| *** nimish has joined #buildstream | 01:31 | |
| *** nimish has quit IRC | 01:45 | |
| *** nimish has joined #buildstream | 01:48 | |
| *** ajibade_ has joined #buildstream | 01:52 | |
| *** ajibade__ has joined #buildstream | 01:56 | |
| *** ajibade_ has quit IRC | 02:18 | |
| *** ajibade__ has quit IRC | 02:18 | |
| *** nimish has quit IRC | 03:08 | |
| *** bochecha has joined #buildstream | 03:41 | |
| *** bochecha has quit IRC | 03:59 | |
| *** bochecha has joined #buildstream | 04:04 | |
| *** mohan43u has joined #buildstream | 04:51 | |
| *** kapil___ has joined #buildstream | 06:25 | |
| juergbi | doras[m]: the root-read-only flag is a feature of the script element that works for actual (automated) builds, but bst shell --build doesn't know about this flag | 06:55 |
|---|---|---|
| juergbi | would probably be good if bst shell handles it the same way but that's not trivial to fix | 06:56 |
| doras[m] | Well, that explains it. A bit counterintuitive, I have to say. I had always imagined "bst shell --build" to be the exact action that happens when a real build runs (minus the shell part), so as to be able to debug build issues and behavior in a reproducible way. | 07:00 |
| juergbi | doras[m]: yes, it typically is very close. the issue is that the plugin API supports the root read-only flag only for Sandbox.run() which is used to actually execute a build command. That method is not used for bst shell as build commands are not executed | 07:05 |
| *** ZOZO has joined #buildstream | 07:44 | |
| *** ZOZO has quit IRC | 08:57 | |
| *** kapil___ has quit IRC | 09:04 | |
| *** WSalmon_ has joined #buildstream | 09:28 | |
| *** jonathanmaw has joined #buildstream | 09:36 | |
| *** raoul has joined #buildstream | 10:11 | |
| *** jonathanmaw_ has joined #buildstream | 10:36 | |
| *** jonathanmaw has quit IRC | 10:37 | |
| *** bochecha has quit IRC | 10:45 | |
| *** cs-shadow has joined #buildstream | 11:42 | |
| *** bochecha has joined #buildstream | 12:21 | |
| *** nimish has joined #buildstream | 13:07 | |
| *** nimish has quit IRC | 13:11 | |
| *** nimish has joined #buildstream | 13:12 | |
| gitlab-br-bot | raoul.hidalgocharman approved MR !1117 (jennis/warn_for_nonexistent_domains->master: Fail when we explictly try to include/exclude non-existent domains in a filter element) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1117 | 13:35 |
| jennis | thanks for the review rafaelff[m] | 13:59 |
| jennis | raoul * | 13:59 |
| *** kapil___ has joined #buildstream | 14:04 | |
| gitlab-br-bot | jennis merged MR !1117 (jennis/warn_for_nonexistent_domains->master: Fail when we explictly try to include/exclude non-existent domains in a filter element) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1117 | 14:47 |
| gitlab-br-bot | LaurenceUrhegyi closed issue #466 (Optimize `bst build` initialization time) on buildstream https://gitlab.com/BuildStream/buildstream/issues/466 | 15:07 |
| *** toscalix has joined #buildstream | 15:56 | |
| *** nimish has quit IRC | 16:10 | |
| *** kapil___ has quit IRC | 16:24 | |
| *** bochecha has quit IRC | 16:45 | |
| *** toscalix has quit IRC | 17:06 | |
| *** ipatch has joined #buildstream | 17:35 | |
| cs-shadow | I am trying to backport all/most of our tox improvements to bst 1.2 so that it can be tested more reliably. Would be good if someone can have a look at !1111 (I love that number!) | 17:56 |
| cs-shadow | More details in the MR | 17:56 |
| gitlab-br-bot | MR !1111: Run tests using tox https://gitlab.com/BuildStream/buildstream/merge_requests/1111 | 17:56 |
| *** jonathanmaw_ has quit IRC | 18:05 | |
| *** dtf has joined #buildstream | 18:15 | |
| *** dtf has joined #buildstream | 18:17 | |
| *** raoul has quit IRC | 18:19 | |
| gitlab-br-bot | jjardon opened issue #894 (Document roadmap / plans for cache server for buildstream 2.0) on buildstream https://gitlab.com/BuildStream/buildstream/issues/894 | 18:19 |
| juergbi | cs-shadow: 1.2 will still use pytest-runner. that shouldn't be an issue, right? | 18:23 |
| juergbi | I'm not sure about some remaining differences in requirements/ between your branch and master | 18:24 |
| juergbi | e.g., your branch has pytest=4.1.1 in dev-requirements.txt while master has ==4.0.2. intentional? | 18:24 |
| juergbi | also pytest, pluggy, pytest-forked, wrapt, pytest-xdist have higher req in your 1.2 branch than in master. is this just somewhat random, would it make sense to align? | 18:26 |
| cs-shadow | juergbi: That's right, `pytest-runner` shouldn't be an issue as it is only involved during "setup.py test" and not "pytest" directly. And all the tox stuff uses `pytest` | 18:27 |
| cs-shadow | the versions are more interesting. We could, in theory, keep them the same as master iff we had exactly the same requirements, which we don't | 18:27 |
| cs-shadow | so, what I ended up doing was updating them to their current latest versions (which is what `setup.py test` would have given us anyway) | 18:28 |
| juergbi | some versions are also older, though, e.g., protobuf, pylint | 18:29 |
| juergbi | so it seems a bit odd looking at the numbers, but it's probably not a real issue | 18:29 |
| cs-shadow | the ones that are older, are usually because the 1.2 branch had them pinned. For example, we have `pylint == 2.1.1` in the requirements.in (which was copied from existing setup.py) | 18:30 |
| juergbi | that makes sense | 18:30 |
| juergbi | as long as these version differences don't cause any pain in CI, I don't mind, of course | 18:31 |
| juergbi | just wanted to check that this is not an issue | 18:31 |
| cs-shadow | I definitely agree with your motivation. Not having pain the CI is more than half the reason why I'm doing it :) | 18:32 |
| juergbi | I haven't been following the details of the toxification on master, so no expert in this area. but the MR looks good to me | 18:33 |
| juergbi | anyway no risk on the code side as no actual code was touched | 18:33 |
| cs-shadow | juergbi: thanks very much! | 18:34 |
| gitlab-br-bot | cs-shadow merged MR !1111 (chandan/toxify-1.2->bst-1.2: Run tests using tox) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1111 | 19:54 |
| *** cs-shadow has quit IRC | 22:07 | |
| *** phildawson has quit IRC | 22:31 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!