20:13:40 <CaptainTobin> hey frg_Away
20:13:51 <frg_Away> CaptainTobin hi
20:14:06 <CaptainTobin> https://cdn.discordapp.com/attachments/1192245432982704189/1200561463400607937/gLKCNCcARrVXHDBMhHIXqNUW.png
20:14:14 <CaptainTobin> if you didn't see it in the backlog
20:18:28 <frg_Away> CaptainTobin saw it. Based on central or older branch ?
20:18:36 <CaptainTobin> release
20:20:33 <CaptainTobin> i spoke with emillio in the mozdev matrix channel
20:21:50 <CaptainTobin> my main blockers right now are https://searchfox.org/mozilla-central/source/services/settings/dumps/gen_last_modified.py#52 and https://searchfox.org/mozilla-central/source/python/mozbuild/mozbuild/base.py#962-1023
20:21:58 <CaptainTobin> but there may be a few more
20:22:33 <CaptainTobin> these prevent anything but firefox android thunderbird and heh seamonkey from building
20:23:36 <CaptainTobin> i kinda get why they are currently like that but it sure makes it hard to build especially with rapid release
20:23:45 <CaptainTobin> i am sure you can sympathise frg_Away
20:26:18 <frg_Away> Not sure if the first is needed outside taskcluster. Second one yes. There are other dependencies in too all over the place but mostly for tests.
20:27:13 <CaptainTobin> well i expressed I didn't know if I should just make it assume everyone gets at least the central copy or if i should do a wider patch to make remote settings build conditional
20:29:30 <CaptainTobin> the main issue is ./mach run won't execute if it isn't those things
20:29:49 <CaptainTobin> i kinda would like ./mach run to work
20:29:59 <CaptainTobin> doing it manually is a pain
20:30:29 <CaptainTobin> fix those, I do my experimental example and Mozilla is a platform again.
20:30:33 <CaptainTobin> easy
20:32:20 <CaptainTobin> there is another effort abusing firefox's residual xulrunner functions but it requires node and still requires building the codebase.. it is an interesting idea and approch but I like approching it kinda old school as everyone here well knows
20:34:14 <CaptainTobin> frg_Away: fluent is kinda a strange input format
20:34:33 <CaptainTobin> and i am not sure how I feel about how it rewrites the dom rather than xml entities
20:36:05 <CaptainTobin> I am approching this as not what has been lost or mutated but what is still there and what is the current equivalent
20:37:40 <CaptainTobin> and is it still possible (yes it is confirmed) and if mozilla would permit it and especially if anyone still values the potental albeit not the same as it was but potental none the less contained in a mozilla codebase as a platform
20:39:14 <CaptainTobin> I also want mozilla to build with a qt toolkit not gtk
20:39:28 <frg_Away> Well we have fluent now in 2.53 but it still needs some more work/backports. I think it is another case where it is very flexible but you actually don't need all of it. I prefer stuff a bit more static if you know what I mean.
20:40:00 <CaptainTobin> well you can keep dtds
20:40:44 <CaptainTobin> but maybe I could just write a converter
20:41:12 <CaptainTobin> and it just transforms dtd and properties to ftl which transforms it into its properties format
20:41:23 <CaptainTobin> throw that in as a build step
20:41:26 <CaptainTobin> and there you go
20:42:28 <CaptainTobin> it stands for firefox translation language doesn't it
20:42:29 <CaptainTobin> ffs
20:42:30 <CaptainTobin> lol
20:42:39 <frg_Away> CaptainTobin I think it acutally exists for dtd to fluent
20:43:11 <CaptainTobin> you backporting remote settings?
20:43:16 <CaptainTobin> or did you already
20:46:30 <frg_Away> no remote settings
20:58:10 <CaptainTobin> frg_Away: any idea how I should handle jsdownloads in a simple way?
20:58:36 <CaptainTobin> I shunned it from the start lol
21:00:28 <frg_Away> I forgot half the things about it. I know that I cleaned up our implementation a lot but it took time.