21:00:08 #startmeeting swift 21:00:08 Meeting started Wed Jan 8 21:00:08 2025 UTC and is due to finish in 60 minutes. The chair is timburke. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:08 The meeting name has been set to 'swift' 21:00:15 who's here for the swift meeting? 21:01:25 o/ 21:02:41 i know mattoliver is still out -- kota was around earlier this week, but doesn't seem logged in now 21:03:02 well, may as well start :-) 21:03:20 i actually updated the agenda! 21:03:23 #link https://wiki.openstack.org/wiki/Meetings/Swift 21:03:27 first up 21:03:39 #topic welcome jianjian to the core team 21:04:30 the cores had all been talking about jianjian's good reviews, and wanted to extend an invitation to be on the upstream core team 21:05:00 yesterday i reached out, he accepted, and i see we've already merged some code because of it! 21:05:06 🎉 21:05:42 Thanks, Tim! 😇 I am very excited. 21:05:59 glad to have you with us doing more and more upstream work :-) 21:06:39 i think this may be the first time i've had occasion to update the core list, so i'm excited too :D 21:06:49 after working on swift for more than 2.5 years, I feel the more I work on, the more I found out I need to learn more. will use +2 wisely and look forward to contributing more into swift universe 💪 21:07:13 next up 21:07:19 #topic py2 is dead! 21:07:39 thanks for your kind word, Tim, and thanks other core devs for helping me out all the time! 21:08:08 coming up on the new year, i was a little impulsive and went ahead and merged p 853590 21:08:09 https://review.opendev.org/c/openstack/swift/+/853590 - swift - Drop py2 support (MERGED) - 23 patch sets 21:08:37 yeah, let's use python 3! 21:09:04 it's been a long time coming, i think we're finally ready, and i'm so excited to not have that albatross around our neck for another year 21:09:22 🎉 21:09:47 that patch just updated packaging and gate jobs 21:09:54 #link https://review.opendev.org/c/openstack/swift/+/853697 21:09:55 patch 853697 - swift - Remove py2-only code paths - 13 patch sets 21:10:15 does a lot to get rid of the old py2-only code paths 21:10:35 (including removing our dependency on six) 21:11:31 there's other clean up that can be done, but i figured a net-reduction of ~1400 LoC is a good place to start 21:12:13 though it looks like i need to fix a pep8 error :-/ 21:12:17 lol, it's a huge patch 21:12:54 yeah... sorry everybody! but zaitcev has already +2ed it, and i know acoles is interested in taking a look sooner rather than later, too 21:13:24 (i'm guessing so he can deal with the fallout on feature/mpu sooner rather than later) 21:13:50 seems there is a small flake8 check error, https://zuul.opendev.org/t/openstack/build/125a50c965da48679b126872236a66d2 21:14:17 but there's at least one other project we might want to look at, too! 21:14:18 yeah, we will be happy to rebase our patches on top of it 21:14:30 #link https://review.opendev.org/c/openstack/pyeclib/+/938098 21:14:30 patch 938098 - pyeclib - Drop support for Python 2.7, 3.5-3.8 - 4 patch sets 21:15:24 that one also drops some older py3 versions, mainly just because they're EOLed 21:16:30 nice, so 3.8 is also an old version already 21:16:54 but i feel pretty OK about it, especially given our infrequent updates to pyeclib -- i think anyone still on old python versions will be content to remain on 1.6.4 for a while 21:17:26 py39 will EOL this October 21:17:51 yes, "Python 3.8 reached its end-of-life (EOL) on October 7, 2024" 21:17:52 woww 21:18:14 there's a reason we really need to look at upgrading our py3 again already, despite having only *just* finished the last transition 21:18:47 ack, they are EOL already 21:19:58 there are some other follow-on pyeclib patches stacked on top of that to clear out some cruft from the old versions 21:20:58 big one is p 938101 -- get to take out a whole module that was only included for the sake of py2! 21:20:59 https://review.opendev.org/c/openstack/pyeclib/+/938101 - pyeclib - Get rid of custom enum module - 3 patch sets 21:21:36 next up 21:21:41 #topic swift release 21:21:56 i've been thinking that it's about time to get a new one out! 21:22:30 especially since we merged https://review.opendev.org/c/openstack/swift/+/937493 21:22:30 patch 937493 - swift - xprofile: Stop using eval() (MERGED) - 1 patch set 21:23:35 i'll try to get some release notes together this week, so keep an eye out for the patch! 21:23:50 Nice! 21:23:58 that's all i've got for this week 21:24:03 #topic open discussion 21:24:33 anything else we should bring up? or i can let you go a little early ;-) 21:24:40 big thanks to Tim for working on those py2/py3 upstream patches, it's a lot of work! 21:24:56 i've been itching to do it for *so long* 21:26:01 literally the whole time i've been PTL has been a long slog of trying to make this transition happen, so i'm so glad it's nearly there 21:26:54 would love not to worry about py2 anymore, I will also look at those patches 21:27:01 no other topic from me 21:27:28 oh! but speaking of! i finally figured out a probe test issue that lead me to do this: https://github.com/openstack/swift/blob/master/test/probe/test_reconstructor_rebuild.py#L529-L537 21:28:08 and in the process discovered another long-running py3 http parsing bug :-( 21:28:35 but at least i've got a fix! https://review.opendev.org/c/openstack/swift/+/938631 21:28:36 patch 938631 - swift - Fix unicode-header-name handling - 1 patch set 21:29:41 a long-running bug... wow 21:29:48 if you get a chance, i'd appreciate some eyes on it 21:30:27 yeah, interested to look into it 21:30:29 it's https://bugs.python.org/issue37093 all over again, just that i didn't work around it completely before 21:30:33 thanks! 21:30:50 all right, that's all i've got, too 21:31:12 happy new year! it's already looking like the best year yet for swift :-) 21:31:24 thank you for coming, and thank you for working on swift! 21:31:29 #endmeeting