/srv/irclogs.ubuntu.com/2022/03/18/#ubuntu-devel.txt

dbungertJackFrost: by chance are you running autopkgtest in schroot?  Sometimes tests try to write to home and those tests might work in another backend.00:30
sarnolddoes HOME get set somewhere meaningful on the others?00:32
JackFrostHah, build time tests in pbuilder.00:32
sarnold(that's my first guess, anyway)00:32
=== JackFrost is now known as Unit193
LocutusOfBorgenr0n, armhf is fixed with a little more patch06:38
LocutusOfBorgppc64el when built with -O3 is something I leave to you06:38
LocutusOfBorgas well as updating the upstream pull request once you have the full fix06:38
LocutusOfBorghttps://launchpad.net/~costamagnagianfranco/+archive/ubuntu/locutusofborg-ppa/+sourcefiles/nfstrace/0.4.3.2+git20200805+b220d04-2.1ubuntu1/nfstrace_0.4.3.2+git20200805+b220d04-2.1ubuntu1.dsc06:39
alexghitiHi guys, could someone trigger this job for me please? https://autopkgtest.ubuntu.com/request.cgi?release=jammy&arch=amd64&package=python-readme-renderer&trigger=pygments/2.11.2%2Bdfsg-210:00
alexghitiThanks!10:00
seb128alexghiti, k10:01
alexghitiseb128 thanks10:02
mirespaceHello... I'm packaging a huge project, and i discovered that the compression being used for control et al is zstd and lintian is flagging that as an error ... any way to overcome this? force another compression?10:39
mirespaceI suppose I'll try with -Z option  for dpkg-source10:47
ahasenackmorning12:18
ahasenackHappy Friday12:21
mirespacehappy Friday ahasenack :)12:23
ahasenackhi mirespace12:23
seb128doko, ginggs, is that expected?16:01
seb128# python3.10 -c 'import sysconfig; print(sysconfig.get_path("include"))'16:01
seb128/usr/local/include/python3.1016:01
seb128# python3.9 -c 'import sysconfig; print(sysconfig.get_path("include"))'16:01
seb128/usr/include/python3.916:01
seb128 16:01
seb128that's the reason why brltty rebuild is failing atm16:01
tumbleweedseb128: yes, this changed16:02
* tumbleweed looks at brltty16:02
tumbleweedalso relevant: https://bugs.python.org/issue4444516:03
seb128tumbleweed, https://github.com/brltty/brltty/blob/master/Tools/pythoncmd#L5016:03
seb128tumbleweed, and https://github.com/brltty/brltty/blob/master/Bindings/Python/bindings.m4#L12416:04
seb128tumbleweed, it feels weird that sysconfig doesn't include the distro paths16:12
tumbleweedseb128: it does16:12
tumbleweedthe default scheme changed from unix_prefix to unix_local16:12
tumbleweed(or posix_ as appropriate)16:12
tumbleweedwhich it used to be back in the 2.7 days, FWIW16:12
tumbleweedthe easy patch is to force the use of the unix_prefix scheme in this pythoncmd script16:13
tumbleweedjust seeing what the effects are16:13
seb128tumbleweed, thanks, I'm not going to have time to poke more today but I might pick it up on monday if noone beats me to it16:18
tumbleweedseb128: I'll file a PR upstream now, and a bug in Debian16:21
seb128tumbleweed, thanks16:22
seb128tumbleweed, I confirmed that adding scheme="posix_prefix" to the get_path calls fixes the build16:26
seb128tumbleweed, did you want to upload the fix to Ubuntu or should I do it?16:30
tumbleweedgo for it, if you've got it ready16:31
seb128tumbleweed, I haven't yet but my local try worked so I will clean that up and upload on monday if no one has done it before, thanks again for the help!16:33
tumbleweednp16:33
enr0nLocutusOfBorg: I added a patch to fix the nfstrace build errors on armhf and ppc64el: https://bugs.launchpad.net/ubuntu/+source/nfstrace/+bug/1965155/comments/316:37
ubottuLaunchpad bug 1965155 in nfstrace (Ubuntu) "nfstrace: FTBFS: header_window.cpp:77:82: error: format ‘%d’ expects argument of type ‘int’, but argument 5 has type ‘time_t’ {aka ‘long int’} [-Werror=format=]" [Undecided, New]16:37
LocutusOfBorgenr0n, https://github.com/epam/nfstrace/pull/50/files17:49
ubottuPull 50 in epam/nfstrace "nfstrace: fix format string specifiers detected by ncurses-6.3" [Open]17:49
LocutusOfBorgthere was already a fix...17:49
LocutusOfBorgmeh, uploading your version17:49
LocutusOfBorgI really would like if you could update the pull request you opened17:50
LocutusOfBorgand also please if possible open a PR with the TIRPC debian patch17:50
LocutusOfBorgthey are doing TIRPC search for fedora only just removing that check might work17:50
enr0nLocutusOfBorg: yeah I didn't notice that until I opened my PR. It was a quick enough fix, and I have already updated my PR.17:59
enr0nThere was another armhf failure anyways that wouldn't be fixed by the other PR.'17:59
LocutusOfBorgyes indeed18:01
LocutusOfBorgand about dropping the 4 lines in cmakelists?18:01
LocutusOfBorghttps://github.com/epam/nfstrace/blob/master/CMakeLists.txt18:01
LocutusOfBorgI didn't test but removing: 36 37 38 46 47 should do the trick18:01
LocutusOfBorgalso maybe changing the message at line 4518:02
enr0nI don't have much previous context for the TIRPC patch, but sure if I have time I will look into upstreaming.18:09
Eickmeyer[m]mapreri: Looks like inkscape is depending on libpoppler115, but only 118 is available. Should we do a NCR?19:31
jbichaEickmeyer[m]: check your inkscape version; it doesn't sounds like you have the current jammy release19:39
Eickmeyer[m]jbicha: yeah, something's fishy here. Good point.19:41
Eickmeyer[m]jbicha: Yep, I had an inkscape PPA enabled. Disabling it fixed the problem. #RookieMistake19:47
Eickmeyer[m]mapreri: Disregard.19:47
=== chrisccoulson_ is now known as chrisccoulson
LocutusOfBorgenr0n, I did it21:56
LocutusOfBorghttps://github.com/epam/nfstrace/pull/5221:56
ubottuPull 52 in epam/nfstrace "TIRPC: Make sure its search and found." [Open]21:56
LocutusOfBorgso now we can go hopefully without patches on next upstream release21:56

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!