| *** tristan has quit IRC | 05:30 | |
| *** tristan has joined #buildstream | 05:55 | |
| *** tristan has quit IRC | 06:37 | |
| *** tristan has joined #buildstream | 06:38 | |
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 2 commits (last: testutils: Added optional subdir parameter to repo creation) https://gitlab.com/BuildStream/buildstream/commit/8584c2df94c13be42bcdef0b6687208e11ab0e70 | 07:33 |
|---|---|---|
| *** bochecha has joined #buildstream | 07:37 | |
| bochecha | I've got an element with `kind: autotools`, and builstream seems to insist on wanting to run the autogen step, instead of directly going to the configure script (which exists) | 08:06 |
| tristan | bochecha, the default is to ignore that and attempt to reconfigure from a non-release-tarball state yes | 08:08 |
| *** tiagogomes has quit IRC | 08:08 | |
| bochecha | really? the other autotools element I have just directly runs the configure script :x | 08:08 |
| tristan | http://buildstream.gitlab.io/buildstream/elements/autotools.html#module-elements.autotools | 08:09 |
| bochecha | yeah, I'm looking at that | 08:09 |
| tristan | default is %{autogen} and then %{configure} in configure-commands | 08:09 |
| bochecha | any reason to always redo the autogen step? | 08:10 |
| tristan | Well, this comes from something that has a history I cannot entirely explain | 08:10 |
| tristan | Its possible that it's a good idea to change it | 08:11 |
| bochecha | (flatpak-builder just skips the autogen step if a configure script exists, fwiw) | 08:11 |
| tristan | Mmmmyeahhh not really looking at that as a shining example of greatness but rather an ad-hoc shim, to be honest | 08:11 |
| bochecha | so my other autotools module doesn't have an {autogen,bootstrap}{,.sh} script, and its configure script exists, so the autogen step conditionals just don't do anything | 08:12 |
| tristan | I could speculate if I was not on my way out the door | 08:12 |
| bochecha | it does have a configure.ac and all the rest though, so it would work to just run `autoreconf` on it, but due to the conditionals that doesn't happen | 08:12 |
| bochecha | which makes this thing very inconsistent and surprising... | 08:13 |
| tristan | There is a history of building everything from git, and resorting to normalizing tarballs to git in the off cases where an upstream VCS was unavailable | 08:13 |
| tristan | This might be an artifact of managing downstream patches in an aggregated git repo instead of the (admittedly horrible) practice of managing .patch files against upstreams | 08:13 |
| tristan | bochecha, best is to raise this on the mailing list I think | 08:14 |
| bochecha | alright, I will :) | 08:14 |
| tristan | To be honest I like the unconditional autogen, and would probably prefer to ensure it always tries to run autoregen -ivf in the case that a configure happens to exist | 08:15 |
| tristan | If thats at least reliable | 08:15 |
| tristan | because it ensures the same build commands keep working when you go switching from release tarballs to development VCS branches and then back again | 08:16 |
| tristan | deciding what to do because of what happens to be in your checkout feels more error prone than just hammering it the same way every time | 08:16 |
| bochecha | I don't have an opinion on whether an unconditional autogen is a good or bad thing :) | 08:18 |
| bochecha | but if bst wants to always go through the autogen step, then it should really always happen | 08:18 |
| bochecha | this element entirely skips the autogen steps and starts with the configure one: https://framagit.org/bochecha/org.ideascube.Sdk/blob/devel/elements/python.bst | 08:19 |
| bochecha | without me doing anything special, it just happens this way :) | 08:19 |
| bochecha | hence my surprise | 08:19 |
| *** tristan has quit IRC | 08:20 | |
| *** semanticdesign has joined #buildstream | 09:12 | |
| *** valentind has joined #buildstream | 09:22 | |
| *** semanticdesign has quit IRC | 09:23 | |
| *** bochecha has quit IRC | 10:27 | |
| *** givascu has joined #buildstream | 10:28 | |
| *** bochecha has joined #buildstream | 10:29 | |
| *** bochecha has quit IRC | 11:08 | |
| *** givascu has quit IRC | 13:31 | |
| *** givascu has joined #buildstream | 13:48 | |
| *** tristan has joined #buildstream | 14:56 | |
| *** bochecha has joined #buildstream | 15:04 | |
| valentind | tristan, is there a plan for exporting to ostree directly from buildstream? | 15:10 |
| tristan | valentind, not for the moment no, trying to avoid feature creep for things which can easily be scripted around buildstream | 15:12 |
| tristan | valentind, one thought is maybe we can have an unsafe checkout mode which gives the user hardlinks back into the artifact, i.e. "handle with care" but makes such operations a bit more optimized | 15:13 |
| gitlab-br-bot | push on buildstream@remove-arches (by Tristan Van Berkom): 20 commits (last: _pipeline.py: Adjust to new loader API) https://gitlab.com/BuildStream/buildstream/commit/d970158299f7eeb9c4cc5d1f61d70ccd7c3bcbff | 15:14 |
| gitlab-br-bot | buildstream: merge request (remove-arches->master: WIP: Remove arches) #117 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/117 | 15:14 |
| gitlab-br-bot | push on buildstream@remove-pre-post-commands (by Tristan Van Berkom): 20 commits (last: Adjust commands to multiple targets) https://gitlab.com/BuildStream/buildstream/commit/54f2eb8502851086804a50d67e279cbb58d9a187 | 15:16 |
| gitlab-br-bot | buildstream: merge request (remove-pre-post-commands->master: WIP: Remove pre post commands) #118 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/118 | 15:16 |
| gitlab-br-bot | buildstream: merge request (remove-arches->master: Remove arches) #117 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/117 | 15:17 |
| gitlab-br-bot | buildstream: issue #120 ("Remove 'arches' conditionals") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/120 | 16:25 |
| gitlab-br-bot | buildstream: merge request (remove-arches->master: Remove arches) #117 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/117 | 16:25 |
| gitlab-br-bot | buildstream: Tristan Van Berkom deleted branch remove-arches | 16:25 |
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 7 commits (last: context.py: Removing arches) https://gitlab.com/BuildStream/buildstream/commit/c05eec40cc3471554f5cc2306ac7d62c76056520 | 16:25 |
| gitlab-br-bot | buildstream: merge request (remove-pre-post-commands->master: Remove pre post commands) #118 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/118 | 16:46 |
| gitlab-br-bot | push on buildstream@master (by Tristan Van Berkom): 3 commits (last: buildelement.py: Issue #121 - Remove traces of pre-/post- commands) https://gitlab.com/BuildStream/buildstream/commit/b9b02fcae1bd054faa3a674bd4c9362fe25a4fa8 | 16:46 |
| gitlab-br-bot | buildstream: issue #121 ("Remove pre-/post- command prefix support") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/121 | 16:46 |
| gitlab-br-bot | buildstream: merge request (remove-pre-post-commands->master: Remove pre post commands) #118 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/118 | 16:46 |
| gitlab-br-bot | buildstream: Tristan Van Berkom deleted branch remove-pre-post-commands | 16:46 |
| *** bochecha has quit IRC | 17:26 | |
| *** tristan has quit IRC | 21:19 | |
| *** givascu has quit IRC | 21:22 | |
| *** valentind has quit IRC | 23:13 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!