2017-07-03 18:29:14 <ncopa> hum
2017-07-03 18:29:22 <ncopa> i wonder whats broke with cups-filters
2017-07-03 18:29:28 <ncopa> it built on my dev box earlier
2017-07-10 19:32:03 <ncopa> sorry about those test suites failing
2017-07-10 19:32:55 <ncopa> i looked at some of them today. cmocka is missing on aarch64, also due to test suite failing
2017-07-10 19:33:06 <ncopa> but i could not figure out why
2017-07-11 11:52:10 <fabled> algitbot: retry master
2017-07-11 12:40:33 <fabled> ncopa, why is perl-test-roo not built?
2017-07-11 12:41:53 <fabled> ncopa,  is it lua-aports skipping checkdepends?
2017-07-11 12:42:02 <ncopa> yes
2017-07-11 12:42:16 <fabled> ncopa, how to fix that then?
2017-07-11 12:42:24 <ncopa> that was a problem while building 3.6 release
2017-07-11 12:42:30 <ncopa> was never solved properly
2017-07-11 12:42:40 <fabled> algitbot: retry master
2017-07-11 12:42:57 <ncopa> we could make lua-aports pull it in explicitly
2017-07-11 12:43:03 <fabled> seems to be that if i'm lucky it picks up correct order
2017-07-11 12:43:22 <ncopa> but then we need to make sure that checkdepends does not create any circular deps
2017-07-11 12:43:59 <ncopa> i think the idea with checkdepends was that we could allow circular deps for tests
2017-07-11 12:45:30 <fabled> algitbot: retry master
2017-07-11 12:47:31 <fabled> algitbot: retry master
2017-07-11 12:49:37 <fabled> algitbot: retry master
2017-07-11 12:49:55 <fabled> algitbot: retry master
2017-07-11 12:50:04 <ncopa> oh
2017-07-11 12:50:15 <ncopa> checkdepends is added to lua-aports git
2017-07-11 12:50:32 <ncopa> i suppose we only need to tag new release
2017-07-11 12:54:56 <fabled> algitbot: retry master
2017-07-11 12:57:50 <fabled> algitbot: retry master
2017-07-13 12:29:24 <fabled> how come libreoffice on s390x fails with: unzip: unsupported method 2048
2017-07-13 12:29:35 <fabled> is unzip crippled or buggy on s390x?
2017-07-15 06:05:15 <fabled> jirutka, thanks for pulling it to stable it was in my todo. just got distracted, and i don't usually backport until edge has finished building
2017-07-15 06:05:30 <fabled> perhaps we should mark the commits in commit message that should be backported so we can have script to check which ones are not yet done so we don't accidentally forget them
2017-07-15 10:14:09 <jirutka> fabled: I tag it on GH with backport-it
2017-07-15 10:15:11 <jirutka> fabled: I’m not sure that it can be easily automated, it’s not always reapplying the same commit
2017-07-15 13:35:56 <jirutka> I don’t know how to upload tarball to dev.a.o from snapshot func :/
2017-07-15 23:36:01 <jirutka> **** please someone upload snapshot from testing/libcoro to dev.a.o ****
2017-07-17 10:27:37 <jirutka> algitbot: retry master
2017-07-17 10:41:17 <jirutka> algitbot: retry master
2017-07-17 14:45:51 <ncopa> something interesting is happening
2017-07-17 14:45:55 <ncopa> >>> ldb: Analyzing dependencies...
2017-07-17 14:45:55 <ncopa> Segmentation fault
2017-07-18 07:18:49 <ncopa> i'll fix rlwrap
2017-07-19 06:07:39 <fabled> something is wrong on s390x: 
2017-07-19 06:07:39 <fabled> >>> py-urllib3: Unpacking /var/cache/distfiles/py-urllib3-1.21.1.zip...
2017-07-19 06:07:40 <fabled> unzip: unsupported method 2048
2017-07-19 08:18:15 <ncopa> fabled: yes. somethine is wrong in busybox unzip apparently
2017-07-19 08:18:20 <ncopa> on s390x
2017-07-19 08:23:08 <fabled> hum, seems my bb add-shell patch went ignored
2017-07-19 08:23:43 <fabled> http://lists.busybox.net/pipermail/busybox/2017-July/085612.html
2017-07-19 08:31:33 <ncopa> fabled: maybe add a bug on their bugtracker
2017-07-21 23:14:57 <jirutka> algitbot: retry master
2017-07-22 21:23:10 <jirutka> algitbot: retry master
2017-07-24 19:43:21 <ncopa> jirutka^^^
2017-07-24 19:56:07 <ncopa> jirutka: we should fix lua-aports to handle "exit 0" at the top of apkbuild
2017-07-24 19:56:21 <ncopa> and we should redirect output
2017-07-24 19:56:52 <ncopa> so if someone does: echo "hello there" in global scope, things does not go kaboom
2017-07-24 19:57:26 <ncopa> . ./APKBUILD >/dev/null
2017-07-24 19:57:37 <ncopa> ideally, we should redirect to tmpfile
2017-07-24 19:57:40 <ncopa> and check size of tmpfile
2017-07-24 19:57:49 <ncopa> if its non-zero, abort with error
2017-07-24 19:59:38 <jirutka> i’m currently investigating lua posix module, to have popen that can handle stdout, stderr and status simultaneously
2017-07-24 20:00:16 <ncopa> its non trivial
2017-07-24 20:00:42 <ncopa> you can create pipes
2017-07-24 20:00:50 <ncopa> and redirect
2017-07-24 20:00:58 <ncopa> dup()
2017-07-24 20:01:34 <ncopa> but you need use poll() or select() on 3 different fds
2017-07-24 20:01:42 <ncopa> to read from all 3 at same time
2017-07-24 20:02:05 <ncopa> and popen() will use stdio
2017-07-24 20:02:14 <ncopa> which is buffered
2017-07-24 20:02:31 <ncopa> you'll want use write(2) and read(2)
2017-07-24 20:02:56 <ncopa> might be worth making a dedicated module for that
2017-07-25 13:40:24 <jirutka> rdutra: are you the guy from IBM? I don’t remember nicknames
2017-07-25 13:40:37 <rdutra> jirutka: yes, I am
2017-07-25 13:41:08 <jirutka> rdutra: great, could you please look at http://build.alpinelinux.org/buildlogs/build-edge-ppc64le/community/lua-copas/lua-copas-2.0.2-r1.log ? LuaJIT segfaults in one test
2017-07-25 13:41:25 <jirutka> rdutra: I remember that someone from IBM patched LuaJIT to work on ppc64le
2017-07-25 13:44:06 <rdutra> jirutka: ok I can take a look
2017-07-25 13:45:15 <jirutka> rdutra: thanks!
2017-07-25 17:42:15 <jirutka> algitbot: retry master
2017-07-25 21:19:44 <travis-ci> lua-aports:master |{author}| set up Travis CI to run luacheck >> passed | https://travis-ci.org/alpinelinux/lua-aports/builds/257485212
2017-07-25 21:25:26 <travis-ci> lua-aports:master [passed] |Jakub Jirutka| travis: fix IRC message template | https://travis-ci.org/alpinelinux/lua-aports/builds/257487041
2017-07-25 21:37:51 <travis-ci> mkinitfs:master [errored] |Jakub Jirutka| travis: notify on IRC | https://travis-ci.org/alpinelinux/mkinitfs/builds/257489656
2017-07-25 21:41:43 <travis-ci> apk-tools:master [passed] |Jakub Jirutka| travis: notify on IRC | https://travis-ci.org/alpinelinux/apk-tools/builds/257488946
2017-07-26 11:26:38 <jirutka> algitbot: retry master
2017-07-26 13:31:50 <jirutka> clandmeter: ^
2017-07-26 13:32:39 <rdutra> I think tarantool is not currently supported in ppc64le
2017-07-26 13:33:14 <jirutka> clandmeter: I started with tarantool pkg few weeks ago, there’s more work needed to be done…
2017-07-26 13:33:25 <clandmeter> ?
2017-07-26 13:34:05 <jirutka> clandmeter: you haven’t reviewed it much, did you? there’s a lot of bundled libs and some troubles with tests
2017-07-26 13:34:19 <jirutka> clandmeter: I’ll at least add what I have done to the current pkg
2017-07-26 13:35:00 <clandmeter> its just a start, and it runs.
2017-07-26 13:35:09 <clandmeter> its in testing for a good reason :)
2017-07-26 13:35:16 <jirutka> oh, it runs!
2017-07-26 13:35:18 <jirutka> omg
2017-07-26 13:36:12 <jirutka> also 1.7.4.99 is old…
2017-07-26 13:36:50 <jirutka> the latest version is .308
2017-07-26 13:40:53 <jirutka> omfg!
2017-07-26 13:41:00 <jirutka> grrr
2017-07-26 13:41:52 <clandmeter> seems latest is .335
2017-07-26 13:41:58 <clandmeter> i scrolled too much down
2017-07-26 13:42:10 <clandmeter> they dont officially state the current version?
2017-07-26 13:44:19 <jirutka> eh, pardon, yes, 335
2017-07-26 16:11:33 <jirutka> kaniini: do we _really_ want to start splitting openrc runscripts _now_?
2017-07-26 16:11:58 <kaniini> my paymasters say yes
2017-07-26 16:12:05 <jirutka> kaniini: your WHAT?!
2017-07-26 16:12:25 <jirutka> kaniini: have you discussed it with ncopa?
2017-07-26 16:13:25 <kaniini> yes, actually the concept of splitting was discussed and consensus was reached that it wasn't a terrible idea.
2017-07-26 16:30:53 <kaniini> (to clarify my paymasters say they want openrc split in Adelie, they have no opinion on what Alpine does)
2017-07-26 16:43:55 <jirutka> kaniini: who are your paymasters?
2017-07-26 16:44:33 <kaniini> well technically i am the one doing the paying
2017-07-26 16:44:40 <jirutka> ?
2017-07-26 16:44:53 <jirutka> like you pay yourself…? o.O
2017-07-26 16:44:55 <kaniini> but adelie is basically a distribution we developed for our own internal use
2017-07-26 16:45:04 <kaniini> to support our infrastructure
2017-07-26 16:45:10 <kaniini> because freebsd went completely bonkers
2017-07-26 16:45:12 <jirutka> i thought that it’s one-girl show
2017-07-26 16:45:19 <jirutka> according to commit history in Adélie repository
2017-07-26 16:45:20 <kaniini> that girl works for me :)
2017-07-26 16:45:32 <jirutka> wow
2017-07-26 16:45:41 <jirutka> interesting
2017-07-26 16:45:49 <jirutka> can I work for you too? ;)
2017-07-26 16:45:54 <Xe> so effectively using adelie as an experimental branch for larger, sweeping changes?
2017-07-26 16:45:57 <kaniini> we try to do most work directly in alpine instead of in adelie
2017-07-26 16:46:17 <kaniini> adelie largely is the "backup plan"
2017-07-26 16:46:19 <^7heo> upstream all the fixes.
2017-07-26 16:46:30 <^7heo> which is the attitude of alpine too, afaik
2017-07-26 16:46:53 <kaniini> originally, adelie was based on gentoo as a backup plan to if alpine were to fail
2017-07-26 16:47:07 <kaniini> but gentoo seems to have serious troubles
2017-07-26 16:47:13 <kaniini> so we switched to being an alpine consumer
2017-07-26 16:47:31 <kaniini> which makes more sense anyway because if we need to, we can just crossgrade our alpine servers to adelie (which we plan to do anyway)
2017-07-26 16:48:28 <kaniini> jirutka: maybe in future.  right now i just purged a few tech bros and i'm not looking to hire anyone right now
2017-07-26 16:48:55 <jirutka> kaniini: okay, so let me know when the situation change ;
2017-07-26 16:48:56 <jirutka> ))
2017-07-26 21:54:24 <kaniini> jirutka: ^
2017-07-28 22:29:46 <jirutka> algitbot: retry master
2017-07-29 02:07:11 <travis-ci> apk-tools:master [passed] |A. Wilcox| index: Only rewrite arch if no error is encountered | https://travis-ci.org/alpinelinux/apk-tools/builds/258777048
2017-07-29 19:21:15 <jirutka>  nmeum: could you please fix these broken builds? ^
2017-07-29 19:21:23 <nmeum> you mean libpng?
2017-07-29 19:22:05 <nmeum> jirutka: ^
2017-07-29 19:22:21 <jirutka> nmeum: yes
2017-07-29 19:23:12 <jirutka> it seems that it’s only libpng, s390x failures are unrelated, there’s more broken things for longer time
2017-07-29 19:23:51 <nmeum> libpng builds fine on x86 and x86_64 it only fails on armhf and aarch64
2017-07-29 19:24:08 <Shiz> forgot to join
2017-07-29 19:24:16 <jirutka> yes, but previous version was fine, so there’s probably something wrong in this release
2017-07-29 19:25:14 <jirutka> Assembler messages: arm/../pngpriv.h:476: Error: bad instruction `typedef unsigned long png_ptruint'; Shiz , do you have some idea?
2017-07-29 19:25:32 <Shiz> it's trying to process a C file as if it were assembly
2017-07-29 19:25:38 <Shiz> or rather a header file
2017-07-29 19:25:38 <jirutka> huh
2017-07-29 19:25:48 <nmeum> probably I was looking into it but I don't haven an arm machine at hand currently which makes it hard to fix this
2017-07-29 19:25:57 <jirutka> i understand
2017-07-29 19:26:00 <nmeum> I could just revert the upgrade for now
2017-07-29 19:28:05 <nmeum> libpng 1.6.31 seems to fix this
2017-07-29 19:28:33 <nmeum> however as I said above a) I couldn't test this and b) libpng-apng isn't available for that version yet
2017-07-29 19:28:48 <jirutka> i see notice on http://www.libpng.org/pub/png/libpng.html
2017-07-29 19:30:29 <jirutka> hm, quite confusing, 1.6.31 is already relased, but not mentioned on homepage https://sourceforge.net/projects/libpng/files/libpng16/1.6.31/
2017-07-29 19:30:59 <jirutka> nmeum: what is libpng-apng?
2017-07-29 19:31:22 <nmeum> a patchset for libpng to support the apng format
2017-07-29 19:31:28 <nmeum> apng are animated pngs
2017-07-29 19:31:35 <nmeum> *apngs
2017-07-29 19:31:44 <jirutka> aha
2017-07-29 19:32:03 <jirutka> i see
2017-07-29 19:33:02 <jirutka> well, then it’d be probably better to revert the upgrade until libpng-apng for 1.6.31 is released
2017-07-29 19:33:57 <nmeum> ok, will do so
2017-07-29 19:34:27 <jirutka> thanks
2017-07-29 19:37:07 <nmeum> there you go
2017-07-29 19:37:33 <Shiz> nice desu