| === guiverc2 is now known as guiverc | ||
| tsimonq2 | Hey arraybolt3 look at this...... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/01e19f5935d09fe2ffcdfcc90c5fe1588b08eb77>) | 00:55 |
|---|---|---|
| tsimonq2 | https://git.lubuntu.me/Lubuntu/lubuntu-breeze-config ... I feel bad I didn't even know this existed ugh. | 01:15 |
| lubot | [telegram] <Roberalz> That worked in old versions of lxqt, but it doesn't work anymore. (re @lubuntu_bot: (irc) <tsimonq2> https://git.lubuntu.me/Lubuntu/lubuntu-breeze-config ... I feel bad I didn't even know this existed ugh.) | 01:25 |
| * tsimonq2 sent a code block: https://libera.ems.host/_matrix/media/v3/download/libera.chat/350931083d8eed2ab1533432c2cf8eb9885ca477 | 01:37 | |
| tsimonq2 | <lubot> "[telegram] <Roberalz> That..." <- Oh yeah? | 02:05 |
| arraybolt3[m] | > <@tsimonq2:linuxdelta.com> Hey arraybolt3 look at this...... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/6e1038336dc542a5710fb664db398c9ebe1f0aaf>) | 02:47 |
| tsimonq2 | Yep :) at least for liblxqt | 02:47 |
| arraybolt3[m] | (Sorry to have been gone for a while, I yet again got pulled away.) | 02:48 |
| tsimonq2 | No worries | 02:48 |
| arraybolt3[m] | Ok, so I need to finish setting up the dev env in my VM, then resume work. | 02:52 |
| === IchabodCrane is now known as DalekSec | ||
| tsimonq2 | I need to get home, I'll review stuff in the morning :) | 03:10 |
| arraybolt3[m] | Does anyone remember what doc says which options to set for Lintian in sbuildrc? I *think* it was `['-E', '-v', 'I', 'L', '+pedantic']` but I don't remember for sure. | 03:17 |
| tsimonq2 | I think that's correct | 03:20 |
| arraybolt3[m] | Heh, I guess you are the documentation for a lot of stuff so if that looks right to you, +1. | 03:21 |
| tsimonq2 | This is all off the top of my head. I don't have notes, only reminders :) | 03:21 |
| tsimonq2 | I've said it before, but Ansible is just downright amazing for this sort of thing. Seriously, you can set it up to configure the system just the way you like it, install whatever packages you need, and it just does it. Written in Python on the backend, YAML or INI for all the config files | 03:23 |
| tsimonq2 | I have it set up for my production system, which also happens to be my dev system heh :) | 03:23 |
| tsimonq2 | Anyway... | 03:24 |
| arraybolt3[m] | Sounds cool. Guess that's another thing that needs to be learned. | 03:24 |
| tsimonq2 | Get the list! the earth starts to shake | 03:31 |
| arraybolt3[m] | I don't think you guys want my personal "to be learned" list in the middle of The List: https://notes.lubuntu.me/huOk59_iRSaAMZDl_my8bw | 03:32 |
| tsimonq2 | Let me know if you end up making progress anywhere... would be nice to get lubuntu-meta out of the way and at least in packaging/in the queue but it likely won't be accepted until the archive actually opens | 03:48 |
| arraybolt3[m] | Simon Quigley: I doubt you're up this late, but... uh... where is Lubuntu's seed? :D | 04:34 |
| arraybolt3[m] | nvm found it | 04:35 |
| arraybolt3[m] | Success! https://code.launchpad.net/~arraybolt3/ubuntu-seeds/+git/ubuntu-seeds/+merge/432472 <-- Simon Quigley | 04:45 |
| -queuebot:#lubuntu-devel- Unapproved: accepted lubuntu-default-settings [source] (lunar-proposed) [23.04.1] | 11:32 | |
| -queuebot:#lubuntu-devel- Unapproved: accepted lubuntu-artwork [source] (lunar-proposed) [23.04.1] | 11:32 | |
| arraybolt3[m] | `Kernel panic - not syncing - coffee levels dangerously low!` | 14:40 |
| * genii coredumps arraybolt3[m] to study the coffee problem | 14:40 | |
| * arraybolt3[m] 's core splats all over the floor | 14:41 | |
| arraybolt3[m] | Alright, while things are still peaceful, I should get some coffee and food. | 14:42 |
| genii | heh | 14:43 |
| === arraybolt3_ is now known as arraybolt3 | ||
| tsimonq2 | Morning! | 17:53 |
| arraybolt3[m] | Aloha! | 17:54 |
| tsimonq2 | <arraybolt3[m]> "Success! https://code.launchpad..." <- Merged | 17:54 |
| * arraybolt3[m] seriously considers getting a second cup of coffee to prepare for the incoming storm | 17:55 | |
| arraybolt3[m] | Thank you kindly! Firing up Lunar VM to proceed with work. | 17:55 |
| * arraybolt3[m] is getting in a fight in #libera over browser choices... | 17:56 | |
| arraybolt3[m] | Hey do we want to pick a placeholder wallpaper for Lunar for telling VMs apart like we did last time, but early in the cycle this time? | 17:57 |
| tsimonq2 | arraybolt3[m]: I think that would be a good idea, Dan Simmons what do you think? | 17:58 |
| tsimonq2 | arraybolt3: Is lubuntu-meta the last thing on the new-release list that you know of? | 18:00 |
| tsimonq2 | teward: I'm interested in hearing about Drone and any CI implementation once you have the chance. | 18:00 |
| tsimonq2 | In the meantime, we should probably start with Debian LXQt, once that's all done | 18:00 |
| arraybolt3[m] | Simon Quigley: I don't think so, but I think this is the heaviest part of the job most likely. | 18:00 |
| arraybolt3[m] | Alright, running update script. | 18:01 |
| tsimonq2 | I'd be interested in seeing which packages still need upstreaming to Debian and where the current points of contention are. "Where are we at?" | 18:01 |
| arraybolt3[m] | Simon Quigley: Most of them I believe are done. libfm-qt is the doozy because of symbols. Maybe I can pass that one off on you since you'll be able to defend yourself against Andrew. He was too much for me - some stuff we deliberately did, I had no clue why, and he was grilling me on it, so... :P | 18:02 |
| arraybolt3[m] | Anyway once I'm done with -meta I'll get a list. | 18:10 |
| tsimonq2 | arraybolt3[m]: Thanks; I'll work through `libfm-qt` and other items on the list once we have one :) | 18:10 |
| arraybolt3[m] | Thanks :) | 18:15 |
| arraybolt3[m] | I'll also try to work with stuff, but maybe if you can show me any tricks you use for that, that would be nice, since it looks like a major headache to me. | 18:15 |
| tsimonq2 | No worries, we'll work through it | 18:15 |
| arraybolt3[m] | Alright, the x font error is fixed! \o/ | 18:19 |
| arraybolt3[m] | Whoa what in the world? I just crashed QTerminal... | 18:25 |
| arraybolt3[m] | Oh wait no I didn't, it just hung for some reason. | 18:25 |
| tsimonq2 | arraybolt3[m]: I'd be interested to see how this Qt update impacts our existing applications | 18:29 |
| arraybolt3[m] | Makes sense. For some reason pasting with Ctrl+Shift+V failed by hanging QTerminal for a good while and then allowing it to seemingly function normally again but without pasting anything. | 18:29 |
| arraybolt3[m] | Simon Quigley: Alright, I believe lubuntu-meta is done. THere's only two Lintian lines I'm worried about:... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/9615907118b5a27ddd421baeae56effbff8be259>) | 18:36 |
| tsimonq2 | > <@arraybolt3:matrix.org> Simon Quigley: Alright, I believe lubuntu-meta is done. THere's only two Lintian lines I'm worried about:... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/457020c5365339eeb2db784c50f9f76dfbf6aca3>) | 18:36 |
| tsimonq2 | The answer to that question will give you the answer to this one | 18:37 |
| arraybolt3[m] | I looked this up and remember thinking that these could just be ignored. Lemme check... | 18:37 |
| arraybolt3[m] | "Conflicts should only be used if two packages can never be unpacked at the same time, or for some situations involving virtual packages (where a version clause is not appropriate). In particular, when moving files between packages, use Breaks plus Replaces, not Conflicts plus Replaces." | 18:38 |
| arraybolt3[m] | The problem is I don't know what either of those two old packages do, or if they even still exist. | 18:38 |
| tsimonq2 | Way back in the day when we were first switching to LXQt, we offered two separate metapackages, lubuntu-gtk-desktop and lubuntu-qt-desktop | 18:38 |
| arraybolt3[m] | I see they's considered purely virtual according to apt-cache. | 18:39 |
| tsimonq2 | Right, so let me fill in this piece for you: (Conflicts/Breaks)/Replaces are usually dropped after an LTS if: | 18:40 |
| tsimonq2 | - LTS-1 had those virtual packages | 18:40 |
| tsimonq2 | - LTS has the proper Breaks/Replaces, to move users onto the correct package | 18:40 |
| tsimonq2 | As in... if 20.04 shipped with the lubuntu-qt-desktop metapackage installed by default, we want users upgrading to 22.04 to automatically be moved to lubuntu-desktop, having removed lubuntu-qt-desktop beforehand | 18:41 |
| tsimonq2 | That's what those fields do | 18:41 |
| tsimonq2 | So, if the upgrade path is good, if LTS -> LTS works and Kinetic -> Lunar works, you may consider dropping those lines and then re-testing Kinetic -> Lunar upgrades | 18:41 |
| tsimonq2 | Again, the entire goal is, "we deprecated this binary package, if you try to install it it will automatically install lubuntu-desktop, and once we no longer need to account for an upgrade path where 'pre-upgrade' has the deprecated binary package, you can drop that line" | 18:42 |
| tsimonq2 | I've never had to explain this concept to someone else, so please let me know what your questions are :) | 18:42 |
| arraybolt3[m] | So basically they were just a tool to make upgrades go smoothly and now that the tool has done its job we can test to make sure it really has done its job and then discard it? | 18:43 |
| tsimonq2 | Yeah, basically | 18:43 |
| tsimonq2 | My gut is just to drop any Breaks/Replaces lines on lubuntu-qt-desktop and lubuntu-gtk-desktop | 18:43 |
| arraybolt3[m] | OK that makes sense. And that's the solution I was thinking of also. | 18:43 |
| arraybolt3[m] | So this is one of those tricky situations involving virtual packages that the Lintian page was talking about. | 18:44 |
| tsimonq2 | Yep ;) | 18:44 |
| arraybolt3[m] | OK, I'll try dropping them and see if it builds, then... how do we test if it upgrades right? PPA? | 18:44 |
| arraybolt3[m] | Upload to archive and see what happens? | 18:45 |
| arraybolt3[m] | (The latter seems easier and relatively safe at this point in the cycle.) | 18:45 |
| tsimonq2 | Yeah, I'd throw it in a PPA, try a fresh install of Kinetic, do the aforementioned substitution dance, add the PPA, and *then* `full-upgrade` | 18:45 |
| arraybolt3[m] | Ah, OK. | 18:45 |
| arraybolt3[m] | deep breath now I have to move my GPG key into the VM | 18:46 |
| tsimonq2 | arraybolt3[m]: I mean... no you don't ;) | 18:46 |
| arraybolt3[m] | Well I can move the packaging out of the VM and then sign and upload it but that's even more difficult. | 18:47 |
| tsimonq2 | `debuild -us -uc [...]` to get the files, transfer the files to your host for signing and upload from there? | 18:47 |
| arraybolt3[m] | Getting files out of Boxes is much trickier than getting them in :P | 18:47 |
| tsimonq2 | :P | 18:47 |
| arraybolt3[m] | I dunno though, it may be worth it so that I don't have a bunch of GPG mess everywhere. | 18:47 |
| tsimonq2 | Oh, and in your PPA, I'd enable the proposed pocket explicitly. It's in the PPA settings, but a lot of these initial packages to bootstrap haven't migrated yet | 18:48 |
| arraybolt3[m] | tsimonq2: I thought it was `debuild -S -d -sa`? | 18:48 |
| tsimonq2 | arraybolt3[m]: `-us -uc` skips signing | 18:48 |
| arraybolt3[m] | Oh. OK I get it. | 18:48 |
| arraybolt3[m] | And then I can sign after the fact how? (I know I can look this up but that would possibly take a while.) | 18:48 |
| tsimonq2 | `debsign -k<MYKEYIDHERE> /path/to/source.changes` | 18:49 |
| arraybolt3[m] | Nice! Thank you! | 18:50 |
| tsimonq2 | Of course :) | 18:50 |
| arraybolt3[m] | (Now I remember seeing that tool before I think :P) | 18:50 |
| teward | tsimonq2: I've been trying to reach you about renewing your expired car insurance | 18:54 |
| teward | tsimonq2: i've been a bit too busy to stab Drone into submission, but once we get Drone set up it should be trivial to define CI - except that you have the CI as a definition in the repository so Drone can see it, and then we enable it. I've got a ton to get done today so I can't discuss it right now | 18:55 |
| tsimonq2 | teward: WE HAVE BEEN TRYING TO REACH YOU ABOUT YOUR CAR'S EXTENDED WARRANTY | 18:57 |
| tsimonq2 | lmao | 18:57 |
| tsimonq2 | anyway | 18:57 |
| tsimonq2 | teward: No worries, I'll see if I can play with it At Some Point | 18:57 |
| teward | tsimonq2: i don't think i've enabled it everywhere yet on Gitea. I've got a test repo but i'm busy as shit so i'll take a peek at it later but i'm busy for today | 18:57 |
| tsimonq2 | cool cool :) | 18:58 |
| arraybolt3[m] | Simon Quigley: OK so this is interesting. I removed all of the Breaks/Replaces/Provides from lubuntu-meta, sbuild'ed it, installed it, then rebooted. Somehow that has set the default wallpaper to the original kudu from Unsplash rather than the one that was picked for the final ISO. | 19:36 |
| arraybolt3[m] | I have no clue how on earth installing that managed to mess with lubuntu-artwork, but... it did. | 19:36 |
| tsimonq2 | red herring - I've seen similar wallpaper-related bugs on my own system | 19:37 |
| arraybolt3[m] | Oh OK. | 19:37 |
| arraybolt3[m] | I wonder what on earth is wrong with the wallpaper? I mean the symlinks are literally now pointing to the original pictures. | 19:38 |
| arraybolt3[m] | So something changed them... | 19:38 |
| arraybolt3[m] | OK well I guess I'll finish getting a shared folder set up and then get this built and signed. | 19:38 |
| tsimonq2 | <arraybolt3[m]> "I wonder what on earth is..." <- My *guess* is that it's cached somewhere and the cache regeneration is racey | 19:43 |
| arraybolt3[m] | PPA upload complete, will test once it's finished building and publishing. | 19:49 |
| arraybolt3[m] | <arraybolt3[m]> "I wonder what on earth is..." <- Er... at least one symlink has changed, I didn't check both. | 19:50 |
| tsimonq2 | <arraybolt3[m]> "Er... at least one symlink has..." <- Interesting... a bug worth tracking down either way :) | 19:57 |
| arraybolt3[m] | Alright, back. | 20:53 |
| tsimonq2 | Note: I have to be in bed by 8 PM or soon thereafter to get up in time for my passport appointment. Please get things to me by 6 PM. Thanks! | 21:15 |
| arraybolt3[m] | Prepping the VM now. | 21:16 |
| arraybolt3[m] | In fact I think I'm to the point where I do the upgrade and PPA-add. | 21:16 |
| arraybolt3[m] | Simon Quigley: Alright, everything's working! I'm going to push to Gitea directly onto the ubuntu/lunar branch if that's OK with you. | 21:23 |
| arraybolt3[m] | (Just dawned on me that ubuntu/lunar isn't a branch yet, so push, here we come!) | 21:24 |
| arraybolt3[m] | Simon Quigley: https://git.lubuntu.me/Lubuntu/lubuntu-meta/src/branch/ubuntu/lunar | 21:25 |
| arraybolt3[m] | Builds Lintian-clean, upgrades with zero issues. | 21:25 |
| arraybolt3[m] | Also I think I found the wallpaper problem - there's a Lunar update to lubuntu-artwork. I'm guessing they accidentally reverted something. | 21:26 |
| tsimonq2 | arraybolt3[m]: This changelog message is misleading... Conflicts not Breaks :) https://git.lubuntu.me/Lubuntu/lubuntu-meta/commit/bacb1e67a749a5a0dc1fa73c6b0eb79390e5b61e | 21:27 |
| -ubot93:#lubuntu-devel- Commit bacb1e6 in Lubuntu/lubuntu-meta "Dropped obsolete Breaks/Replaces/Provides" | 21:27 | |
| arraybolt3[m] | :facepalm: Fixing. | 21:27 |
| tsimonq2 | Otherwise LGTM, will update before EOD | 21:28 |
| tsimonq2 | Pop quiz: when do you tag the final upload? | 21:28 |
| tsimonq2 | (With Git) | 21:28 |
| arraybolt3[m] | Alright, finished pushing. | 21:28 |
| arraybolt3[m] | tsimonq2: Looks like at the very end from what I'm seeing previously in the Git repo, so my last commit should be tagged. | 21:29 |
| arraybolt3[m] | So at the end of preparing for an new release. | 21:29 |
| arraybolt3[m] | Forgot to do that. | 21:29 |
| tsimonq2 | arraybolt3[m]: Right, but when? | 21:29 |
| arraybolt3[m] | Vague wording, sorry. When you make the final commit for a new release. | 21:29 |
| tsimonq2 | The mantra I've been "raised on" is, wait to push the tag until the upload is accepted into the archive | 21:30 |
| arraybolt3[m] | Oh, OK, that makes sense. | 21:30 |
| tsimonq2 | Makes it easier to iterate on if your sponsor/the queue reviewer wants changes made :) | 21:30 |
| arraybolt3[m] | Because there could be more changes you need to make. | 21:31 |
| arraybolt3[m] | So I was right, but also am premature in going to create the tag. | 21:31 |
| tsimonq2 | I mean, you can create the tag locally, that isn't an issue | 21:31 |
| tsimonq2 | I'd just wait to push the tag :) | 21:31 |
| arraybolt3[m] | And also maybe fix my mangled tag mess that I had locally earlier. | 21:32 |
| arraybolt3[m] | (That git fetch bit me hard, thankfully I didn't have to push the tags and so now I can just reclone the repo and have all that junk fall off.) | 21:32 |
| arraybolt3[m] | Which does indeed appear to have worked. | 21:32 |
| arraybolt3[m] | Hey I did create an ubuntu/kinetic branch - should I be tagging and pushing that too? | 21:33 |
| tsimonq2 | arraybolt3[m]: Perfect :) | 21:36 |
| tsimonq2 | arraybolt3[m]: Here's the most logical workflow I've been able to come up with... | 21:37 |
| tsimonq2 | There is a point, somewhere in the Git history for all of our packages, a point at which it is completely in-sync with the archive | 21:38 |
| tsimonq2 | Before tagging is done, you should be checking to make sure that the archive exactly matches your local copy, minus the .git folder | 21:39 |
| tsimonq2 | I never wrote that down I don't think, so it never became policy for us all, so we've let that slide | 21:39 |
| tsimonq2 | When you're importing from the archive, keep that in mind. Any change a human commits to the repo should be self-contained. You aren't personally making any other changes besides importing, but you want that tag history for every upload that's ever hit the archive | 21:40 |
| tsimonq2 | I would find the "last known good point". If it's an older release, start with that branch and import uploads one-by-one. Launchpad always has copies of diffs and filterdiff is a thing, so this could (and probably should) be automated | 21:42 |
| tsimonq2 | When you've hit the end of a release, create a new branch from the one you've just finished. Work your way up again | 21:42 |
| tsimonq2 | Repeat until you get to $now | 21:42 |
| tsimonq2 | That is the most correct way to do it | 21:43 |
| tsimonq2 | As for the quickest way to do it? As long as it roughly follows that logic, get creative :) | 21:44 |
| arraybolt3[m] | Sounds like pretty much what I did except for the tagging. I made sure I was perfectly exactly in sync with the archive for all of the Kinetic commits. | 21:47 |
| arraybolt3[m] | So now I can just tag as needed and push the branch and the tags. | 21:47 |
| tsimonq2 | Correct, except for the latest tag :) | 21:47 |
| arraybolt3[m] | In Kinetic? | 21:48 |
| tsimonq2 | Did all of that make sense? | 21:48 |
| arraybolt3[m] | The latest tag in Lunar, yes, hold off on, but the latest in Kinetic is already said and done. | 21:48 |
| tsimonq2 | arraybolt3[m]: If it's said and done, yeah, tag and push :) | 21:48 |
| arraybolt3[m] | <tsimonq2> "I would find the "last known..." <- OK and I started from Jammy, imported what needed it, so I also need to add a Jammy tag. And I made sure that I had the archive and Git repo in perfect sync. | 21:50 |
| arraybolt3[m] | Alrighty, here goes! | 21:50 |
| tsimonq2 | arraybolt3[m]: Remember the scripts I posted last night :) | 21:51 |
| arraybolt3[m] | Nice. I took a slightly more "sledgehammer" approach to it and made sure that all the changes from the archive were applied properly by just nuking everything locally and then copying the right part of the archive to it, then using gitk to make sure it looked sane. | 21:52 |
| arraybolt3[m] | But I should save those scripts somewhere for future work. | 21:52 |
| arraybolt3[m] | So weird. QTerminal just randomly stops letting paste functionality work out of nowhere. | 21:55 |
| arraybolt3[m] | Ssssssshhoooooott. I just accidentally did "git push --tags" from within the tag-mangled repo. | 21:58 |
| arraybolt3[m] | It doesn't look like all the garbage tags are even visible, but crud. Do I just delete them and then redo git push --tags? | 21:58 |
| tsimonq2 | arraybolt3[m]: Nope, because that won't delete the remote copy... | 21:59 |
| arraybolt3[m] | Dagnabbit. This is what I get for having two repos with otherwise identical data in the same folder. | 21:59 |
| tsimonq2 | Remember the `git push origin :master` trick? (DO NOT RUN THIS, EVER, DEAR READERS) | 21:59 |
| tsimonq2 | Yeah that works with tags too | 21:59 |
| tsimonq2 | And when you pushed it, git should have given you an output of what tags it pushed, yeah? | 22:00 |
| arraybolt3[m] | Yes. | 22:00 |
| tsimonq2 | Alright, give me a second to find some info on querying remote tags from the CLI to see if you have any weirdness going on | 22:00 |
| tsimonq2 | https://stackoverflow.com/questions/25984310/how-to-see-remote-tags | 22:01 |
| arraybolt3[m] | All that went wrong is when I did a "git fetch upstream" to get the necessary info from Launchpad, it pulled in a WHOPPING LOAD of tags from Launchpad. Now they all just got pushed to Gitea. Tons of "imported" and "accepted" things. | 22:01 |
| tsimonq2 | Nice job ;) don't worry, totally fixable | 22:02 |
| arraybolt3[m] | The output of "git ls-remote --tags origin" | 22:02 |
| arraybolt3[m] | Hold on, it was too big to paste into Matrix. | 22:03 |
| arraybolt3[m] | https://termbin.com/hzv6 | 22:03 |
| tsimonq2 | First things first, the import tags will be easy | 22:04 |
| tsimonq2 | https://stackoverflow.com/questions/45818817/delete-multiple-git-remote-tags-and-push-once | 22:04 |
| tsimonq2 | We don't have any import/ tags, I'd use a variation of that to get rid of all of them in GiTea | 22:05 |
| arraybolt3[m] | So `git push -d origin $(git tag -l "import/*")`? | 22:05 |
| arraybolt3[m] | (DANGEROUS COMMAND ABOVE FOR BACKLOG READERS) | 22:05 |
| tsimonq2 | If you want practice beforehand, you can always make a copy of the directory, add the copy path as a remote to your original Git repo, and test the deletions there | 22:05 |
| tsimonq2 | arraybolt3[m]: That seems about right, go for it | 22:06 |
| arraybolt3[m] | That's smart. I can also just run the git tag command to see what it's going to try and remove. | 22:06 |
| arraybolt3[m] | Alright, looks good, command is running. | 22:07 |
| tsimonq2 | I have to use that trick for $dayjob :P | 22:07 |
| arraybolt3[m] | YAY it worked. | 22:07 |
| tsimonq2 | NICE | 22:07 |
| arraybolt3[m] | So now rinse and repeat with "applied/", right? After making sure we don't use that prefix? | 22:08 |
| arraybolt3[m] | (I don't see that prefix in use in gitk.) | 22:08 |
| tsimonq2 | Yep :) | 22:08 |
| tsimonq2 | Also, check the ubuntu/ tags. You don't want to remove those tags if they're pointing to the right commits | 22:09 |
| arraybolt3[m] | I think every ubuntu/ tag is correct. | 22:09 |
| tsimonq2 | You do want to remove them if they're pointing to Launchpad refs | 22:09 |
| arraybolt3[m] | I will double-check to see if they need to be removed. | 22:09 |
| arraybolt3[m] | "applied/*" has now been nuked, *whew!*. Thank you for saving me from that and for not tearing me to pieces for botching like that on production. | 22:10 |
| kc2bez[m] | <tsimonq2> "I think that would be a good..." <- I concur, excellent idea. | 22:11 |
| tsimonq2 | arraybolt3[m]: Like I said before, if you're honest about your mistakes I have zero issue helping you fix them :) | 22:11 |
| arraybolt3[m] | Thank you. :) I have no problem with being honest with a mistake. Better to make the whole team work than to make ones' self look good (not to mention the fallout if it was discovered later and the fact that stuff like that is pretty difficult to hide...). | 22:12 |
| tsimonq2 | Agreed :) | 22:12 |
| tsimonq2 | Speaking of agreeing, I think we should definitely get that wallpaper stuff on the list ;) | 22:13 |
| kc2bez[m] | <arraybolt3[m]> "I wonder what on earth is..." <- The symlinks are in 2 different packages I think. Default settings and artwork. | 22:13 |
| arraybolt3[m] | kc2bez[m]: When I worked on artwork I saw both symlinks in there IIRC. | 22:13 |
| arraybolt3[m] | And it's a lubuntu-artwork update that causes the wallpapers to go wonky. | 22:14 |
| arraybolt3[m] | tsimonq2: Will add 👍️ | 22:14 |
| tsimonq2 | arraybolt3[m]: May be 16 year old Simon's fault :P | 22:16 |
| kc2bez[m] | tsimonq2: TBF I could have dove into it deeper too. | 22:17 |
| tsimonq2 | Ideally, the symlinks would only be in the artwork package, I think | 22:17 |
| kc2bez[m] | I agree. | 22:17 |
| arraybolt3[m] | Simon Quigley: OK what is this? `daa59645570c0c7a24aec21268e7706779e6089c refs/tags/ubuntu/20.04.3^{}` | 22:17 |
| arraybolt3[m] | I know that ubuntu/20.04.3 is ,and I *think* the ^ is "before", but the {} has me stumped. | 22:17 |
| arraybolt3[m] | Is that a tag that needs removed, or what even is it? | 22:18 |
| arraybolt3[m] | (It's one of only two tags that I didn't see in gitk.) | 22:18 |
| tsimonq2 | Maybe... a pointer of some kind? Try just doing `git tag` no arguments | 22:18 |
| arraybolt3[m] | I don't see it in there. | 22:19 |
| tsimonq2 | Is it a different commit ref than the one without the suffix? | 22:19 |
| arraybolt3[m] | What I did find out from the Launchpad tag intrusion nightmare is that we missed a tag in Focal, though. | 22:19 |
| arraybolt3[m] | Ooo, good question, one moment... | 22:19 |
| arraybolt3[m] | Simon Quigley: No. The commit SHA1's are identical. | 22:20 |
| tsimonq2 | Perfect, so I say ignore it | 22:21 |
| arraybolt3[m] | (As determined by doing "git checkout ubuntu/20.04.3" then using gitk to find the commit SHA1.) | 22:21 |
| arraybolt3[m] | OK, that leaves only ubuntu/20.04.7, which actually has a spot it belongs, so I guess I'll get that in carefully. | 22:21 |
| arraybolt3[m] | deep sigh of relief All the bad tags are gone. | 22:22 |
| tsimonq2 | Sweeeeet :) | 22:22 |
| arraybolt3[m] | And the one missing tag has been added. So hey, after nearly giving myself a heart attack and having to be bailed out by an expert developer, we actually fixed a minor issue! 🥴 | 22:26 |
| arraybolt3[m] | Simon Quigley: OK, so just so I make sure I know what else I need to do, I saw you already did the Calamares desktop file, so I'm guessing you probably also finished the rest of the list while I was fighting with lubuntu-meta. Is there anything left on the list I need to do? | 22:28 |
| arraybolt3[m] | Like, did we already update the default seed branch in Launchpad and Gitea? ANd whatever needed done with repo-list and the setup-phab script? And the Github default branches for the mirrored repositories? (I don't think I even can do that last one.) | 22:30 |
| arraybolt3[m] | (I don't know if the use of setup-phab is still a thing.) | 22:30 |
| tsimonq2 | arraybolt3[m]: Nope, it isn't | 22:32 |
| arraybolt3[m] | IT needs rewritten for Gitea doesn't it? | 22:32 |
| arraybolt3[m] | It looks to me like it goes through all the repos and makes ubuntu/RELEASE branches (in our case, ubuntu/lunar). | 22:32 |
| tsimonq2 | Yeah :) | 22:33 |
| tsimonq2 | arraybolt3[m]: I think that's about it, sure | 22:33 |
| arraybolt3[m] | K. That looks to me like the last thing I can do that needs done as far as bootstrapping, except for maybe lintian and devscripts. | 22:34 |
| arraybolt3[m] | I don't know how to change the default branch for the seeds if it hasn't been changed already, and I don't think I have proper access to GitHub to change branches (and I have no idea what I would be doing there and probably shouldn't have that access yet), but the creation of the ubuntu/lunar branches looks like something that's doable. | 22:34 |
| tsimonq2 | arraybolt3[m]: Those are all done :) | 22:36 |
| tsimonq2 | arraybolt3[m]: Do what you can and let me know what's left :) | 22:36 |
| kc2bez[m] | The seed branch is usually done by a release team member. If you look in phab or LP you can see it changed already. | 22:37 |
| arraybolt3[m] | kc2bez[m]: Makes sense. Where exactly would I look to see that change? | 22:38 |
| kc2bez[m] | It isn't set as default but the branch is created | 22:39 |
| arraybolt3[m] | Simon Quigley: lubuntu-meta has been completed. I believe it is ready for sponsoring. (You probably already knew that but just in case, I thought I should probably say it so that nothing got missed on account of me.) | 22:39 |
| kc2bez[m] | https://phab.lubuntu.me/source/seed/browse/lunar/ | 22:40 |
| kc2bez[m] | You will see if you clone that it actually clones the LP repo | 22:41 |
| arraybolt3[m] | Ah, OK. So in LP it needs changed. So I'm guessing that means I need to make another MP. | 22:41 |
| arraybolt3[m] | Actually, it looks like "default branch" isn't something handled by Git AFAICT. I don't think I have access to change that, and I probably shouldn't change it in Gitea since that should be automatically changes when LP changes, right? | 22:43 |
| kc2bez[m] | I don't think it is a function of git, no. | 22:43 |
| kc2bez[m] | I've always adjusted it in the web interface. | 22:43 |
| arraybolt3[m] | OK. | 22:43 |
| kc2bez[m] | Maybe Simon Quigley knows some git-fu | 22:44 |
| arraybolt3[m] | Then I guess the last thing to be done that I can do is the setup-gitea script. | 22:44 |
| tsimonq2 | refs/heads/head ? | 22:44 |
| tsimonq2 | git symbolic-ref HEAD refs/heads/ubuntu/lunar && git push origin HEAD:HEAD ??? | 22:46 |
| arraybolt3[m] | https://stackoverflow.com/questions/51274430/change-from-master-to-a-new-default-branch-git That looks somewhat like the right idea to me. | 22:46 |
| arraybolt3[m] | Though I don't know if an MP is going to modify that since that's not a part of any git branch I don't think. | 22:46 |
| arraybolt3[m] | I might modify the default branch of my own repo but I don't think that will carry over. Plus the web UI sounds way easier. | 22:47 |
| tsimonq2 | arraybolt3[m]: True, true :) | 22:48 |
| tsimonq2 | One of these days you'll be in ~lubuntu-dev so you can Just Do It :P | 22:48 |
| kc2bez[m] | tsimonq2: This ^ | 22:48 |
| arraybolt3[m] | 👍️ That was my question, whether I had access to do it or not and how to do it. If I don't have access, problem solved. | 22:48 |
| arraybolt3[m] | OK, life calls again. I'll be back later (probably tonight) to do the setup-gitea script, and then I think everything I can do will have been done. All that I think is left is the default branch for the seeds and the Github branches for the mirrored repositories. | 22:50 |
| * arraybolt3[m] goes afk, thank you guys and be back soon! | 22:50 | |
| kc2bez[m] | Take care, I have to go to. | 22:50 |
| === guiverc2 is now known as guiverc | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!