[04:30] <vpa1977> bdmurray: java-common migrated, thanks again !!!!!
[09:02] <Skuggen> mdeslaur: Hi! Sorry, I hadn't seen that before. I'll ask
[09:04] <Skuggen> coreycb: Great work on the router issue. The full revert should be reasonably safe as the router code is quite independent of server. Dev may have found the underlying issue, so we might also be able to get a targeted patch out
[10:07] <slyon> juliank: is there any way to use 'apt search ..' to only search for packages in the 'main' component?
[10:09] <juliank> No
[10:09] <juliank> slyon: You can use apt list with patterns perhaps
[10:09] <juliank> slyon: but you can't search descriptions in there
[10:10] <slyon> ok.. I'd like to search descriptions from 'main'
[10:10] <juliank> To find all packages in main, you could use ?not(?section(/))
[10:10] <slyon> searching just package names should already help, though.
[10:10] <slyon> let me try that
[10:11] <juliank> I need to implement ?description
[10:13] <juliank> Why is there no component pattern too
[10:13] <juliank> I want to search ?component(main), not ?not(?section(/))
[10:13] <juliank> or ?component(restricted) instead of ?section(restricted/.*)
[10:14] <juliank> I know I know we should just implement sql on top of the apt cache
[10:14] <slyon> right, a component pattern would be nice and searching the description, too.
[10:14] <slyon> but I think the ?not(?section(/)) does the trick for me now. thanks a lot!
[10:15] <slyon> my output is relatively small, to I can filter the rest manually
[10:16] <juliank> Can also use apt-cache show ?not(?section(/)) | grep-dctrl -rFDescription regex
[10:19] <juliank> ok there is a bug *somewhere*
[10:20] <juliank> oh yeah apt show splits on / first and hence parses ?not(?section( as the pattern, and )) as the release.
[11:49] <mdeslaur> Skuggen: thanks!
[11:58] <Skuggen> mdeslaur: coreycb: I have another patch for the router bug (basically, it couldn't handle 20+ long passwords over unencrypted channels). Just checking the build, and I can upload to the bug (Jan was able to reproduce the issue in the bug and tested against that)
[12:06] <mdeslaur> Skuggen: great!
[13:39] <coreycb> Skuggen: great, thank you
[13:41] <coreycb> Skuggen: that would probably be worth cutting a new patch release to include the bug fix
[19:07] <arraybolt3> Could I ask for some autopkgtest retries for lubuntu-update-notifier? I see it's stuck in PyQt5, but when I do an autopkgtest locally with all of -proposed in use, the test passes. List of URLs: https://paste.ubuntu.com/p/dzVYBFB5Br/
[19:08] <arraybolt3> (I added the "&all-proposed=1" to the end of each one, I think that's the way to do that but I don't know for sure, so please review and tell me if I messed that up.)
[19:10] <vorlon> arraybolt3: running
[19:10] <arraybolt3> Thanks!
[19:10] <vorlon> fwiw I'm just running the following pipeline rather than copying the above pastebin: retry-autopkgtest-regressions --blocks pyqt5 --all-proposed | xargs -rn1 -P10 curl --cookie ~/.cache/autopkgtest.cookie -o /dev/null --silent --head --write-out '%{url_effective}: %{http_code}\n'
[19:12] <arraybolt3> Wow, that looks complicated :P
[19:12] <arraybolt3> I think I can see the general idea of how that works, though. Thanks for sharing!
[19:13] <vorlon> the bit after the pipe is documented in the help output of retry-autopkgtest-regressions
[19:14] <arraybolt3> That's part of ubuntu-archive-tools (or something similarly named, can't remember it rn), right?
[19:15] <arraybolt3> Nevermind, found it.
[19:17] <vorlon> yah
[19:20] <MacSlow> Greetings everybody!
[19:21] <sarnold> wb MacSlow
[19:21] <MacSlow> hey sarnold
[20:33] <MacSlow> bbl