[07:16:57] Krinkle: yes. T226797#8458653 (the "next step" part) would be the low-effort fix (maybe; depends on the exact rules used by browsers to differentiate between tracking and SSO). But we might be better off just rewriting how SUL authentication works. [07:16:58] T226797: CentralAuth login session and auto-login no longer work across wikis in Safari and Firefox - https://phabricator.wikimedia.org/T226797 [12:19:10] Can someone +2 a patch fixing some chmod of config relates files in master please? https://gerrit.wikimedia.org/r/961372 [12:20:22] done [12:20:26] ta [12:21:01] I can't obviously see when they changed [12:27:10] the yaml file was in https://gerrit.wikimedia.org/r/c/mediawiki/core/+/828553 it seems [12:27:31] (found with `git log -p` and then reverse searching for 755 in less… mode changes don’t seem to be covered by any of the --diff-filter letters) [12:28:09] ok, the other three files also happened in that commit apparently [12:28:42] wait, you commented on that change already last year and fixed it in https://gerrit.wikimedia.org/r/c/mediawiki/core/+/840243 [12:28:44] did they change again since then? [12:28:58] ummm [12:29:00] yeah [12:29:01] lmao [12:29:09] yup, https://gerrit.wikimedia.org/r/c/mediawiki/core/+/900640 [12:29:33] ta [12:29:44] (where it was apparently pointed out but not followed up on 🤷) [12:30:01] someone™ should probably look into why it keeps happening *ducks* [12:30:07] lol [12:30:15] something something dev envs something something [12:30:27] * Lucas_WMDE blames windows just in case [17:12:10] James_F: want to finish https://gerrit.wikimedia.org/r/q/topic:T343994 today? :) [17:12:10] T343994: OutputPage::setPageTitle() should call Message::escaped() when given a Message - https://phabricator.wikimedia.org/T343994 [17:53:07] MatmaRex: Despite your non-compliant commit messages, I guess I'll merge them. :-) [17:56:18] heh, thanks [17:57:09] MatmaRex: C+2'ed the lot for getDescription(); https://gerrit.wikimedia.org/r/c/mediawiki/core/+/947895 probably needs more patches though? [17:57:51] oh, does it? :( [17:59:56] yeah, it does. i haven't checked it before, i didn't really expect a lot of uses [18:01:13] https://codesearch.wmcloud.org/deployed/?q=setPageTitle%5C%28.%2A%5Cb%28msg%7CwfMessage%29%5Cb [21:07:14] TimStarling: when you have a couple of minutes, could you reply to https://phabricator.wikimedia.org/T281325#9193098 ? thanks [21:24:04] Forwarding tgr|away 's message on Slack: "Can we change H425 from "project tags include" to "project tags added"? It's really confusing that years old tasks land in our inbox whenever e.g. someone subscribes to them." [21:25:15] Krinkle, I also realized when anyone in MW platform comments on a task on phab (hence subscribed), irrespective of the task age, our board tag gets added by default. Maybe we want to revisit this per Gergo's request? [22:59:02] I'm not aware of tagging happening based on who comments or is subscribed. That may be coincidence. The tagging of old tasks is intentional on my part, but I'm fine revisiting it for sure. The reason for it is to avoid relying on individuals like myself to take care of old tasks when they become relevant, and instead gradually and automatically triage them as people interact with tasks for areas in areas we maintain. We could vary it and [22:59:02] eg do that for RL and BagO only, and for Auth limit it to new tasks. [23:53:27] Lucas_WMDE: one would think minus-x would catch it. isn't that why we have that? that way at least it fails CI.