[00:00:04] RoanKattouw: Can you put the Thanks extension on it whilst you're at it? :-D [00:00:24] ori-l: do you have some time to review https://gerrit.wikimedia.org/r/#/c/87311/ ? (API module for MassMessage) [00:00:39] Sure [00:01:17] RoanKattouw: oh right, was distracted by your other (and everyone eles's) requests [00:01:39] legoktm: OK; by when? [00:01:45] (03PS4) 10Bsitu: auto-expand text-area as content grows [extensions/Flow] - 10https://gerrit.wikimedia.org/r/94285 (owner: 10Matthias Mullie) [00:01:51] RoanKattouw: wed at 2pm? [00:02:02] Sure [00:02:16] ori-l: before next wmf branch is made, so thursday [00:02:32] OK, I'll do it this weekend, probably Sunday [00:02:45] thanks [00:02:52] feel free to nag if you see me on IRC [00:02:56] it'd be appreciated in fact [00:02:58] (03PS5) 10CSteipp: Improve user-facing error messages [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/93121 [00:03:09] will do :) [00:03:16] (03CR) 10CSteipp: "PS5: manual rebase" [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/93121 (owner: 10CSteipp) [00:03:24] ori-l: Speaking of nagging, did you have to do any changes on your end for all those new and renamed ve.track events? [00:03:26] (03CR) 10Bsitu: "PS4: Hide the vertical scrollbar" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/94285 (owner: 10Matthias Mullie) [00:04:25] RoanKattouw: added, thanks man [00:04:37] Awesome [00:07:50] (03CR) 10Kaldari: [C: 04-1] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/92668 (owner: 10MaxSem) [00:10:28] (03PS1) 10GWicke: Update Parsoid to 3c18560f0 [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94450 [00:11:57] (03PS6) 10CSteipp: Improve user-facing error messages [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/93121 [00:13:03] gwicke: Is Parsoid now being pushed as part of the wmfN train? [00:13:32] James_F: normally it is [00:13:41] gwicke: Ah. [00:13:45] RoanKattouw: is there anything controversial about this patch? > https://gerrit.wikimedia.org/r/#/c/94248/ [00:13:51] we have a fix for the job runner that we'd like to get out asap though [00:14:01] the reason for the suspiciously low load [00:14:01] gwicke: The things that pass me by due to the private deploy. :-( [00:14:08] Ah. [00:14:10] Yeah, that's bad. [00:14:12] hm? [00:14:21] this was deployed yesterday with the normal train [00:14:28] (03PS3) 10Jdlrobson: Create prepareQuickTemplate function [core] - 10https://gerrit.wikimedia.org/r/94248 [00:14:28] hard to figure out because of the time lag [00:14:53] * James_F nods. [00:15:03] James_F: this is the php code, not the parsoid service [00:15:15] Yeah, I see. [00:15:23] we plan to move the service out of the extension, just waiting for a repo to finally be created for us [00:15:47] James_F: also, SAL has all the hashes for the service [00:15:52] just grep for 'parsoid' [00:16:19] gwicke: I know. [00:17:18] so not very private. [00:17:39] "Private" == no ping for James_F in IRC when it happens. :-) [00:17:42] jdlrobson: Probably not, it's just that it reminds people of how horrible that code is [00:17:51] Not secret. :-) [00:18:05] gwicke: If you want a repo I can give you one [00:18:08] (03PS1) 10Aaron Schulz: Made cleanupUploadStash use batching for the last two loops [core] - 10https://gerrit.wikimedia.org/r/94451 [00:18:14] RoanKattouw: we have requests in already [00:18:23] With whom? [00:18:30] there is a mw request page [00:18:59] right [00:19:17] https://www.mediawiki.org/wiki/Git/New_repositories/Requests [00:19:39] ah, I see that QChris replied there and complained about services being too generic [00:20:14] (03CR) 10Jdlrobson: [C: 031] Adding a quotation marks message to core. [core] - 10https://gerrit.wikimedia.org/r/93880 (owner: 10Kaldari) [00:20:29] (03PS1) 10Bartosz Dziewoński: SpecialRecentChanges: Remove '@todo Uses radio buttons (HASHAR)' [core] - 10https://gerrit.wikimedia.org/r/94452 [00:20:30] RoanKattouw: if you can convince him or just do it then that would be great ;) [00:20:39] gwicke: I think services/ is a good name. [00:21:11] yeah, the reason is explicitly that it is more generally useful code [00:21:16] gwicke: Alternatively it could be mediawiki/services/parsoid and mediawiki/services/rashomon? [00:21:34] rashomon is relatively independent of mw [00:21:34] gwicke: So new tree under mediawiki/ [00:21:34] gwicke: I don't usually create repos, but when I do, I create them for my team ;) [00:21:53] (03PS1) 10Aaron Schulz: Made cleanupUploadStash use batching for the last two loops [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94453 [00:21:55] (03PS6) 10MaxSem: DifferenceEngine cleanup [core] - 10https://gerrit.wikimedia.org/r/92668 [00:22:00] (03CR) 10Aaron Schulz: [C: 032] Made cleanupUploadStash use batching for the last two loops [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94453 (owner: 10Aaron Schulz) [00:22:02] gwicke: Sure, like VE - but primary use is the MW use, at least at first? [00:22:09] * James_F shrugs. [00:22:11] So I'd be happy to create these but I'll want Chad to sign off on the naming [00:22:25] James_F: it should be in a separate repo for access right etc reasons [00:22:33] and separate packaging [00:22:51] mediawiki can pull it in as a subrepo, just as with extensions [00:22:58] gwicke: Separate packaging, sure; for access rights, though, those are normally common across repos regardless, unless there are good reasons. [00:23:06] if that makes sense (which I doubt) [00:23:09] ACLs inherit by default but can be overridden [00:23:34] we don't want to deploy rashomon to the PHP cluster [00:24:44] James_F: in gerrit rights are per repo [00:25:04] gwicke: Err. Yes? [00:25:24] I think you guys are talking past each other [00:25:24] gwicke: But observe how they inherit from All Projects except in a few, exceptional cases. [00:25:32] When James said "new tree under mediawiki/" he meant a tree of repos [00:25:32] yup, merge rights in Parsoid don't imply merge rights in core etc [00:25:43] Indeed. [00:26:21] if they are separate then they also will have separate rights [00:26:30] unless those are all defined at the root of the tree [00:28:12] I guess in the Parsoid I could see the reason for preferring a placement in mediawiki/services, as it does parse MediaWiki markup and calls back into the API [00:28:23] Rashomon otoh is fairly generic [00:28:56] I hope that we'll develop more generic services too [00:29:36] (03Merged) 10jenkins-bot: Made cleanupUploadStash use batching for the last two loops [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94453 (owner: 10Aaron Schulz) [00:30:17] (03PS1) 10Aaron Schulz: Made cleanupUploadStash use batching for the last two loops [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94454 [00:30:21] gwicke: Go with mediawiki/services/rashomon for now, and move it later once you can convince people of the wider need? It's easy enough to shift, and saves time now that could be more profitably spent on coding than arguing. :-) [00:31:11] James_F: it should be on the same level as our squid repos for example [00:31:26] gwicke: It's not me you have to convince. :-) [00:31:33] analytics have their own top-level [00:31:47] as has pywikibot [00:31:58] so I don't see a reason really [00:32:58] well, 'services' as a top level is a bit unclear [00:33:07] what is it a service for? [00:33:48] well, since we're talking about git and gerrit-y things...who handles https://www.mediawiki.org/wiki/Gerrit/Project_ownership requests nowadays, if anyone? or do I have to wait until my requests are old enough to legally drive? [00:34:02] legoktm: what is 'analytics' for? [00:34:21] projects related to the analytics team [00:34:23] (03PS7) 10CSteipp: Improve user-facing error messages [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/93121 [00:34:30] ashley: ^d does. [00:34:31] i'm guessing [00:34:37] (03PS1) 10Aaron Schulz: Speed up thumbnail purging a bit for stash script [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94455 [00:34:51] legoktm: yeah, similarly with services it would primarily be services [00:34:59] there's a services team? [00:35:15] legoktm: Yes, Platform. [00:35:17] * James_F gdrs. [00:35:26] heh [00:35:29] * ashley pokes ^d with a link to https://www.mediawiki.org/wiki/Gerrit/Project_ownership [00:35:37] James_F: cheers :-) [00:36:05] ashley: Always. :-) [00:36:15] gwicke: I doubt its worth bikeshedding over, but services isn't a great top level name. if its easy to move repos, then just create it and let people argue afterwards :P [00:36:27] <^d> It's not easy to move repos. [00:36:32] <^d> And I said I didn't like "services" [00:36:38] <^d> But nobody listens to the gerrit admin :) [00:36:46] well, then bikeshed first! [00:36:59] ^d: I suggested mediawiki/services/parsoid for the Parsoid MediaWiki service. [00:37:01] ^d: what is your preference for non-mw-specific web services? [00:37:18] <^d> "services" is so incredibly vague. [00:37:28] ^d: Unlike "Analytics". :-) [00:37:34] (03CR) 10Aaron Schulz: [C: 032] Speed up thumbnail purging a bit for stash script [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94455 (owner: 10Aaron Schulz) [00:37:37] <^d> James_F: Didn't like that either ;-) [00:37:40] gwicke: what are the repos for? [00:37:45] it is not much more precise than 'extensions', yeah ;) [00:37:47] ^d: Didn't veto it, however. :-) [00:37:57] * James_F would have. [00:38:12] <^d> Meh, why have structure at all. [00:38:17] <^d> Let's just put everything in the top level. [00:38:24] (03PS1) 10Bartosz Dziewoński: SpecialWatchlist: Minor cleanup [core] - 10https://gerrit.wikimedia.org/r/94456 [00:38:32] Eww. [00:39:04] well, I think I could live with mediawiki/services too [00:39:40] for some external user having a separate repository matters more than its path [00:40:00] Yeah. [00:40:37] <^d> ashley: https://www.mediawiki.org/w/index.php?title=Gerrit%2FProject_ownership&diff=816738&oldid=816728 [00:41:12] (03Merged) 10jenkins-bot: Speed up thumbnail purging a bit for stash script [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94455 (owner: 10Aaron Schulz) [00:41:36] ^d: muchas gracias :) and the other request, too, pretty please? :D [00:41:50] <^d> Oh dur, missed that. [00:41:51] <^d> Sec. [00:42:35] (03CR) 10Bartosz Dziewoński: [C: 04-1] SpecialWatchlist: Minor cleanup [core] - 10https://gerrit.wikimedia.org/r/94456 (owner: 10Bartosz Dziewoński) [00:42:41] <^d> ashley: Actually, you should've already had that one. ACL was fucked up. Fixed. [00:42:51] RoanKattouw || ^d: if you'd like to create those two repos then that would be great of course ;) [00:43:10] ^d: you're my hero (once again)! [00:43:15] see 'rashomon' and 'parsoid' in https://www.mediawiki.org/wiki/Git/New_repositories/Requests [00:43:44] I'll do that later today [00:44:08] (03PS2) 10Bartosz Dziewoński: SpecialWatchlist: Minor cleanup [core] - 10https://gerrit.wikimedia.org/r/94456 [00:44:17] gwicke: I take it you'll want to move the js/ dir out of the Parsoid extension into this new repo? [00:44:28] s/later today/probably on Tuesday/ , actually [00:44:56] <^d> ashley: You're welcome :) [00:46:00] (03CR) 10Legoktm: [C: 032] SpecialWatchlist: Minor cleanup [core] - 10https://gerrit.wikimedia.org/r/94456 (owner: 10Bartosz Dziewoński) [00:46:51] RoanKattouw: yes [00:47:03] but with history, so just import the full repo [00:47:12] we can then rename stuff in the copy [00:47:20] gwicke: Import and then rename and prune? [00:47:22] * James_F nods. [00:47:56] I figured I would import the history, filter-branch it, then move it [00:48:15] But I can do full history as well, that's probably a good idea [00:48:38] It wouldn't have been appropriate for the other split I did recently (oojs-ui) but it's appropriate here [00:49:03] RoanKattouw: we don't want to change all the hashes [00:49:07] (03PS4) 10Chad: Give the search page a little UI love [core] - 10https://gerrit.wikimedia.org/r/94276 [00:49:10] <^d> MatmaRex: ^ [00:49:18] <^d> Fixed your last comment about module styles. [00:49:24] Right [00:49:38] Also I guess we were creating a new project in oojs-ui, you're not really [00:49:45] (03Merged) 10jenkins-bot: SpecialWatchlist: Minor cleanup [core] - 10https://gerrit.wikimedia.org/r/94456 (owner: 10Bartosz Dziewoński) [00:50:46] * RoanKattouw is just thinking out loud about all the other repo splits he has to do [00:51:31] * gwicke is thinking about adding a request for mediawiki/services/mathoid, although that really has nothing to do with mediawiki [00:51:58] (03PS1) 10CSteipp: Update Gender [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/94458 [00:53:43] (03CR) 10Legoktm: [C: 032] Expose iw_trans in siteinfo API interwikimap prop [core] - 10https://gerrit.wikimedia.org/r/91522 (owner: 10Bartosz Dziewoński) [00:53:51] <^d> gwicke: I had a patch earlier today that mass-converted all of "MediaWiki" to "Wikioid" but I decided not to push to master ;-) [00:54:00] (03CR) 10Bartosz Dziewoński: [C: 032] "OK." [core] - 10https://gerrit.wikimedia.org/r/94276 (owner: 10Chad) [00:54:09] ^d: hehe ;) [00:54:34] <^d> \o/ [00:54:38] <^d> Thanks for the merge MatmaRex! [00:54:48] <^d> Now we have a slightly sexier search screen :) [00:54:58] ^d: my pleasure. and byyy the way, i have some patches waiting for review [00:55:04] around 25, to be precise [00:55:19] would be nice if some of them got reviewed :P [00:56:19] (03PS1) 10GWicke: Update Parsoid to 3c18560f0 [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94459 [00:56:55] (03Merged) 10jenkins-bot: Expose iw_trans in siteinfo API interwikimap prop [core] - 10https://gerrit.wikimedia.org/r/91522 (owner: 10Bartosz Dziewoński) [00:57:41] (03CR) 10Chad: "lgtm, but let's at least mention it in the release notes. There's bound to be somebody using this behavior" [core] - 10https://gerrit.wikimedia.org/r/94150 (owner: 10Bartosz Dziewoński) [00:57:59] (03CR) 10Legoktm: "(1 comment)" [extensions/Echo] - 10https://gerrit.wikimedia.org/r/93113 (owner: 10Bartosz Dziewoński) [00:58:07] (03CR) 10Bartosz Dziewoński: [C: 04-1] "Hm, yeah, good point." [core] - 10https://gerrit.wikimedia.org/r/94150 (owner: 10Bartosz Dziewoński) [00:58:52] (03Merged) 10jenkins-bot: Give the search page a little UI love [core] - 10https://gerrit.wikimedia.org/r/94276 (owner: 10Chad) [00:58:53] MatmaRex: A while back I removed section edit links from the edit conflict page. As in, not section edit links attached to actual sections, but to the "Their text" and "Your text" headers. Those links obviously didn't go anywhere useful. That bug had gone unnoticed for 5.5 years :) [00:58:57] (03CR) 10Aaron Schulz: [C: 032] Improve user-facing error messages [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/93121 (owner: 10CSteipp) [00:59:06] I'd like to deploy a fix for the parsoid job runner which is currently broken in prod [00:59:22] (03CR) 10Bartosz Dziewoński: "(1 comment)" [extensions/Echo] - 10https://gerrit.wikimedia.org/r/93113 (owner: 10Bartosz Dziewoński) [00:59:41] gwicke: Poke greg-g about that [01:00:00] greg-g: ping [01:00:13] MatmaRex: heh https://gerrit.wikimedia.org/r/#/c/86318/ [01:00:28] (03PS1) 10Aaron Schulz: filebackend: improved "adviseStat" performance [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94460 [01:00:39] (03PS1) 10Aaron Schulz: filebackend: improved "adviseStat" performance [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94461 [01:00:44] (03CR) 10Aaron Schulz: [C: 032] filebackend: improved "adviseStat" performance [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94461 (owner: 10Aaron Schulz) [01:01:08] (03CR) 10Aaron Schulz: [C: 032] filebackend: improved "adviseStat" performance [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94460 (owner: 10Aaron Schulz) [01:01:09] (03Merged) 10jenkins-bot: Improve user-facing error messages [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/93121 (owner: 10CSteipp) [01:01:21] (03CR) 10Aaron Schulz: [C: 032] Made cleanupUploadStash use batching for the last two loops [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94454 (owner: 10Aaron Schulz) [01:02:23] (03PS2) 10Bartosz Dziewoński: Suppress section edit links with action=render [core] - 10https://gerrit.wikimedia.org/r/94150 [01:02:24] (03PS1) 10Bartosz Dziewoński: Add missing full stop to 7127c179 release notes [core] - 10https://gerrit.wikimedia.org/r/94462 [01:02:25] (03CR) 10Kaldari: [C: 032] Refactor diffs [core] - 10https://gerrit.wikimedia.org/r/92588 (owner: 10MaxSem) [01:02:30] * gwicke guesses that greg-g has left already [01:02:35] legoktm: yeah… i was trying out stuff and forgot to rebase and kill that commit :P [01:02:57] (03CR) 10Kaldari: [C: 032] DifferenceEngine cleanup [core] - 10https://gerrit.wikimedia.org/r/92668 (owner: 10MaxSem) [01:03:00] legoktm: and i've already got it ingrained that `git review` should be followed by `-y` [01:03:10] (03PS1) 10Aaron Schulz: Speed up thumbnail purging a bit for stash script [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94463 [01:03:14] (03CR) 10Aaron Schulz: [C: 032] Speed up thumbnail purging a bit for stash script [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94463 (owner: 10Aaron Schulz) [01:03:21] gwicke: here now [01:03:36] eek, what's up? [01:03:40] greg-g: so I'd like to push out a fix for the parsoid job runner [01:03:49] it currently crashes in prod [01:04:01] nope still here [01:04:22] greg-g: any objections? [01:04:52] gwicke: nope, go forth and fix it [01:05:13] alright [01:05:21] (03CR) 10GWicke: [C: 032] Update Parsoid to 3c18560f0 [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94459 (owner: 10GWicke) [01:05:23] <^d> AaronSchulz: I'm thinking of just removing those debug statements entirely from cdb stuff. [01:05:36] to be explicit, just the needed changes to fix the problem, no new other code on Fridays ;) [01:05:37] not sure how useful they are [01:05:44] (03CR) 10GWicke: [C: 032] Update Parsoid to 3c18560f0 [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94450 (owner: 10GWicke) [01:05:58] <^d> I don't think they're useful at all. And they're likely to just spam the hell out of your debug log. [01:06:01] grrrit-wm: this is only a fix, and only in the job runner [01:06:13] <^d> I'm going to amend the patch to remove them. [01:06:14] eh, greg-g [01:06:39] cool [01:07:28] greg-g: do you want to be asked for each such fix from now on? [01:07:31] (03CR) 10Aaron Schulz: [C: 032] Update Gender [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/94458 (owner: 10CSteipp) [01:08:31] gwicke: yeah, how often is production broken like that needed an emergency fix? [01:08:33] (03Merged) 10jenkins-bot: filebackend: improved "adviseStat" performance [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94461 (owner: 10Aaron Schulz) [01:08:38] needing* [01:08:59] from my side, very rarely [01:09:25] so is the general policy to ask you before pushing a fix? [01:09:29] MatmaRex: what is -y? [01:09:32] gwicke: yeah [01:09:43] especially on a Friday :) [01:10:13] (03CR) 10Legoktm: [C: 032] Add missing full stop to 7127c179 release notes [core] - 10https://gerrit.wikimedia.org/r/94462 (owner: 10Bartosz Dziewoński) [01:10:22] ok, was so far not under the impression that this happens too often [01:10:23] legoktm: it answers 'yes' when git-review ask 'for real?' [01:10:27] (03Merged) 10jenkins-bot: filebackend: improved "adviseStat" performance [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94460 (owner: 10Aaron Schulz) [01:10:35] (when submitting multiple commits) [01:10:54] heh [01:12:13] gwicke: usually there's not many production emergencies, and it's usually just a sanity check before trying to fix, so, obviously, don't block on me responding to fix something with the least amount of diff nedded [01:12:28] (usually, except for these past two weeks ;) ) [01:12:37] (03Merged) 10jenkins-bot: Made cleanupUploadStash use batching for the last two loops [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94454 (owner: 10Aaron Schulz) [01:12:40] (03Abandoned) 10GICodeWarrior: PHP [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94097 (owner: 10Sakshi) [01:12:59] greg-g: ok, thanks [01:13:17] gwicke: so, lemme rephrase: just check-in/check the deploy calendar when fixing broken prod [01:15:48] * AaronSchulz wishes jenkins was fast [01:17:04] ok, syncing to wmf2 [01:18:10] (03Merged) 10jenkins-bot: Refactor diffs [core] - 10https://gerrit.wikimedia.org/r/92588 (owner: 10MaxSem) [01:18:52] (03CR) 10Aaron Schulz: [V: 032] Speed up thumbnail purging a bit for stash script [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94463 (owner: 10Aaron Schulz) [01:19:29] (03Merged) 10jenkins-bot: DifferenceEngine cleanup [core] - 10https://gerrit.wikimedia.org/r/92668 (owner: 10MaxSem) [01:20:54] (03PS7) 10Chad: Clean up CDB classes [core] - 10https://gerrit.wikimedia.org/r/94258 [01:20:55] (03PS8) 10Chad: Don't use wfIsWindows() in CDB writers [core] - 10https://gerrit.wikimedia.org/r/94268 [01:21:38] <^d> AaronSchulz: Ready for review again :p [01:21:40] (03PS7) 10Chad: Update docs in CdbPHP.php [core] - 10https://gerrit.wikimedia.org/r/94267 (owner: 10Siebrand) [01:22:34] * legoktm pokes ^d about the RfC again :) [01:23:43] ^d: or can i just bug you with specific questions that I have so far? [01:24:13] (03Merged) 10jenkins-bot: Update Gender [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/94458 (owner: 10CSteipp) [01:28:17] ^d: https://gerrit.wikimedia.org/r/#/c/94451/1 [01:29:23] <^d> legoktm: So, initial thoughts I had.... [01:30:39] <^d> 1) If we're enumerating all default settings (metadata section), I don't see the need for $wgHiddenConfigOptions. [01:30:59] <^d> I think we can define that info in the metadata. [01:31:02] Oh, just set 'hidden' => true [01:31:04] good point [01:31:50] <^d> 2) I'm not quite as big a fan of singletons as I was when I wrote my draft. My idea these days is putting it in RequestContext & friends (cf: https://gerrit.wikimedia.org/r/#/c/92004/) [01:32:10] right [01:32:13] I added in: [01:32:15] "Accessible via context: $conf = $this->getContext()->getConf();" [01:32:28] <^d> Ah, missed that :) [01:33:48] Right now my main concerns are: [01:34:03] 1) Is JSON considered "fast enough" for the WMF needs, or should it also be implemented in CDB? [01:34:29] <^d> I'd like to benchmark it, but my assumption is the latter is faster. [01:34:39] <^d> hhvm may make it not really matter either way, but I digress :) [01:34:53] 2) How to sanely implement crosswiki stuff [01:35:14] <^d> That was always one of the things I struggled with. [01:35:21] <^d> interwiki stuff is hard :) [01:35:35] Right now a lot of settings are by dblist, so I was thinking of implementing CentralAuth's wikisets in core, but that would get a lot more complicated than it needs to. [01:35:44] (03CR) 10Aaron Schulz: [C: 031] "What is the use of the new exception subtype?" [core] - 10https://gerrit.wikimedia.org/r/94258 (owner: 10Chad) [01:36:37] (03CR) 10Chad: "A) It's a little more explicit for your error logs, while still being a MWException so normal catching & so forth works" [core] - 10https://gerrit.wikimedia.org/r/94258 (owner: 10Chad) [01:37:05] Otherwise I'm thinking of just doing 'value' : {'default: 'something', 'enwiki':'something else', 'wikisource': 'something for the whole family'} [01:37:10] and we'd keep dblists for now [01:37:21] (03CR) 10GICodeWarrior: [C: 04-2] "(2 comments)" [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94353 (owner: 10Sakshi) [01:37:25] <^d> Yeah let's keep dblists :) [01:37:48] (03Abandoned) 10GICodeWarrior: R [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94096 (owner: 10Sakshi) [01:38:01] (03Abandoned) 10GICodeWarrior: A [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94095 (owner: 10Sakshi) [01:38:12] (03Abandoned) 10GICodeWarrior: F [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94094 (owner: 10Sakshi) [01:38:43] <^d> legoktm: Whatever structure you choose, make sure the Cdb-backed version can do something to break it down further. It doesn't store arrays or fancy things, just key => value [01:39:01] <^d> (But then again, if json is fine it may be no big deal and we won't have to diverge) [01:39:05] hm [01:39:31] <^d> I'll try to make up some numbers on json vs cdb reads. [01:39:45] <^d> s/make up/benchmark/ ;-) [01:39:46] What if we use JSON for human readable and actual master storage [01:39:59] and then CDB can just be 'xxwiki-wgName' => 'actual real value'? [01:40:00] <^d> That's what I'd prefer for MW by default. [01:40:08] <^d> I think drop the database-backed version entirely. [01:40:17] <^d> (Then everything can be configured :)) [01:40:30] except for $wgEnableOnWikiConfiguration! [01:41:00] <^d> Why not? Presumably MW would just read that from the LS.json like any other setting? [01:41:40] Technically it would work. Just not for sanity [01:42:08] change all the settings to stupid values, then disable the onwiki configuration so no one can fix it ;) [01:42:08] <^d> Well, that's an implementation detail, no need to worry about it in the RFC. [01:42:11] yeah [01:42:19] <^d> So yeah, here's the best course of action. [01:42:32] <^d> 1) Remove the remaining references to a database, we're going json-only (and maybe cdb) [01:43:02] <^d> 2) Clean up that proposed json structure (and heck, let's make DefaultSettings a .json too :)) [01:43:36] If its a php file, you can do cool stuff like $wgAddGroups &= User::getAllRights() [01:43:51] or...something like that. [01:43:57] <^d> I think a design goal of this should be removing config from code :) [01:44:11] <^d> (which we're doing nicely, minus the DefaultSettings) [01:44:27] hrm. [01:44:29] <^d> Anyway, think on it, make a few more edits, and I think we'll have a fine RFC for wider discussion. [01:44:51] <^d> Good work so far! [01:45:15] thanks [01:51:59] (03PS4) 10Jdlrobson: Create prepareQuickTemplate function [core] - 10https://gerrit.wikimedia.org/r/94248 [01:52:06] (03PS5) 10Jdlrobson: Create prepareQuickTemplate function [core] - 10https://gerrit.wikimedia.org/r/94248 [01:52:14] (03CR) 10GICodeWarrior: [C: 031] "Looks great. Thanks! :-)" [extensions/CategoryTree] - 10https://gerrit.wikimedia.org/r/94099 (owner: 10Navdeep Bagga) [01:56:26] (03CR) 10Catrope: [C: 032] Create prepareQuickTemplate function [core] - 10https://gerrit.wikimedia.org/r/94248 (owner: 10Jdlrobson) [01:59:39] (03Merged) 10jenkins-bot: Create prepareQuickTemplate function [core] - 10https://gerrit.wikimedia.org/r/94248 (owner: 10Jdlrobson) [02:06:31] (03CR) 10EBernhardson: [C: 031] Give extensions a chance to hide their (unrelated) log entries [core] - 10https://gerrit.wikimedia.org/r/92504 (owner: 10Matthias Mullie) [02:08:58] (03PS3) 10Catrope: Category Tree Extension: Fix CSS according to the vector layout [extensions/CategoryTree] - 10https://gerrit.wikimedia.org/r/94099 (owner: 10Navdeep Bagga) [02:09:00] RoanKattouw: I clearly broke something on my local development machine; but I'm wondering if you know why chrome/resourceloader would fight like this? "Resource interpreted as Script but transferred with MIME type text/html: "http://donate.dev.local/load.php?debug=false&lang=en&modules=startup&only=scripts&skin=vector&*". Special:CentralNoticeBanners:17" [02:09:16] mwalker: When you visit that URL what do you see? [02:09:30] a partially broken page [02:09:43] ^d: so you would want to get rid of the Conf::get('blah') lookups right? [02:09:47] like; some of the RL css failed to load [02:09:59] mwalker: No, no, that load.php URL [02:10:07] and turn it into $conf->get('blah') [02:10:24] ah; good call [02:10:29] bug in my code somewhere [02:10:44] it's throwing a bunch of notices and other junk [02:11:17] ... which makes sense because now I recall I was messing with some CN resourceloading stuff earlier that is incomplete [02:12:19] (03PS2) 10Dereckson: Asserts if a stream wrapper 'uploadsource' already exists [core] - 10https://gerrit.wikimedia.org/r/94351 (owner: 10Mwjames) [02:14:14] (03CR) 10Legoktm: "* global wiki is whatever $wgCentralAuthGlobalUserModule['wiki'] is set to. For WMF it would be 'metawiki'" [extensions/CentralAuth] - 10https://gerrit.wikimedia.org/r/7274 (owner: 10Szymon Świerkosz) [02:14:35] <^d> legoktm: Yeah, that's the idea :) [02:14:47] ok. /me adjusts [02:15:03] <^d> I'm going to start my weekend now [02:15:17] (03CR) 10Legoktm: [C: 032] ext.echo.overlay: Various minor consistency fixes and cleanup [extensions/Echo] - 10https://gerrit.wikimedia.org/r/93113 (owner: 10Bartosz Dziewoński) [02:15:28] ^d: enjoy [02:15:37] (03Merged) 10jenkins-bot: ext.echo.overlay: Various minor consistency fixes and cleanup [extensions/Echo] - 10https://gerrit.wikimedia.org/r/93113 (owner: 10Bartosz Dziewoński) [02:17:33] (03CR) 10Dereckson: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/94351 (owner: 10Mwjames) [02:18:48] (03CR) 10Catrope: [C: 032] Fix CSS according to the vector layout [extensions/CategoryTree] - 10https://gerrit.wikimedia.org/r/94099 (owner: 10Navdeep Bagga) [02:18:50] (03Merged) 10jenkins-bot: Fix CSS according to the vector layout [extensions/CategoryTree] - 10https://gerrit.wikimedia.org/r/94099 (owner: 10Navdeep Bagga) [02:33:19] (03Abandoned) 10JGonera: Don't cache LESS files with errors when debug=1 [core] - 10https://gerrit.wikimedia.org/r/88677 (owner: 10JGonera) [02:37:43] (03PS9) 10CSteipp: Password Expiration [core] - 10https://gerrit.wikimedia.org/r/92037 [02:43:22] (03CR) 10Spage: [C: 04-1] "(3 comments)" [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/94369 (owner: 10Phuedx) [02:44:15] (03PS1) 10MarkAHershberger: Ready RC1 * Rearrange release notes * Bump version [core] - 10https://gerrit.wikimedia.org/r/94473 [02:44:26] thanks spagewmf [02:44:51] (03CR) 10jenkins-bot: [V: 04-1] Ready RC1 * Rearrange release notes * Bump version [core] - 10https://gerrit.wikimedia.org/r/94473 (owner: 10MarkAHershberger) [02:45:01] (03Abandoned) 10MarkAHershberger: Ready RC1 * Rearrange release notes * Bump version [core] - 10https://gerrit.wikimedia.org/r/94473 (owner: 10MarkAHershberger) [02:46:27] (03PS1) 10MarkAHershberger: Ready RC1 [core] (REL1_22) - 10https://gerrit.wikimedia.org/r/94474 [02:58:52] (03CR) 10MarkAHershberger: [C: 032] "self-review for tarball" [core] (REL1_22) - 10https://gerrit.wikimedia.org/r/94474 (owner: 10MarkAHershberger) [03:46:36] (03PS1) 10MarkAHershberger: Put diff files in the right place. [tools/release] - 10https://gerrit.wikimedia.org/r/94478 [03:47:51] (03PS2) 10MarkAHershberger: Put diff files in the right place. [tools/release] - 10https://gerrit.wikimedia.org/r/94478 [03:48:15] (03CR) 10MarkAHershberger: [C: 032 V: 032] "this is needed so it will work" [tools/release] - 10https://gerrit.wikimedia.org/r/94478 (owner: 10MarkAHershberger) [04:01:00] (03PS3) 10leucosticte: Set $wgNoFollowLinks to false by default [core] - 10https://gerrit.wikimedia.org/r/94401 [04:01:25] oh for crying out loud, I did it again. anyone around that knows what I need to do if I accidentally did 'git commit --all --amend' on the *first* commit (hence, 'ammending' an unrelated previous commit?) [04:04:03] Do you want to keep your changes? [04:04:18] if not git reset HEAD~1 --hard && git pull [04:08:58] I want to keep my changes, yes [04:09:13] I'm actually not sure if there's a real problem now that I think of it [04:09:33] --amend loaded the message of a previous commit.. if I erase all of it and write my own -- my commit will be independent, right? [04:09:51] err, yeah [04:09:54] or will it produce a commit that's dependent on whatever previous commit I worked with [04:09:55] but not based off master [04:09:59] right [04:10:02] that's the problem :\ [04:10:04] You could just take a copy of the altered files [04:10:09] meh. [04:10:10] put them somewhere save [04:10:13] clean up the working copy [04:10:16] paste them back in [04:10:18] commit properly [04:10:19] ;) [04:10:21] hehe [04:10:42] well. eh. What about rebasing? Won't things work if I rebase onto master? [04:11:07] I wanted to avoid having to save/paste the files. [04:11:11] might have to, though [04:11:25] rebase might work.. [04:11:49] mrr. I wanted to avoid the trouble, the files are all over the place. [04:11:59] might as well bite the bullet though [04:12:30] Err [04:12:41] You should be able to get git to make a diff for the commit [04:12:46] save and reapply that [04:13:23] so here's a question, then [04:13:29] I did the mistaken --amend [04:13:53] but i'm still inside the editor... is there a way for me to abort the commit now? or should I just finish the commit and then save the files and reset, etc [04:14:16] hmm [04:14:30] sorry, I'm a bit new to git. I am hoping... :p [04:14:33] I know if you don't enter anything on a new commit you can exit and it will abort [04:14:54] I know that works for interactive rebases [04:14:54] But on an amend, you've already got the file changes so you don't need to amend the commit summary [04:15:00] * mooeypoo nods [04:15:07] damn. okay, well [04:15:12] hang on [04:15:30] i think i'll just have to check out 'git show HEAD', make sure i have all the files + diffs, then reset and redo [04:15:38] http://stackoverflow.com/questions/1459150/how-to-undo-git-commit-amend-done-instead-of-git-commit [04:15:43] oh! [04:15:50] hey, I looked in stackoverflow..... [04:15:55] how did I miss that [04:16:22] okay, I see 'git reset --soft' and panic. The 'soft' means I don't lose files, I assume? [04:16:45] yeah [04:17:09] awesomesauce [04:17:10] thanks! [04:22:53] woot, thanks Reedy ! that seemed to work :) [04:37:37] * Aaron|home reads https://www.facebook.com/notes/facebook-engineering/presto-interacting-with-petabytes-of-data-at-facebook/10151786197628920 [04:45:16] (03Abandoned) 10MarkAHershberger: Clean up the release notes and reorder them a bit. [core] (REL1_22) - 10https://gerrit.wikimedia.org/r/94314 (owner: 10MarkAHershberger) [05:16:03] (03PS2) 10TTO: Update default contents of interwiki table [core] - 10https://gerrit.wikimedia.org/r/84303 [05:24:45] (03PS1) 10Aaron Schulz: Fixed loading of stat cache entries into the wrong key [core] - 10https://gerrit.wikimedia.org/r/94480 [05:24:59] (03PS1) 10Aaron Schulz: Fixed loading of stat cache entries into the wrong key [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94481 [05:25:11] (03PS1) 10Aaron Schulz: Fixed loading of stat cache entries into the wrong key [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94482 [05:25:42] (03CR) 10Aaron Schulz: [C: 032] Fixed loading of stat cache entries into the wrong key [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94482 (owner: 10Aaron Schulz) [05:26:01] (03CR) 10Aaron Schulz: [C: 032] Fixed loading of stat cache entries into the wrong key [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94481 (owner: 10Aaron Schulz) [05:26:33] * Aaron|home wonders when Reedy sleeps [05:30:35] Hrm. [05:31:42] That was me. [05:33:29] (03Merged) 10jenkins-bot: Fixed loading of stat cache entries into the wrong key [core] (wmf/1.23wmf2) - 10https://gerrit.wikimedia.org/r/94482 (owner: 10Aaron Schulz) [05:34:57] (03Merged) 10jenkins-bot: Fixed loading of stat cache entries into the wrong key [core] (wmf/1.23wmf3) - 10https://gerrit.wikimedia.org/r/94481 (owner: 10Aaron Schulz) [05:38:55] there we go, fast finally [05:38:57] * Aaron|home throws https://gerrit.wikimedia.org/r/#/c/94480/ at ori-l [06:15:20] (03CR) 10Aaron Schulz: [C: 032] xml dump maintenance scripts should use the 'dump' db group [core] - 10https://gerrit.wikimedia.org/r/91213 (owner: 10ArielGlenn) [06:16:15] (03CR) 10ArielGlenn: [C: 032] Fixed loading of stat cache entries into the wrong key [core] - 10https://gerrit.wikimedia.org/r/94480 (owner: 10Aaron Schulz) [06:19:06] (03Merged) 10jenkins-bot: xml dump maintenance scripts should use the 'dump' db group [core] - 10https://gerrit.wikimedia.org/r/91213 (owner: 10ArielGlenn) [06:19:36] (03PS1) 10Kaldari: Updating extension authors and documentation [extensions/ArticleCreationWorkflow] - 10https://gerrit.wikimedia.org/r/94483 [06:20:22] (03CR) 10Kaldari: [C: 032] Updating extension authors and documentation [extensions/ArticleCreationWorkflow] - 10https://gerrit.wikimedia.org/r/94483 (owner: 10Kaldari) [06:20:39] (03CR) 10Kaldari: [V: 032] Updating extension authors and documentation [extensions/ArticleCreationWorkflow] - 10https://gerrit.wikimedia.org/r/94483 (owner: 10Kaldari) [07:55:18] hmpf, always growing https://www.mediawiki.org/w/index.php?title=Gerrit/Reports/Open_changesets_by_newbie_owner&diff=0&oldid=816051 [08:34:12] (03CR) 10Spage: [C: 031] "Wikis shouldn't assume mediawiki.ui is available but do, e.g. fr-wiki use it to style [Explorez Wikipédia !] on https://fr.wikipedia.org/w" [core] - 10https://gerrit.wikimedia.org/r/87296 (owner: 10Mattflaschen) [08:41:28] (03CR) 10Spage: "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/91116 (owner: 10Spage) [08:52:38] (03CR) 10Nemo bis: "If you use the same keys, translatewiki.net should automatically re-add the translations to the file where you moved the keys/messages. Be" [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/94370 (owner: 10Phuedx) [09:17:29] (03CR) 10Raimond Spekking: [C: 032 V: 032] Localisation updates from http://translatewiki.net. [core] - 10https://gerrit.wikimedia.org/r/94428 (owner: 10L10n-bot) [09:20:32] (03Merged) 10jenkins-bot: Localisation updates from http://translatewiki.net. [core] - 10https://gerrit.wikimedia.org/r/94428 (owner: 10L10n-bot) [09:50:39] (03PS1) 10Gerrit Patch Uploader: Add support for gradients in Internet Explorer 10 and 11. Remove non-exixtent -ms-border-radius and -o-border-radius properties. Updated deprecated -moz-border-radius-XX (pre FF 3.0) properties. [core] - 10https://gerrit.wikimedia.org/r/94489 [09:50:40] (03CR) 10Gerrit Patch Uploader: "This commit was uploaded using the Gerrit Patch Uploader [1]." [core] - 10https://gerrit.wikimedia.org/r/94489 (owner: 10Gerrit Patch Uploader) [09:53:03] (03PS1) 10TTO: Clean up some old readme files and remove references to SVN [core] - 10https://gerrit.wikimedia.org/r/94490 [10:02:35] (03CR) 10Peachey88: "(3 comments)" [core] - 10https://gerrit.wikimedia.org/r/94490 (owner: 10TTO) [10:10:49] (03CR) 10Daniel Kinzler: [C: 04-1] "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/93611 (owner: 10Aude) [10:13:54] (03CR) 10Daniel Kinzler: "What's the point of the setters if the members remain public, and there are no getters?" [core] - 10https://gerrit.wikimedia.org/r/94360 (owner: 10Aude) [10:22:24] (03CR) 10Edokter: [C: 04-1] "Cscott keeps erroniously asuming some magic CSS will fix hlist. I have spent two years trying ands I can guarantee there is no CSS solutio" [core] - 10https://gerrit.wikimedia.org/r/94443 (owner: 10Cscott) [10:22:26] (03CR) 10TTO: "Some might prefer gerrit as it presents the projects in alphabetical order (gitblit seems to only be able to show them in order of activit" [core] - 10https://gerrit.wikimedia.org/r/94490 (owner: 10TTO) [10:23:15] (03Abandoned) 10Aude: Add visibility and setters for RCCacheEntry variables [core] - 10https://gerrit.wikimedia.org/r/94360 (owner: 10Aude) [10:23:39] (03PS2) 10TTO: Clean up some old readme files and remove references to SVN [core] - 10https://gerrit.wikimedia.org/r/94490 [10:29:04] (03PS1) 10Shirayuki: Minor fix for consistency [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/94491 [10:35:35] (03CR) 10Aude: "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/93611 (owner: 10Aude) [11:50:26] (03PS1) 10Sakshi: PHP script [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94497 [11:52:38] (03PS2) 10Sakshi: PHP script [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94497 [12:13:00] (03PS2) 10Bartosz Dziewoński: Add missing full stop to 7127c179 release notes [core] - 10https://gerrit.wikimedia.org/r/94462 [12:13:15] (03CR) 10Bartosz Dziewoński: [C: 032] Add missing full stop to 7127c179 release notes [core] - 10https://gerrit.wikimedia.org/r/94462 (owner: 10Bartosz Dziewoński) [12:18:16] (03Merged) 10jenkins-bot: Add missing full stop to 7127c179 release notes [core] - 10https://gerrit.wikimedia.org/r/94462 (owner: 10Bartosz Dziewoński) [12:21:31] (03CR) 10Bartosz Dziewoński: [C: 04-1] "Would be +1 per my older comments on the other patch, but release notes need updating for 1.23. But this definitely feels like a good idea" [core] - 10https://gerrit.wikimedia.org/r/87296 (owner: 10Mattflaschen) [12:23:47] (03CR) 10Bartosz Dziewoński: "This will conflict with https://gerrit.wikimedia.org/r/#/c/93584/" [core] - 10https://gerrit.wikimedia.org/r/94489 (owner: 10Gerrit Patch Uploader) [12:24:03] (03CR) 10Bartosz Dziewoński: "This will conflict with https://gerrit.wikimedia.org/r/#/c/94489/ , please rebase one on top of the other." [core] - 10https://gerrit.wikimedia.org/r/93584 (owner: 10Mattflaschen) [12:33:03] (03CR) 10Raimond Spekking: "+1 for i18n as long as this patchset is merged prior to 94369 which adds the messages. The scripts of translatewiki.net will add the exist" [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/94370 (owner: 10Phuedx) [12:35:50] (03PS5) 10Rillke: Messages with links in it must be parsed [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/77491 [12:36:26] (03CR) 10Rillke: "Rebased to master's HEAD" [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/77491 (owner: 10Rillke) [13:13:09] (03PS1) 10Rillke: Set message and group used on [[Special:SpecialPages]] [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/94499 [13:23:43] (03CR) 10Edokter: "I did rebase, but as this has not been merged, I could not have known about this patch. My changes are rather trivial, someone working on " [core] - 10https://gerrit.wikimedia.org/r/93584 (owner: 10Mattflaschen) [13:48:28] (03CR) 10Bartosz Dziewoński: "Yes, definitely, I cross-referenced the patches exactly so that people working on one know about the existence of the other, and so that s" [core] - 10https://gerrit.wikimedia.org/r/93584 (owner: 10Mattflaschen) [14:00:50] (03PS1) 10Tinaj1234: Removed indentation button from WikiEditor toolbar. [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/94500 [14:20:10] (03CR) 10Cscott: "Please let's discuss the *bug* over on bugzilla, and discuss just this *patch* here. Comments here should be limited to whether or not th" [core] - 10https://gerrit.wikimedia.org/r/94443 (owner: 10Cscott) [15:30:33] Hello Mr. Dereckson! [15:30:41] Hi andre___ [15:30:59] hi arav93 [15:31:20] There's this problem with bug 43956 [15:32:22] every bug is a problem, by definition. :P [15:32:49] * Reedy hands andre__ some more underscores [15:33:12] Reedy: stop being jealous because you're under-underscored! [15:33:13] A problem? You mean an enhancement, lowest priority! [15:35:54] arav93: in case you expected anything here, you might want to ask a question, or describe "this problem". :) [15:36:56] arav93: plus please read https://www.mediawiki.org/wiki/Gerrit/Tutorial#Amending_a_change [15:39:47] (03PS6) 10Rillke: Messages with links in it must be parsed [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/77491 [15:43:07] (03CR) 10Rillke: "After re-basing, I discovered that someone changed .html() to .text in member function ``recoverFromError``. So this little change grew a " [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/77491 (owner: 10Rillke) [15:57:35] andre__, I have submitted a diff file in the bugzilla page itself, gerrit seems to have way too much problems. [15:57:49] arav93, where exactly is a Gerrit problem? [15:57:52] I don't see one [15:57:59] did you read https://www.mediawiki.org/wiki/Gerrit/Tutorial#Amending_a_change ? [15:58:03] Previously, I submitted https://gerrit.wikimedia.org/r/#/c/94503/. [15:58:37] Would the diff file I have submitted in https://bugzilla.wikimedia.org/show_bug.cgi?id=43956 solve the problem? [15:59:09] (03PS1) 10IAlex: Sync RELEASE-NOTES-1.22 between master and REL1_22 [core] - 10https://gerrit.wikimedia.org/r/94507 [16:00:09] arav93, which "problem"? [16:00:26] arav93, which of the two diff files? [16:00:30] could you please be exact? [16:01:14] Build failed. https://integration.wikimedia.org/ci/job/operations-mw-config-phplint/6738/console : FAILURE in 2s [16:01:30] This is the comment I recieved after I upladed the patch for the third time. [16:02:36] arav93, yes, probably because your checkout wasn't up-to-date [16:02:48] and it was not possible to automatically merge the changes that happened in the meantime. [16:03:07] arav93, Hence the "Please rebase your change". [16:03:23] Ok. [16:03:59] There was yet another variable to be renamed, I renamed that. Made a diff file. And uploaded it. [16:04:12] did you read https://www.mediawiki.org/wiki/Gerrit/Tutorial#Prepare_to_push_your_change_set_to_Gerrit ? [16:04:23] it recommends to rebase :) [16:04:46] arav93, you normally don't make a diff file for Gerrit. [16:04:56] if you uploaded a diff to Bugzilla, then yes, probably. [16:05:06] we don't use Bugzilla for patch review though, but Gerrit [16:05:36] This is the first bug I'm trying to fix, as it would have been obviously been clear to you. So, I jus figured I'd upload to bugzilla. Hence the diff file. [16:05:56] I know that, but how do I know if the patch I uploaded is right or not. [16:06:10] arav93, did you read the test output in https://integration.wikimedia.org/ci/job/operations-mw-config-phplint/6738/console actually? [16:06:25] Stuff like "<<<<<<< HEAD" should not be in a patch [16:07:13] arav93, and https://www.mediawiki.org/wiki/Gerrit/Tutorial#Make_and_commit_your_change explains to use "git diff" to check your patch manually. To spot such things. [16:07:40] Ok, could you help me make a file to upload to bugzilla right now? [16:07:51] I have made all possible changes to the code. [16:08:43] arav93, it should go to Gerrit. [16:08:50] there's no sense in uploading patches to Bugzilla. [16:09:03] Gerrit is the tool which is used for code review. [16:09:22] I know. Is it wrong to upload there? [16:10:26] arav93: it is, IMHO. [16:10:59] arav93, you can upload patches to Bugzilla (or any other random website out there), but they will rot there and not get attention. [16:11:13] gerrit.wikimedia.org is the place for patches. [16:11:24] This is my first patch. So some exceptions? [16:12:41] well, I cannot manually test here if the patch will build, because I don't have a copy of Wikimedia's server setup around. [16:12:51] that's why the automatic testing is very welcome in Gerrit. [16:13:01] as you see, your patch failed because it wasn't clean. [16:13:28] Not sure what you want to achieve by uploading it to Bugzilla instead of Gerrit. If the patch does not work, it does not work, no matter where you've out it before. [16:13:41] err, "put" [16:15:06] so you'll still need to fix your patch. [16:15:32] Ok. [16:19:12] ain't easy... [16:33:15] (03Restored) 10Umherirrender: (bug 40667) deprecate rc_cur_time of recentchanges [core] - 10https://gerrit.wikimedia.org/r/42278 (owner: 10Umherirrender) [16:45:17] (03PS2) 10Umherirrender: Deprecate rc_cur_time of recentchanges [core] - 10https://gerrit.wikimedia.org/r/42278 [16:45:54] (03CR) 10Umherirrender: "Patch Set 2: Rebased and updated version numbers to 1.23/1.24" [core] - 10https://gerrit.wikimedia.org/r/42278 (owner: 10Umherirrender) [16:46:58] (03CR) 10Hoo man: [C: 031] "Kill it with fire!" [core] - 10https://gerrit.wikimedia.org/r/42278 (owner: 10Umherirrender) [16:48:19] (03CR) 10jenkins-bot: [V: 04-1] Deprecate rc_cur_time of recentchanges [core] - 10https://gerrit.wikimedia.org/r/42278 (owner: 10Umherirrender) [17:10:31] (03PS3) 10Umherirrender: Deprecate rc_cur_time of recentchanges [core] - 10https://gerrit.wikimedia.org/r/42278 [17:11:11] (03CR) 10Umherirrender: "Patch Set 3: Ignore column on sqlite tests, because it will be dropped in a later version." [core] - 10https://gerrit.wikimedia.org/r/42278 (owner: 10Umherirrender) [17:12:11] (03CR) 10GICodeWarrior: [C: 04-2] "Thanks for the update Sakshi." [extensions/CSS] - 10https://gerrit.wikimedia.org/r/94497 (owner: 10Sakshi) [17:13:32] (03CR) 10jenkins-bot: [V: 04-1] Deprecate rc_cur_time of recentchanges [core] - 10https://gerrit.wikimedia.org/r/42278 (owner: 10Umherirrender) [17:22:09] (03PS4) 10Umherirrender: Remove recentchanges.rc_cur_time from sql statements [core] - 10https://gerrit.wikimedia.org/r/42278 [17:23:22] (03CR) 10Umherirrender: "Patch Set 4: Only remove it from the sql statements as first step, leave it in tables.sql, but at hint when it should be dropped." [core] - 10https://gerrit.wikimedia.org/r/42278 (owner: 10Umherirrender) [18:31:12] (03PS1) 10Bartosz Dziewoński: Rewrite CSSMin::remap to support multiple url() values in one rule [core] - 10https://gerrit.wikimedia.org/r/94511 [18:44:43] (03PS1) 10Adamw: New hook to instrument page save failures [core] - 10https://gerrit.wikimedia.org/r/94512 [18:45:38] (03PS2) 10Adamw: New hook to instrument page save failures [core] - 10https://gerrit.wikimedia.org/r/94512 [18:48:42] (03PS1) 10Umherirrender: Add content model to the page information [core] - 10https://gerrit.wikimedia.org/r/94513 [18:55:01] (03PS1) 10Umherirrender: Add used message keys to ContentHandler::getLocalizedName [core] - 10https://gerrit.wikimedia.org/r/94515 [19:00:08] (03PS1) 10Adamw: Log attempted page saves [extensions/WikimediaEvents] - 10https://gerrit.wikimedia.org/r/94516 [19:13:37] (03CR) 10jenkins-bot: [V: 04-1] Localisation updates from http://translatewiki.net. [extensions/Survey] - 10https://gerrit.wikimedia.org/r/94560 (owner: 10L10n-bot) [19:14:10] (03PS1) 10UltrasonicNXT: Expand content to fill width [extensions/SocialProfile] - 10https://gerrit.wikimedia.org/r/94566 [19:15:22] (03PS1) 10Umherirrender: Remove duplicate entries in messages.inc [core] - 10https://gerrit.wikimedia.org/r/94576 [19:27:25] quick question: how do I save a comment in code in gerrit? [19:27:40] it only lets me save as 'draft'. [19:30:25] (03PS9) 10Aaron Schulz: filebackend: Added supported for retrieving file metadata/headers [core] - 10https://gerrit.wikimedia.org/r/86642 [19:30:29] Edokter: if you are doing a code-review, then after you save your comment as "draft", they will get published when you give the score on the change main page. [19:31:27] hmm, what if I don't want to review? [19:31:57] you can always put a score of 0 [19:32:27] I am not aware of any other workarounds, so you can wait for someone more experienced. :-) [20:05:25] (03PS2) 10IAlex: Sync RELEASE-NOTES-1.22 between master and REL1_22 [core] - 10https://gerrit.wikimedia.org/r/94507 [20:06:00] (03PS3) 10IAlex: Sync RELEASE-NOTES-1.22 between master and REL1_22 [core] - 10https://gerrit.wikimedia.org/r/94507 [20:14:39] (03PS3) 10EBernhardson: Require core edit permission to perform write actions [extensions/Flow] - 10https://gerrit.wikimedia.org/r/94103 [20:14:43] (03CR) 10jenkins-bot: [V: 04-1] Require core edit permission to perform write actions [extensions/Flow] - 10https://gerrit.wikimedia.org/r/94103 (owner: 10EBernhardson) [20:15:33] (03PS1) 10Andrew Bogott: Implement ValidateNewUser hook to check shell name. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/94580 [20:24:01] (03CR) 10Kipcool: [C: 032 V: 032] "nice :)" [extensions/WikiLexicalData] - 10https://gerrit.wikimedia.org/r/93922 (owner: 10Hiong3-eng5) [20:25:07] (03PS1) 10Andrew Bogott: Add a hook to validate account creation fields. [core] - 10https://gerrit.wikimedia.org/r/94581 [20:25:25] (03CR) 10Andrew Bogott: "This patch will be harmless but ineffective until core implements the ValidateNewUser hook, as in this patch:" [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/94580 (owner: 10Andrew Bogott) [20:26:33] * andrewbogott would appreciate a review of https://gerrit.wikimedia.org/r/94581 [20:26:59] …that being outside my comfort zone, probably needs some corrections. [20:30:07] (03CR) 10coren: [C: 032] "Straightforward hook." [core] - 10https://gerrit.wikimedia.org/r/94581 (owner: 10Andrew Bogott) [20:30:42] coren, thx! [20:36:24] (03PS1) 10Aaron Schulz: BacklinkCache performance tweaks [core] - 10https://gerrit.wikimedia.org/r/94582 [20:54:49] (03PS10) 10EBernhardson: Moderate a post or topic [extensions/Flow] - 10https://gerrit.wikimedia.org/r/91135 [20:55:02] (03CR) 10jenkins-bot: [V: 04-1] Moderate a post or topic [extensions/Flow] - 10https://gerrit.wikimedia.org/r/91135 (owner: 10EBernhardson) [21:07:59] (03CR) 10Jack Phoenix: [C: 04-1] "The "gap" you're seeing is there because some space on the right (or on the left for RTL languages, I guess) has been reserved for the rat" [extensions/SocialProfile] - 10https://gerrit.wikimedia.org/r/94566 (owner: 10UltrasonicNXT) [21:52:19] (03CR) 10UltrasonicNXT: "Okay, no prob. I was wondering if there was something that would fill the space, but I couldn't find anything..." [extensions/SocialProfile] - 10https://gerrit.wikimedia.org/r/94566 (owner: 10UltrasonicNXT) [21:54:57] (03PS2) 10Adamw: Log attempted page saves [extensions/WikimediaEvents] - 10https://gerrit.wikimedia.org/r/94516 [22:20:05] (03CR) 10MarkAHershberger: [C: 032] Sync RELEASE-NOTES-1.22 between master and REL1_22 [core] - 10https://gerrit.wikimedia.org/r/94507 (owner: 10IAlex) [22:23:01] (03Merged) 10jenkins-bot: Sync RELEASE-NOTES-1.22 between master and REL1_22 [core] - 10https://gerrit.wikimedia.org/r/94507 (owner: 10IAlex) [22:30:04] (03PS1) 10Adamw: (bug 56849) Pass true oldid for atomic page save [core] - 10https://gerrit.wikimedia.org/r/94584 [22:33:07] (03CR) 10jenkins-bot: [V: 04-1] (bug 56849) Pass true oldid for atomic page save [core] - 10https://gerrit.wikimedia.org/r/94584 (owner: 10Adamw) [22:38:17] hey addshore :D [22:39:10] (03CR) 10Jack Phoenix: "No worries, I'm sorry that your first SocialProfile patch didn't make it in, but better luck next time!" [extensions/SocialProfile] - 10https://gerrit.wikimedia.org/r/94566 (owner: 10UltrasonicNXT) [22:39:53] (03PS3) 10Adamw: New hook to instrument page save failures [core] - 10https://gerrit.wikimedia.org/r/94512 [22:40:30] (03PS3) 10Adamw: Log attempted page saves [extensions/WikimediaEvents] - 10https://gerrit.wikimedia.org/r/94516 [22:42:59] (03CR) 10jenkins-bot: [V: 04-1] New hook to instrument page save failures [core] - 10https://gerrit.wikimedia.org/r/94512 (owner: 10Adamw) [22:50:42] (03PS1) 10Legoktm: Whitelist YuviPanda [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/94586 [22:57:00] (03PS1) 10Inchikutty: OPW Microtask:adding button to show osm map [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/94587 [23:03:22] hey ori-l, since you became a fan of ULS lately, wanna review https://gerrit.wikimedia.org/r/#/c/94511/ sometime? [23:03:24] "Rewrite CSSMin::remap to support multiple url() values in one rule" [23:03:45] (i'm writing some tests for that right now) [23:04:31] I'm totally swamped [23:04:45] I owe Krinkle & legoktm reviews from weeks ago [23:04:53] heh [23:04:58] maybe if I clear those up this weekend [23:05:17] * legoktm waves [23:05:26] hey [23:06:18] MatmaRex: ori-l is a fan of ULS? [23:06:28] how things change in the span of an intercontinental flight... [23:07:18] I think MatmaRex meant a "fan" [23:07:40] I am a fan of it when it works [23:07:46] :) [23:07:48] true true [23:07:54] (03CR) 10MaxSem: [C: 032] Whitelist YuviPanda [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/94586 (owner: 10Legoktm) [23:08:00] (03Merged) 10jenkins-bot: Whitelist YuviPanda [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/94586 (owner: 10Legoktm) [23:08:09] MaxSem: woot. [23:40:15] (03CR) 10Legoktm: [C: 031] "Tested, works like expected." [core] - 10https://gerrit.wikimedia.org/r/94513 (owner: 10Umherirrender) [23:52:46] (03PS1) 10Bartosz Dziewoński: More tests for CSSMin::remap [core] - 10https://gerrit.wikimedia.org/r/94588 [23:54:07] (03CR) 10jenkins-bot: [V: 04-1] More tests for CSSMin::remap [core] - 10https://gerrit.wikimedia.org/r/94588 (owner: 10Bartosz Dziewoński) [23:54:33] * MatmaRex stabs PHPUnit [23:54:45] not that i expected it to work the first time 'round