11:03:34 <richard> #startmeeting Tor Browser Weekly Meeting 2022-07-11
11:03:34 <MeetBot> Meeting started Mon Jul 11 11:03:34 2022 UTC.  The chair is richard. Information about MeetBot at http://wiki.debian.org/MeetBot.
11:03:34 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
11:03:43 <richard> pad: https://pad.riseup.net/p/tor-tbb-keep
11:04:38 <richard> as usual, please update your sections and tidy up your gitlab boards :)
11:10:31 <msim> brb 1 min
11:12:43 <msim> back :)
11:14:19 <richard> well ok then
11:15:46 <richard> looks like we don't have any discussion points or announcements this week, so let's go through the bolded points
11:16:13 <richard> let's start with msim's and then we can talk about android
11:16:29 <msim> okay so, goxsys doesn't actually have a release tag/branch for go 1.18
11:16:45 <msim> there's an "internal" branch: https://github.com/golang/sys/tree/internal-branch.go1.18-vendor
11:16:54 <msim> although i'm not entirely sure what the "internal" designation entails
11:17:30 <PieroV> msim: have you tried to diff and see what it changes?
11:18:08 <PieroV> Also, it is a dependency of something else... We may be able to just get rid of it
11:18:28 <msim> well, a lot
11:19:03 <msim> git diff has >200k lines of output .-.
11:19:14 <msim> (since the last commit hash used, fae7ac547cb717d141c433a2a173315e216b64c4)
11:19:43 <richard> 2019
11:19:44 <richard> oof
11:21:00 <msim> i'm not entirely sure we can remove it PieroV, since it's in the dep chain for obfs4
11:21:24 <PieroV> Okay. Then are we talking of https://github.com/golang/sys, right?
11:21:29 <msim> (obfs4 -> goutls -> goxcrypto -> goxsys)
11:21:33 <msim> PieroV: yes
11:21:46 <PieroV> oh, goutls, now I see.
11:22:13 <richard> you could just use the current HEAD commit on that branch no?
11:22:32 <PieroV> Why don't we take master, directly?
11:22:56 <msim> well we can do that, but what concerns me is the "internal" part of the branch name
11:23:04 <boklm> we can check in goxcrypto go.mod or go.sum file which commit it requires
11:23:05 <msim> i.e. there's no official release of goxsys for 1.18
11:23:42 <PieroV> There aren't branches for 1.17 either
11:23:45 <PieroV> +1 for what bolkm suggests
11:23:53 <boklm> it seems other go projects have internal-branch
11:24:03 <boklm> https://go.googlesource.com/net/+/refs/heads/internal-branch.go1.18-vendor/go.mod
11:24:08 <richard> v0.0.0-20190412213103-97732733099 in go.mod of goxcrypto
11:24:28 <msim> diff between current and 97732733099d is 28k lines, a bit more manageable
11:24:45 <boklm> maybe we need to update goxcrypto too, and a few other go modules
11:25:16 <msim> oh hm, 97732733099d comes before the current commit hash
11:25:46 <msim> oh right, the old version was causing the build failure for go 1.18 on macos
11:26:03 <PieroV> golang.org/x/sys v0.0.0-20210615035016-665e8c7367d1 in the current head of crypto
11:26:45 <msim> if it compiles then wfm :D
11:27:07 <msim> once i'm finished rebasing i'll try with 665e8c7367d1
11:28:45 <richard> kind of a pita to get their release versions to match up with git commits >:[
11:29:01 <richard> but anyway
11:29:48 <PieroV> Why are we using obfs4proxy 0.0.12 instead of 0.0.13?
11:29:54 <PieroV> IIRC there was some reason
11:30:19 <PieroV> But 0.0.13 removes the dependency on yawning's goutls fork, if I see correctly
11:31:13 <richard> I do not know
11:32:23 <PieroV> I think we could ask anticensorship and update if needed, and only then fix the remaining problems with go deps
11:32:51 <boklm> I think cohosh has been using some script to update modules, which can maybe help with this
11:33:26 <richard> for dependency resolution?
11:33:40 <boklm> yes
11:35:09 <richard> ok sounds like a start
11:35:56 <richard> msim: do you mind taking the lead on working w/ anticensorship to figure out the story here?
11:36:19 <msim> sounds good to me :)
11:37:11 <richard> alright then lots move on to Android
11:37:41 <PieroV> I think I could do some rebases to 102 if I have some free time
11:38:14 <PieroV> And then to the reorder also on Android, before a possible sync with destkop
11:38:33 <PieroV> I think it's better if we do this before new persons start working on it
11:39:23 <richard> that is a good point
11:40:38 <richard> well, we have a couple of weeks to get our house in order there
11:41:17 <PieroV> Yes, sorta. I think it's a long thing
11:41:29 <PieroV> I could do that after the most important things for S131
11:42:39 <richard> ok
11:43:06 <richard> I was planning on rebasing android 102 this week as well
11:43:35 <PieroV> Oh, okay. You could do the rebase starting from v100, even though it was not reviewed
11:43:47 <PieroV> I can check if it's missing some commits, if needed
11:43:58 <PieroV> I think we should also squash a pair of commits in Fenix
11:44:31 <richard> yeah I'll ping you if I get stuck, but should be an interesting time regardless
11:47:06 <richard> boklm: can you prioritize looking into/streamlining the Android signing process this week?
11:47:14 <boklm> richard: yes
11:47:29 <richard> if that wasn't already your plan
11:47:50 <boklm> I will start looking at this tomorrow
11:48:23 <richard> ok then, sounds like we're sorted for this week
11:49:01 <richard> boklm: perfect
11:49:32 <richard> donuts is working on drafting a special blog post for the 11.5 release, so it should be a bit more interesting then our usual terse Changelog prose :)
11:49:42 <boklm> ok
11:50:03 <PieroV> I have another thing for the changelogs
11:50:12 <richard> overall I'm really happy with this release and excited things to come
11:50:15 <PieroV> Nothing very important, just a issue I should've closed for 91.10
11:50:24 <richard> oh?
11:50:26 <PieroV> (so, for the old version)
11:50:40 <PieroV> Yes, we've missed that we switched to Firefox's patch for window.name
11:50:45 <PieroV> And dropped ours
11:51:37 <richard> ah I see
11:52:08 <richard> well not the end of the world, we can amend the changelog when we release 12.a1
11:52:22 <PieroV> okay, sounds good to me
11:52:25 <richard> if there's not anything else then we can call it here
11:52:33 <PieroV> yes, nothing from me
11:52:41 <boklm> nothing from me
11:53:23 <richard> ok then, see you all on IRC
11:53:23 <richard> #endmeeting