/srv/irclogs.ubuntu.com/2020/05/19/#smooth-operator.txt

jammorning all06:25
MarkMaglanamorning jam !06:35
* Chipaca pokes the internal irc08:42
* Chipaca presents footage of said event: https://www.youtube.com/watch?v=JMJXvsCLu6s08:42
jammorning Chipaca08:45
Chipacajam: how's things?08:48
jamgoing alright here.08:58
jamChipaca, they just called to tell me my coffee machine should be fixed by tomorrow, whi08:58
jamwhich is pretty good news.08:58
jamThough the repair cost is ~ the same as an aeropress :)08:58
ChipacaI'm probably going to jinx it, but my aeropress is 6+ years old and still working just fine :)09:00
Chipacaniemeyer: hiya! two questions about 'ops.lib.use': one is if you have a strong preference for LIB vs LIB_. The other is whether API/PATCH are the right names for what semantic versioning (which we encourage) is calling MAJOR and MINOR09:53
Chipacapatch in particular seems strange as it's the name of the third element (the z in x.y.z)09:54
* Chipaca is not blocked by these questions and goes back to writing tests09:54
niemeyerChipaca: major and minor mean nothing to most people.. API and patch versions has a stronger semantic09:57
niemeyerChipaca: My preference is obviously for the existing names otherwise I'd not have used them, but it's certainly subjective09:59
niemeyer(in terms of the underline, that is)09:59
Chipacaniemeyer: ok10:01
Chipacaniemeyer: sometimes it feels like it's too mashed together without the _, but sometimes it's fine10:01
Chipacait's *very* subjective :)10:01
niemeyerThese names will be always seem next to each other..10:02
niemeyerJust type them out and look at them for a moment10:02
Chipacaniemeyer: yeah, i think they feel weird when writing the tests, but not too weird in the init file itself, which is probably the right way around for weird10:02
niemeyerLooked better without, since the whole prefix is mostly noise in that case.. but I won't argue too much in either direction :)10:03
Chipacaniemeyer: another q: what about API or PATCH being 0?10:41
Chipacathat's ok, right?10:41
niemeyerChipaca: Yeah, sounds fine10:44
facubatistaMuy buenos días a todos!11:12
jammorning facubatista11:27
facubatistahola jam11:27
Chipacajam: we should add a test for dispatch + independent action13:21
jamChipaca, indeed. Though it should follow the same logic as the hooks/ independent script (AFAICT)13:23
vgrevtsevhi team, a quick q: do we have a way to alter the "Ports" in the `juju status` output after the charm is deployed?13:25
jamvgrevtsev, that is generally populated from calling "open-port", IIRC13:28
vgrevtsevI can't find any reference to the `open-port` in the framework code, so I'm wondering how it's getting done and how can I invoke it within the charm code13:30
vgrevtsevI doubt that the `open-port` is available using the k8s charms.. at least I tried to run it inside the operator pod and got nothing13:30
Chipacavgrevtsev: https://github.com/canonical/operator/issues/17913:46
Chipacaso no surprise you don't find it yet :)13:47
Chipacai don't know about the k8s side of your question though13:47
vgrevtsevOk, so at least something! ;D13:48
* facubatista brb13:50
facubatistajam, which is the minimum set we need? (as the OF itself adds other hooks, right?)18:29
jamSo a minimum to be forward and backward compatible is probably dispatch + start + upgrade-charm18:49
jamI think those are all the entry points that you could hit18:49
facubatistajam, no "install"?19:23
Chipacafacubatista: yes, also install20:01
facubatistaI included it just in case20:01
Chipaca:)20:02
facubatistaSo, First version of the build command: https://github.com/canonical/charmcraft/pull/420:03
facubatistaand with that, /me eods20:03
Chipacafacubatista: huzzah20:04
Chipacafacubatista: have a good'un!20:05

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!