_message (action=action@entry=do_abort, fmt=fmt@entry=0x7f53928cee31 "%s\n") at ../sysdeps/posix/libc_fatal.c:155 #3 0x7f53927c7b1a in malloc_printerr (str=str@entry=0x7f53928cd041 "corrupted double-linked list") at malloc.c:5347 #4 0x7f53927c884c in unlink_chunk (p=p@entry=0x557cb9422af0, av=0x7f5392900b80 ) at malloc.c:1460 #5 0x7f53927c89b7 in malloc_consolidate (av=av@entry=0x7f5392900b80 ) at malloc.c:4502 #6 0x7f53927c9180 in _int_free (av=0x7f5392900b80 , p=0x557cb939ca90, have_lock=) at malloc.c:4400 #7 0x7f538bff364b in CUDTUnited::~CUDTUnited() () at /usr/lib/x86_64-linux-gnu/libsrt-gnutls.so.1 #8 0x7f5392780b96 in __cxa_finalize (d=0x7f538c069000) at cxa_finalize.c:83 #9 0x7f538bff09e3 in () at /usr/lib/x86_64-linux-gnu/libsrt-gnutls.so.1 #10 0x7ffc4c1a0980 in () #11 0x7f5396594343 in _dl_fini () at dl-fini.c:138 Disabling the parallelism during test runs works for me at least on amd64, so I will do it as a workaround to get 18.8 built properly for all architectures. Building 19.0 with current ffmpeg and libsrt-gnutls works fine for me at least on buster-bpo/amd64. I will test sid/amd64 before I upload 19.0 Alpha1 to experimental. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#968339: Accept
Control: owner -1 ! Hi Gianfranco! Thanks for reporting this! I will investigate the issue and file it upstream. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다
Bug#936805: kodi: Python2 removal in sid/bullseye
Dear colleagues, I made a mistake versioning Kodi packages in the unofficial repo. Packages were versioned 19.0+git instead of 19.0~git so if there are other people using my Github repo, I advise removing Kodi with "apt autoremove kodi*" and re-installing it with "apt update && apt install kodi" to get the future updates and later to be able to migrare to Kodi 19.0 from the main Debian repository. The fixed binary packages have been pushed to https://basilgello.github.io/kodi-nightly-debian-repo and the README has been updated with an appropriate warning. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#964591: kodi FTBFS with libmicrohttpd 0.9.71
Control: owner -1 = Hi Adrian! This bug was already reported upstream by Gentoo maintainer Chris Andrews: https://github.com/xbmc/xbmc/pull/18131 I will notify the upstream that Leia is also affected so once the fix is merged to master, there will be Leia backport closing this bug. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#962221: Fixes for CVE-2020-13696 (#962221)
Mattia, July 7, 2020 2:42:20 PM UTC, Vasyl Gello написав(-ла): >Got it! OK, let me do a quick fix for both issues and push additional commit. Commit is pushed, please try rebuilding the package! -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#962221: Fixes for CVE-2020-13696 (#962221)
Hi Mattia! July 7, 2020 2:25:37 PM UTC, Mattia Rizzolo написав(-ла): >..however it fails to build :) > > dh_auto_install > install -d /build/xawtv-3.107/debian/tmp > make -j4 install DESTDIR=/build/xawtv-3.107/debian/tmp > AM_UPDATE_INFO_DIR=no >make[1]: Entering directory '/build/xawtv-3.107' >/usr/bin/install -c -d -m 755 /build/xawtv-3.107/debian/tmp/usr/bin >/usr/bin/install -c console/dump-mixers console/record console/showriff >console/showqt console/streamer console/webcam console/scantv console/ttv >console/radio console/fbtv console/v4l-info >/build/xawtv-3.107/debian/tmp/usr/bin >/usr/bin/install -c -m4755 -o root console/v4l-conf >/build/xawtv-3.107/debian/tmp/usr/bin >/usr/bin/install: cannot change ownership of >'/build/xawtv-3.107/debian/tmp/usr/bin/v4l-conf': Operation not permitted >make[1]: *** [console/Subdir.mk:100: install] Error 1 >make[1]: Leaving directory '/build/xawtv-3.107' >dh_auto_install: error: make -j4 install DESTDIR=/build/xawtv-3.107/debian/tmp >AM_UPDATE_INFO_DIR=no returned exit code 2 >make: *** [debian/rules:6: binary] Error 25 >dpkg-buildpackage: error: debian/rules binary subprocess returned exit status 2 > > >this is related to the addition of Rules-Requires-Root. When run >without fakeroot it's not possible to run such `chmod` commands. In >fact, they are most likely always wrong to run them anyway… I of course built the package but my buildsetup always uses fakeroot exactly to get rid of chown() calls. I will update Rules-Requires-Root to yes then. >In d/copyright, that boilerplate-y thing you copied into the Comment >field, IMHO you should just get rid of it. Also, it's missing many of >the years in the copyright claims: a copyright claim without a year is >at most an legal headache and at worst invalid. Got it! OK, let me do a quick fix for both issues and push additional commit. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#962221: Fixes for CVE-2020-13696 (#962221)
Hi Mattia! July 6, 2020 6:58:05 PM UTC, Mattia Rizzolo написав(-ла): >but could either of you do a bunch of housekeeping work as well, like: > * bumping dh compat > * drop --dbgsym-migration > * drop the .menu files > * would be awesome to have the copyright file rewrote using dep-5 > * I pushed the modernized package however if you check resulting Lintian report, there are two errors claiming two libs are not compiled against libc and several others missing requured prerequisites. I have not figured yet how to fix these, maybe you know? -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#962221: Fixes for CVE-2020-13696 (#962221)
Hi Mattia! By partial I understood that upstream fixed the core part but the Debian patch sjould have been adapted to reflect new changes. Jeremy, can you please correct me if I am wrong? -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 July 6, 2020 6:58:05 PM UTC, Mattia Rizzolo написав(-ла): >On Mon, Jul 06, 2020 at 05:10:30AM +0000, Vasyl Gello wrote: >> Thanks for contributing the security release! I checked your changes and >> pushed them to the team repo. >> I do not have an upload rights, so CCing Sebastian and Mattia. > >Sure, > >but could either of you do a bunch of housekeeping work as well, like: > * bumping dh compat > * drop --dbgsym-migration > * drop the .menu files > * would be awesome to have the copyright file rewrote using dep-5 > * > >Also, the commit adding the CVE patch mentions "partial fix", as does >the sec-tracker page. Can anybody explain shortly what's with that, >where is the full fix (if there is), and how come the LTS upload claims >this to be fully fixed instead (CCing the LTS team and the uploader for >this). > >-- >regards, >Mattia Rizzolo > >GPG Key: 66AE 2B4A FCCF 3F52 DA18 4D18 4B04 3FCD B944 4540 .''`. >More about me: https://mapreri.org : :' : >Launchpad user: https://launchpad.net/~mapreri `. `'` >Debian QA page: https://qa.debian.org/developer.php?login=mattia `- signature.asc Description: PGP signature
Bug#962221: Fixes for CVE-2020-13696 (#962221)
Hi Jeremy! Thanks for contributing the security release! I checked your changes and pushed them to the team repo. I do not have an upload rights, so CCing Sebastian and Mattia. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#936805: kodi: Python2 removal in sid/bullseye
Hi Nicholas! I joined Debian to package Kodi 19.0 and full archive of binary addons and I already made a significant progress on the way: 1. The build dependencies for kodi not present in Debian at the moment of my join included: - dav1d (accepted to unstable 2 days ago, maintained by Dylan Aissi), - libudfread (awaiting approval from FTP team, maintained by me), - shairplay (awaiting approval from FTP team, maintained by me) So far, after libudfread & shairplay are accepted into unstable, we'll have all dependencies within Debian. 2. The build dependencies requiring additional work were fixed: - flatbuffers (1.11.0 fixed, waiting for upstream to tag 1.12.1 closing GCC-10 build failures) - libcdio (libcdio++ / libiso9660++ added, waiting for Gabriel T. Gomez to upload the package to unstable), - libsrt (patches proposed for review) 3. The unofficial binary repository targeting buster-backports/amd64 has been published on https://basilgello.github.io/kodi-nightly-debian-repo containing no-change rebuilds of kodi build dependencies, kodi itself and binary addons gradually added as prepared. The accompanying source code is hosted on Salsa: https://salsa.debian.org/basilgello-guest 4. After Kodi upstream declares the EOL of 18.x "Leia" branch, the repositories will be pushed to multimedia team's space and on release, the whole set will be uploaded to unstable. Uploading to experimental was considered excessive by Balint Reczey as every build occupies 4GB of Debian snapshot server space forever on. On Sun, 5 Jul 2020 11:48:39 -0400 Nicholas D Steeves wrote: > Hi, > > On Fri, Aug 30, 2019 at 07:22:18AM +, Matthias Klose wrote: > > Package: src:kodi > > Version: 2:17.6+dfsg1-4 > > Severity: normal > > Tags: sid bullseye > > User: debian-pyt...@lists.debian.org > > Usertags: py2removal > > > > Python2 becomes end-of-live upstream, and Debian aims to remove > > Python2 from the distribution, as discussed in > > https://lists.debian.org/debian-python/2019/07/msg00080.html > > > > Your package either build-depends, depends on Python2, or uses Python2 > > in the autopkg tests. Please stop using Python2, and fix this issue > > by one of the following actions. > > > > [snip] > > This bug will be solved when updating to Kodi ≥ 19, which prominently > declares it migrated to Python 3. Of course some plugins might not be > py3 ready, so it's probably time to stage the kodi-without-py2 > packages in experimental and report bugs upstream. > > The Python Team is moving ahead with making at py2 dep RC for low > popcon leaf packages this week, and while there isn't a roadmap > (afaik), I suspect this bug will become serious this fall (2020). > > Thanks, > Nicholas -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 signature.asc Description: PGP signature
Bug#963798: srt: Documentation is not installed into libsrt-doc and causes FTBFS if pbuilder is used
Hi Sebastian! I know that Build-Depends must be installed prior to clean target, however git-pbuilder introduced the way to skip installing build-dependencies if pbuilder chroot is used. I like this idea, because host system is not bloated by unaccountable dependencies. I had a recent discussion with Mattia on that matter, and he recommended me to use dh-sequence-XXX under Build-Depends-Arch (and/or Build-Depends-Indep if needed) to extend this behaviour to packages requiring 'special' dependencies. I think it won't hurt anyone to apply this technique to libsrt which I did in proposed patches. Anyway, more robust build and more proper usage of chroot (on my strong belief) is worth the additional work. -- Vasyl Gello == Certified SolidWorks Expert Mob.:+380 (98) 465 66 77 E-Mail: vasek.ge...@gmail.com Skype: vasek.gello == 호랑이는 죽어서 가죽을 남기고 사람은 죽어서 이름을 남긴다 July 3, 2020 7:07:36 AM UTC, Sebastian Ramacher написав(-ла): >Control: tags -1 - ftbfs >Control: retitle -1 srt: empty documentation package > >On 2020-06-27 13:04:06 +, Vasyl Gello wrote: >> Source: srt >> Version: 1.4.1-5 >> Severity: serious >> Tags: patch ftbfs >> Justification: fails to build from source (but built successfully in the >> past) >> >> Dear colleagues, >> >> Doing a no-change rebuild of srt for buster-backports/amd64, I noticed that >> 'gbp buildpackage' fails on 'dh clean --with sphinxdoc': >> >> dpkg-checkbuilddeps: error: Unmet build dependencies: cmake chrpath help2man >> libgnutls28-dev libssl-dev pkg-config tclsh >> .[1;33mW: Unmet build-dependency in source.[0m >> dpkg-source: info: using patch list from debian/patches/series >> dpkg-source: info: applying 001-multiarch-rpath.patch >> dh clean --with sphinxdoc >> dh: error: unable to load addon sphinxdoc: Can't locate >> Debian/Debhelper/Sequence/sphinxdoc.pm in @INC (you may need to install the >> Debian::Debhelper::Sequence >> BEGIN failed--compilation aborted at (eval 14) line 1. >> >> make: *** [debian/rules:21: clean] Error 25 >> >> The core issue is that sphinxdoc is installed in chroot while dh_clean is >> first >> called on host system. > >Build-Depends and Build-Conflicts must be satisfied for the clean >target. When you run that outside of the pbuilder chroot, you have to >install Build-Depends there as well. See >https://www.debian.org/doc/debian-policy/ch-relationships.html#relationships-between-source-and-binary-packages-build-depends-build-depends-indep-build-depends-arch-build-conflicts-build-conflicts-indep-build-conflicts-arch > >Cheers >-- >Sebastian Ramacher signature.asc Description: PGP signature
Bug#963821: dpkg-source error on repacked tarballs initially signed by upstream signing key
Package: src:dpkg Version: 1.20.1 Severity: serious Tags: ftbfs Justification: fails to build from source (but built successfully in the past) Dear colleagues, Today the dpkg was upgraded from 1.19.7 to 1.20.1 and I noticed that packages started failing to build. For example, the package with the following uscan configuration: opts="component=libdvdread-embedded, \ repack, \ compression=xz, \ pgpsigurlmangle=s/$/.asc/" \ https://download.videolan.org/pub/videolan/libdvdread/([\d][\d\.]+[a-z]?)/libdvdread-([\d][\d\.]+[a-z]?)\.tar\.(?:gz|bz2|xz) \ ignore throws the following error: dpkg-source: error: upstream signing key but no upstream tarball signature However, the use case is perfectly legitimate here: first, uscan checks the upstream GPG signature using debian/upstream/signing-key and then repacks the tarball to match the specified compression format (the compression format has to be one for all tarballs in the package because gbp does not recognize tarballs with different extension). I found out that the following upstream commit: From ca1cb131d8945d9d47871110f6a3010a501cd03a Mon Sep 17 00:00:00 2001 From: Guillem Jover Date: Sun, 22 Mar 2020 23:32:56 +0100 Subject: [PATCH] Dpkg::Source::Package: Check missing expected tarball signatures When the source package provides an upstream signing key, it is expected that the source package provides upstream tarball signatures. If not, then error out, to avoid building packages with the missing files, which tends to be very easy to get into. introduced the bug. What I'd expect the resolved bug is either: - soften the error to warning, or - introduce a local-option to suppress the new behavior. The local-option is better here because the error enforces the maintainer to think about the root cause and either fix the watchfile or override the local-options documenting the expected outcome. Downgrading the dpkg version to 1.19.7 is a temporary workaround as well. Vasyl -- Package-specific info: -- System Information: Debian Release: bullseye/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 4.15.0-108-generic (SMP w/6 CPU cores) Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=C, LC_CTYPE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8), LANGUAGE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8) Shell: /bin/sh linked to /bin/dash Init: unable to detect Versions of packages dpkg depends on: ii libbz2-1.0 1.0.8-3 ii libc62.30-8 ii liblzma5 5.2.4-1+b1 ii libselinux1 3.0-1+b3 ii tar 1.30+dfsg-7 ii zlib1g 1:1.2.11.dfsg-2 dpkg recommends no packages. Versions of packages dpkg suggests: ii apt2.1.6 ii debsig-verify 0.22 -- no debconf information
Bug#963798: srt: Documentation is not installed into libsrt-doc and causes FTBFS if pbuilder is used
Source: srt
Version: 1.4.1-5
Severity: serious
Tags: patch ftbfs
Justification: fails to build from source (but built successfully in the past)
Dear colleagues,
Doing a no-change rebuild of srt for buster-backports/amd64, I noticed that
'gbp buildpackage' fails on 'dh clean --with sphinxdoc':
dpkg-checkbuilddeps: error: Unmet build dependencies: cmake chrpath help2man
libgnutls28-dev libssl-dev pkg-config tclsh
.[1;33mW: Unmet build-dependency in source.[0m
dpkg-source: info: using patch list from debian/patches/series
dpkg-source: info: applying 001-multiarch-rpath.patch
dh clean --with sphinxdoc
dh: error: unable to load addon sphinxdoc: Can't locate
Debian/Debhelper/Sequence/sphinxdoc.pm in @INC (you may need to install the
Debian::Debhelper::Sequence
BEGIN failed--compilation aborted at (eval 14) line 1.
make: *** [debian/rules:21: clean] Error 25
The core issue is that sphinxdoc is installed in chroot while dh_clean is first
called on host system. Figuring this out, I noticed that libsrt-doc has no files
except of changelog.gz and copyright.
I fixed the issues and provided a doc-base registration of HTML manual.
Please review the attached patches and upload a backportvto buster once fixed.
Vasyl
-- System Information:
Debian Release: bullseye/sid
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Kernel: Linux 4.15.0-108-generic (SMP w/6 CPU cores)
Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=C, LC_CTYPE=C (charmap=UTF-8) (ignored: LC_ALL set to
en_US.UTF-8), LANGUAGE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: unable to detect
>From ec7f2e365f90ea8d2d9e7b7f730f902d9b59b6b8 Mon Sep 17 00:00:00 2001
From: Vasyl Gello
Date: Fri, 19 Jun 2020 08:09:12 +
Subject: [PATCH 1/4] debian: move python3-sphinx to Build-Depends-Indep
Fixes FTBFS edge case using pbuilder chroot, when dh_clean invoked
on host system to build source package fails to find debhelper sequence
provided by a package listed in Build-Depends and installed later inside
the chroot.
The solution is adapted from pynfft package:
https://salsa.debian.org/science-team/pynfft/-/commit/4c4c47acc0dba6ee0f3fa6cf4c3fc6ffbe41
The "override_dh_auto_build" conflicts with
"override_dh_auto_build-{arch,indep}", so
renaming "override_dh_auto_build" to "override_dh_auto_build-arch".
Thanks-To: to Mattia Rizzolo
Signed-off-by: Vasyl Gello
---
debian/control | 2 +-
debian/rules | 6 --
2 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/debian/control b/debian/control
index 063cae9..58c9ae8 100644
--- a/debian/control
+++ b/debian/control
@@ -8,8 +8,8 @@ Build-Depends: debhelper-compat (= 13),
libgnutls28-dev,
libssl-dev,
pkg-config,
- python3-sphinx,
tclsh
+Build-Depends-Indep: python3-sphinx
Standards-Version: 4.5.0
Section: libs
Homepage: https://github.com/Haivision/srt
diff --git a/debian/rules b/debian/rules
index 7a832ed..081541a 100755
--- a/debian/rules
+++ b/debian/rules
@@ -18,7 +18,7 @@ include /usr/share/dpkg/default.mk # provides DEB_VERSION
export USE_ENCLIB=gnutls
%:
- dh $@ --with sphinxdoc
+ dh $@
override_dh_auto_clean:
dh_clean
@@ -33,12 +33,14 @@ override_dh_auto_configure:
cd build-openssl && cmake .. $(CMAKE_OPTS) -DUSE_ENCLIB=openssl
cd build-gnutls && cmake .. $(CMAKE_OPTS) -DUSE_ENCLIB=gnutls
-DTARGET_srt=srt-gnutls
-override_dh_auto_build:
+override_dh_auto_build-arch:
#dh_auto_build --builddirectory=debian/build/openssl
#dh_auto_build --builddirectory=debian/build/gnutls
mkdir -p build-openssl build-gnutls
cd build-openssl && $(MAKE)
cd build-gnutls && $(MAKE)
+
+override_dh_auto_build-indep:
http_proxy='http://127.0.0.1:9/' python3 -m sphinx -N -c debian -bhtml
docs debian/html
override_dh_auto_install:
--
2.27.0
>From 6a0ca791c0c832dda75f6f00b0453b34933cf4ab Mon Sep 17 00:00:00 2001
From: Vasyl Gello
Date: Sat, 27 Jun 2020 10:49:24 +
Subject: [PATCH 2/4] debian: install HTML documents to libsrt-doc
Documents generated by sphinx were installed nowhere,
and libsrt-doc had no document files at all.
Fixing it by providing debian/libsrt-doc.docs
Signed-off-by: Vasyl Gello
---
debian/libsrt-doc.docs | 1 +
1 file changed, 1 insertion(+)
create mode 100644 debian/libsrt-doc.docs
diff --git a/debian/libsrt-doc.docs b/debian/libsrt-doc.docs
new file mode 100644
index 000..82dbdaf
--- /dev/null
+++ b/debian/libsrt-doc.docs
@@ -0,0 +1 @@
+debian/html
--
2.27.0
>From de168b1ed490295a7fde12fc5db79bee44992044 Mon Sep 17 00:00:00 2001
From: Vasyl Gello
Date: Sat, 27 Jun 2020 11:48:23 +
Subject: [PATCH 3/4] debian: spare Sphinx doctrees into debian/doctrees
* Fixes Lintisn warning 'package-contains-python-doctree-file'
Signed-off-by: Vasyl Gello
---
debian/rules | 4 +
Bug#963534: flatbuffers: FTBFS on alpha and hurd
Source: flatbuffers Version: debian/1.11.0+dfsg1-1.5 Severity: serious Tags: patch ftbfs Justification: fails to build from source (but built successfully in the past) Dear colleagues, My previous symbol adjustment for buster made flatbuffers to show in my personal package page. Exploring it, I reached buildd logs and found out that alpha, hurd and sh4 builds silently fail. Alpha build fails on symbols and hurd on missing PATH_MAX definition. I fixed both issues and filed another MR against flatbuffers on Salsa. The sh4 failure is trickier because it is basically a gtest failing with bus error, so I can not investigate it straight away because I do not posess any sh4 hardware. I also fixed the Lintian warnings and build-tested it against buster-bpo and sid on amd64. Please review and merge the patches if they are good to go. After that, we can bump the upstream to 1.12.0 released 3 months ago. In the meantime, I would like to have the fixed build uploaded to buster-backports because of kodi package depending in it. Vasyl -- System Information: Debian Release: bullseye/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 4.15.0-106-generic (SMP w/6 CPU cores) Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=C, LC_CTYPE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8), LANGUAGE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8) Shell: /bin/sh linked to /bin/dash Init: unable to detect
Bug#963481: FTBFS: Missing symbols on buster-backports/amd64
Source: spdlog Version: 1.5.0+ds-4 Severity: serious Tags: patch ftbfs Justification: fails to build from source (but built successfully in the past) Dear colleagues, Trying to perform a no-change rebuild of spdlog 1.5.0+ds-3 and -4 for buster-backports, I encountered the FTBFS on missing symbols. I opened a MR on Salsa against -3 fixing the build but it was somehow overlooked, maybe because I did not file the accompanying bug. Now I rebased the MR against med-team/spdlog master and I want to have it reviewed and merged. After the fixed build hits testing, I also would like to ask the maintainers to upload the backport for buster. This is required for Kodi package I am working on. -- System Information: Debian Release: bullseye/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 4.15.0-106-generic (SMP w/6 CPU cores) Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=C, LC_CTYPE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8), LANGUAGE=C (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8) Shell: /bin/sh linked to /bin/dash Init: unable to detect Regards, Vasyl
Bug#961550: flatbuffers FTBFS: MISSING symbols for libflatbuffers1
Source: flatbuffers Version: 1.11.0+dfsg1-1.3 Severity: serious Tags: patch ftbfs Dear colleagues, I tried to build flatbuffers for buster/amd64, and got the following symbols missing: * _ZN11flatbuffers14IntToStringHexB5cxx11Eii@Base 1.11.0+dfsg This is an inline symbol, g++ 8.3.0 inlined it properly instead of making it weak. * _ZN11flatbuffers9GetFieldSERKNS_5TableERKN10reflection5FieldE@Base 1.11.0+dfsg1 Also inline symbol * _ZN11flexbuffers7Builder5AlignENS_8BitWidthE@Base 1.11.0+dfsg1 * _ZN11flexbuffers7Builder6EndMapEm@Base 1.11.0+dfsg1 * _ZN11flexbuffers7Builder8WriteAnyERKNS0_5ValueEh@Base 1.11.0+dfsg1 * _ZN11flexbuffers7BuilderD1Ev@Base 1.11.0+dfsg1 * _ZN11flexbuffers7BuilderD2Ev@Base 1.11.0+dfsg1 * _ZNK10reflection5Field6VerifyERN11flatbuffers8VerifierE@Base 1.11.0+dfsg1 * _ZNK11flexbuffers9Reference8AsUInt64Ev@Base 1.11.0+dfsg1 * _ZNSt6vectorIhSaIhEE17_M_realloc_insertIJhEEEvN9__gnu_cxx17__normal_iteratorIPhS1_EEDpOT_@Base 1.11.0+dfsg1 NOT inlined symbols but missing nonetheless despute of "optional" tagging. I have created a PR on Salsa to fix the missing symbols and also the lintian errors on newly-added symbols: https://salsa.debian.org/debian/flatbuffers/-/merge_requests/4 Building the same fixed git snapshot for buster and sid on amd64 completed successfully, the build artifacts were used to rebuild Kodi whuch runs successfully on buster. Please review the PR and push the updated revision to unstable. Pushing the same release to stable would be ideal as well. -- Vasyl Gello signature.asc Description: PGP signature
