| *** traveltissues has quit IRC | 07:11 | |
| *** traveltissues has joined #buildstream | 08:40 | |
| *** bochecha has joined #buildstream | 08:57 | |
| gitlab-br-bot | coldtom opened issue #1258 (Plugin tests accessing private API) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1258 | 09:38 |
|---|---|---|
| gitlab-br-bot | coldtom opened MR !1806 (coldtom/testing-api->master: Stop plugin tests using private API) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1806 | 09:41 |
| *** tiagogomes has joined #buildstream | 09:48 | |
| *** tme5 has joined #buildstream | 09:50 | |
| *** tpollard has joined #buildstream | 09:56 | |
| coldtom | ack mypy isn't happy about that because i'm doing imports from a cython file | 09:57 |
| benschubert | doa #type: ignore on the line fo rthis | 09:58 |
| benschubert | We should provide typings for those files, especially since we are opening them up now | 09:58 |
| benschubert | but it's quite complex to write correctly :/ | 09:59 |
| coldtom | ty benschubert | 10:00 |
| *** phildawson has joined #buildstream | 10:11 | |
| *** jonathanmaw has joined #buildstream | 10:21 | |
| tme5 | does anyone have any ideas on how to test what a git source actually downloads? | 10:55 |
| tme5 | since the repo gets deleted when it's staged i can't look at that | 10:56 |
| benschubert | > since the repo gets deleted when it's staged i can't look at that | 10:57 |
| benschubert | All sources keep a local copy of the source. So in that case, you keep the .git. | 10:57 |
| benschubert | Opening a workspace would also ensure you get the .git folder. | 10:57 |
| benschubert | But ultimately, what is it exactly that you are interested in? Ensuring that a particular file/directory exists? you can 'source checkout' and assert on that for example | 10:57 |
| tme5 | i'm writing a test for the feature of fetching with depth=1 when an exact tag is given | 11:00 |
| tme5 | so i think i need some way of testing the repo created by the fetch | 11:00 |
| tme5 | a workspace is a good idea | 11:01 |
| benschubert | jjardon: around the tar removal, which breaks many tests. I was wondering if we should go for an approach where we tag tests requiring external plugins (@pytest.mark.plugins) and then run them at the same time as the source test plugins? That would allow us to have more complete and easy tests setup (especially if tar/git move away, that's a lot of tests that rely on them) | 11:11 |
| coldtom | benschubert: would this approach also allow at least some test of an external element in the bst test suite too? that seems to be something missing at the moment | 11:13 |
| benschubert | coldtom: depends where the test lives: if it's in buildstream/tests, yes. if it's in the plugin repo no (but that can be fixed by triggering other pipelines through our tests, which is a separate thread I want to start) | 11:14 |
| coldtom | of the two approaches, i definitely prefer bst triggering pipelines on bst-plugins-* | 11:15 |
| benschubert | Yep, I however think we might still need the first one too (e.g: the example tests) | 11:16 |
| benschubert | so I think we might require both | 11:16 |
| coldtom | ah, yeah, it seems difficult to make useful examples without needing plugins that will soon be external | 11:17 |
| coldtom | could someone take a look at https://gitlab.com/BuildStream/bst-plugins-experimental/merge_requests/61 and https://gitlab.com/BuildStream/bst-plugins-experimental/merge_requests/72 if they have time please? | 11:36 |
| coldtom | i've also opened https://gitlab.com/BuildStream/bst-plugins-experimental/merge_requests/73, which i'd appreciate opinions on | 11:51 |
| benschubert | I like that :D | 11:53 |
| *** santi has joined #buildstream | 11:55 | |
| *** santi has quit IRC | 11:56 | |
| *** santi has joined #buildstream | 11:57 | |
| benschubert | coldtom: actually we should also probably remove the pylint parts around code formatting just in case :) | 12:02 |
| coldtom | benschubert: good point, i just removed pycodestyle due to conflicts , i can only assume pylint formatting will be a similar issue | 12:04 |
| benschubert | yep. I believe we had a commit that did that in BuildStream itself | 12:04 |
| *** santi has quit IRC | 12:51 | |
| *** santi has joined #buildstream | 12:53 | |
| *** pointswaves has joined #buildstream | 12:53 | |
| *** phildawson has quit IRC | 12:59 | |
| *** phildawson has joined #buildstream | 12:59 | |
| *** pointswaves has quit IRC | 13:02 | |
| *** narispo has quit IRC | 13:31 | |
| *** narispo has joined #buildstream | 13:31 | |
| *** traveltissues has quit IRC | 13:40 | |
| *** phildawson has quit IRC | 13:47 | |
| *** phildawson has joined #buildstream | 13:48 | |
| *** phildawson has quit IRC | 14:01 | |
| *** lachlan has joined #buildstream | 14:03 | |
| *** lachlan has quit IRC | 14:12 | |
| *** narispo has quit IRC | 14:23 | |
| *** narispo has joined #buildstream | 14:23 | |
| *** lachlan has joined #buildstream | 14:24 | |
| *** lachlan has quit IRC | 14:33 | |
| *** lachlan has joined #buildstream | 14:35 | |
| *** lachlan has quit IRC | 14:39 | |
| *** lachlan has joined #buildstream | 14:48 | |
| *** lachlan has quit IRC | 14:51 | |
| *** lachlan has joined #buildstream | 14:57 | |
| *** phildawson has joined #buildstream | 15:06 | |
| coldtom | just ooi, is there any idea of when bst 2 will be released as stable at all? | 15:07 |
| *** lachlan has quit IRC | 15:09 | |
| *** pointswaves has joined #buildstream | 15:23 | |
| juergbi | coldtom: there is no timeline yet. however, Chandan posted a tentative list of 2.0 blockers to the ML a few months back | 15:25 |
| bochecha | was it decided what the plugin story will be like for 2.0? (there were plans of splitting everything out, making bst capable to install the plugins required by a project, eventually one venv per plugin…) | 15:32 |
| coldtom | bochecha: i _think_ it's going to be a few domain-related repositories, although what exactly this means is TBD | 15:33 |
| coldtom | i.e. presumably something like bst-plugins-container, bst-plugins-scm, bst-plugins-buildsystems or the like | 15:34 |
| bochecha | ok, so it's still not entirely decided | 15:41 |
| juergbi | right, deciding and implementing this is a blocker for 2.0 | 15:42 |
| *** lachlan has joined #buildstream | 15:46 | |
| *** lachlan has quit IRC | 15:52 | |
| *** lachlan has joined #buildstream | 16:00 | |
| *** santi has quit IRC | 16:03 | |
| *** phildawson has quit IRC | 16:10 | |
| *** lachlan has quit IRC | 16:10 | |
| *** phildawson has joined #buildstream | 16:10 | |
| *** lachlan has joined #buildstream | 16:14 | |
| *** pointswaves has quit IRC | 16:17 | |
| *** lachlan has quit IRC | 16:30 | |
| *** phildawson has quit IRC | 16:36 | |
| *** phildawson has joined #buildstream | 16:41 | |
| tme5 | should i denote cache key breaks in the NEWS file? | 16:48 |
| *** bochecha has quit IRC | 16:48 | |
| tpollard | I don't think it's needed, have we done it before? | 16:51 |
| coldtom | on a per-plugin basis i don't think so, but personally i'd like NEWS entries for general cache key breaks | 16:52 |
| * coldtom would like to know if he has to build everything again before updating bst | 16:52 | |
| *** lachlan has joined #buildstream | 17:07 | |
| *** phildawson has quit IRC | 17:07 | |
| *** santi has joined #buildstream | 17:10 | |
| *** lachlan has quit IRC | 17:22 | |
| *** santi has quit IRC | 17:25 | |
| *** santi has joined #buildstream | 17:30 | |
| *** lachlan has joined #buildstream | 17:30 | |
| *** xjuan has joined #buildstream | 17:32 | |
| *** xjuan has joined #buildstream | 17:33 | |
| *** lachlan has quit IRC | 17:34 | |
| *** lachlan has joined #buildstream | 17:37 | |
| *** lachlan has quit IRC | 17:45 | |
| *** tme5 has quit IRC | 17:46 | |
| *** lachlan has joined #buildstream | 17:46 | |
| *** lachlan has quit IRC | 17:49 | |
| *** xjuan has quit IRC | 17:56 | |
| *** xjuan has joined #buildstream | 17:57 | |
| *** lachlan has joined #buildstream | 18:01 | |
| *** lachlan has quit IRC | 18:04 | |
| *** lachlan has joined #buildstream | 18:11 | |
| *** lachlan has quit IRC | 18:26 | |
| *** pointswaves has joined #buildstream | 18:33 | |
| *** santi has quit IRC | 18:34 | |
| *** xjuan has quit IRC | 18:34 | |
| *** santi has joined #buildstream | 18:35 | |
| *** pointswaves has quit IRC | 18:37 | |
| *** pointswaves has joined #buildstream | 18:37 | |
| *** lachlan has joined #buildstream | 18:39 | |
| *** jonathanmaw has quit IRC | 18:40 | |
| *** pointswaves has quit IRC | 18:40 | |
| *** lachlan has quit IRC | 18:44 | |
| *** lachlan has joined #buildstream | 18:47 | |
| *** lachlan has quit IRC | 18:50 | |
| *** pointswaves has joined #buildstream | 19:30 | |
| *** santi has quit IRC | 19:59 | |
| *** santi has joined #buildstream | 19:59 | |
| *** delli3_ has joined #buildstream | 20:01 | |
| *** delli3 has quit IRC | 20:03 | |
| *** dellie__ has joined #buildstream | 20:10 | |
| *** delli3_ has quit IRC | 20:13 | |
| *** santi has quit IRC | 20:32 | |
| *** pointswaves has quit IRC | 21:40 | |
| *** pointswaves has joined #buildstream | 22:40 | |
| *** pointswaves has quit IRC | 23:17 | |
| *** tristan has quit IRC | 23:43 | |
| *** tristan has joined #buildstream | 23:43 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!