| *** locallycompact has quit IRC | 00:21 | |
| *** rjek has quit IRC | 02:05 | |
| *** rjek has joined #baserock | 02:08 | |
| *** rjek has quit IRC | 02:13 | |
| *** rjek has joined #baserock | 02:14 | |
| *** rjek has quit IRC | 02:19 | |
| *** rjek has joined #baserock | 02:20 | |
| *** rjek has quit IRC | 02:28 | |
| *** rjek has joined #baserock | 02:30 | |
| *** thrace has quit IRC | 03:34 | |
| *** persia_ has quit IRC | 03:34 | |
| *** thrace has joined #baserock | 03:37 | |
| *** edcragg has quit IRC | 03:53 | |
| *** edcragg has joined #baserock | 03:54 | |
| *** persia_ has joined #baserock | 03:55 | |
| *** locallycompact has joined #baserock | 07:45 | |
| *** anahuelamo has joined #baserock | 07:57 | |
| *** ctbruce has joined #baserock | 08:08 | |
| *** jude_ has joined #baserock | 08:09 | |
| *** toscalix has joined #baserock | 08:44 | |
| *** paulwaters_ has joined #baserock | 09:15 | |
| *** anahuelamo has quit IRC | 09:40 | |
| *** gtristan has joined #baserock | 10:00 | |
| *** locallycompact has quit IRC | 10:06 | |
| *** anahuelamo has joined #baserock | 10:19 | |
| *** locallycompact has joined #baserock | 10:40 | |
| *** anahuelamo has quit IRC | 10:52 | |
| *** tiagogomes has quit IRC | 10:55 | |
| *** tiagogomes has joined #baserock | 10:55 | |
| *** anahuelamo has joined #baserock | 11:00 | |
| locallycompact | so | 11:02 |
|---|---|---|
| locallycompact | when the artifact split is done in a stratum like build-essential-minimal, then that ends up in the final system | 11:02 |
| locallycompact | but what are the DEFAULT stratum splits for? nothing specifies to use them | 11:02 |
| locallycompact | has this ever done anything? | 11:03 |
| paulsherwood | i was attempting to allow situation where user specifies [minimal-] or [runtime-] as default, rather than having to go into systems and define specific artifacts there | 11:04 |
| paulsherwood | s/as default/as global default in their .conf/ | 11:04 |
| locallycompact | this feels wrong | 11:05 |
| paulsherwood | it is wrong | 11:05 |
| paulsherwood | splitting only really makes sense for systems... so it should not involve any prep during chunk/stratum creation | 11:06 |
| locallycompact | So atm I have turned splitting off in V10 and then run build minimal it goes away and builds and makes a full fat tarball and puts it away in the cache, while omitting build-mode: bootstrap chunks and include-mode: build-time assembs | 11:08 |
| locallycompact | So we have that thing that could be split | 11:08 |
| locallycompact | What decides that | 11:08 |
| locallycompact | or is it better to just have two systems one for runtime and one for debug | 11:08 |
| locallycompact | or maybe just have something that has nothing to do with this tool whose job it is to only split filesystem tarballs against some regex rules | 11:10 |
| SotK | what is wrong with caching the split artifacts themselves rather than one massive unsplit artifact? | 11:11 |
| paulsherwood | SotK: in my world view a chunk contains everything, and splitting creates a subset of it. this seems more logical than having a chunk as 5 artifacts | 11:12 |
| paulsherwood | (and allows custom splitting, after builds have happened) | 11:13 |
| locallycompact | build :: (Repo, Ref, Instructions) -> Artifact | 11:13 |
| locallycompact | split :: Artifact -> List<Artifact> | 11:13 |
| locallycompact | actually that's too oversimplified | 11:14 |
| * paulsherwood was thinking the same | 11:15 | |
| locallycompact | split is more like, split :: Artifact -> Rules -> List<Artifact> | 11:17 |
| locallycompact | but if those never actually exist then it's more just like a strip against a system | 11:18 |
| locallycompact | maybe we could just add a mask: or filter: or strip: field to assemblages in a contents | 11:20 |
| locallycompact | with a list of regexes | 11:20 |
| locallycompact | and not have the full dictionary of -bins, -libs | 11:20 |
| locallycompact | why do they need labels | 11:20 |
| locallycompact | just filter: [regexes] | 11:21 |
| locallycompact | so production-x86_64 has one element in its contents which is debug-system-x86-64 with an additional filter/strip | 11:21 |
| locallycompact | then it stores in the cache as what it is when it gets deploy | 11:21 |
| paulsherwood | sadly, i don't fully understand the requirements. my goto expert would be radiofree | 11:25 |
| *** richard_maw has joined #baserock | 11:47 | |
| *** ChanServ sets mode: +v richard_maw | 11:47 | |
| *** richard_maw has left #baserock | 11:54 | |
| *** jude_ has quit IRC | 12:06 | |
| *** jude_ has joined #baserock | 12:09 | |
| *** locallycompact has quit IRC | 12:21 | |
| *** locallycompact has joined #baserock | 12:26 | |
| *** tiagogomes has quit IRC | 13:19 | |
| *** tiagogomes has joined #baserock | 13:21 | |
| *** anahuelamo has quit IRC | 13:27 | |
| *** anahuelamo has joined #baserock | 13:28 | |
| jmacs | Hi all! Was a mustard document ever made for the Baserock GENIVI baseline? | 14:24 |
| paulsherwood | yes, it was, and we hosted it for genivi members to review. the content technically should not be exposed to the wild internet (would need a password) | 14:30 |
| jmacs | Ah, I see | 14:31 |
| * pedroalvarez finds baserock mustard | 14:37 | |
| pedroalvarez | not GENIVI, just Baserock | 14:38 |
| jmacs | pedroalvarez: Where's that? | 14:38 |
| pedroalvarez | it wasn't open either | 14:38 |
| jmacs | The *baserock* mustard isn't open source? | 14:39 |
| pedroalvarez | very long time ago | 14:39 |
| * paulsherwood is happy for it to be opened... but can't vouch for the quality of any of its content | 14:43 | |
| pedroalvarez | I'd see its contents first :P | 14:43 |
| *** toscalix_ has joined #baserock | 14:47 | |
| *** toscalix has quit IRC | 14:49 | |
| tiagogomes | oh, it ws the description of artifact splitting. This was documented after all | 14:49 |
| tiagogomes | s/ws/has | 14:49 |
| paulsherwood | interesting. maybe if we'd published it, we'd not have forgotten about it! :) | 14:50 |
| pedroalvarez | is it something that we are interested on having? mustard.baserock.org? | 14:55 |
| paulsherwood | i would say so... particularly if we could update the mustard content via gitlab workflow | 14:58 |
| *** toscalix_ has quit IRC | 15:32 | |
| *** toscalix_ has joined #baserock | 15:32 | |
| *** toscalix_ has quit IRC | 15:36 | |
| *** toscalix_ has joined #baserock | 15:43 | |
| *** toscalix_ has quit IRC | 15:48 | |
| *** toscalix_ has joined #baserock | 15:48 | |
| *** toscalix_ has quit IRC | 15:48 | |
| * JPohlman1 remembers creating a GENIVI baseline mustard in... 2013? ;) | 16:54 | |
| *** JPohlman1 is now known as JPohlmann | 16:54 | |
| *** JPohlmann has joined #baserock | 16:54 | |
| JPohlmann | I'm on my way to Manchester btw, meeting with Stephen around 8.30 if anyone fancies joining us | 16:55 |
| jmacs | Ah, I was wondering when you were landing! Great! | 16:56 |
| JPohlmann | 7.20 local time according to the schedule | 16:57 |
| *** tiagogomes has quit IRC | 17:20 | |
| *** tiagogomes has joined #baserock | 17:26 | |
| *** jude_ has quit IRC | 17:28 | |
| *** jude_ has joined #baserock | 17:28 | |
| *** locallycompact has quit IRC | 17:56 | |
| *** faybrocklebank has quit IRC | 18:01 | |
| *** anahuelamo has quit IRC | 18:04 | |
| *** gtristan has quit IRC | 18:04 | |
| *** jude_ has quit IRC | 18:11 | |
| *** jude_ has joined #baserock | 18:53 | |
| *** ctbruce has quit IRC | 20:50 | |
| *** ctbruce has joined #baserock | 20:50 | |
| *** ctbruce has quit IRC | 21:31 | |
| *** jude_ has quit IRC | 22:22 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!