IRC logs for #buildstream for Tuesday, 2018-11-06

*** Prince781 has quit IRC00:04
*** Prince781_ has joined #buildstream00:04
*** Prince781_ is now known as Prince78100:05
*** janeric has joined #buildstream00:58
*** bochecha has quit IRC01:27
*** dineshdb[m] has quit IRC02:14
*** inigomartinez has quit IRC02:14
*** waltervargas[m] has quit IRC02:15
*** segfault3[m] has quit IRC02:15
*** ssssam[m] has quit IRC02:15
*** rafaelff[m] has quit IRC02:15
*** mattiasb has quit IRC02:15
*** albfan[m] has quit IRC02:15
*** alatiera has quit IRC02:15
*** cgmcintyre[m] has quit IRC02:15
*** krichter[m] has quit IRC02:15
*** kailueke[m] has quit IRC02:15
*** tlater[m] has quit IRC02:15
*** awacheux[m] has quit IRC02:15
*** jjardon[m] has quit IRC02:15
*** doras[m] has quit IRC02:15
*** asingh_[m] has quit IRC02:15
*** oknf[m] has quit IRC02:15
*** Demos[m] has quit IRC02:15
*** connorshea[m] has quit IRC02:15
*** pro[m] has quit IRC02:15
*** theawless[m] has quit IRC02:15
*** m_22[m] has quit IRC02:16
*** abderrahim[m] has quit IRC02:16
*** xjuan has quit IRC02:22
*** Prince781 has quit IRC02:26
*** dineshdb[m] has joined #buildstream02:38
*** waltervargas[m] has joined #buildstream02:41
*** inigomartinez has joined #buildstream02:52
*** segfault3[m] has joined #buildstream02:52
*** Prince781 has joined #buildstream02:57
*** jjardon[m] has joined #buildstream03:02
*** asingh_[m] has joined #buildstream03:07
*** oknf[m] has joined #buildstream03:07
*** doras[m] has joined #buildstream03:08
*** Prince781 has quit IRC03:09
*** connorshea[m] has joined #buildstream03:13
*** Demos[m] has joined #buildstream03:18
*** eldritch has joined #buildstream03:23
*** inigomartinez has quit IRC03:37
*** jjardon[m] has quit IRC03:37
*** connorshea[m] has quit IRC03:37
*** doras[m] has quit IRC03:37
*** waltervargas[m] has quit IRC03:37
*** Demos[m] has quit IRC03:37
*** dineshdb[m] has quit IRC03:37
*** oknf[m] has quit IRC03:37
*** segfault3[m] has quit IRC03:37
*** asingh_[m] has quit IRC03:38
*** theawless[m] has joined #buildstream04:29
*** pro[m] has joined #buildstream04:31
*** abderrahim[m] has joined #buildstream04:46
*** m_22[m] has joined #buildstream04:52
*** albfan[m] has joined #buildstream05:18
*** ssssam[m] has joined #buildstream05:32
*** mattiasb has joined #buildstream05:55
*** rafaelff[m] has joined #buildstream05:56
*** cgmcintyre[m] has joined #buildstream06:15
*** alatiera has joined #buildstream06:38
*** krichter[m] has joined #buildstream06:48
*** kailueke[m] has joined #buildstream06:51
*** tlater[m] has joined #buildstream07:01
*** awacheux[m] has joined #buildstream07:04
*** asingh_[m] has joined #buildstream07:14
*** jjardon[m] has joined #buildstream07:20
*** segfault3[m] has joined #buildstream07:23
*** oknf[m] has joined #buildstream07:24
*** connorshea[m] has joined #buildstream07:25
*** inigomartinez has joined #buildstream07:35
*** doras[m] has joined #buildstream07:57
*** waltervargas[m] has joined #buildstream08:05
*** Demos[m] has joined #buildstream08:05
*** dineshdb[m] has joined #buildstream08:08
gitlab-br-botjonathanmaw approved MR !931 (jmac/remote-sandbox-errors->master: _sandboxremote.py: Use the standard SandboxError) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/93108:40
*** toscalix has joined #buildstream09:02
valentindJust to be sure. If I need some extra python dependencies in a branch where the tests depend on that dependency. I need first to make a merge with just the dependency added so that the when it is merged, the docker images are generated with the dependency. And then I can merge the rest.09:03
valentindAm I right?09:03
valentindAh, no there is a dev-requirements.txt in the buildstream-docker-images itself.09:05
jmacI've finished the performance test on !911 and it looks roughly equivalent. If anyone's got time to review it, that would be welcome.09:40
gitlab-br-botMR !911: Direct CAS-to-CAS import https://gitlab.com/BuildStream/buildstream/merge_requests/91109:40
juergbijmac: I will take a look09:41
*** benschubert has joined #buildstream09:48
*** tiagogomes has joined #buildstream09:50
*** raoul has joined #buildstream10:03
WSalmonI found that if i import stream in to the frontend/cli.py then the auto completion brakes but the tests do not capture this, is this know acceptable behavour or should we have a issue to catch this sort of error or have we tried and found it too hard? the issue seems to be a import of a import of stream doing 'from ... import foo' and python not getting the loaction right when run to give auto complete10:14
WSalmonthe test would be to check for this failure mode of auto completion rather than to allow stream to be imported10:15
*** bochecha has joined #buildstream10:16
juergbiWSalmon: if you can think of a simple way to improve/extend the completion tests to cover this, that would certainly be great10:17
juergbihowever, if it requires a large effort, there are probably more important tasks to tackle first10:18
WSalmoni havent really thought about testing, i just had a panic that one of my MR that was close to landing was going to brake auto complete and thought this might help some one else not have my same panic, (i have fixed my MR now)10:19
WSalmoni will have a think about testing10:19
juergbiyes, it would definitely be good to catch that. one option might also be to change the way imports are handled for completion to actually allow this (just a random thought, don't know whether this makes sense)10:20
*** alatiera_ has joined #buildstream10:33
*** lachlan has quit IRC10:46
*** lachlan has joined #buildstream11:15
*** lachlan has quit IRC11:23
*** lachlan has joined #buildstream11:36
KinnisonAre Provenance objects ever altered?11:36
* Kinnison is trying to understand why they have a clone() method11:37
*** lachlan has quit IRC11:39
*** TacoThief has joined #buildstream11:45
gitlab-br-botwillsalmon opened issue #751 (test for import failures for auto completes) on buildstream https://gitlab.com/BuildStream/buildstream/issues/75111:48
jennisDoes anyone have the link for our IRC logs? :)11:50
jmacThe one in the topic? Why the smiley?11:50
jennisJust being friendly11:51
jennisand thanks11:51
* jennis did not notice that before11:51
jmacAh, OK, I wondered if I was missing a joke11:51
*** lachlan has joined #buildstream11:54
*** jonathanmaw has joined #buildstream12:15
*** lachlan has quit IRC12:19
*** lachlan has joined #buildstream13:00
*** lachlan has quit IRC13:10
*** lachlan has joined #buildstream13:19
*** lachlan has quit IRC13:24
tpollardjuergbi: I've rebased !768 on top of !922 that landed in master, hopefully it's acceptable13:33
gitlab-br-botMR !768: Retries should fail Backport 1.2 https://gitlab.com/BuildStream/buildstream/merge_requests/76813:33
gitlab-br-botMR !922: Split up artifact cache and CAS cache https://gitlab.com/BuildStream/buildstream/merge_requests/92213:33
tpollarderm, that should be !78613:33
gitlab-br-botMR !786: Don't pull artifact buildtrees by default https://gitlab.com/BuildStream/buildstream/merge_requests/78613:33
juergbitpollard: will take a quick look at the cas/artifactcache part13:34
tpollardcheers :)13:35
juergbitpollard: looks fine at quick glance. just two tiny nitpicks13:43
tpollardjuergbi: cool13:45
tpollardjuergbi: do you think initial or original could be used in place of element in that variable name?13:47
juergbioriginal sounds fine to me13:48
tpollardty!13:50
alatierahello, I recall a while ago I was talking with someone about a plugin that generates oci/docker images13:53
alatieradoes this exist, and is it functional?13:53
alatieraor was it just a concept/future problem, I have a hard time recalling13:53
WSalmon^cs-shadow finn tlater[m]13:55
finndocker image plugin13:58
finnI don't think I ever used it to generate docker images13:58
finnI've used the docker source plugin13:58
WSalmondose it live in bst-external?13:59
Kinnisonalatiera: It was mostly conceptual though there is a prototype which uses bst checkout and docker's tooling13:59
Kinnisonalatiera: https://gitlab.com/BuildStream/buildstream/blob/master/contrib/bst-docker-import13:59
alatieraKinnison, Nice seems straight forward overall14:01
alatieraKinnison: so it can be done with just making a tarball of the bst checkout? Pretty cooL!14:02
Kinnisonalatiera: I'll admit I'm not a docker/oci expert, I just remember Chandan merging that14:03
alatieraI am intrested in using Buildstream to generate images for the gitlab CI in gnome mainly, docker is a pain to deal with14:06
alatieralooks doable, thanks a lot for the helP!14:06
gitlab-br-botrichardmaw-codethink approved MR !799 (Qinusty/unit-test-utils->master: Fix issue with _pretty_size with large numbers of bytes) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/79914:12
gitlab-br-botrichardmaw-codethink merged MR !799 (Qinusty/unit-test-utils->master: Fix issue with _pretty_size with large numbers of bytes) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/79914:12
*** lachlan has joined #buildstream14:22
*** jonathanmaw_ has joined #buildstream14:26
*** jonathanmaw has quit IRC14:26
*** lachlan has quit IRC14:37
*** lachlan has joined #buildstream14:39
*** Prince781 has joined #buildstream14:52
*** Prince781 has quit IRC14:54
*** lachlan has quit IRC14:57
*** lachlan has joined #buildstream15:00
gitlab-br-bottpollard approved MR !931 (jmac/remote-sandbox-errors->master: _sandboxremote.py: Use the standard SandboxError) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/93115:07
*** lachlan has quit IRC15:12
*** jonathanmaw_ has quit IRC15:16
*** jonathanmaw has joined #buildstream15:16
*** lachlan has joined #buildstream15:20
*** lachlan has quit IRC15:28
*** lachlan has joined #buildstream15:32
*** lachlan has quit IRC15:36
*** lachlan has joined #buildstream15:45
*** lachlan has quit IRC15:48
*** lachlan has joined #buildstream16:16
*** lachlan has quit IRC16:24
*** ctolentino has quit IRC16:26
*** lachlan has joined #buildstream16:46
*** toscalix has quit IRC16:46
gitlab-br-botjennis opened MR !932 (jennis/docs_patch->master: Update our instructions/examples regarding how we declare remote caches) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/93216:48
jennisA somewhat lengthy docs MR ^ I would appreciate if anyone could spare the time to review16:49
jennisPlease try and understand the MR description before reviewing16:49
adds68skullman, hey what is the method for printing to stdout during a build again?16:56
skullmanif it's in a plugin it's probably self.status16:56
skullmanor self.debug16:56
adds68skullman, that's it! Thanks16:57
*** lachlan has quit IRC17:14
*** Prince781 has joined #buildstream17:15
gitlab-br-botjmacarthur closed issue #746 (SandboxRemote causes bugs instead of errors) on buildstream https://gitlab.com/BuildStream/buildstream/issues/74617:16
gitlab-br-botjmacarthur merged MR !931 (jmac/remote-sandbox-errors->master: _sandboxremote.py: Use the standard SandboxError) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/93117:16
gitlab-br-botdanielsilverstone-ct opened MR !933 (danielsilverstone-ct/further-optimisations->master: WIP: Some more microoptimisations) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/93317:20
KinnisonFor those who might care, https://gitlab.com/BuildStream/buildstream/merge_requests/933 carries two more possible optimisations for pre-scheduler time in `bst build` totalling around 28s improvement in my test case17:21
Kinnisonit's WIP because I'm trying to find further improvements17:21
*** lachlan has joined #buildstream17:23
adds68When using a mirror for a bst source, where would i find it once it has been resolved?17:23
adds68I have a list of sources for a given element, calling source.mirror.url still returns the mirror name, rather than the resolved url17:24
*** abderrahim2 has joined #buildstream17:26
*** abderrahim1 has quit IRC17:27
ikerperezHow long will last "bst source-bundle" on Buildstream? is it worth to start working with it or should I wait for this request be approved https://gitlab.com/BuildStream/buildstream/merge_requests/82017:31
*** lachlan has quit IRC17:41
persiaikerperez: Probably not worth working with it.17:50
persiaThat said, if you have a use case that it uniquely solves, that is worth discussion.  Maybe the use case can be addressed another way, or maybe there are reasons not to deprecate it.17:50
persiaIn the absence of argument, I'm expecting it to be announced as deprecated in the next release.17:51
ikerperezok, thanks persia17:57
gitlab-br-botvalentindavid opened MR !934 (valentindavid/script-artifact-corruption->master: Fix cache corruption by scripts when layout and integration commands are used) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/93417:58
*** finn has quit IRC18:03
*** tpollard has quit IRC18:03
*** tiagogomes has quit IRC18:03
*** jonathanmaw has quit IRC18:03
*** phildawson_ has quit IRC18:03
*** WSalmon has quit IRC18:03
*** raoul has quit IRC18:03
*** rdale has quit IRC18:04
*** toscalix has joined #buildstream18:04
*** toscalix has quit IRC18:05
*** toscalix has joined #buildstream18:06
*** toscalix has quit IRC18:07
*** tiagogomes has joined #buildstream18:22
*** tpollard has joined #buildstream18:22
*** WSalmon has joined #buildstream18:23
*** Prince781 has quit IRC18:29
*** lachlan has joined #buildstream18:39
*** finn has joined #buildstream18:40
*** bethw has quit IRC18:40
*** lachlan has quit IRC18:40
*** lachlan has joined #buildstream18:40
*** finn has quit IRC18:40
*** benbrown has quit IRC18:40
*** benbrown has joined #buildstream18:40
*** aiden has quit IRC18:40
*** lachlan has quit IRC18:40
*** aiden has joined #buildstream18:40
*** adds68 has quit IRC18:40
*** coldtom has quit IRC18:40
*** coldtom has joined #buildstream18:41
*** adds68 has joined #buildstream18:41
*** bethw has joined #buildstream18:41
*** Prince781 has joined #buildstream19:24
*** xjuan has joined #buildstream19:38
*** greppy has joined #buildstream20:14
*** Prince781 has quit IRC20:42
*** benschubert has quit IRC20:53
*** bjurr has joined #buildstream21:36
*** alatiera_ has quit IRC23:50

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!