[10:41:13] <gehel>	 lunch + errand
[13:00:04] <inflatador>	 <o/
[14:15:50] <ebernhardson>	 \o
[14:24:15] <ebernhardson>	 i suppose back to integration testing migration...annoyingly it's a fairly big change,  we are using synchronous browser and it needs to all migrate to async/await
[14:28:17] <inflatador>	 .o/
[15:03:08] <dcausse>	 I thought we already migrated to async wbdio
[15:03:24] <dcausse>	 o/
[15:05:02] <dcausse>	 I see we call things like 'await text.getText()'
[15:06:40] <ebernhardson>	 indeed some of it is migrated, i'm just looking at the world right now, anything with Promise.coroutine has to go away, the client.rawRequest returns promises, pretty sure that has to go await/async.
[15:06:52] <ebernhardson>	 part i guess is i don't know enough about how it's supposed to work, maybe it is more interoperable...
[15:07:12] <ebernhardson>	 basically anthing using `.then` has to migrate iiuc
[15:07:39] <dcausse>	 ah you mean bluebird has to go away?
[15:08:05] <ebernhardson>	 i think so, although i could be misunderstanding. My reading is that bluebird is an "old style", replaced by native await/async
[15:08:25] <dcausse>	 yes definitely
[15:09:06] <dcausse>	 getting rid of bluebird is certainly a consequent change I'm afraid
[15:09:35] <ebernhardson>	 i suppose on the upside...all the features are written in cucumber so it's just migrating world and some steps
[15:09:45] <ebernhardson>	 at least it's not changing each indiviual test function
[15:09:48] <dcausse>	 yes I hope so
[15:10:32] <dcausse>	 and "class Page" are not using bluebird hopefully and they should be the ones mainly interacting with wdio 
[15:12:05] <dcausse>	 needs to afk for a while
[16:28:11] <ebernhardson>	 randomly interesting...have to replace restify (it's abandoned) in the test suite,  fastify seems the suggested replcaement. They have a define LTS/support range, and then offer paid "never-ending support" for whatever version you use
[16:29:06] <ebernhardson>	 i'm sure other things do that too, i suppose i've just not seen in phrased that way
[16:50:57] <dcausse>	 :/
[16:53:34] <ebernhardson>	 so much abandonware...the other side of the tracker.js which makes the http requests is also abandoned :P
[16:53:50] <ebernhardson>	 err, the http request library is abandoned
[16:54:45] <ebernhardson>	 mwbot last release was 3 years ago...i'm going to just hope that one is fine :P
[16:55:38] <ebernhardson>	 except...i guess the reason we used that http client is that mwbot uses that http client :( maybe it will be fine for now...
[16:59:57] <dcausse>	 sigh...
[17:00:18] <dcausse>	 this mwbot is handy
[17:00:29] <ebernhardson>	 indeed it is, an poking codesearch i suspect 20+ extensions use it
[17:04:57] <dcausse>	 well I guess that's a good reason to help maintaining it if needs be :)
[17:05:27] <ebernhardson>	 there looks to be an mwbot-ts thing that is maintained, but it's not a direct successor.  Can't tell if it's a spiritual successor
[19:03:14] <inflatador>	 back in ~90
[19:03:59] <dcausse>	 dinner
[19:21:23] <ebernhardson>	 surprisingly, thats a succesfull run on wdio 9 and nodejs 20.  (and i was partially wrong about browser sync access, we had already gotten rid of it)
[19:22:50] <ebernhardson>	 i had to compile a custom version of the fresh container with a few more dependencies though, will see about upstreaming.
[19:41:49] * ebernhardson spoke too soon...maybe only api tests work :P
[20:04:36] <ebernhardson>	 ahh, it's always the silly things.  If the browser window is too small the search input is hidden, and the new version (and maybe the old? not sure if default window size changed) wont let you interact with non-visible elements
[20:31:49] <inflatador>	 back
[20:48:12] <ebernhardson>	 not terrible: Spec Files:      43 passed, 1 skipped, 44 total (100% completed) in 00:12:43
[20:48:41] <ebernhardson>	 the skipped one has @expect_failure tagged. Basically never runs.
[21:57:05] * ebernhardson wonders if cirrus should be the one specifying the fresh image version...would be easier that way