| *** Prince781 has quit IRC | 00:28 | |
| *** alatiera_ has quit IRC | 00:46 | |
| *** Prince781 has joined #buildstream | 01:08 | |
| *** bochecha has quit IRC | 01:22 | |
| *** slaf has quit IRC | 03:00 | |
| *** slaf has joined #buildstream | 03:02 | |
| *** Havvy has joined #buildstream | 04:29 | |
| *** Prince781 has quit IRC | 07:14 | |
| *** Prince781 has joined #buildstream | 07:25 | |
| *** alatiera_ has joined #buildstream | 08:26 | |
| *** benschubert has joined #buildstream | 08:41 | |
| *** toscalix has joined #buildstream | 08:53 | |
| *** finn has joined #buildstream | 08:54 | |
| *** toscalix_ has joined #buildstream | 09:10 | |
| *** WSalmon has joined #buildstream | 09:14 | |
| *** toscalix has quit IRC | 09:36 | |
| *** toscalix has joined #buildstream | 09:38 | |
| *** toscalix has quit IRC | 09:41 | |
| *** jonathanmaw has joined #buildstream | 09:45 | |
| *** tiagogomes has joined #buildstream | 09:47 | |
| *** tpollard has joined #buildstream | 09:57 | |
| *** tiagogomes has quit IRC | 10:00 | |
| *** raoul has joined #buildstream | 10:29 | |
| *** lachlan has joined #buildstream | 10:31 | |
| *** bochecha has joined #buildstream | 10:39 | |
| *** jonathanmaw_ has joined #buildstream | 10:45 | |
| *** jonathanmaw has quit IRC | 10:47 | |
| juergbi | jmac: as you'll likely do work on top of !922, it may also make sense for you to take a look at my changes in that MR. the main commit is a bit tedious to really review but maybe check that the new API of the two classes makes sense to you as well | 10:49 |
|---|---|---|
| gitlab-br-bot | MR !922: Split up artifact cache and CAS cache https://gitlab.com/BuildStream/buildstream/merge_requests/922 | 10:49 |
| laurence | lachlan, here's the initial discussions on performance monitoring from earlier this year - https://mail.gnome.org/archives/buildstream-list/2017-December/msg00012.html | 10:50 |
| laurence | https://mail.gnome.org/archives/buildstream-list/2018-February/msg00012.html | 10:50 |
| jmac | Ok, I'll consider that my review task for the day | 10:50 |
| laurence | https://mail.gnome.org/archives/buildstream-list/2018-March/msg00018.html | 10:50 |
| juergbi | ta | 10:50 |
| laurence | A few links there for some helpful context, lachlan | 10:50 |
| lachlan | Thanks laurence | 10:51 |
| *** tiagogomes has joined #buildstream | 10:54 | |
| lachlan | I will look at incorporating this in !917 | 10:55 |
| gitlab-br-bot | MR !917: Add trigger for benchmarking https://gitlab.com/BuildStream/buildstream/merge_requests/917 | 10:55 |
| gitlab-br-bot | jonathanmaw approved MR !928 (juerg/remote-execution->master: element.py: Pass bare_directory also to SandboxRemote()) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/928 | 11:31 |
| gitlab-br-bot | juergbi merged MR !928 (juerg/remote-execution->master: element.py: Pass bare_directory also to SandboxRemote()) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/928 | 11:33 |
| Nexus | benschubert: Would you be able to have a look at my reply to https://gitlab.com/BuildStream/buildstream/issues/732 please and get back to me so i can look into this issue? :) | 11:53 |
| benschubert | Nexus: done! | 11:54 |
| Nexus | thanks :D | 11:55 |
| benschubert | realized I forgot an important detail (wsl) -_- sorry! | 11:55 |
| Nexus | benschubert: Which commit of master was that? | 11:56 |
| benschubert | any should do, let me check | 11:57 |
| benschubert | https://gitlab.com/BuildStream/buildstream/blob/master/buildstream/_fuse/fuse.py#L77 that's the offending line | 11:58 |
| benschubert | I just realized there is a "pylint: skip-file" so it might actually be a pylint issue -_-' | 11:58 |
| Nexus | i believe Kinnison posted a pylint fix a short time back? I'm wondering if that is why i don't get a linting error | 12:00 |
| benschubert | the last change is 2 months ago | 12:01 |
| benschubert | I had a later version | 12:01 |
| benschubert | I'll give it a more thorough look :/ | 12:01 |
| Kinnison | We may be running with a more modern pytest-pylint version than you | 12:01 |
| Kinnison | it may be worth clearing your .eggs out | 12:01 |
| benschubert | I\m running pip install -r dev-requirements.txt. Would expect that to be up to date | 12:02 |
| gitlab-br-bot | valentindavid opened MR !929 (valentindavid/limit_fetchers_in_overnight_build->master: Use less fetchers in overnight build to work-around limitations from servers.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/929 | 12:16 |
| *** paulsherwood has quit IRC | 12:20 | |
| *** paulsherwood has joined #buildstream | 12:20 | |
| lachlan | I think one of the requirements under the benchmarking repo is wrong - yaml does not resolve. | 12:22 |
| * skullman grumbles about tests now failing because pytest can't find the tmp_path fixture | 12:27 | |
| gitlab-br-bot | jjardon approved MR !929 (valentindavid/limit_fetchers_in_overnight_build->master: Use less fetchers in overnight build to work-around limitations from servers.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/929 | 12:27 |
| tpollard | well, our test setup allowed it to slip through | 12:29 |
| skullman | hm, use of tmp_path was added as part of "_frontend/app.py: Set correct element-path in interactive bst-init", and tmp_path was added as part of https://github.com/pytest-dev/pytest/issues/3985. looks like a version bump should fix it | 12:29 |
| tpollard | yep, was discussed on friday | 12:29 |
| skullman | ah, I wasn't around then, that'll explain it | 12:30 |
| * tpollard nods | 12:30 | |
| tpollard | it did however capture something relying on 3.6+, because the deb runner is on 3.5 | 12:32 |
| skullman | looks like I need 3.8 in dev-requirements.txt for it to work | 12:34 |
| tpollard | sorry I was referring to python directly there | 12:35 |
| skullman | np | 12:35 |
| gitlab-br-bot | valentindavid opened issue #749 (google.protobuf.message.DecodeError: Error parsing message) on buildstream https://gitlab.com/BuildStream/buildstream/issues/749 | 12:39 |
| *** jonathanmaw_ has quit IRC | 12:41 | |
| *** jonathanmaw has joined #buildstream | 12:41 | |
| gitlab-br-bot | jjardon closed issue #742 (Overnigth tests are failing: too many client connections to a specific repo) on buildstream https://gitlab.com/BuildStream/buildstream/issues/742 | 12:41 |
| gitlab-br-bot | jjardon merged MR !929 (valentindavid/limit_fetchers_in_overnight_build->master: Use less fetchers in overnight build to work-around limitations from servers.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/929 | 12:41 |
| *** xPsycho has joined #buildstream | 13:20 | |
| jjardon | valentind: hi, did you see my comment at https://gitlab.com/BuildStream/buildstream/merge_requests/745#note_114559346 ? I think it would be nice to merge that before 1.4 | 13:25 |
| valentind | Oh, so I need to make a merge request of my changes. | 13:26 |
| valentind | jjardon, I assigned the ticket to myself. will do that. | 13:27 |
| jjardon | valentind: thanks! | 13:31 |
| gitlab-br-bot | richardmaw-codethink opened MR !930 (richardmaw/pytest-dep-update->master: dev-requirements: Update pytest dependency to 3.8) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/930 | 13:37 |
| skullman | \o/ movement on https://github.com/projectatomic/bubblewrap/pull/293 | 13:39 |
| *** jonathanmaw_ has joined #buildstream | 14:00 | |
| *** jonathanmaw has quit IRC | 14:01 | |
| Nexus | Currently, if we have a typo in the "element-path" cariable of a project.conf, rather than a yaml error, it just sets the element path to the project root. This looks to me like we're just checking if "element-path" is there, rather than checking if the variable matches the list of acceptable variables. Does that seem accurate? | 14:36 |
| juergbi | Nexus: we normally use node_validate() to check for invalid keys. not sure what's going wrong in that case | 14:42 |
| juergbi | see Project._load_second_pass() | 14:42 |
| Nexus | kk, in Stream._load(), Project._load_second_pass is called waaaay too late | 15:09 |
| Nexus | it attempts to use the element-path on line _stream.py:876, and Project._load_second_pass isn't called until line 931 | 15:11 |
| juergbi | ah, maybe we can actually move node_validate to _load(). not sure | 15:23 |
| *** lachlan has quit IRC | 15:38 | |
| ikerperez | Hi, I am trying to apply fossology to a project built using buildstream. Fossology it's a licence compliance tool and to be able to operate it needs the source code of all the elements needed the project. Is there any way to get the source code of the elements and their dependencies in Buildstream? | 15:39 |
| ikerperez | https://www.fossology.org/ | 15:39 |
| *** lachlan has joined #buildstream | 15:42 | |
| Nexus | juergbi: that would be a large change :/ | 15:45 |
| tiagogomes | valentind what's the status of https://gitlab.com/BuildStream/buildstream/merge_requests/908? It is needed for our project | 15:45 |
| WSalmon | there are other licence checker programs already in the external plugins repo you might be able to tweak, maybe adds68 could help ikerperez, i think he was working on it recently | 15:45 |
| *** lachlan has quit IRC | 15:45 | |
| juergbi | ikerperez: there is also a pending MR for source-checkout, which could help here: https://gitlab.com/BuildStream/buildstream/merge_requests/820 | 15:46 |
| tpollard | ikerperez: are you expecting to use fossology to create spdx reports? | 15:46 |
| juergbi | Nexus: what speaks against simply moving that single node_validate() call to other method? (i haven't looked into it) | 15:46 |
| ikerperez | tpollard, probably in the future will be of interest. But for now I am just trying to make it work | 15:47 |
| WSalmon | ikerperez, that said if its not suitable we have things like "bst source-bundle" could help but we are currently working to remove it and replace it with something very similar | 15:48 |
| ikerperez | juergbi I think that will make possible to use fossology | 15:48 |
| adds68 | ikerperez, are you wanting to do this during a Ci stage or as a plugin to buildstream? | 15:48 |
| ikerperez | thanks WSalmon I will have a look to that and to the tools you mentioned | 15:48 |
| phildawson_ | ikerperez, the current implementation source-bundle doesn't work for multiple elements so you'd need to use 'bst show' (or something) to get your dependency list then do source-checkout on each one individually | 15:48 |
| ikerperez | during the CI | 15:49 |
| ikerperez | ok, thanks phildawson_ | 15:49 |
| ikerperez | I will try your suggestions | 15:50 |
| tpollard | adds68: there's quite a few modes in fossology, but the most common usage is 'here's a tarball of my sources please scan them all' | 15:51 |
| jmac | juergbi: Why does ArtifactCache need a push_message method? (In !922) | 15:52 |
| gitlab-br-bot | MR !922: Split up artifact cache and CAS cache https://gitlab.com/BuildStream/buildstream/merge_requests/922 | 15:52 |
| juergbi | jmac: this is temporary, did you see my comment in the MR? | 15:53 |
| juergbi | SandboxRemote needs this, afaict | 15:53 |
| juergbi | as it currently uses the project's remotes, it doesn't have its own CASRemote | 15:53 |
| jmac | Oh, I see, because SandboxRemote is using ArtifactCache at the moment | 15:54 |
| jmac | Yes, that makes sense | 15:54 |
| Nexus | juergbi: the issue as i see it is that node_validate requires the projct ot be know, the project can only be figured out when the elements are returned, which happens after pipeline.load | 16:08 |
| Nexus | Which means currently, you can only check the yaml AFTER that point | 16:09 |
| Nexus | but it needs to be confirmed before then to prevent this issue | 16:09 |
| tpollard | ikerperez: I'm interested in source/license handling in buildstream going forward, so if any of your work is public please keep us updated! | 16:11 |
| ikerperez | tpollard I will do it | 16:11 |
| juergbi | Nexus: we might need node_validate() in both methods but can't we already call it in _load() using pre_config_node? that would obviously not validate includes as they haven't been processed yet but it might allow us to validate the rest | 16:11 |
| Nexus | juergbi: sorry i'm not familiar with the pre_config node | 16:12 |
| juergbi | valentind may be able to provide some more hints as he implemented include support / added pre_config_node | 16:15 |
| *** lachlan has joined #buildstream | 16:15 | |
| juergbi | (I assume commit 171e803f5 introduced this issue) | 16:16 |
| Nexus | ah juergbi, that's another thing in Project, not _stream | 16:19 |
| *** lachlan has quit IRC | 16:22 | |
| *** bochecha has quit IRC | 16:22 | |
| *** lachlan has joined #buildstream | 16:32 | |
| *** jonathanmaw has joined #buildstream | 16:32 | |
| *** jonathanmaw_ has quit IRC | 16:32 | |
| jmac | juergbi: Sorry, I've got another comment issue on !922, I should have batched these up. Feel free to leave it. | 16:33 |
| gitlab-br-bot | MR !922: Split up artifact cache and CAS cache https://gitlab.com/BuildStream/buildstream/merge_requests/922 | 16:33 |
| jmac | Otherwise, I'm happy with the MR | 16:33 |
| gitlab-br-bot | jmacarthur approved MR !922 (juerg/cas->master: Split up artifact cache and CAS cache) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/922 | 16:33 |
| juergbi | no worries, thanks for the review | 16:33 |
| *** toscalix_ is now known as toscalix | 16:40 | |
| gitlab-br-bot | tpollard approved MR !930 (richardmaw/pytest-dep-update->master: dev-requirements: Update pytest dependency to 3.8) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/930 | 16:40 |
| *** toscalix has quit IRC | 16:41 | |
| gitlab-br-bot | jmacarthur opened issue #750 (Alter SandboxRemote to use separate CAS store) on buildstream https://gitlab.com/BuildStream/buildstream/issues/750 | 16:41 |
| *** toscalix has joined #buildstream | 16:41 | |
| toscalix | Kinnison: your objection/comment to https://gitlab.com/BuildStream/buildstream/merge_requests/908 is it something solved, so the MR can be approved? | 16:41 |
| *** juanalday has joined #buildstream | 16:42 | |
| *** juanalday has quit IRC | 16:46 | |
| skullman | The bubblewrap change https://gitlab.com/BuildStream/buildstream/merge_requests/868 depends on was merged. Strictly it should probably depend on a release being made, but it looks like 0.3.2 will be next, so I'm going to un-mark it as WIP. | 16:46 |
| *** juanalday has joined #buildstream | 16:46 | |
| Kinnison | toscalix: That seems like a very superficial argument along the lines of "I can't test all of this so I refuse to test any of it" but I won't block the approval | 16:47 |
| * persia agrees that it makes sense to test at least some of it, if only to avoid potential future regressions | 16:48 | |
| toscalix | Kinnison: thanks | 16:50 |
| toscalix | Kinnison: thanks for adding the comment on the MR. You saved me a ping | 16:55 |
| jjardon | ikerperez: please, if you have some time, update https://gitlab.com/BuildStream/buildstream/issues/441 with your findings | 17:05 |
| jjardon | tpollard: ^ | 17:05 |
| gitlab-br-bot | richardmaw-codethink merged MR !930 (richardmaw/pytest-dep-update->master: dev-requirements: Update pytest dependency to 3.8) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/930 | 17:08 |
| tpollard | cheers jjardon | 17:09 |
| ikerperez | jjardon I will do it | 17:18 |
| *** lachlan has quit IRC | 17:19 | |
| *** jonathanmaw_ has joined #buildstream | 17:24 | |
| *** jonathanmaw has quit IRC | 17:24 | |
| *** abderrahim1 has joined #buildstream | 17:25 | |
| *** toscalix has quit IRC | 17:26 | |
| *** abderrahim has quit IRC | 17:27 | |
| gitlab-br-bot | jmacarthur opened MR !931 (jmac/remote-sandbox-errors->master: _sandboxremote.py: Use the standard SandboxError) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/931 | 17:37 |
| valentind | Nexus, juergbi, I was away. Is there anything I can help on related with include support? | 17:40 |
| *** lachlan has joined #buildstream | 17:41 | |
| *** juanalday has quit IRC | 17:42 | |
| *** aibo has joined #buildstream | 17:43 | |
| *** juanalday has joined #buildstream | 17:46 | |
| jmac | I've managed a "Cognitive complexity" of 51 in one of my MRs | 17:47 |
| gitlab-br-bot | juergbi closed issue #659 (Split up artifact cache and cas cache) on buildstream https://gitlab.com/BuildStream/buildstream/issues/659 | 17:51 |
| gitlab-br-bot | juergbi merged MR !922 (juerg/cas->master: Split up artifact cache and CAS cache) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/922 | 17:51 |
| *** tiagogomes has quit IRC | 17:52 | |
| tpollard | eek, dreaded can't me merged email | 17:54 |
| * tpollard will rebase in the morning | 17:54 | |
| *** lachlan has quit IRC | 18:02 | |
| *** xjuan has joined #buildstream | 18:12 | |
| juergbi | tpollard: that might be my fault. let me know if it's unclear how to rebase | 18:18 |
| *** xjuan has quit IRC | 18:19 | |
| *** juanalday has quit IRC | 18:21 | |
| *** lachlan has joined #buildstream | 18:32 | |
| *** juanalday has joined #buildstream | 18:34 | |
| *** Juan_ has joined #buildstream | 18:37 | |
| *** juanalday has quit IRC | 18:40 | |
| *** Prince781 has joined #buildstream | 18:43 | |
| *** lachlan has quit IRC | 18:46 | |
| *** Juan__ has joined #buildstream | 18:47 | |
| *** Juan_ has quit IRC | 18:49 | |
| *** Juan__ has quit IRC | 18:50 | |
| *** jonathanmaw_ has quit IRC | 18:51 | |
| *** Prince781 has quit IRC | 18:56 | |
| *** cohn has joined #buildstream | 19:06 | |
| *** xjuan has joined #buildstream | 19:11 | |
| *** raoul has quit IRC | 19:17 | |
| *** xjuan has quit IRC | 19:56 | |
| *** xjuan has joined #buildstream | 20:08 | |
| *** xjuan has quit IRC | 20:42 | |
| *** benschubert has quit IRC | 20:45 | |
| *** lachlan has joined #buildstream | 20:50 | |
| *** Prince781 has joined #buildstream | 21:23 | |
| *** xjuan has joined #buildstream | 21:31 | |
| *** Prince781 has quit IRC | 21:58 | |
| *** Prince781 has joined #buildstream | 21:59 | |
| *** bochecha has joined #buildstream | 22:14 | |
| *** Prince781 has quit IRC | 22:26 | |
| *** alatiera_ has quit IRC | 22:44 | |
| *** Prince781 has joined #buildstream | 23:30 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!