IRC logs for #buildstream for Friday, 2019-01-11

« Thursday, 2019-01-10 Index Saturday, 2019-01-12 »
*** xjuan has joined #buildstream00:53
*** xjuan has quit IRC01:48
*** tristan has joined #buildstream03:24
*** nimish has joined #buildstream03:29
*** kapil___ has joined #buildstream04:20
gitlab-br-bottristanvb merged MR !1058 (tristan/element-processing-order->master: Make build plan element list stable) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/105804:42
*** skullman has quit IRC05:16
*** tristan has quit IRC05:44
*** nimish has quit IRC05:56
*** nimish has joined #buildstream06:59
*** kapil___ has quit IRC07:09
*** alatiera has joined #buildstream07:36
*** tpollard has joined #buildstream08:40
*** toscalix has joined #buildstream08:49
*** tpollard has quit IRC09:22
*** CTtpollard has joined #buildstream09:22
*** bochecha has joined #buildstream09:42
jmac!1056 is a ten-line change if anyone has time to review it09:45
gitlab-br-botMR !1056: _sandboxremote.py: Create marked directories before running https://gitlab.com/BuildStream/buildstream/merge_requests/105609:45
*** nimish has quit IRC09:47
bochechato whoever maintains the buildstream docker images and hasn't updated them to Fedora 29 yet: thank you very much :)09:49
bochecha(Fedora 29 breaks our ABI checker script in Freedesktop-Sdk in a way that it silently ignores 98% of libraries, and always reports ABI compatibility as a result, even when the ABI was actually broken 😱️)09:50
jmacvalentind: I've reproduced the RESOURCE_EXHAUSTED problem you saw earlier.09:50
valentindjmac, there is a branch fixing it.09:51
valentindhttps://gitlab.com/BuildGrid/buildgrid/merge_requests/13809:52
jmacAh yes, thanks09:53
*** benschubert has joined #buildstream10:06
*** jonathanmaw has joined #buildstream10:11
*** lachlan has joined #buildstream10:28
*** alatiera_ has joined #buildstream10:35
*** alatiera has quit IRC10:37
*** alatiera_ is now known as alatiera10:37
*** CTtpollard is now known as tpollard10:39
tpollardhas anyone seen bst delete entries in a project.conf after running?10:39
valentindtpollard, sounds weird. I thought only elements, project.refs and junction.refs could be rewritten.10:44
tpollardyes it is weird10:45
tpollardI seem to have a test case in which an entry is removed from a project.conf after a build has run10:46
tpollardI would have thought if it was a malformed/parsing issue it would just complain about it10:48
valentindtpollard, has it happened only once? or can you reproduce it?10:48
tpollardvalentind: I will push it to a branch soon, it's a test case for https://gitlab.com/BuildStream/buildstream/merge_requests/1024/10:49
tpollardI'm trying to complete a testcase where multiple project.confs (i.e top level and subproject/junction) define the same remote url but with different push config10:52
tpollardand part of that requires me to write the urls to the files at runtime, which seems to succeed but after bst has ran the entries seemingly get removed10:54
*** nimish has joined #buildstream10:55
*** juergbi has quit IRC11:03
*** juergbi has joined #buildstream11:19
*** nimish has quit IRC11:25
*** flatmush has quit IRC11:34
*** flatmush has joined #buildstream11:34
*** finn_ has joined #buildstream12:07
*** phildawson_ has quit IRC12:43
*** bochecha has quit IRC12:53
tpollardvalentind: I've pushed the failing test to https://gitlab.com/BuildStream/buildstream/merge_requests/1024/13:48
*** nimish has joined #buildstream14:01
*** nimish has quit IRC14:11
*** nimish has joined #buildstream14:11
jennisI'm looking to return information to the user without using the stream API, what's the general process to do this?14:26
*** nimish has quit IRC14:26
*** nimish has joined #buildstream14:27
lachlanHi, I have a patch to fix a benchmarking issue which is currently preventing one of the key benchmarks from running successfully. If someone could take a look with respects to merging I would be most obliged. https://gitlab.com/BuildStream/benchmarks/commits/lachlanmackenzie/Fix_bst_fetching14:27
jennisContext: I'm adding the `bst artifact delete` command, if we specify an artifact that is NOT in the cache, we get a misleading Exception, instead, in cli.py, for each artifact we try to delete with `bst artifact delete`, I want to output to users artifacts we haven't explicitly deleted *because they weren't there in the first place*14:28
tpollardjennis: check if it's cached at cli.py and return the warning there?14:30
jennisYes, but I'm asking how we return such warnings?14:30
tpollardbut I suppose you have to steam.load it to be able to query if it's cached first14:31
tpollard*stream14:31
jennisexactly, which I don't do for the case of explicit artifact deletion14:31
jennisWe have examples of `click.echo('WARNING:...')` in cli.py, but I'm not convinced this is suitable14:32
jennis"WARNING: something you're explictly trying to delete is not here..." <--- is that ok?14:32
jennisahh, no, this is what I want, as I can direct that to stderr :)14:33
jennisthanks tpollard14:33
tpollardI think shell has an app error raised in cli.py if a buildtree is not found for a specific condition14:34
jennisYeah, I don't want to raise an error though, because if we specify multiple artifact refs for arguments, I don't want latter arguments to *not* be deleted14:35
tpollardmaybe a canditate for a configurable fatal warning14:36
tpollardi.e their might be a situation in which you'd want it to error out if an artifact was missing14:37
tpollardthat was expected to be there14:37
tpollardbut by default continue with warning makes sense yeh14:37
jennisTrue, I'll leave that for MR discussion though14:37
jennisbecause if we accept the fact that if we're deleting, and it's already not there, that's good news14:38
jennisSeems sensible to me14:38
tpollardyeh I think that's a fine default presumption14:38
*** lachlan has quit IRC14:47
*** nimish has quit IRC14:52
*** bochecha has joined #buildstream14:52
*** nimish has joined #buildstream14:53
*** nimish has quit IRC14:58
*** nimish has joined #buildstream14:58
*** lachlan has joined #buildstream15:03
laurencelachlan, it's the two latest commits to your branch there that you want folks to have a look at?15:04
laurenceNot sure who has an understanding of that code,was a fair while back that ssssam[m] added the baserock tests there15:05
laurencenot that that should stop others having a look15:05
*** nimish_ has joined #buildstream15:13
*** nimish has quit IRC15:14
*** nimish_ is now known as nimish15:14
*** nimish has quit IRC15:33
*** nimish has joined #buildstream15:34
gitlab-br-botaevri opened MR !1059 (aevri/tox_no_cov->master: pytest-cov==2.6.1, contributing: '--last-failed' and '--no-cov' tips) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/105915:40
*** nimish has quit IRC15:44
*** nimish has joined #buildstream15:44
*** nimish has quit IRC15:48
*** lachlan has quit IRC16:05
*** lachlan has joined #buildstream16:20
*** toscalix has quit IRC16:21
*** lachlan has quit IRC16:28
benschubertIs anyone running python 3.7 with BuildStream by any chance?16:30
benschubertFor me {tox,pytest} --no-cov ends up failing all the time16:31
*** lachlan has joined #buildstream16:33
*** nimish has joined #buildstream16:37
*** lachlan has quit IRC16:37
gitlab-br-botcs-shadow approved MR !1059 (aevri/tox_no_cov->master: pytest-cov==2.6.1, contributing: '--last-failed' and '--no-cov' tips) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/105916:43
*** nimish has quit IRC16:46
*** nimish has joined #buildstream16:47
*** lachlan has joined #buildstream16:53
*** nimish has quit IRC16:57
*** nimish has joined #buildstream16:57
*** juergbi has quit IRC16:58
*** juergbi has joined #buildstream16:58
*** jonathanmaw has quit IRC17:02
*** tristan has joined #buildstream17:05
*** ChanServ sets mode: +o tristan17:05
*** sambishop has quit IRC17:07
*** tpollard has quit IRC17:14
gitlab-br-botaevri merged MR !1059 (aevri/tox_no_cov->master: pytest-cov==2.6.1, contributing: '--last-failed' and '--no-cov' tips) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/105917:19
*** nimish has quit IRC17:22
*** nimish has joined #buildstream17:23
*** nimish has joined #buildstream17:23
*** nimish has quit IRC17:28
*** nimish has joined #buildstream17:29
*** finn_ has quit IRC17:33
*** nimish has quit IRC17:33
*** nimish has joined #buildstream17:34
gitlab-br-botjennis opened MR !1060 (jennis/new_artifact_subcommands->master: Introduce new artifact subcommand `delete`) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/106017:35
*** nimish has quit IRC17:54
*** nimish has joined #buildstream17:54
gitlab-br-bottristanvb opened issue #857 (test_pull_missing_blob spuriously crashes) on buildstream https://gitlab.com/BuildStream/buildstream/issues/85717:56
*** alatiera_ has joined #buildstream18:27
*** alatiera has quit IRC18:28
*** alatiera_ is now known as alatiera18:28
*** nimish has quit IRC18:40
*** lachlan has quit IRC18:43
*** nimish has joined #buildstream18:49
*** nimish has joined #buildstream18:50
*** tristan has quit IRC19:05
*** tristan has joined #buildstream19:22
*** tristan_ has joined #buildstream19:23
*** nimish has quit IRC19:32
*** bochecha has quit IRC20:08
*** nimish has joined #buildstream20:21
*** alatiera_ has joined #buildstream20:27
*** alatiera has quit IRC20:28
*** alatiera_ is now known as alatiera20:29
*** nimish has quit IRC20:45
*** benschubert has quit IRC21:11
*** alatiera has quit IRC21:15
*** alatiera has joined #buildstream21:16
*** tristan_ has quit IRC21:24
*** tristan has joined #buildstream22:24
*** alatiera has quit IRC22:41
*** tristan has quit IRC23:53
« Thursday, 2019-01-10 Index Saturday, 2019-01-12 »

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