| *** Prince781 has quit IRC | 00:27 | |
| *** Prince781 has joined #buildstream | 00:47 | |
| *** bochecha has quit IRC | 01:14 | |
| *** xjuan has quit IRC | 01:16 | |
| *** Prince781 has quit IRC | 04:09 | |
| *** Prince781 has joined #buildstream | 04:11 | |
| *** Prince781 has quit IRC | 04:12 | |
| *** Prince781 has joined #buildstream | 04:29 | |
| *** catonano has joined #buildstream | 05:09 | |
| *** catonano has quit IRC | 06:08 | |
| *** Prince781 has quit IRC | 06:11 | |
| *** catonano has joined #buildstream | 06:28 | |
| *** catonano has quit IRC | 06:32 | |
| *** catonano has joined #buildstream | 06:32 | |
| *** catonano has quit IRC | 06:36 | |
| *** catonano has joined #buildstream | 06:36 | |
| *** catonano has quit IRC | 06:47 | |
| *** catonano has joined #buildstream | 06:48 | |
| *** Prince781 has joined #buildstream | 07:09 | |
| *** catonano has quit IRC | 07:17 | |
| *** catonano has joined #buildstream | 07:17 | |
| *** catonano has quit IRC | 07:21 | |
| *** catonano has joined #buildstream | 07:21 | |
| *** catonano has quit IRC | 07:27 | |
| *** catonano has joined #buildstream | 07:31 | |
| *** Prince781 has quit IRC | 07:39 | |
| *** tristan has joined #buildstream | 07:52 | |
| *** lankanmon has joined #buildstream | 08:18 | |
| *** catonano has quit IRC | 08:30 | |
| *** catonano has joined #buildstream | 08:30 | |
| *** solid_black has quit IRC | 08:31 | |
| *** solid_black has joined #buildstream | 08:31 | |
| *** solid_black has joined #buildstream | 08:32 | |
| *** catonano has quit IRC | 08:34 | |
| *** catonano has joined #buildstream | 08:34 | |
| *** catonano has quit IRC | 08:56 | |
| *** catonano has joined #buildstream | 08:56 | |
| *** catonano has quit IRC | 09:00 | |
| *** catonano has joined #buildstream | 09:00 | |
| *** catonano has quit IRC | 09:04 | |
| *** catonano has joined #buildstream | 09:04 | |
| *** catonano has quit IRC | 09:10 | |
| *** catonano has joined #buildstream | 09:10 | |
| *** catonano has quit IRC | 09:14 | |
| *** catonano has joined #buildstream | 09:14 | |
| *** bochecha has joined #buildstream | 09:15 | |
| *** toscalix has joined #buildstream | 09:16 | |
| *** toscalix has quit IRC | 09:18 | |
| *** toscalix has joined #buildstream | 09:21 | |
| *** benschubert has joined #buildstream | 09:43 | |
| *** catonano has quit IRC | 09:44 | |
| *** tristan has quit IRC | 09:46 | |
| *** tristan has joined #buildstream | 09:47 | |
| *** catonano has joined #buildstream | 09:51 | |
| *** raoul has joined #buildstream | 09:56 | |
| *** tristan has quit IRC | 10:03 | |
| *** jonathanmaw has joined #buildstream | 10:17 | |
| *** autoprime has joined #buildstream | 10:29 | |
| *** lachlan has joined #buildstream | 10:31 | |
| laurence | tpollard, re this comment - https://gitlab.com/BuildStream/buildstream/merge_requests/786#note_113096093 | 10:31 |
|---|---|---|
| laurence | ncie work - if we can get into the habit of doing this for all tickets it'd be an ideal workflow, so thanks for starting us off | 10:32 |
| tpollard | I didn't know to do it for the mr or the issue, so decided on mr | 10:32 |
| laurence | couold i ask that the associated gitlab issue gets moved inot 'Verify' instead of 'Doing' ? | 10:32 |
| tpollard | Ok | 10:33 |
| laurence | there are no boards for MRs on gitlab, which would be a better method. | 10:33 |
| tpollard | although I thought verify was post merge? | 10:33 |
| Kinnison | Yes, verify is once the MR is merged | 10:34 |
| * tpollard will hold off on that then | 10:43 | |
| laurence | ah, good point, so it is... | 10:45 |
| * laurence gets more coffee | 10:45 | |
| laurence | and considers adding to the policy adjustments MR | 10:45 |
| toscalix | tpollard: we are not pushing any policy related with labels on MR at this point | 10:46 |
| toscalix | the main goal for pushing several scales of labels is to have a clear WIP view that help to us understand what is being done or what we should pay attentio to | 10:46 |
| toscalix | since there are no boards for MR, the need for pushing labels is reduced | 10:47 |
| toscalix | so we haven't. It is better to improve the current state of issues so at least we have a clear WIP view there than have issues and MR in a nonoptimal state | 10:47 |
| toscalix | the win on using labels is when everybody use them | 10:48 |
| toscalix | the more labels... the bigger chance to have them used only partially | 10:49 |
| adds68 | The meaning of labels is not documented anywhere, so nobody will ever use them | 10:50 |
| adds68 | Unless the label itself is clear, like "Documentation" for EG | 10:51 |
| laurence | most of them are documented, there is a small note next to them to explain, when you see the list of all labels (which the policy points you too) | 10:55 |
| laurence | to* | 10:55 |
| adds68 | laurence, to get to that list though, you have to scroll down to the bottom of the issue template > "view contribution guidelines" > "visit policies guide" > then visit "Global milestones on Gitlab: https://gitlab.com/groups/BuildStream/-/milestones" | 11:05 |
| adds68 | sorry * Check the global scale labels: https://gitlab.com/groups/BuildStream/-/labels | 11:06 |
| jmac | https://gitlab.com/BuildStream/buildstream/labels lists them both (group and project labels) | 11:12 |
| toscalix | adds68: they are https://gitlab.com/BuildStream/nosoftware/alignment/blob/master/proposal_to_apply_Gitlab_features_to_BuildStream.md | 11:16 |
| toscalix | it is complete , but they are | 11:17 |
| toscalix | that was the original proposal, and the policy is here: https://gitlab.com/cip-project/cip-testing/imported-kernel-tests | 11:18 |
| toscalix | unfinished, but is there | 11:19 |
| *** jonathanmaw has quit IRC | 11:20 | |
| adds68 | toscalix, could you please review: https://gitlab.com/BuildStream/nosoftware/alignment/merge_requests/8 | 11:24 |
| adds68 | jmac thanks for that link, that seems the best place to go :) | 11:24 |
| *** jonathanmaw has joined #buildstream | 11:25 | |
| *** mohan43u has quit IRC | 11:26 | |
| jmac | I think that's where most people will find the label descriptions - I agree with Sander that the nosoftware repo is a bit obscure at the moment | 11:27 |
| *** mohan43u has joined #buildstream | 11:27 | |
| *** tristan has joined #buildstream | 11:29 | |
| *** mohan43u has quit IRC | 11:38 | |
| *** mohan43u has joined #buildstream | 11:39 | |
| toscalix | adds68: thanks for the commit, I appreciate it | 11:40 |
| adds68 | toscalix, no problem, just trying to help where i can :) | 11:41 |
| *** lachlan has quit IRC | 11:44 | |
| toscalix | adds68: I added before the wrong link, sorry. This is the policy: https://gitlab.com/BuildStream/nosoftware/alignment/blob/master/BuildStream_policies.md | 11:45 |
| *** mohan43u has quit IRC | 11:46 | |
| *** mohan43u has joined #buildstream | 11:46 | |
| *** lachlan has joined #buildstream | 11:47 | |
| *** ctolentino has joined #buildstream | 11:48 | |
| *** tristan has quit IRC | 12:21 | |
| *** tristan has joined #buildstream | 12:25 | |
| *** bochecha has quit IRC | 12:35 | |
| *** ctolentino has quit IRC | 12:56 | |
| jonathanmaw | tristan: I've finished making changes in response to your review of https://gitlab.com/BuildStream/buildstream/merge_requests/869 | 12:58 |
| *** tiagogomes has quit IRC | 13:03 | |
| *** tiagogomes has joined #buildstream | 13:04 | |
| *** alatiera_ has joined #buildstream | 13:07 | |
| alatiera | regarding this mail, https://mail.gnome.org/archives/buildstream-list/2018-October/msg00058.html | 13:09 |
| alatiera | > 3.) Additional dependencies for testing an element should not be required while building it. | 13:09 |
| alatiera | Would the above mean that I won't be able to declare a dep on asan.bst module while building my tests? | 13:09 |
| *** ChanServ sets mode: +o tristan | 13:10 | |
| tristan | Ahh, looks like I duplicated that in my added (6) :) | 13:10 |
| *** chill has joined #buildstream | 13:10 | |
| * alatiera looks | 13:10 | |
| tristan | alatiera, I'm not sure what asan.bst is, but I think the point is that you would indeed | 13:11 |
| tristan | alatiera, I.e. let's say that asan.bst provides something that you only need for testing your module, something like xvfb is a decent example (a fake X server required for running tests for a lot of GUI apps) | 13:11 |
| alatiera | I brought it up as an example, its an adress sanitizer | 13:11 |
| tristan | alatiera, the point is that you don't implicitly bring it in to the module's dependencies that you are building, just because you need it to test; because you test in a separate element | 13:12 |
| alatiera | tristan: but will I be able to include it for the tests with the proposal above or would the tests use the build deps? | 13:13 |
| tristan | jonathanmaw, Interesting... so what I was not sure about in my comment is why cli.py remains unchanged - looking at your test case it looks like this implicitly changes the behavior of `bst shell --build` | 13:13 |
| tristan | alatiera, it's easier to consider the whole thing without the "conditions" feature | 13:14 |
| tristan | alatiera, Just imagine you have a separate element which depends on the build, and it's purpose is to test the build | 13:14 |
| tristan | alatiera, it can depend on other things as well | 13:14 |
| alatiera | tristan: oh I think I understand it now, thanks! | 13:15 |
| tristan | alatiera, The "conditions" feature is mostly there to express a dependency graph which does this with less verbosity in your pipeline declaration :) | 13:15 |
| alatiera | fwiw, I was experiementing with using bst as the CI environment for some GNOME libraries the other day | 13:16 |
| alatiera | the idea was to reuse the gnome-build-meta module set, but we did not want to have every element depend on valgrind for example just for the tests | 13:17 |
| tristan | jonathanmaw, I'll update the issue, sec... | 13:17 |
| tristan | alatiera, right, that is precisely the kind of thing this is about :) | 13:17 |
| alatiera | hence my intrest in it | 13:18 |
| alatiera | overall bst has been really nice for me so far, amazing work! | 13:18 |
| tristan | alatiera, it's still only one half of the recipe to make things very nice though | 13:18 |
| tristan | For instance, right now every element needs to be declared in a separate bst file | 13:19 |
| tristan | And this means that we require a lot of files to express a more complicated pipeline, an enhancement to that would be to focus on ways to make it more convenient to express the testing element in the same bst file as the build element | 13:20 |
| *** jonathanmaw_ has joined #buildstream | 13:20 | |
| *** jonathanmaw has quit IRC | 13:21 | |
| benschubert | Hey, could I get some feedback on https://gitlab.com/BuildStream/buildstream/issues/736 ? I'm slightly worried that we might be leaking too many things between our tests :/ | 13:45 |
| toscalix | benschubert: we have scheduled on thursday a meeting to go over the backlog of tickets opened the last 3 weeks, where many of us has been at events. We will go over this one | 13:52 |
| *** jonathanmaw_ is now known as jonathanmaw | 13:57 | |
| jonathanmaw | tristan: also, I've picked up and fixed https://gitlab.com/BuildStream/buildstream/merge_requests/743 again, iirc there was some interest in this during the gathering | 13:58 |
| tristan | jonathanmaw, Ah right ! | 13:59 |
| benschubert | toscalix: thanks! | 14:24 |
| adds68 | Can you print to the console from inside a plugin? | 14:36 |
| adds68 | ( for debugging purposes ) | 14:36 |
| skullman | self.status() etc. | 14:36 |
| skullman | self.debug and run with --debug would be more appropriate for debugging | 14:36 |
| adds68 | skullman so just calling that from inside the plugin class will allow me to print? | 14:38 |
| skullman | yep, sends messages to the parent process to log to console if it's in a subprocess etc. | 14:39 |
| adds68 | skullman oh nice thanks :) | 14:40 |
| tiagogomes | hmm, options of type string are not supported? | 14:43 |
| tristan | tiagogomes, that is intentional, to strictly limit the possible outputs of a given project | 14:46 |
| tristan | The same is true of integers | 14:46 |
| tiagogomes | Right | 14:56 |
| *** Prince781 has joined #buildstream | 14:57 | |
| *** catonano has quit IRC | 14:59 | |
| jonathanmaw | adds68: last time I tried it, though, I found that it doesn't work from within Plugin.configure. Dunno if it's changed since, that was a *long* while ago. | 14:59 |
| *** catonano has joined #buildstream | 15:01 | |
| *** Prince781 has quit IRC | 15:06 | |
| *** tristan has quit IRC | 15:06 | |
| tiagogomes | I am using the include directive to include a file from a junctioned project. I am always getting a 'var' undefined to matter the places where I define that 'var' in the subproject | 15:18 |
| *** lachlan has quit IRC | 15:19 | |
| *** lachlan has joined #buildstream | 15:21 | |
| *** bochecha has joined #buildstream | 15:46 | |
| adds68 | jonathanmaw, yes it doesn't seem to work for me either =/ | 15:52 |
| *** brool has joined #buildstream | 15:56 | |
| lachlan | Hi, I'm trying to use a runner for the benchmarking page and I'm getting "ERROR: Preparation failed: Error response from daemon: Get https://registry-1.docker.io/v2/buildstream/buildstream-fedora/manifests/latest: received unexpected HTTP status: 503 Service Unavailable" | 15:57 |
| lachlan | Is anybody else experiencing a similar problem? | 15:57 |
| *** chiaratolentino has joined #buildstream | 15:58 | |
| *** ctolentino has joined #buildstream | 16:03 | |
| *** ctolentino has quit IRC | 16:04 | |
| *** ctolentino has joined #buildstream | 16:08 | |
| tiagogomes | Should be working now | 16:08 |
| *** ctolentino has quit IRC | 16:10 | |
| adds68 | skullman seems self.debug and self.status do not print anything when building | 16:14 |
| adds68 | skullman, is there a specific place i should put them? I am extending the Element class | 16:14 |
| skullman | you'd know as much as me there, try jonathanmaw | 16:15 |
| adds68 | skullman, thanks jonathanmaw ^ :D | 16:15 |
| *** tristan has joined #buildstream | 16:17 | |
| jonathanmaw | adds68: odd, not sure what causes that | 16:17 |
| WSalmon | lachlan, did you fix it? i had a ci funny earlier, i clicked try again and all was fine the second go... | 16:17 |
| jonathanmaw | adds68: it seems status messages are dropped if buildstream is not being run in verbose mode | 16:18 |
| cs-shadow | lachlan: WSalmon: Docker Hub (and the CDNs in front of it) have been known to cause such failures from time to time. Not sure if there is an easy fix | 16:18 |
| adds68 | jonathanmaw ah ok, so i need to pass --verbose as well as --debug ? | 16:18 |
| *** solid_black has quit IRC | 16:18 | |
| *** ctolentino has joined #buildstream | 16:19 | |
| jonathanmaw | adds68: I think it should be verbose for status | 16:19 |
| jonathanmaw | not sure of the exact behaviour for debug | 16:19 |
| *** chiaratolentino has quit IRC | 16:21 | |
| jonathanmaw | yep, looks like you should just need --debug to get debug messages. | 16:22 |
| WSalmon | jonathanmaw, do you need it before the command tho? bst [bst options] command [comand options] ? eg, bst --debug workspace open? | 16:23 |
| adds68 | WSalmon yea that is how i am running it atm | 16:24 |
| jonathanmaw | yeah, bst --verbose --debug workspace open, I think | 16:24 |
| adds68 | jonathanmaw, this is where i have put self.debug("test message") | 16:24 |
| adds68 | https://gitlab.com/freedesktop-sdk/freedesktop-sdk/blob/valentindavid/experimental-manifest/bootstrap/plugins/collect-manifest.py#L32 | 16:24 |
| jonathanmaw | they don't work in configure, afaict, even with special command-line stuff | 16:25 |
| jonathanmaw | When I was fiddling around trying to get stuff to work, I ended up using 'print' | 16:25 |
| jonathanmaw | though I never tried it, maybe self._get_context().message("test message") would work, but I haven't tried it. | 16:26 |
| WSalmon | i have had issues with even prints not getting out, the only sure fire way i have found to get stuff to stdout is to raise a exception however that is o"bviously suboptimal" | 16:26 |
| WSalmon | * "obviously suboptimal" | 16:27 |
| WSalmon | lol | 16:27 |
| adds68 | jonathanmaw i shall try that | 16:28 |
| adds68 | WSalmon, yes print statements don't work for me either =/ | 16:28 |
| adds68 | I'm not sure how people are expected to debug =/ | 16:29 |
| *** tristan has quit IRC | 16:30 | |
| *** tristan has joined #buildstream | 16:30 | |
| *** zack12 has joined #buildstream | 16:31 | |
| coldtom | when i was trying to get stuff to print i wrote to a logfile :P | 16:31 |
| ctolentino | Hello! Could I have developer permissions for the BuildStream project please (username: chiaratolentino)? Thanks! | 16:35 |
| ctolentino | Btw, the links to the IRC logs doesn't seem to be working | 16:35 |
| adds68 | coldtom, eugh serious, that was like the one thing i didn't want to do :P | 16:39 |
| adds68 | I get debug messages from every single element, but for some reason nothing from the plugin 0_0 | 16:40 |
| jonathanmaw | hrm, I'm looking at buildstream/_frontend/cli.py and wondering whether I could generalise out ensure_project_dir (since I would like to be able to search for files other than project.conf in the same way) | 16:50 |
| juergbi | hi ctolentino, done | 16:51 |
| ctolentino | Thanks! | 16:51 |
| jonathanmaw | AUIU that's there for bash completions, so I don't know whether importing functions from utils would be permitted | 16:51 |
| jonathanmaw | or whether importing specific functions from utils would make that better (but is proscribed by CONTRIBUTING.bst) | 16:53 |
| *** Prince781 has joined #buildstream | 17:02 | |
| *** catonano has quit IRC | 17:03 | |
| *** catonano has joined #buildstream | 17:08 | |
| adds68 | toscalix, https://gitlab.com/BuildStream/buildstream/merge_requests/913 This is a second MR for the issue template change, could you please review | 17:11 |
| *** MindlessDrone has joined #buildstream | 17:14 | |
| *** rdale has quit IRC | 17:15 | |
| toscalix | adds68: added comments | 17:15 |
| *** rdale has joined #buildstream | 17:17 | |
| *** abderrahim has joined #buildstream | 17:22 | |
| *** t_acko has joined #buildstream | 17:22 | |
| adds68 | toscalix, done | 17:23 |
| *** abderrahim4 has quit IRC | 17:24 | |
| *** xjuan has joined #buildstream | 17:28 | |
| *** Prince781 has quit IRC | 17:28 | |
| *** brlogger has joined #buildstream | 17:30 | |
| *** Prince781 has joined #buildstream | 17:30 | |
| *** tiagogomes has quit IRC | 17:47 | |
| *** toscalix has quit IRC | 17:53 | |
| *** slaf has quit IRC | 18:03 | |
| *** jonathanmaw has quit IRC | 18:08 | |
| *** raoul has quit IRC | 18:09 | |
| *** slaf has joined #buildstream | 18:19 | |
| *** Prince781 has quit IRC | 18:25 | |
| *** lachlan has quit IRC | 18:34 | |
| *** catonano has quit IRC | 18:39 | |
| *** catonano has joined #buildstream | 18:42 | |
| *** Prince781 has joined #buildstream | 19:34 | |
| *** catonano has quit IRC | 20:02 | |
| *** Prince781 has quit IRC | 21:04 | |
| *** Prince781 has joined #buildstream | 21:06 | |
| *** sircmpwn has joined #buildstream | 21:35 | |
| *** SirPNut has joined #buildstream | 22:19 | |
| *** tristan has quit IRC | 22:31 | |
| *** catonano has joined #buildstream | 22:46 | |
| *** Prince781 has quit IRC | 22:55 | |
| *** alatiera_ has quit IRC | 22:58 | |
| *** Prince781 has joined #buildstream | 23:01 | |
| *** slaf has quit IRC | 23:31 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!