| [18:41] <jlosito> testing bot | |
| [18:47] <jlosito> 1 more test | |
| [18:54] <Chipaca> Dmitrii-Sh: welcome! | |
| [18:54] <Chipaca> ubuntulog3: also welcome | |
| [18:54] <Chipaca> jlosito: thanks | |
| [18:55] <jlosito> np | |
| === Chipaca changed the topic of #smooth-operator to: general discussion of canonical's operator framework || github.com/canonical/operator | |
| [19:11] <Chipaca> phew | |
| [19:12] <Chipaca> woop, everything works ;-) | |
| [19:13] * Chipaca goes to make dinner | |
| [19:13] <niemeyer> I'd just say "on the operator framework" | |
| [19:13] <niemeyer> :P | |
| === Chipaca changed the topic of #smooth-operator to: general discussion of the operator framework || github.com/canonical/operator | |
| [19:13] <Chipaca> niemeyer: like so? | |
| [19:13] <niemeyer> Thanks | |
| [19:14] <Chipaca> i'm sure there's a way of just telling chanserv to do that | |
| [19:14] <Chipaca> instead of op/topic/deop | |
| [19:14] <Chipaca> anywhoo, dinner-making | |
| [19:15] <niemeyer> Have fun! | |
| [19:15] <niemeyer> It's already beer o'clock here.. | |
| [19:15] <niemeyer> and debugging something.. but at this stage any findings are a bonus :) | |
| [19:22] <Chipaca> no beer here because it's a light pre-yoga dinner | |
| [19:22] <Chipaca> might have a beer after yoga; can't be _too_ healthy | |
| [19:22] <niemeyer> I'm sure your zeneth capabilities would be much improved | |
| [19:23] <niemeyer> Not so sure about balance though | |
| [19:23] <Chipaca> i could pretend it's about that, but it's just trying to keep my back un-messed-up | |
| [20:16] <Facu> niemeyer, did we think already about the following situation? 1. the event is deferred, it's `.snapshot` is called; 2. the charm is upgraded, the event code itself changes! 3. the event is restored, a new code needs to handle previous "format" | |
| [20:16] <Facu> I mean, are we just leaving that consideration to the developer, providing any guide? any structure? | |
| [20:19] <niemeyer> Facu: Yeah, we don't expect devs to be using snapshot directly very often | |
| [20:19] <niemeyer> Facu: We have the StoredState mechanism for that | |
| [20:19] <niemeyer> People can just do _stored = StoredState() at class level | |
| [20:20] <niemeyer> And only store there the data they want to persist | |
| [20:20] <Facu> but everytime you define an Event you need to provide its snapshot/restore pairs, right? | |
| [20:20] <niemeyer> Naming is important precisely to highlight that this is persisted | |
| [20:20] <niemeyer> (thus the choice of "_stored") | |
| [20:20] <niemeyer> Facu: Every time you define an event *type* that needs parameters | |
| [20:21] <Facu> perfect | |
| [20:22] <niemeyer> Facu: We should probably make EventBase a child of Object, so StoredState would work there as well | |
| [20:22] <niemeyer> But I haven't thought through the consequences of that | |