| *** semanticdesign has quit IRC | 01:29 | |
| *** semanticdesign has joined #buildstream | 01:29 | |
| *** semanticdesign has quit IRC | 01:49 | |
| *** semanticdesign has joined #buildstream | 01:50 | |
| *** semanticdesign has quit IRC | 02:45 | |
| *** semanticdesign has joined #buildstream | 02:46 | |
| *** semanticdesign has joined #buildstream | 02:47 | |
| *** semanticdesign has quit IRC | 03:37 | |
| *** tristan has quit IRC | 05:41 | |
| *** tristan has joined #buildstream | 05:47 | |
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 1 commit (last: Ensure that artifact file permissions are set in the right order) https://gitlab.com/BuildStream/buildstream/commit/a968bab76201856a67c876be25745fee912c136b | 06:58 |
|---|---|---|
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 1 commit (last: element.py: Force remove artifact assembly directory on failure to commit) https://gitlab.com/BuildStream/buildstream/commit/446875be50996de0c1447af43de75d1f7b037219 | 07:15 |
| gitlab-br-bot | buildstream: merge request (81-non-empty-read-only-directories-not-handled-during-bst-build-and-others->master: Ensure that artifact file permissions are set in the right order) #100 changed state ("closed"): https://gitlab.com/BuildStream/buildstream/merge_requests/100 | 07:15 |
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 1 commit (last: _plugincontext.py: Fix third party plugin loading) https://gitlab.com/BuildStream/buildstream/commit/58a1dd4ba9aef43a47d7dd24deee6eb9e9e45b30 | 08:36 |
| gitlab-br-bot | buildstream: merge request (external_plugin_errors->master: _plugincontext.py: Fix third party plugin loading) #107 changed state ("closed"): https://gitlab.com/BuildStream/buildstream/merge_requests/107 | 08:37 |
| gitlab-br-bot | buildstream: issue #110 ("Stacktrace when loading plugin installed in 'develop' mode") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/110 | 08:37 |
| *** jonathanmaw has joined #buildstream | 08:48 | |
| *** bochecha has joined #buildstream | 09:05 | |
| *** ssam2 has joined #buildstream | 09:07 | |
| *** adds68 has quit IRC | 09:22 | |
| *** adds68 has joined #buildstream | 09:22 | |
| ssam2 | tristan, i'm getting confused by the term "canonical push/pull urls" | 09:29 |
| ssam2 | as in https://gitlab.com/BuildStream/buildstream/issues/112 | 09:29 |
| ssam2 | does that just mean, the issue that the ssh:// push protocol requires a corresponding http URL to actually do anything ? | 09:30 |
| tristan | ssam2, it basically means; it would be nice to just give buildstream one url | 09:31 |
| tristan | for one artifact server | 09:31 |
| ssam2 | right; but user authentication makes that more complex | 09:31 |
| tristan | hopefully the scheme also supports anything we'd need, like port number | 09:31 |
| tristan | yes | 09:31 |
| tristan | It doesnt necessarily mean that all users need to authenticate and push, and we could have multiple url schemes | 09:32 |
| ssam2 | ok, that makes sense then | 09:33 |
| tristan | ssam2, so one simpler-that-writing-a-whole-deamon approach might be... supporting pull with an ssh url ? | 09:33 |
| ssam2 | yeah, that seems much easier | 09:33 |
| tristan | So then, people could use http in read-only | 09:33 |
| tristan | But... it doesnt yet consider that people will want to push and pull tarballs | 09:33 |
| ssam2 | indeed. i think that's a tangent from #111 and the multiple artifact caches feature, though | 09:34 |
| tristan | So, whatever is done for that; should hopefully consider that using the same configuration API, we can eventually support interacting with tarball using clients | 09:34 |
| ssam2 | to push and pull tarballs we need to add a daemon with its own protocol that can receive tarballs | 09:34 |
| tristan | I.e. one doesnt need to be implemented for the other | 09:34 |
| ssam2 | yeah | 09:34 |
| tristan | But one needs to consider that the other will have to happen | 09:34 |
| tristan | So clueless-about-networking me... wants to hope that using an ssh login to an artifact server and running a remote script... might allow that remote script to direct the client to another http url ? | 09:37 |
| tristan | I.e. pull-over-ssh might be just like pull-over-http with one extra roundtrip ? | 09:37 |
| ssam2 | hmmmmmm | 09:37 |
| tristan | heh | 09:38 |
| ssam2 | ostree-push does implement a protocol | 09:38 |
| ssam2 | so no reason it couldn't respond with a http URL | 09:38 |
| ssam2 | good idea | 09:38 |
| tristan | ostree-push is something else... look at our fork | 09:38 |
| ssam2 | yeah | 09:38 |
| tristan | ah I misread you | 09:38 |
| tristan | anyway yeah | 09:38 |
| ssam2 | and since there's a protocol, we could extend it to receive tarballs in future if we wanted | 09:39 |
| tristan | right, certainly... | 09:39 |
| tristan | also download of packed up ostree stuff via TarStream might make sense for pull | 09:40 |
| ssam2 | yeah | 09:40 |
| ssam2 | but we presumably don't want that hidden behind ssh, so we'd need to have something running as a cgi script in the http server, or listenening on a different port, to handle the tarball requests | 09:40 |
| ssam2 | still, it's easy enough to write something in python that does such | 09:41 |
| ssam2 | and can run as a standalone http daemon, or gated behind a real one, as per sysadmin preferences | 09:41 |
| *** tlater has joined #buildstream | 09:41 | |
| tristan | nod, if we do, we'll want to use twisted or flask or such | 09:41 |
| tristan | the web stuff baked into the standard library is the farthest thing from secure | 09:42 |
| tristan | also a goal would be to have something you can install and run | 09:42 |
| tristan | instead of the ad-hoc stuff we have right now | 09:42 |
| tristan | well, maybe not run | 09:42 |
| tristan | I dont know... it gets complicated and then you need a daemon ? | 09:42 |
| tristan | anyway, simplifying that setup so that it's *easy* is important | 09:43 |
| ssam2 | flask is nice in this regard | 09:43 |
| ssam2 | as I think most python web frameworks are | 09:43 |
| ssam2 | in that you can run the webapp as a standalone httpd server for testing / local setups | 09:43 |
| ssam2 | and integrate it with apache or whatever using wsgi for production | 09:44 |
| jonathanmaw | tlater: I've had a look, and it seems the .yaml files weren't installed to the right place | 10:33 |
| jonathanmaw | mystery solved | 10:33 |
| tlater | jonathanmaw: Oh, alright. This probably requires some better documentation/errors | 10:33 |
| jonathanmaw | tlater: hmm, Element should be raising errors if the file is missing, anyway. | 10:35 |
| tlater | jonathanmaw: Do all plugins require defaults? | 10:36 |
| tristan | I think so | 10:37 |
| tristan | It might not need to be required, but I think it currently expects the adjacent yaml file | 10:37 |
| tristan | tlater, not only because they need to *set* defaults, but otherwise comments in the YAML file are also the defacto way to document plugins | 10:37 |
| tristan | so you'll have your sphinx stuff loading those too | 10:38 |
| jonathanmaw | ok, looks like the current behaviour is to allow plugins with missing defaults files https://gitlab.com/BuildStream/buildstream/blob/master/buildstream/element.py#L1456 | 10:47 |
| tristan | ssam2, so I noticed the other day... there is one thing that is missing, or rather a convenience we dont have - I'm thinking of leaving it out | 10:51 |
| tristan | ssam2, that is; with the older arches approach, it looks like the arch is automatically fed to variables | 10:51 |
| tristan | as in %{bst-arch}, %{bst-target-arch} | 10:52 |
| ssam2 | oh yeah | 10:52 |
| tristan | With the current implementation, you'd need to define a variable conditionally depending on the arch type project option | 10:52 |
| ssam2 | how would that look ? | 10:52 |
| tristan | which is, annoying; a feature to solve that would be allowing a variable to be declared with the value of an option | 10:52 |
| * tristan writes psuedo yaml | 10:53 | |
| ssam2 | should we not just expose all the options as variables ? | 10:53 |
| tristan | ssam2, like this: https://bpaste.net/show/f69aa3087184 | 10:55 |
| tristan | Ehhhh | 10:55 |
| tristan | ssam2, declaratively, maybe | 10:56 |
| tristan | that's the feature I was thinking | 10:56 |
| tristan | Add some base option to the option definition, like export-variable: True | 10:56 |
| ssam2 | ok, i guess that's sensible to make it optional | 10:56 |
| tristan | namespaces and unwanted surprises and all | 10:57 |
| tristan | ssam2, or rather: export-variable: "variable-name" | 11:00 |
| ssam2 | better yeah | 11:00 |
| gitlab-br-bot | push on buildstream@jonathan/all-elements-need-yaml (by Jonathan Maw): 1 commit (last: Make all element loading require a yaml file) https://gitlab.com/BuildStream/buildstream/commit/58fc5628c4fdb6911b3ea57cdc8e0ad7790dd447 | 11:04 |
| gitlab-br-bot | buildstream: merge request (jonathan/all-elements-need-yaml->master: Make all element loading require a yaml file) #108 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/108 | 11:05 |
| tlater | Before I forget about this again, tristan, what exactly is your idea for https://gitlab.com/BuildStream/buildstream/merge_requests/93 ? | 11:06 |
| tristan | didn't I comment on that yesterday ? | 11:08 |
| tristan | tlater, did you pay attention to the changes I made on your other branch before landing ? | 11:10 |
| tristan | I got one part wrong apparently, fixing comment and code around that | 11:10 |
| tlater | Yes, I just read through all of them | 11:10 |
| tristan | tlater, the reason you had 'try / finally:' in element.py because of tmpfile mechanics trying to remove things unsafely yes ? | 11:10 |
| tlater | This is two weeks ago - but I think it was a permission issue. tmpfile mechanics won't change the permissions and error out | 11:11 |
| tristan | and... why the hell do we use tmpfile at all to create the directory for assembling the artifact anyway ? | 11:11 |
| tristan | https://gitlab.com/BuildStream/buildstream/issues/81#note_43329110 <-- tlater yes that is the reason | 11:11 |
| tristan | juergbi, you did that right ? any rationale for tmpdir of assemblydir ? | 11:12 |
| tlater | Right, yeah, and I'm pretty sure that the try/finally just ensures that the permissions are correct | 11:12 |
| * tlater looks at tristan's changes again | 11:13 | |
| juergbi | tristan: was this because of the introduction of subdirectories? | 11:13 |
| juergbi | i.e., we have to construct the artifact structure, otherwise the high-level ostree commit support won't work? | 11:13 |
| tristan | juergbi, we added assemblydir I believe because yes, logs/ meta/ files/ thing | 11:13 |
| tristan | juergbi, So, the whole sandbox itself is a tmpdir | 11:14 |
| tristan | which is why I wonder | 11:14 |
| tristan | :) | 11:14 |
| tristan | looks like a tmpdir in a tmpdir | 11:14 |
| * tristan thinks, instead of try/finally, just use a hard coded dir name | 11:14 | |
| tristan | and remove the extra tmpdir context manager thing | 11:15 |
| tristan | win + win + win | 11:15 |
| * tlater thinks that makes sense | 11:15 | |
| tlater | :) | 11:15 |
| tlater | tristan: Also, yes, you did comment on !93, but I haven't made any real changes since I put it up originally | 11:16 |
| tlater | It's still floating in "let's discuss this on IRC" | 11:16 |
| tristan | 93 93 93 93.... | 11:16 |
| tlater | The one I linked earlier :) | 11:16 |
| tlater | Warnings for overwriting things | 11:17 |
| tristan | tlater, fix the comments I made related to the sources and we're good to go | 11:17 |
| tlater | You're fine with just warnings there then? | 11:17 |
| tlater | Ah, no, you did comment about errors | 11:18 |
| tlater | Alright, ta | 11:18 |
| tristan | looks like that's working | 11:24 |
| gitlab-br-bot | push on buildstream@fix-artifact-cleanup (by Tristan Van Berkom): 1 commit (last: element.py: Fix cleanup after collection of artifact with read-only directories) https://gitlab.com/BuildStream/buildstream/commit/b1552ea657a0d9da7ee8733334a02a2f92dc7f97 | 11:28 |
| gitlab-br-bot | buildstream: merge request (fix-artifact-cleanup->master: element.py: Fix cleanup after collection of artifact with read-only directories) #109 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/109 | 11:29 |
| tristan | So I also move the artifacts dir from the sandboxroot, to the rootdir which the sandboxroot is inside of | 11:34 |
| tristan | not sure why it was nested that far, but CI looks to be working | 11:34 |
| gitlab-br-bot | buildstream: merge request (fix-artifact-cleanup->master: element.py: Fix cleanup after collection of artifact with read-only directories) #109 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/109 | 11:45 |
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 1 commit (last: element.py: Fix cleanup after collection of artifact with read-only directories) https://gitlab.com/BuildStream/buildstream/commit/b1552ea657a0d9da7ee8733334a02a2f92dc7f97 | 11:45 |
| gitlab-br-bot | buildstream: Tristan Van Berkom deleted branch fix-artifact-cleanup | 11:45 |
| gitlab-br-bot | buildstream: issue #81 ("Non-empty read-only directories not handled during 'bst build' and others") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/81 | 11:46 |
| *** tristan has quit IRC | 11:51 | |
| *** tristan has joined #buildstream | 12:46 | |
| gitlab-br-bot | push on buildstream@sam/no-install-python-deps (by Sam Thursfield): 1 commit (last: .gitlab-ci.yml: Use `pip3 install --no-index` when installing BuildStream) https://gitlab.com/BuildStream/buildstream/commit/a4a323857a2958c4c24dd625ee289bdb2a293e30 | 12:52 |
| gitlab-br-bot | buildstream: merge request (sam/no-install-python-deps->master: .gitlab-ci.yml: Use `pip3 install --no-index` when installing BuildStream) #110 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/110 | 12:53 |
| jonathanmaw | tristan: looking at all the test failures as a result of making it mandatory to have an accompanying .yaml file with every element, it seems the "stack" element has been happily existing without a .yaml | 12:56 |
| jonathanmaw | given that stacks haven't been hugely undocumented as a result, I'm inclined to rethink my position | 13:02 |
| jonathanmaw | so I'll leave my changes to make them mandatory alone for now. | 13:03 |
| jjardon[m] | hi, for the documentation at https://buildstream.gitlab.io/buildstream/install.html#installing; is there a git repo somewhere to send a improvement/fix? | 13:03 |
| jjardon[m] | ah, I think I found it: https://gitlab.com/BuildStream/buildstream/blob/master/doc/source/install.rst | 13:05 |
| gitlab-br-bot | push on buildstream@jjardon/arch (by Javier Jardón): 1 commit (last: doc/source/install.rst: Add instructions for Arch systems) https://gitlab.com/BuildStream/buildstream/commit/3bf6c9d027efb8c8f1074041d52570e7212a598a | 13:21 |
| gitlab-br-bot | buildstream: merge request (jjardon/arch->master: doc/source/install.rst: Add instructions for Arch systems) #111 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/111 | 13:21 |
| gitlab-br-bot | buildstream: merge request (jjardon/arch->master: doc/source/install.rst: Add instructions for Arch systems) #111 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/111 | 13:22 |
| gitlab-br-bot | push on buildstream@jjardon/arch (by Javier Jardón): 20 commits (last: tests/testutils/runcli.py: Provide the last raised exception in the result) https://gitlab.com/BuildStream/buildstream/commit/c986ab45d9325dbe5dd242291e40d157f238eea9 | 13:22 |
| gitlab-br-bot | buildstream: merge request (jjardon/arch->master: doc/source/install.rst: Add instructions for Arch systems) #111 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/111 | 13:22 |
| tristan | jjardon[m], looks good, you sure that is python3 right ? | 13:49 |
| tristan | and you dont just happen to have it installed anyway ? | 13:49 |
| jjardon[m] | tristan: yup, in Arch python = python3 | 13:51 |
| tristan | jjardon[m], set to merge anyway :) | 13:52 |
| tristan | jjardon[m], I think for the GNOME wiki page we can link back to our docs as much as possible, so that's good to have another distro, fedora instructions would be nice too | 13:53 |
| tristan | I'll try to cook up that wiki page if I can... | 13:53 |
| tristan | seems might be a good thing just to get the rest of the release-team to actually try it out | 13:54 |
| tristan | and move on from there | 13:54 |
| jjardon[m] | tristan: yes, the link is already there, see the install section at https://wiki.gnome.org/jjardon/BuildSystemComponentBuildstream#Setting_up_BuildStream | 13:55 |
| tristan | Sweet | 13:55 |
| jjardon[m] | tristan: can you remind me where the jhbuild conversions are? | 13:56 |
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 1 commit (last: doc/source/install.rst: Add instructions for Arch systems) https://gitlab.com/BuildStream/buildstream/commit/96af5cdb25fa0272957e297a7585962f824c97a7 | 14:01 |
| gitlab-br-bot | buildstream: merge request (jjardon/arch->master: doc/source/install.rst: Add instructions for Arch systems) #111 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/111 | 14:01 |
| gitlab-br-bot | buildstream: Javier Jardón deleted branch jjardon/arch | 14:01 |
| gitlab-br-bot | push on buildstream@74-prevent-artifacts-from-containing-files-in-buildstream-build (by Tristan Maat): 6 commits (last: exceptions.py: Added USER_ASSERTION load error) https://gitlab.com/BuildStream/buildstream/commit/bf679177ffd3041ad72600a87086bee19807753c | 14:07 |
| gitlab-br-bot | buildstream: merge request (74-prevent-artifacts-from-containing-files-in-buildstream-build->master: Add warnings when staging to /buildstream/build) #93 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/93 | 14:07 |
| jjardon[m] | tristan: thanks! how the updated docs reach https://buildstream.gitlab.io/buildstream/install.html ? | 14:18 |
| jonathanmaw | found the cause of my missing .yaml files, I was missing include_package_data in setup.py | 14:36 |
| gitlab-br-bot | push on buildstream@74-prevent-artifacts-from-containing-files-in-buildstream-build (by Tristan Maat): 1 commit (last: Add warnings when staging to /buildstream/build) https://gitlab.com/BuildStream/buildstream/commit/5992fcf93bff0ca5ec5177482c35d4b80a5b289e | 14:37 |
| tlater | jonathanmaw: Yes, that's an important one | 14:37 |
| * tlater checks buildstream documentation on plugins | 14:37 | |
| gitlab-br-bot | buildstream: merge request (74-prevent-artifacts-from-containing-files-in-buildstream-build->master: Add warnings when staging to /buildstream/build) #93 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/93 | 14:37 |
| jonathanmaw | tlater: it's theren | 14:37 |
| jonathanmaw | though it wasn't in the setup.py file that your repo had | 14:38 |
| jonathanmaw | hence my confusion | 14:38 |
| jonathanmaw | all sorted now | 14:38 |
| gitlab-br-bot | buildstream: merge request (102-run-ci-as-non-root-user->master: WIP: Resolve "Run CI as non-root user") #104 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/104 | 14:55 |
| gitlab-br-bot | push on buildstream@102-run-ci-as-non-root-user (by Tristan Maat): 20 commits (last: _frontend/main.py: Added -o/--option main CLI params) https://gitlab.com/BuildStream/buildstream/commit/9dbeffa291f3eb3a951ae785f7b42b211f708470 | 14:55 |
| gitlab-br-bot | push on buildstream@non-sandbox-builds (by Tristan Maat): 7 commits (last: Ensure that artifact file permissions are set in the right order) https://gitlab.com/BuildStream/buildstream/commit/a968bab76201856a67c876be25745fee912c136b | 14:59 |
| gitlab-br-bot | buildstream: merge request (non-sandbox-builds->master: Add %{script} format to `buildstream show`) #102 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/102 | 14:59 |
| jjardon[m] | hi, is it not possible to build several elements with the same command line? | 15:00 |
| jjardon[m] | something like | 15:00 |
| jjardon[m] | bst build --track core/adwaita-icon-theme.bst core/dconf.bst core/glib-networking.bst core/gvfs.bst core/libcanberra.bst | 15:00 |
| jjardon[m] | currently bst errors with "Error: Got unexpected extra arguments (core/dconf.bst core/glib-networking.bst core/gvfs.bst core/libcanberra.bst)" | 15:00 |
| *** bochecha has quit IRC | 15:03 | |
| gitlab-br-bot | buildstream: issue #115 ("Failure when trying to compile core/adwaita-icon-theme.bst from https://gnome7.codethink.co.uk/gnome-modulesets.git") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/115 | 15:14 |
| ssam2 | jjardon[m], not possible at the moment i think | 15:22 |
| ssam2 | jjardon[m], you could create a stack element that includes all of them... | 15:22 |
| jjardon[m] | ssam2: yep, thanks | 15:23 |
| *** givascu has joined #buildstream | 15:30 | |
| *** givascu has quit IRC | 16:01 | |
| *** tiagogomes has quit IRC | 16:20 | |
| *** jonathanmaw has quit IRC | 16:35 | |
| *** ssam2 has quit IRC | 16:57 | |
| *** tlater has quit IRC | 17:01 | |
| *** jude has quit IRC | 17:07 | |
| *** jude has joined #buildstream | 18:34 | |
| *** jude has quit IRC | 20:11 | |
| *** jude has joined #buildstream | 20:22 | |
| *** jude has quit IRC | 20:28 | |
| *** jude has joined #buildstream | 22:21 | |
| *** jude has quit IRC | 22:52 | |
| *** jude has joined #buildstream | 23:08 | |
| *** jude has quit IRC | 23:12 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!