IRC logs for #buildstream for Wednesday, 2019-03-27

« Tuesday, 2019-03-26 Index Thursday, 2019-03-28 »
*** xjuan has quit IRC00:05
*** gokcennurlu has quit IRC00:09
*** gokcennurlu_ has quit IRC00:09
*** nimish2711 has joined #buildstream00:53
*** nimish2711 has quit IRC01:03
*** nimish2711 has joined #buildstream01:09
*** alatiera has quit IRC01:12
*** nimish2711 has quit IRC01:59
*** nimish2711 has joined #buildstream02:05
*** nimish2711 has quit IRC02:32
*** rdale has quit IRC03:56
*** mohan43u has quit IRC05:32
*** mohan43u has joined #buildstream05:37
*** mohan43u has quit IRC05:43
*** mohan43u has joined #buildstream07:28
*** tristan has quit IRC07:45
*** tristan has joined #buildstream08:18
*** rdale has joined #buildstream08:26
*** toscalix has joined #buildstream08:37
gitlab-br-botaevri closed issue #960 (Better error messages for `bst track` for cross-junction elements) on buildstream https://gitlab.com/BuildStream/buildstream/issues/96009:23
*** raoul has joined #buildstream09:41
gitlab-br-botwillsalmon closed MR !1260 (willsalmon/simpletest->master: Refactor local source to use the shinny new code in utils) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/126009:43
*** jonathanmaw has joined #buildstream09:45
gitlab-br-botjjardon opened issue #976 (Regression: buildstream master cannot build fdsdk anymore) on buildstream https://gitlab.com/BuildStream/buildstream/issues/97609:48
*** ChanServ sets mode: +o tristan09:53
tristanbenschubert, looks like you beat me to it, I was pretty sure it was a regression from !1254 no ?09:53
gitlab-br-botMR !1254: Rework Element.dependencies to be more efficient https://gitlab.com/BuildStream/buildstream/merge_requests/125409:53
benschuberttristan: exact09:54
tristanI see, yeah but it's technically not09:54
benschubertbecause bst-external was using undocumented/private api on dependencies()09:54
benschubertI'm curious as to why we had this. Was it just to exlude ourselves in https://gitlab.com/BuildStream/bst-external/blob/master/bst_external/elements/collect_manifest.py#L176 ?09:55
tristanSo we need to go ahead with splitting up bst-external I suppose09:55
tristanI don't know why it was using that actually09:55
benschubertsplitting up?09:56
tristanProbably we can make it work with both APIs, but I dont expect it to hold water for long09:56
tristanbenschubert, I think this came up in the last IRC team meeting09:56
tristan(or I might be misremembering and we discussed it elsewhere)09:57
*** toscalix has quit IRC09:57
benschubertoh you mean creating other plugin repos with more topic-focused plugins?09:57
*** toscalix has joined #buildstream09:57
tristanbenschubert, Essentially we need to make sure bst-external remains BuildStream 1, and that we fork it for BuildStream 2, so any changes we need to make to those plugins for BuildStream 2 dont effect production09:57
tristanI think we decided the first step is to make it bst-experimental09:58
benschuberttristan: could we instead promote alpha versions? if installed through pip they would not be picked unless explicitely asked for09:58
tristan(For BuildStream 2), until a more clear plan of where plugins will end up is complete09:58
benschubertah ok09:58
benschubertsure09:58
juergbidid anyone get a chance to take a look at !123210:09
gitlab-br-botMR !1232: Initial partial CAS support for remote execution https://gitlab.com/BuildStream/buildstream/merge_requests/123210:09
raoulI can have a look over that given I looked over the first partial cas MR10:15
juergbithanks10:15
csorianohi, this is the definition of the fdo runtime right? https://gitlab.com/freedesktop-sdk/freedesktop-sdk/blob/18.08/elements/desktop-platform.bst10:29
csorianoah, and this https://gitlab.com/freedesktop-sdk/freedesktop-sdk/blob/18.08/elements/base-platform.bst10:30
csorianonow makes more sense...10:30
tristancsoriano, fwiw, people on freenode in #freedesktop-sdk will be more knowledgeable about how that project is constructed10:33
tristancsoriano, I'm currently figuring out it's structure right now hehe :)10:34
csorianotristan: oh, thanks :)10:34
*** lachlan has joined #buildstream10:38
*** lachlan has quit IRC10:53
*** lachlan has joined #buildstream10:54
*** nimish2711 has joined #buildstream11:01
*** gokcennurlu has joined #buildstream11:01
gitlab-br-botphildawson opened (was WIP) MR !1261 (phil/expose-templated-tests->master: Expose templated source tests) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/126111:11
*** alatiera has joined #buildstream11:14
*** lachlan has quit IRC11:19
*** lachlan has joined #buildstream11:51
*** alatiera has quit IRC11:53
*** alatiera has joined #buildstream12:01
gitlab-br-botaevri opened MR !1263 (aevri/copyright->master: testutils/yaml.py: add copyright notice) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/126312:29
gitlab-br-botcs-shadow approved MR !1263 (aevri/copyright->master: testutils/yaml.py: add copyright notice) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/126312:51
*** lachlan has quit IRC13:02
*** xjuan has joined #buildstream13:25
*** xjuan has quit IRC13:28
*** tristan has quit IRC13:47
gitlab-br-botmarge-bot123 merged MR !1263 (aevri/copyright->master: testutils/yaml.py: add copyright notice) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/126313:50
gitlab-br-botphildawson opened issue #977 (ostree plugin's presence in BuildStream core is painful when developing buildstream on platforms without ostree) on buildstream https://gitlab.com/BuildStream/buildstream/issues/97713:53
*** lachlan has joined #buildstream14:08
*** lachlan has quit IRC14:17
*** tristan has joined #buildstream14:22
*** lachlan has joined #buildstream14:25
*** lachlan has quit IRC14:31
benschuberttristan: actually for the bug in bst-external, we could fix it to still work with both variants :)15:18
tpollardheh, it seems when running the tests gnome likes to intermittently create a mount icon on my desktop15:26
tpollardunless I'm going crazy, but I'm pretty sure it's happening15:27
juergbitpollard: yes, I see this as well15:35
juergbidue to FUSE15:35
tpollardjuergbi: yeh I guess so, 'tests/integration/script.py' being one of them15:50
*** tpollard has quit IRC15:52
*** tpollard has joined #buildstream15:57
gitlab-br-botadds68 closed issue #655 (Update CONTRIBUTING.rst to point to website) on buildstream https://gitlab.com/BuildStream/buildstream/issues/65516:15
jmacI'm writing a BuildElement-derived plugin and I'd like to have a variable %{build-id} which is set after staging. But if I use %{build-id} in the plugin yaml, it won't resolve - it fails at "Resolving elements" as build-id isn't defined yet16:19
jmacIs there any way to set a variable later?16:19
*** lachlan has joined #buildstream16:22
juergbijmac: BuildElement uses the resolved commands to calculate the cache key, which is needed before staging16:26
juergbican you elaborate on what build-id is or how it's generated during/after staging?16:27
jmacIt's the contents of a file in the staging directory. We use it as the build number parameter for Python wheels16:29
jmacE.g. `setup.py bdist_wheel --build-num %{build-id} --dist-dir %{install-root}`16:30
jmacAn equivalent which does work would be 'setup.py bdist_wheel --build-num `cat /etc/buildid`', but that doesn't give me much control16:31
juergbijmac: what extra control do you need, do you need to modify/process the buildid?16:37
jmacWell, for example, I'd like to omit --build-num entirely if /etc/buildid doesn't exist. I can do that using subprocess & bash scripts, but it's not pretty16:38
juergbiI see. yes, it's not particularly pretty but I don't see a good alternative right now16:39
jmacFair enough, it's useful to know there isn't a cleaner way at least16:39
juergbia helper script might be slightly cleaner than doing it in the .ayml16:40
jmacI'd like the logic for it to be contained in the plugin though16:41
juergbiyes, I think ugly shell commands in the yaml is overall preferable16:41
* jmac nods16:44
*** lachlan has quit IRC17:01
*** lachlan has joined #buildstream17:05
*** lachlan has quit IRC17:18
*** phildawson_ has joined #buildstream17:34
*** phildawson has quit IRC17:35
*** swick has quit IRC17:49
*** lachlan has joined #buildstream17:50
*** lachlan has quit IRC18:03
*** jonathanmaw has quit IRC18:09
*** lachlan has joined #buildstream18:17
*** alatiera has quit IRC18:29
*** alatiera has joined #buildstream18:30
*** raoul has quit IRC18:33
gitlab-br-botmarge-bot123 merged MR !1232 (juerg/partial-cas-remote->master: Initial partial CAS support for remote execution) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/123218:49
*** phildawson_ has quit IRC19:03
*** lachlan has quit IRC19:04
*** toscalix has quit IRC19:05
*** lachlan has joined #buildstream19:12
*** lachlan has quit IRC19:46
*** lachlan has joined #buildstream19:48
*** lachlan has quit IRC21:05
*** alatiera_ has joined #buildstream21:50
*** alatiera has quit IRC21:50
*** alatiera_ is now known as alatiera21:51
gitlab-br-botaevri opened (was WIP) MR !1262 (aevri/platform_uname->master: use uname in a win32-compatible way) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/126221:55
*** alatiera_ has joined #buildstream21:57
*** alatiera has quit IRC21:57
*** alatiera_ is now known as alatiera21:58
*** nimish2711 has quit IRC22:00
*** alatiera has quit IRC22:02
*** gokcennurlu has quit IRC22:58
« Tuesday, 2019-03-26 Index Thursday, 2019-03-28 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!