[00:04:28] ^d: core is again extremely slow/large to clone -- is there anything that could be done? [00:05:38] i've downloaded 75 Mb and I'm only 50% done [00:06:29] 100 Mb, 53%. Wow. [00:16:28] ori: last time i asked, the answer was "probably yes" [00:20:20] "will be automatically used if the server is MySQL 4.0" ... :P [00:23:23] (03CR) 10Daniel Friesen: [C: 04-1] "+1 To mediawiki.skinbase.* that does sound like a reasonable alternative to mediawiki.skins.common.*" (031 comment) [core] - 10https://gerrit.wikimedia.org/r/114984 (owner: 10Bartosz Dziewoński) [00:25:15] Receiving objects: 100% (474942/474942), 437.15 MiB | 164.00 KiB/s, done. [00:25:30] (03PS1) 10Legoktm: Add 'mw-anonuserlink' class for anonymous users [core] - 10https://gerrit.wikimedia.org/r/115095 [00:34:42] That was really odd. I clicked on [Save page] and it asked me if I wanted to leave the page warning me that any unsaved data would be lost... [00:35:10] Technical_13: where? [00:35:22] since it was a minimal change (I added a "g" to a replace(//, '') on my common.js [00:35:37] I clicked okay and it did save... just odd that it warned me. [00:41:05] (03CR) 10Tim Starling: [C: 032] Make SiteStats (re)initializing more sane [core] - 10https://gerrit.wikimedia.org/r/114994 (owner: 10Hoo man) [00:45:01] (03Merged) 10jenkins-bot: Make SiteStats (re)initializing more sane [core] - 10https://gerrit.wikimedia.org/r/114994 (owner: 10Hoo man) [00:45:36] huh: Okay, so it just did it again... [00:45:53] And clicking on "stay on page" this time saved it and took me away.. [00:46:02] so that is even more wierd... [00:46:13] Can you debug it? [00:46:39] Technical_13: which page did you modify? [00:46:57] It seems to be just my common.js [00:47:07] http://en.wikipedia.org/wiki/User:Technical_13/common.js [00:57:01] (03PS1) 10Tim Starling: [1.23wmf14] Make SiteStats (re)initializing more sane [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/115096 [00:57:20] (03PS1) 10Tim Starling: [1.23wmf15] Make SiteStats (re)initializing more sane [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115097 [00:57:37] (03CR) 10Tim Starling: [C: 032] [1.23wmf14] Make SiteStats (re)initializing more sane [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/115096 (owner: 10Tim Starling) [00:57:41] (03CR) 10Tim Starling: [C: 032] [1.23wmf15] Make SiteStats (re)initializing more sane [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115097 (owner: 10Tim Starling) [01:02:55] (03Merged) 10jenkins-bot: [1.23wmf14] Make SiteStats (re)initializing more sane [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/115096 (owner: 10Tim Starling) [01:03:09] (03Merged) 10jenkins-bot: [1.23wmf15] Make SiteStats (re)initializing more sane [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115097 (owner: 10Tim Starling) [01:03:43] what ever it was huh, it's fixed itself now... [01:16:49] Technical_13: please don't close bugs as WORKSFORME if you can write a user script for them, thats not actually fixed [01:22:53] legoktm: just following the prescedent set in https://bugzilla.wikimedia.org/show_bug.cgi?id=46412 [01:23:39] if there is script to accomplish it, it can be closed... [01:23:47] no, totally not [01:23:48] that can be said of many many bugs [01:23:54] yikes [01:24:54] Technical_13: That's not a precedent. [01:25:06] Sometimes the answer to a bug report is a user script. But each case is different. [01:25:21] Fair enough. [01:25:43] In that case, it actually is a weird subset of bugs where the reporter finds a workaround and summarily closes the ticket. [01:25:59] And we usually give some deference to the reporter. [01:26:54] marked the bug linked above as wontfix as that is by far more correct: It's not fixed, but we also don't want it to be fixed... [01:27:23] I'm sure that Scott will let me know if it doesn't work for him... I'll happily reopen if that is the case. [01:28:05] apparently legoktm already reopened. [01:28:17] No biggie, I'm less than offended as is usually the case. [01:28:41] Technical_13: what bug are you talking about? [01:28:59] Since Anomie isn't around, and legoktm is usually too busy, is there anyone else who knows a bunch about Lua who can help me learn something? [01:29:06] huh: https://bugzilla.wikimedia.org/show_bug.cgi?id=42357 [01:29:41] I agree, JS is not a proper solution [01:30:04] What I wan't to know is, is it possible to set a table in a lua call at the top of the page and have that table be usable by subsequent calls to the same module? It would be like page global variables. [01:30:37] Your code only works on English language wikis [01:30:50] This is definitely not a valid solution [01:31:00] especially as it's JS :P [01:31:35] Technical_13: No, that's not possible, you can only inject wikitext in the place the {{#invoke...}} was [01:32:28] Technical_13: that would be a bad idea, wouldn't it? [01:33:30] What I want to be able to do is make a list with hundreds (or more) module calls and set a min and max limit on all of them. [01:33:53] mh? [01:34:07] Technical_13: each #invoke call should be in an isolated state, so there should be no ability for different #invoke's to communicate [01:34:11] I don't want to have to declare that min and max on every call, so I was hoping to be able to do it as a glabal variable for the page. [01:34:38] Technical_13: please, answer bug reqs with PHP [01:34:44] there's a bug open about this, so it is technically possible but shouldn't be relied upon and will go away [01:34:58] legoktm: I'm aware of that bug [01:35:03] huh: Sometimes the answer isn't PHP. [01:35:05] Often, probably. [01:35:14] Yes, it's possible to leak global state, but doing that is bad... [01:35:30] lua unit tests yet catch that, but production doesn't [01:35:34] legoktm: Which? [01:35:47] Gloria: WP:BEANS [01:35:51] :p [01:35:53] Gloria: https://bugzilla.wikimedia.org/show_bug.cgi?id=61268 [01:36:14] Yep, same one... [01:36:23] Don't abuse bugs [01:36:50] Gloria: most MediaWiki feature requests should be in PHP [01:37:06] Many toolserver/labs features could also be extensions, IMO [01:37:25] Of course, bug requests explicitly about JS should be JS [01:38:04] huh: That's not true. [01:38:18] Your "should be" is nonsense. [01:38:30] The rule is to use the best tool for the job. [01:38:43] Any guidelines for that? [01:39:26] Not apart from that one-liner, no. :-) [01:39:52] Use common sense or ask someone experienced in that area ;) [01:40:06] usuually it's a pretty obvious desicion :P [01:53:42] Gloria: is there already a mediawiki message for "hide anons" and one for "hide logged in" ? [01:53:52] I know there are rc-specific ones [01:54:32] ?uselang=qqx [01:54:56] My live test instance of MediaWiki is apparently very behind. [01:55:03] 134 MB of objects received. [01:58:16] (03PS1) 10Yaron Koren: Replaced wfMsg() calls with wfMessage() (finally) [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115098 [02:01:33] (03CR) 10Yaron Koren: [C: 032 V: 031] Replaced wfMsg() calls with wfMessage() (finally) [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115098 (owner: 10Yaron Koren) [02:01:35] (03Merged) 10jenkins-bot: Replaced wfMsg() calls with wfMessage() (finally) [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115098 (owner: 10Yaron Koren) [02:46:18] Gloria: what is the correct way to access IP::$IP_ADDRESS_STRING ? [02:46:20] that does not work [02:46:30] maybe because it is using define() [02:46:46] What are you trying to do? [02:47:06] SQL - limit page titles to IPs [02:47:57] there is already a regex, so didn't want to reinvent the wheel [02:50:31] huh: just use IP_ADDRESS_STRING [02:50:47] if this is PHP [02:51:48] (03PS1) 10Gnosygnu: v1.2.4.1 [xowa] - 10https://gerrit.wikimedia.org/r/115099 [02:51:51] thanks [02:52:41] apparently is not a valid SQL regexp :S [02:52:59] legoktm: I'm probably taking a wrong approach anyway, but would you mind taking a look? [02:53:08] http://abusefilter-global-main.instance-proxy.wmflabs.org/index.php?limit=50&tagfilter=&title=Special%3AContributions&contribs=user&target=Test+account+please+ignore&namespace=3&hideAnon=1&year=2014&month=-1 [02:53:34] too many \ ? [02:53:39] um [02:53:44] what is it you're trying to do? [02:54:05] legoktm: https://bugzilla.wikimedia.org/show_bug.cgi?id=42357 [02:54:17] oops, wrong one [02:54:25] I was like.... :P [02:54:45] https://bugzilla.wikimedia.org/show_bug.cgi?id=43362 [02:55:28] I'm fairly sure my approach would work in theory, in practice it might not be the most efficient method (IDK) [02:55:49] a) I don't think we should fix that "bug", and b) I don't think your approach will work since it only handles IPs and not registered users [02:56:53] legoktm: b) to handle registered users, I'm just flipping the regexp. I don't think we should treat usernames that are not registered different from registered ones [02:57:10] page_title REGEXP IP thing [02:57:12] then...how would that fix the bug? [02:57:13] page_title NOT REGEXP IP thing [02:57:36] "allow the selective display of either edits to the pages of registered accounts or the pages of anonymous (IP address) editors" [02:57:42] oh [02:57:43] I see [02:57:43] or [02:57:52] replace "registered accounts" with "username" and this works in theory [02:58:12] a) why not fix this bug? [02:58:18] It could be useful [02:58:31] I misread it [02:58:44] I guess that would work, I'm not sure how efficient REGEXP in mysql is though [02:59:00] Why are you using regex to determine whether a user is an anon? [02:59:08] Gloria: is there another way in SQL? [02:59:14] Which table are you dealing with? [02:59:20] rev_user_id = 0 ? [02:59:28] page :/ [02:59:44] If this is Special:Contributions, it's revision, not page. [02:59:52] Well, it's both. [02:59:55] the page_title is involved somehow [02:59:58] Sure. [03:00:04] User:127.0.0.1 would be matched, not User:Example [03:00:08] revision.rev_page maps to page.page_id. [03:00:21] page.page_id gives you page.page_title and page.page_namespace [03:00:50] But for each revision, you get rev_user_text and rev_user. [03:00:53] Yeah, but I'm checking the target page as a IP talk page, not whether the edit was by an IP [03:00:55] (Sorry, not rev_user_id.) [03:01:02] Ah. [03:01:09] See https://bugzilla.wikimedia.org/show_bug.cgi?id=43362 [03:01:16] (maybe WONTFIX would be better? :P ) [03:01:24] But I thought I'd give it a shot [03:01:36] I'm not sure what the use-case is. [03:01:48] * Gloria looks at bug 43179... [03:02:48] I left a comment on bug 43362. [03:03:00] He's proposing an implementation without providing a use-case or problem. [03:05:04] (03CR) 10Gnosygnu: [C: 032 V: 032] "approving own commit" [xowa] - 10https://gerrit.wikimedia.org/r/115099 (owner: 10Gnosygnu) [03:19:10] (03PS13) 10Lekshmi: Changed deprecated $wgCopyrightIcon [core] - 10https://gerrit.wikimedia.org/r/109903 [03:19:34] (03PS9) 10Catrope: Add VisualEditor support [extensions/Math] - 10https://gerrit.wikimedia.org/r/105647 [03:24:51] (03PS1) 10Physikerwelt: LaTeXML literal workaround [extensions/Math] (dev) - 10https://gerrit.wikimedia.org/r/115100 [03:24:53] (03CR) 10jenkins-bot: [V: 04-1] LaTeXML literal workaround [extensions/Math] (dev) - 10https://gerrit.wikimedia.org/r/115100 (owner: 10Physikerwelt) [03:25:29] (03PS2) 10Physikerwelt: LaTeXML literal workaround [extensions/Math] (dev) - 10https://gerrit.wikimedia.org/r/115100 [03:27:17] (03CR) 10Physikerwelt: [C: 032] LaTeXML literal workaround [extensions/Math] (dev) - 10https://gerrit.wikimedia.org/r/115100 (owner: 10Physikerwelt) [03:27:25] (03Merged) 10jenkins-bot: LaTeXML literal workaround [extensions/Math] (dev) - 10https://gerrit.wikimedia.org/r/115100 (owner: 10Physikerwelt) [03:32:13] How can I do a check on whether a particular extension is installed on someone´s local wiki or not? [03:36:58] Niharika: look at Special:Version [03:39:38] (03CR) 10Springle: [C: 031] "Am going ahead and scheduling the schema change as it seems necessary and can be done easily on a slave-by-slave basis." [extensions/AbuseFilter] - 10https://gerrit.wikimedia.org/r/110881 (owner: 10Se4598) [03:42:09] MatmaRex: I think Niharika was asking how to check programmatically. [03:42:16] Though maybe not. [03:42:53] springle: I'm not a trusted user/patch submitter, and there's some issue with resubmitting by others where it doesn't trigger for whatever reason [03:42:55] Niharika: What are you trying to do? [03:49:35] (03PS2) 10Prtksxna: Make padding consistent for extract and timestamp [extensions/Popups] - 10https://gerrit.wikimedia.org/r/114920 [03:51:15] surely there is a corresponding API [03:54:12] (03PS13) 10Prtksxna: Add Triangle/Pokey to the NavigationPopups [extensions/Popups] - 10https://gerrit.wikimedia.org/r/113951 [03:54:36] (03PS4) 10Prtksxna: Truncate text for portrait popups [extensions/Popups] - 10https://gerrit.wikimedia.org/r/114917 [04:05:35] Gloria: Sorry, was away. Yes, I´m trying to check programmatically. I have a code that should run if a specific extension is installed. [04:05:45] (03CR) 10Yuvipanda: [C: 032 V: 032] Truncate text for portrait popups [extensions/Popups] - 10https://gerrit.wikimedia.org/r/114917 (owner: 10Prtksxna) [04:05:54] If it isn´t installed, it gives an error for now. [04:06:02] (03CR) 10Yuvipanda: [C: 032 V: 032] Make padding consistent for extract and timestamp [extensions/Popups] - 10https://gerrit.wikimedia.org/r/114920 (owner: 10Prtksxna) [04:06:07] You probably need to check for the existence of a specific PHP hook, Niharika. [04:06:08] I want to avoid displaying that error. [04:06:17] Okay. [04:33:47] (03CR) 10Springle: API: Add prop=redirects and list=allredirects (031 comment) [core] - 10https://gerrit.wikimedia.org/r/104764 (owner: 10Anomie) [04:41:43] (03PS14) 10Prtksxna: Add Triangle/Pokey to the NavigationPopups [extensions/Popups] - 10https://gerrit.wikimedia.org/r/113951 [04:49:50] (03CR) 10Gergő Tisza: [C: 031] "Everything works now. This will be good for prototyping; before it gets merged, some UX issues need to be solved:" [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/109263 (owner: 10Inchikutty) [04:50:47] (03CR) 10Gergő Tisza: "One more thing for this patchset: could you make "OpenStreetMap" in the " Map data by OpenStreetMap contributors" link to the OSM website?" [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/109263 (owner: 10Inchikutty) [04:57:52] rdwrer: ping [05:27:59] (03CR) 10Gergő Tisza: "I realize this is still WIP, but a few comments about the UI:" [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/112342 (owner: 10Inchikutty) [05:53:09] (03PS1) 10Physikerwelt: WIP: Identification of identifiers [extensions/MathSearch] - 10https://gerrit.wikimedia.org/r/115107 [05:58:39] (03CR) 10Prtksxna: [C: 04-1] "Check out the MediaWiki code conventions for CSS and JS" (035 comments) [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/107892 (owner: 10Inchikutty) [06:01:39] I am trying to push some changes to gerrit but on running 'git review' its telling me that there are no changes to push :\ http://pastie.org/pastes/8763923/text [06:09:22] (03PS1) 10Matthias Mullie: Third level of nesting [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115110 [06:12:32] (03CR) 10Prtksxna: [C: 04-1] "Check out the MediaWiki code conventions for JS - https://www.mediawiki.org/wiki/Manual:Coding_conventions/JavaScript" (034 comments) [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/109263 (owner: 10Inchikutty) [06:16:11] (03PS8) 10Legoktm: Make abstract Config class truly implementation-agnostic [core] - 10https://gerrit.wikimedia.org/r/109850 (owner: 10Ori.livneh) [07:09:19] (03PS13) 1001tonythomas: Ticking search in all namespaces should tick all the remaining ones. [core] - 10https://gerrit.wikimedia.org/r/110610 [07:10:39] (03PS14) 1001tonythomas: Ticking search in all namespaces should tick all the remaining ones. [core] - 10https://gerrit.wikimedia.org/r/110610 [07:15:25] (03CR) 10Nikerabbit: "Please read https://www.mediawiki.org/wiki/Manual:Messages_API" (0321 comments) [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115098 (owner: 10Yaron Koren) [07:53:09] (03PS1) 10Matthias Mullie: Non-hacky way to feed title to Parsoid [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115112 [07:53:14] (03PS1) 10Matthias Mullie: [WIP] Add preview of new comment to watchlist/rc/history [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115113 [07:53:41] (03CR) 10jenkins-bot: [V: 04-1] Non-hacky way to feed title to Parsoid [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115112 (owner: 10Matthias Mullie) [07:54:08] (03CR) 10jenkins-bot: [V: 04-1] [WIP] Add preview of new comment to watchlist/rc/history [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115113 (owner: 10Matthias Mullie) [07:59:08] (03PS1) 10Ori.livneh: Use NFS by default on non-Windows hosts [vagrant] - 10https://gerrit.wikimedia.org/r/115114 [08:10:03] Any ideas on http://pastie.org/pastes/8763923/text ? I tried adding another commit and I am still unable to push any changes :( [08:53:33] prtksxna: pong [08:53:40] prtksxna: Hello! When you're on IRC it's usually good practice to jump right in and ask a question, as it saves time for everyone involved :) how can we help you? [08:53:58] Oh! rdwrer I had pinged you because you had asked me to [08:54:14] Oh, what about? [08:54:32] rdwrer: The triangle commit https://gerrit.wikimedia.org/r/#/c/113951/ [08:55:34] rdwrer: (off-topic) I've been trying to push some changes to gerrit but its been failing for some time - http://pastie.org/pastes/8763923/text - added more commits - http://pastie.org/pastes/8764520/text - still failing [08:56:41] Huh. [08:57:35] That's mighty funky [08:57:52] (03PS2) 10Nemo bis: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 (owner: 10Mjnovice) [08:58:04] (03PS3) 10Nemo bis: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 (owner: 10Mjnovice) [08:58:07] prtksxna: Actually that actively makes no sense. [08:59:09] rdwrer: lol [08:59:20] rdwrer: Should I remove the change from gerrit or something? [08:59:24] Like, the commit hash is different [08:59:27] (03CR) 10Nemo bis: [C: 04-1] "Thanks for your first patch and welcome." [core] - 10https://gerrit.wikimedia.org/r/114966 (owner: 10Mjnovice) [08:59:29] There are clearly changes [08:59:30] rdwrer: Anything I can try? [08:59:35] The error message is just idiotic [08:59:36] rdwrer: Yes, there are! [08:59:49] prtksxna: I vote you rebase locall. [08:59:50] y. [09:00:01] prtksxna: git fetch --all && git rebase origin/master [09:00:11] I doubt it will help [09:00:53] rdwrer: I tried doing that, I did g fetch origin && g rb origin/master [09:01:08] rdwrer: You think --all would help? I know there is only one branch? [09:01:19] s/?/! [09:01:21] Hrmflargles. [09:01:33] prtksxna: --all wouldn't help, I do it out of habit [09:01:38] oic [09:02:10] rdwrer: I even made useless changes to the commits to make sure they are different :P [09:03:12] Hah. [09:04:05] prtksxna: That's super weird [09:04:18] I don't understand what's going on, I assume your next step is to bother ^d about it [09:05:38] Ok, will you look at that commit anyway, I dont think much has changed there [09:06:21] rdwrer: But don't +2 it, I'll have to check if there are any local changes [09:08:32] rdwrer: I got disconnected for a bit, did I miss anything? [09:09:07] (03PS1) 10Hashar: Add jobs for Persona and OpenBadges extensions [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115125 [09:09:08] No. [09:09:16] prtksxna: I think Chad's your next stop [09:09:19] We'll see what happens [09:09:25] Bother me when I can review to merge [09:09:34] rdwrer: Chad? For what? [09:09:40] For Gerrit help [09:09:46] rdwrer: nick? [09:09:46] Seems like Gerrit's fuckin' you over [09:09:54] It does :( [09:09:55] ^d; I already pinged him [09:10:00] (03CR) 10Hashar: [C: 032] "Jenkins jobs added with https://gerrit.wikimedia.org/r/115125" [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115092 (owner: 10Parent5446) [09:10:09] (03Merged) 10jenkins-bot: Add projects for Persona and OpenBadges extensions [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115092 (owner: 10Parent5446) [09:10:11] Oh, ok, thanks rdwrer o/ [09:10:41] (03CR) 10Hashar: [C: 032] Add jobs for Persona and OpenBadges extensions [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115125 (owner: 10Hashar) [09:10:59] (03Merged) 10jenkins-bot: Add jobs for Persona and OpenBadges extensions [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115125 (owner: 10Hashar) [09:11:34] ^d: Hey Chad o/ I am unable to push changes to gerrit - http://pastie.org/pastes/8763923/text - http://pastie.org/pastes/8764520/text [09:16:54] (03PS1) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/OpenBadges] - 10https://gerrit.wikimedia.org/r/115126 [09:16:56] (03CR) 10jenkins-bot: [V: 04-1] Jenkins job validation (DO NOT SUBMIT) [extensions/OpenBadges] - 10https://gerrit.wikimedia.org/r/115126 (owner: 10Hashar) [09:17:06] (03PS1) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/Persona] - 10https://gerrit.wikimedia.org/r/115127 [09:18:00] (03Abandoned) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/Persona] - 10https://gerrit.wikimedia.org/r/115127 (owner: 10Hashar) [09:19:31] (03CR) 10Hashar: "The test job for Persona works, example commit is https://gerrit.wikimedia.org/r/#/c/115127/" [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115092 (owner: 10Parent5446) [09:19:35] (03Abandoned) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/OpenBadges] - 10https://gerrit.wikimedia.org/r/115126 (owner: 10Hashar) [09:29:37] (03PS1) 10Shirayuki: i18n: use "int:" for consistency [core] - 10https://gerrit.wikimedia.org/r/115128 [09:42:01] (03PS4) 10Mjnovice: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 [09:46:08] (03PS5) 10Mjnovice: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 [10:43:54] (03CR) 10Ori.livneh: [C: 032] Use NFS by default on non-Windows hosts [vagrant] - 10https://gerrit.wikimedia.org/r/115114 (owner: 10Ori.livneh) [10:44:12] (03Merged) 10jenkins-bot: Use NFS by default on non-Windows hosts [vagrant] - 10https://gerrit.wikimedia.org/r/115114 (owner: 10Ori.livneh) [10:44:27] (03PS1) 10Ori.livneh: HHVM role: use "hhvm-fastcgi" package [vagrant] - 10https://gerrit.wikimedia.org/r/115134 [10:45:07] (03CR) 10Ori.livneh: [C: 032] HHVM role: use "hhvm-fastcgi" package [vagrant] - 10https://gerrit.wikimedia.org/r/115134 (owner: 10Ori.livneh) [10:45:09] (03Merged) 10jenkins-bot: HHVM role: use "hhvm-fastcgi" package [vagrant] - 10https://gerrit.wikimedia.org/r/115134 (owner: 10Ori.livneh) [10:56:56] (03PS17) 10Physikerwelt: Coverage tests for the Math extension [extensions/Math] - 10https://gerrit.wikimedia.org/r/112353 [10:59:42] (03CR) 10Hashar: [C: 031] "\O/" [extensions/Math] - 10https://gerrit.wikimedia.org/r/112353 (owner: 10Physikerwelt) [11:14:24] (03CR) 10Addshore: [C: 031] Coverage tests for the Math extension [extensions/Math] - 10https://gerrit.wikimedia.org/r/112353 (owner: 10Physikerwelt) [11:14:35] hashar: hello [11:20:30] ori: good evening :-] [11:20:32] hashar: which service in prod do you think would make the most sense as the initial hhvm migration target? [11:20:32] are there any that do not depend on memcached? [11:21:02] cause memcached is broken in hhvm ? [11:21:10] I guess that is because we rely on the php extension dont we? [11:21:15] memcached is fine, but there is no igbinary support yet [11:21:17] we could always fallback to the pure PHP implementation if it works [11:21:30] it works, but igbinary serialization / deserialization does not [11:22:05] at the same time, it is probably not a lot of work to port it -- maybe a few days at the most [11:22:09] might compile it without igbinary :] [11:22:19] as for which service, I have no idea [11:22:24] but then the contents of memcached are useless [11:22:33] I think Chad proposed to make the multi version scripts to use hhvm first [11:22:58] I would love to have a contint slave to run under hvvm as well (the full mw core test suite) [11:23:11] I think he proposed the l10n scripts, but actually the multiversion scripts would be even easier, that's a good idea [11:23:49] with hhvm, maybe we will no more have to rely on cdb files [11:24:46] for contint, I need an instance in labs using Ubuntu 14.04. That depends on puppet passing on that ubuntu version (I think faidon worked on that already) [11:25:03] then having an image in labs, that depends on labs migration from pmtpa to eqiad [11:25:31] "standard" with 14.04 should work [11:25:33] then it is all about creating the 14.04 instance, figure out how to get hhmv installed, register the instance has a slave in Jenkins and write a tiny job that run there :] [11:25:41] but I haven't tested anything more complicated than that [11:25:44] but I don't expect huge troubles [11:26:01] hopefully :-] [11:26:47] morning paravoid [11:26:59] good morning ori [11:27:05] I suppose it's morning for you too :) [11:27:47] paravoid: what would it take to get hhvm packages in prod? what sort of timeline do you think is reasonable? [11:28:20] the packages need to be redone imho [11:28:35] and the person who volunteered at that bug report to do it hasn't acted on it [11:28:51] (which bug?) [11:28:58] the Debian PHP team responded to me that they're very willing to collaborate [11:29:24] so I guess it falls on me to package it up [11:29:53] http://bugs.debian.org/727085 [11:29:55] (03CR) 10Hashar: "Workaround for https://bugzilla.wikimedia.org/show_bug.cgi?id=55668 "Jenkins: jsduck test is sometimes passing when the build contains war" [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/113033 (owner: 10MarkTraceur) [11:30:02] plus a bunch of private threads [11:30:42] plus http://lists.alioth.debian.org/pipermail/pkg-php-maint/2014-January/012988.html [11:31:43] (03CR) 10Phuedx: [C: 032] Log page creations with the PageCreation schema [extensions/WikimediaEvents] - 10https://gerrit.wikimedia.org/r/114905 (owner: 10Mattflaschen) [11:31:45] (03Merged) 10jenkins-bot: Log page creations with the PageCreation schema [extensions/WikimediaEvents] - 10https://gerrit.wikimedia.org/r/114905 (owner: 10Mattflaschen) [11:32:48] paravoid: do we have the hhmv /debian/ sources somewhere ? [11:33:00] paravoid: last time I looked I could not find them :-( [11:33:46] paul doesn't build packages like that [11:33:52] like... a package [11:33:54] paravoid: what would be the practical negative consequences of using facebook's packages for cli-invoked cron-like tasks? [11:34:04] he just packs a source tree up with dpkg-deb [11:34:14] i.e. very wrong [11:38:38] paravoid: ? [11:40:22] (03PS13) 10Hashar: tests to compare En, Qqq and messages.inc [core] - 10https://gerrit.wikimedia.org/r/5550 [11:42:28] (03CR) 10jenkins-bot: [V: 04-1] tests to compare En, Qqq and messages.inc [core] - 10https://gerrit.wikimedia.org/r/5550 (owner: 10Hashar) [11:47:24] my opinion is that we should disregard those packages as they're very poorly written [11:48:25] what's your roadmap looking like? [11:49:38] if we don't show substantial progress w/hhvm (where "substantial progress" likely means getting some period tasks running under hhvm, not necessarily serving web requests) we likely won't get rob's ok to continue working on it full-time [11:49:45] *by the end of march [11:50:42] the hard blockers are igbinary serialization support (which i could do, and which could be bundled separately as an .so file) and packaging [11:54:30] paravoid: do you think that's viable? [11:55:21] well, as much as I sympathise with you, I think that cutting corners to show progress "or else" is kind of the wrong approach :) [11:55:33] (03PS1) 10Hashar: Jobs for mw/ext/FlaggedRev [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115140 [11:55:45] we're a bit overloaded as a team right now (again); I'd happy to take the fall for the delay [11:55:50] (03CR) 10Hashar: [C: 032] Jobs for mw/ext/FlaggedRev [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115140 (owner: 10Hashar) [11:56:02] I have my hands full for the next couple of weeks at least [11:56:07] (03PS1) 10Aude: Remove usage of wikiGetlink deprecated function [extensions/Echo] - 10https://gerrit.wikimedia.org/r/115141 [11:56:12] (03Merged) 10jenkins-bot: Jobs for mw/ext/FlaggedRev [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115140 (owner: 10Hashar) [11:56:22] I've done some preliminary work for hhvm in a local tree, but it's not a piece of cake, there are some complications [11:56:28] the "or else" isn't to motivate corner-cutting; rob's rationale is that there are a lot of other things that are asked of platform and that we couldn't defer them without something substantial to show for it [11:57:52] (03PS1) 10Hashar: Triggers jobs for mw/ext/FlaggedRev [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115142 [11:58:08] paravoid: OK, not trying to rush you or anything, but what sort of rough timeline do you think _would_ be sensible? [11:58:45] (03CR) 10Hashar: [C: 032] Triggers jobs for mw/ext/FlaggedRev [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115142 (owner: 10Hashar) [11:59:07] (03PS2) 10Hashar: Triggers jobs for mw/ext/FlaggedRev [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115142 [11:59:11] (03CR) 10Hashar: [C: 032] Triggers jobs for mw/ext/FlaggedRev [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115142 (owner: 10Hashar) [11:59:15] (03Merged) 10jenkins-bot: Triggers jobs for mw/ext/FlaggedRev [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115142 (owner: 10Hashar) [11:59:51] not for the next 2-3 weeks at least; probably I can give it some time the week of the 17th? [12:00:06] tentative, I'd have to check with mark as well to be sure [12:00:29] (03PS1) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/FlaggedRevs] - 10https://gerrit.wikimedia.org/r/115143 [12:00:35] sure, whatever you can manage. is there anything i can take off your plate to make things easier? [12:00:36] (03CR) 10jenkins-bot: [V: 04-1] Jenkins job validation (DO NOT SUBMIT) [extensions/FlaggedRevs] - 10https://gerrit.wikimedia.org/r/115143 (owner: 10Hashar) [12:00:58] (don't say 'jar deployments'! :P) [12:01:51] not really, but thanks [12:02:01] we've been focusing a bit on internal priorities these past few weeks [12:02:18] ulsfo, eqiad expansion, tampa migration [12:02:34] which is blocking other stuff or bleeding us money [12:02:40] * ori nods [12:03:04] but yeah, this makes us less available to other teams unfortunately :( [12:04:56] (03Abandoned) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/FlaggedRevs] - 10https://gerrit.wikimedia.org/r/115143 (owner: 10Hashar) [12:06:56] (03PS1) 10Hashar: mwext-FlaggedRevs-testextensions-master is not passing [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115147 [12:07:22] (03CR) 10Hashar: [C: 032] mwext-FlaggedRevs-testextensions-master is not passing [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115147 (owner: 10Hashar) [12:07:28] (03Merged) 10jenkins-bot: mwext-FlaggedRevs-testextensions-master is not passing [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115147 (owner: 10Hashar) [12:10:04] hashar: do you know what depends on the GeoIP PHP extension? [12:10:19] centralauth I think [12:10:24] at least that :) [12:10:53] but implicitly, I think the actual function is in the config, not in the extension itself (iirc, I'm not sure) [12:11:03] morning [12:11:16] hello Nikerabbit [12:13:59] (03PS1) 10Hashar: Make some extension jobs voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115148 [12:14:11] (03CR) 10Hashar: [C: 032] Make some extension jobs voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115148 (owner: 10Hashar) [12:14:15] (03Merged) 10jenkins-bot: Make some extension jobs voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115148 (owner: 10Hashar) [12:14:48] ori: no idea :/ [12:17:51] (03Restored) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/SideBarMenu] - 10https://gerrit.wikimedia.org/r/105852 (owner: 10Hashar) [12:17:54] (03PS2) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/SideBarMenu] - 10https://gerrit.wikimedia.org/r/105852 [12:18:38] (03Abandoned) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/SideBarMenu] - 10https://gerrit.wikimedia.org/r/105852 (owner: 10Hashar) [12:50:39] (03PS7) 10Physikerwelt: WIP: phpunit test for XQuery generation [extensions/MathSearch] - 10https://gerrit.wikimedia.org/r/111468 (owner: 10Etienne) [13:11:18] (03PS19) 10Inchikutty: Embedding Static Map to UploadWizard [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/107892 [13:23:20] (03PS14) 10Inchikutty: Adding Leaflet scripts to UploadWizard [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/109063 [13:23:28] back [13:57:06] prtksxna: no ther is no such class, and yes there is one error.. [13:57:16] ds406: What's the error? [13:57:18] http://localhost/xtra/core/load.php?debug=false&lang=en&modules=startup&only=scripts&skin=vector&* [13:57:36] o sorry, copied wrong one.. [13:58:48] Hey zz_yuvipanda is there a role for popups on vagrant, can you tell ds406? [13:59:22] Failed to load resource: the server responded with a status of 500 (Internal Server Error) "and the link that followed was above;^^" [13:59:52] ds406: try vagrant enable-role popups and then vagrant provision [14:07:37] prtksxna: ran those two cmds w/o any errors but now no links are wrking :( wait, i'll send you through localtunnel, see if can fix it... [14:07:58] ds406: sure [14:13:57] (03PS1) 10Manybubbles: Remove bad instructions [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115163 [14:17:00] (03PS1) 10Hashar: Browser tests jobs for CirrusSearch [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115164 [14:18:38] (03PS2) 10Manybubbles: Remove bad instructions [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115163 [14:29:19] (03CR) 10Chad: [C: 032] Remove bad instructions [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115163 (owner: 10Manybubbles) [14:29:24] (03Merged) 10jenkins-bot: Remove bad instructions [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115163 (owner: 10Manybubbles) [14:35:26] (03CR) 10Thiemo Mättig (WMDE): [C: 031] "Tested in Opera 12, Chromium 32 and Firefox 27 (all Ubuntu). All work as expected. Enter does what it normally does. Shift+Enter opens a n" [core] - 10https://gerrit.wikimedia.org/r/48252 (owner: 10Bartosz Dziewoński) [14:37:55] (03PS2) 10Hashar: Browser tests jobs for CirrusSearch [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115164 [14:47:04] (03PS15) 10Bartosz Dziewoński: Ticking 'Search in all namespaces' in prefs should disable other checkboxes [core] - 10https://gerrit.wikimedia.org/r/110610 (owner: 1001tonythomas) [14:48:22] (03CR) 10Bartosz Dziewoński: [C: 032] "I fixed some whitespace and updated the commit message to match what the patch actually does now. Thanks!" [core] - 10https://gerrit.wikimedia.org/r/110610 (owner: 1001tonythomas) [14:50:08] (03PS3) 10Umherirrender: Fix possible error list for action=query&list=blocks [core] - 10https://gerrit.wikimedia.org/r/112734 [14:52:34] (03CR) 10Bartosz Dziewoński: [C: 032] Fix possible error list for action=query&list=blocks [core] - 10https://gerrit.wikimedia.org/r/112734 (owner: 10Umherirrender) [14:53:06] (03Merged) 10jenkins-bot: Ticking 'Search in all namespaces' in prefs should disable other checkboxes [core] - 10https://gerrit.wikimedia.org/r/110610 (owner: 1001tonythomas) [14:54:51] (03CR) 10Chad: "CCing myself after the fact because search. Guess nobody remembered :)" [core] - 10https://gerrit.wikimedia.org/r/110610 (owner: 1001tonythomas) [14:56:05] (03Merged) 10jenkins-bot: Fix possible error list for action=query&list=blocks [core] - 10https://gerrit.wikimedia.org/r/112734 (owner: 10Umherirrender) [14:57:30] (03PS3) 10Umherirrender: Exclude prop=uploadwarning from allimages and stashimageinfo [core] - 10https://gerrit.wikimedia.org/r/112341 [15:01:27] (03PS1) 10Hashar: fetch-mw-ext: fallback to git.wikimedia.org zip [integration/jenkins] - 10https://gerrit.wikimedia.org/r/115169 [15:01:44] (03CR) 10Hashar: [C: 032] fetch-mw-ext: fallback to git.wikimedia.org zip [integration/jenkins] - 10https://gerrit.wikimedia.org/r/115169 (owner: 10Hashar) [15:01:47] (03Merged) 10jenkins-bot: fetch-mw-ext: fallback to git.wikimedia.org zip [integration/jenkins] - 10https://gerrit.wikimedia.org/r/115169 (owner: 10Hashar) [15:04:08] (03CR) 10Chad: "No no no, dear god no. These are *not* cached and are expensive to generate. Services needing tarballs should use something reliable like " [integration/jenkins] - 10https://gerrit.wikimedia.org/r/115169 (owner: 10Hashar) [15:08:32] (03PS1) 10Cmcmahon: Permissions on test2wiki make this test troublesome [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115170 [15:09:32] prateek: are you trying to run load.php file? [15:09:44] prtksxna: are you trying to run load.php file? [15:09:51] (03Abandoned) 10Cmcmahon: Permissions on test2wiki make this test troublesome [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115170 (owner: 10Cmcmahon) [15:10:01] (03PS4) 10Umherirrender: Document $wgResourceLoaderMaxage in DefaultSettings.php [core] - 10https://gerrit.wikimedia.org/r/110637 [15:10:33] hey people, i need someone to merge https://gerrit.wikimedia.org/r/#/c/48252/9 and its dependency, it's been reviewed by thiemo so far [15:10:57] the patch is over a year old :( [15:14:17] (03PS11) 10Bartosz Dziewoński: ChangesListSpecialPage: Implement two new hooks superseding 4 old ones [core] - 10https://gerrit.wikimedia.org/r/106279 [15:16:26] (03CR) 10Chad: Make abstract Config class truly implementation-agnostic (031 comment) [core] - 10https://gerrit.wikimedia.org/r/109850 (owner: 10Ori.livneh) [15:16:37] (03CR) 10Chad: "Other than my final nit, I'm happy with this as-is. If we've got something approaching consensus, let's get this in. Followups always poss" [core] - 10https://gerrit.wikimedia.org/r/109850 (owner: 10Ori.livneh) [15:16:38] hey MatmaRex_ [15:16:38] testing it now [15:17:10] (03CR) 10Bartosz Dziewoński: [C: 04-1] "OK, let's deprecate the stuff." [core] - 10https://gerrit.wikimedia.org/r/106279 (owner: 10Bartosz Dziewoński) [15:17:17] thanks YuviPanda [15:17:30] * YuviPanda waits for vagrant to boot [15:17:37] prtksxna: there is a popups role, yes [15:19:17] morning YuviPanda [15:19:22] hello nike [15:20:38] i mean [15:20:38] hi Nikerabbit [15:22:02] (03PS5) 10Bartosz Dziewoński: jquery.suggestions: Do not duplicate keypress logic [core] - 10https://gerrit.wikimedia.org/r/82814 [15:22:22] MatmaRex: so they look sane enough to me. I'm going to merge. [15:22:41] MatmaRex: it's also only monday, so we'll hopefully hear screams before that if it breaks something suble [15:22:53] MatmaRex: also, why do those patches have to be dependent? [15:23:17] (03PS12) 10Bartosz Dziewoński: ChangesListSpecialPage: Implement two new hooks superseding 4 old ones [core] - 10https://gerrit.wikimedia.org/r/106279 [15:23:23] YuviPanda: heh [15:24:53] YuviPanda: because preventing default handling in the keydown handler causes keydown handler not to fire, iirc [15:24:58] ugh [15:25:06] causes keypress event* not to fire [15:25:08] keypress? [15:25:09] right [15:25:21] or some other one [15:25:24] (03CR) 10Yuvipanda: [C: 032] "Looks sane, tests fine." [core] - 10https://gerrit.wikimedia.org/r/82814 (owner: 10Bartosz Dziewoński) [15:25:33] (03PS10) 10Bartosz Dziewoński: jquery.suggestions, mediawiki.searchSuggest: Fix form submission [core] - 10https://gerrit.wikimedia.org/r/48252 [15:25:35] i'm not sure right now, it didn't work anyway [15:26:01] actually, i think it was preventing the submit event [15:26:12] "accuracy : accuracy of lat/lng from 1=estimated to 6=centroid" wth does 'centroid' mean? [15:26:37] (03CR) 10Yuvipanda: [C: 032] jquery.suggestions, mediawiki.searchSuggest: Fix form submission [core] - 10https://gerrit.wikimedia.org/r/48252 (owner: 10Bartosz Dziewoński) [15:26:37] nm me, just random venting. [15:26:38] Amgine: context? [15:26:39] MatmaRex: there. [15:26:45] YuviPanda: yay [15:26:52] MatmaRex: :) [15:26:57] Geonames, MatmaRex. [15:27:15] by "context", i meant "link". :P [15:27:20] centroid is https://en.wikipedia.org/wiki/Centroid [15:27:43] but i can't tell if that link is relevant [15:27:54] http://download.geonames.org/export/dump/readme.txt [15:28:59] Mmm, is relevant, but sounds like something a google marketer might come up with. [15:30:26] (03Merged) 10jenkins-bot: jquery.suggestions: Do not duplicate keypress logic [core] - 10https://gerrit.wikimedia.org/r/82814 (owner: 10Bartosz Dziewoński) [15:31:14] (03Merged) 10jenkins-bot: jquery.suggestions, mediawiki.searchSuggest: Fix form submission [core] - 10https://gerrit.wikimedia.org/r/48252 (owner: 10Bartosz Dziewoński) [15:35:43] <^d> Amgine_: The -oid suffix makes it sound like an Android app. [15:35:43] * ^d continues to gripe about this [15:40:41] ^d: hehe... the suffix -oid means "-like", so in this case it means "center-like". Which just as annoying, although in a somewhat different way. [15:41:28] (it's why I suggested 'parsoid' as a name) [15:42:14] (03PS1) 10Phuedx: Make the toolbar respond to a content overflow [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/115176 [15:43:13] <^d> Amgine_: Yes, I know what it's supposed to mean. [15:43:56] <^d> :) [15:44:14] d�monoids, for example... [15:45:41] <^d> dæmons for android, of course! [15:45:49] <^d> parsoid -> parsing for android [15:45:54] <^d> mathoid -> math for android [15:45:55] <^d> :) [15:46:12] (03CR) 10Phuedx: "I might work on an alternative patch that uses matchMedia and the matchMedia.js polyfill (see https://github.com/paulirish/matchMedia.js/)" [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/115176 (owner: 10Phuedx) [15:49:40] hey everyone [15:49:40] average: yeah ping manybubbles and ^d [15:49:40] both in US [15:49:41] manybubbles , ^d hi [15:49:41] average: what is up? [15:49:42] <^d> WHY ME?!? [15:49:51] ^d: because pinging is fun [15:49:51] guruoid [15:49:51] I was just curious how you guys are dealing with packaging jars for deploying on production [15:49:53] I recently heard about archiva [15:49:57] but I'm like... a total n00b with archiva, I have no idea about it.. [15:49:58] do you use debian packages ? [15:49:59] <^d> I'm a noob when it comes to packaging :p [15:49:59] do you stuff everything into one big jar ? [15:50:04] how do you deploy Java stuff ? [15:50:11] <^d> For gerrit, I stuff a jar into a .deb [15:50:16] <^d> Then apt-get install [15:50:36] I can answer [15:50:40] <^d> But the package is awful, so there's a lot of manual wrangling and puppet magic. [15:50:49] so there are two issues: [15:50:53] <^d> Gerrit's a bad example to follow :) [15:51:06] how do you dump jars onto the file system [15:51:08] and [15:51:18] how are you sure that the jar you pull from central is good [15:51:29] we're paranoid about central [15:51:49] so the plan for that last one is to bless dependencies by name and hash [15:52:10] that lets us use whatever java build tool we like when we build our package [15:52:13] ottomata: he can help! [15:52:20] haha, average is asking on my behalf [15:52:26] oh fucks [15:52:29] i want to do that, but don't thikn other opsen would let me get away with it [15:52:34] how'd you get away with that? :p [15:52:53] what is "that"? [15:52:55] There is a binary tarball of archiva that I can build a .deb from [15:52:58] its got all the .jars in [15:52:59] it [15:53:07] i could easily make a deb that installed that [15:53:11] I think we had this conversation on the email [15:53:15] yeah [15:53:20] not much resolution though, right? [15:53:21] in this case a deb out of the binary is the best solution [15:53:26] I think we had resolution [15:53:27] deb [15:53:29] deb [15:53:32] deb [15:53:34] I thought [15:53:45] <^d> The other option is not packaging and just having puppet wget a bunch of crap from central and saying f'it ;-) [15:53:55] <^d> [15:53:58] <^d> [15:54:04] haha, i mean [15:54:19] isn't the deb from binary problem licensing? [15:54:24] if we put the .deb in our apt [15:54:26] ? [15:54:30] we might be doing illegal things? [15:55:07] I'm reasonably sure archiva is all apache licensed so we're ok. [15:55:24] manybubbles: including aaaaal dependencies? [15:55:40] let me check [15:55:50] https://gist.github.com/ottomata/9006927 [15:56:37] that is WIP [16:00:15] (03CR) 10Aklapper: [C: 031] "Deployed on boogs.wmflabs.org and I can still edit bug reports. Can go live. :)" [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/114168 (owner: 10Aklapper) [16:01:37] (03PS5) 10Anomie: API: Add prop=redirects and list=allredirects [core] - 10https://gerrit.wikimedia.org/r/104764 [16:02:00] (03PS6) 10Anomie: API: Add prop=redirects and list=allredirects [core] - 10https://gerrit.wikimedia.org/r/104764 [16:02:50] (03CR) 10Anomie: "PS5: Rebase" [core] - 10https://gerrit.wikimedia.org/r/104764 (owner: 10Anomie) [16:04:02] (03CR) 10Yaron Koren: "Yikes! Thanks for all these fixes. I didn't know whether to go with text(), parse() or escaped(), and it looks like I guessed wrong most o" [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115098 (owner: 10Yaron Koren) [16:05:26] (03CR) 10Anomie: API: Add prop=redirects and list=allredirects (031 comment) [core] - 10https://gerrit.wikimedia.org/r/104764 (owner: 10Anomie) [16:08:06] (03PS1) 10Hashar: Trigger mwext-browsertests-CirrusSearch-phantomjs non voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115179 [16:08:41] (03CR) 10Manybubbles: [C: 031] Trigger mwext-browsertests-CirrusSearch-phantomjs non voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115179 (owner: 10Hashar) [16:08:44] (03CR) 10Hashar: [C: 032] Trigger mwext-browsertests-CirrusSearch-phantomjs non voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115179 (owner: 10Hashar) [16:08:47] (03Merged) 10jenkins-bot: Trigger mwext-browsertests-CirrusSearch-phantomjs non voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115179 (owner: 10Hashar) [16:09:06] I want the Beta Feature hook to run if the Beta Feature extension is installed in a Wiki. Any way to check if an extension is installed? [16:09:12] (03CR) 10Anomie: API: Add prop=flow to detect if Flow is enabled on a page (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/111255 (owner: 10Legoktm) [16:09:44] (03PS1) 10Yaron Koren: Fixes to last commit, based on comments from Nikerabbit [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115180 [16:10:09] (03Restored) 10Hashar: Jenkins job validation (DO NOT SUBMIT).. [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/70205 (owner: 10Hashar) [16:10:15] (03PS7) 10Hashar: Jenkins job validation (DO NOT SUBMIT).. [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/70205 [16:11:50] Niharika: ' if ( ( ( class_exists( 'BetaFeatures' ) && BetaFeatures::isFeatureEnabled( $skin->getUser(), 'popups' ) ) ' [16:12:04] Niharika: the first conditional checks if BetaFeatures is installed [16:12:22] YuviPanda: Okay and the second? [16:12:31] <^d> Niharika: In Cirrus we check if class_exists( 'BetaFeatures' ) [16:12:40] Niharika: that just checks if a particular betafeature (in this case, 'popups') has been enabled [16:12:44] Yeah, that's important. [16:12:59] Niharika: yeah, what ^d said. That's good enough to make sure you are accessing BF only when it is installed [16:13:04] Niharika: If BF isn't enabled you can do helpful testing things like always enable it. [16:13:26] s/isn't enabled/isn't installed/ s/enable it/enable the feature/ [16:13:45] YuviPanda, rdwrer_: Okay. Got it. Thanks. [16:13:52] :) [16:14:09] (03CR) 10Aklapper: [C: 04-1] "I get:" [wikimedia/bugzilla/triagescripts] - 10https://gerrit.wikimedia.org/r/105618 (owner: 10Merlijn van Deen) [16:15:59] Niharika: If its enabled [16:15:59] (as a BetaFeature) [16:16:11] hi prtksxna [16:16:30] prtksxna: did you get your gerrit issue fixed? [16:16:38] Hi prtksxna. Yeah, got it. [16:16:42] prtksxna: you should also poke rdwrer_ about the pokey [16:17:04] hey ^d can you help me with my gerrit issue? http://pastie.org/pastes/8763923/text - http://pastie.org/pastes/8764520/text [16:17:43] There is a strange lag I am getting on IRC :\ Messages are coming in bursts :\ [16:18:32] YuviPanda: No, the gerrit issue is not fixed, rdwrer_ called it "An idiotic error message", asking ^d now… [16:18:33] <^d> Meh, git-review. [16:18:34] <^d> I'd rebase the branch against master. [16:18:34] <^d> `git pull -r` [16:21:18] (03CR) 10Yaron Koren: [C: 032 V: 031] Fixes to last commit, based on comments from Nikerabbit [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115180 (owner: 10Yaron Koren) [16:21:18] (03Merged) 10jenkins-bot: Fixes to last commit, based on comments from Nikerabbit [extensions/ApprovedRevs] - 10https://gerrit.wikimedia.org/r/115180 (owner: 10Yaron Koren) [16:21:39] ^d: I tried rebasing a couple of times, didn't help :\ [16:23:22] prtksxna: if the patch is identical to some other one, gerrit will reject it [16:23:23] prtksxna: you actually can't revert to an older patchset [16:23:23] because it ignores all metedata [16:23:25] i filed a bug about it… [16:23:37] https://bugzilla.wikimedia.org/show_bug.cgi?id=53662 [16:24:16] MatmaRex: But the hash is different, so the patch must be too? [16:24:22] http://pastie.org/pastes/8767912/text [16:24:49] prtksxna: Have you tried squashing commits into one another in funky ways? Maybe that will work. [16:25:11] (03PS1) 10Andrew Bogott: Add createInstance maintenance script for use during migration [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/115185 [16:25:19] Oh, wait. [16:25:31] (03PS15) 10Prtksxna: Add Triangle/Pokey to the NavigationPopups [extensions/Popups] - 10https://gerrit.wikimedia.org/r/113951 [16:25:33] prtksxna: Maybe one of the dependent commits is an old version, so it's complaining [16:25:33] prtksxna: no, the sha1 hash includes all metadata [16:25:33] (03PS1) 10Prtksxna: Limit image size for portrait popups [extensions/Popups] - 10https://gerrit.wikimedia.org/r/115186 [16:25:35] (03PS1) 10Prtksxna: [WIP] Flip the starting point of the popup if its on the right side of the screen [extensions/Popups] - 10https://gerrit.wikimedia.org/r/115187 [16:25:37] (03PS1) 10Prtksxna: [WIP] Flip the arrow position if the popup is flipped [extensions/Popups] - 10https://gerrit.wikimedia.org/r/115188 [16:25:37] \o/ [16:25:38] Oh! [16:25:39] Woo! [16:25:41] prtksxna: even things like committing date and time [16:26:04] just doing `git commit --amend` with no changes will produce a different sha1 afaik [16:26:25] Except it won't let you, IIRC? [16:26:32] MatmaRex: Right, so that should've done it? [16:26:42] rdwrer: You can review https://gerrit.wikimedia.org/r/113951 now [16:26:48] rdwrer: why not? [16:27:09] rdwrer: you normally can't create an empty commit, with no differences [16:27:18] rdwrer: Yup, it didn't, but I made a small change and tried and it still didn't work. I tried that again (removed a new line) now and it worked [16:27:18] but you can suppress that too [16:27:21] MatmaRex: ^ [16:27:35] gerrit is really fucking nasty [16:27:51] can somebody remind me why we're using it? [16:28:20] If only I knew… [16:28:34] <^d> MatmaRex: Phabricator wasn't ready enough 2 years ago. [16:28:38] <^d> Let's migrate already. [16:31:05] Dpp_coder: You should give your correct mysql username and password to complete the setup [16:31:11] Are we sure Phabricator's what we want? [16:31:13] ^d: Ooh, phabricator looks nice [16:31:54] tonythomas: ohhk done it :) [16:32:02] Dpp_coder: :) [16:32:17] ^d: GitLab is maturing, too - I keep meaning to look into it [16:32:17] (03PS1) 10Gerrit Patch Uploader: Add 'newonly' option to Special:Contributions and related i18n message [core] - 10https://gerrit.wikimedia.org/r/115189 [16:32:24] (03CR) 10Gerrit Patch Uploader: "This commit was uploaded using the Gerrit Patch Uploader [1]." [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [16:33:17] <^d> rdwrer: Rails? :( [16:33:36] Well. [16:33:37] Yeah. [16:33:43] There are downsides to everything [16:34:24] The upside is, people who have been whinging about the workflow not being like GitHub will shut up - it's basically a GitHub clone. [16:34:43] So we get most of the benefits of GitHub, minus network. [16:35:06] Why not just use Github? [16:35:13] * prtksxna runs and hides [16:35:57] prtksxna: i'd say that github's pull requests won't scale to what we need [16:36:21] some pretty large projects manage somehow, i guess, but it looks crazy to me [16:36:43] I'd actually have faith in GitHub's structure [16:36:45] it's basically one huge unstructured list [16:36:51] of things that need doing [16:36:54] MatmaRex: Why, whats wrong with it? How is gerrit handling it better? [16:37:16] <^d> I like pre and post commit review. [16:37:19] <^d> What tools have that? :) [16:37:29] <^d> (Emphasis on the *and*) [16:38:12] Tall order... [16:38:17] prtksxna: one thing that is apparently a major blocker is that WMF currently needs per-branch permissions [16:38:25] <^d> rdwrer: Phabricator does ;-) [16:38:40] Oh! [16:38:41] prtksxna: as in, i am not allowed to merged to wmf/* branches in core, even thought i can merge to master [16:38:47] Well, maybe we should be mucking around with it then. [16:38:53] <^d> MatmaRex: We don't, strictly speaking. We could get by without :) [16:38:59] <^d> rdwrer: If only we were ;-) [16:39:12] ^d: i have been told that was the primary reason wy we stuck with the monstrosity we're using now [16:39:13] Well [16:39:22] I'll poke around and see if there's a project I can dump on it. [16:39:25] <^d> MatmaRex: wmf/* branches are mostly an artifact of convenience & getting things moved from svn -> git with minimal disruption for deploys. [16:39:33] <^d> We could very much so live without. [16:39:54] MatmaRex: understood :) [16:39:56] i'm sure you could, but that would require effort, wouldn't it? :P [16:40:19] prtksxna: but there's more things that are really useful in gerrit that github doesn't have [16:40:32] https://www.irccloud.com/pastebin/lfokkgOR [16:40:51] <^d> MatmaRex: Indeed. But I'm saying if it came down to it, ACLs per-branch aren't a strict requirement if we're reevaluating. [16:40:58] tonythomas: [16:41:00] Yeah, we could use separate repos instead, or something [16:41:02] prtksxna: you can't add reviewers to github pull reqs without commenting, you can't "star" a pull requests afaik, the pull reqs are entirely separete for each repository, … [16:41:06] tonythomas: what should I do next? [16:41:22] prtksxna: you can't have dependencies between pull requests [16:41:36] Dpp_coder: now mediawiki is running in your localhost/mediawiki-1.21/ ? [16:41:45] <^d> rdwrer: There's some interesting thoughts I have on that. I'm curious if it would help wmf deploy'd repo sizes as a result [16:41:45] prtksxna: there's no way to express a -1 / +1 review, other than (again) commenting [16:41:53] <^d> (Since we could truncate most of history we don't care about there) [16:41:55] prtksxna: there's no review dashboard [16:42:03] Huh, maybe. [16:42:03] MatmaRex: brb - dinner [16:42:04] Dpp_coder: see, the link can change according to your folder name in /var/www/ [16:42:10] there is basically no structure [16:42:24] and that would drive me mad [16:42:32] tonythomas: its running in http://localhost/mediawiki-1.22.2/index.php/Main_Page [16:42:42] <^d> rdwrer: ie: we care about REL1_13, but not so much when deploying ;-) [16:42:51] oh, it's not customizable too, so we can't have "bug NNN" link to bugzilla [16:43:01] Dpp_coder: looks like your successfully setup mediawiki [16:43:30] tl;dr github would suck for me [16:43:54] Dpp_coder: now, inorder to start working with the contribution process, you will need the developement mediawiki. Follow www.mediawiki.org/wiki/Gerrit/Tutorial [16:43:55] tonythomas: now what?? [16:44:03] it might suck less than gerrit for new contributors, until their patches get even fewer reviews than now because no one can find them in the mess [16:44:33] tonythomas: gerrit?? [16:45:05] <^d> MatmaRex: Github's slogan is "social coding" -- you should be sharing your pull requests with friends. [16:45:10] Dpp_coder: yeah. Gerrit it the code review tool used by mediawiki. You push your patches into here [16:45:13] <^d> ;-) [16:45:17] heh [16:45:29] i do share my pending changesets with friends [16:45:33] pitythey don't review them ;_; [16:45:52] i think i have the most open changes in core again [16:46:04] some are trivial, you should totally review them ^d [16:46:06] <^d> gerrit needs a facebook share button [16:46:37] ^d: facebook share ? :) [16:46:46] (03PS2) 10Bartosz Dziewoński: mediawiki.api: Remove misleading comment about JSONP [core] - 10https://gerrit.wikimedia.org/r/110662 [16:46:48] (03PS2) 10Bartosz Dziewoński: Vector: Delete edit-icon.png [core] - 10https://gerrit.wikimedia.org/r/112301 [16:46:50] (03PS2) 10Bartosz Dziewoński: Create Parser::stripInline to avoid code duplication [core] - 10https://gerrit.wikimedia.org/r/114942 [16:46:53] ^d: these are all trivial ^ [16:47:24] and i've self-merged too many things this month already :P [16:47:38] (03PS3) 10Bartosz Dziewoński: mediawiki.notification: Hide #mw-notification-area when it's empty [core] - 10https://gerrit.wikimedia.org/r/109124 [16:48:28] (03CR) 1001tonythomas: "Should something more be done for this PS ?" [core] - 10https://gerrit.wikimedia.org/r/107360 (owner: 1001tonythomas) [16:54:34] (03PS6) 10Bartosz Dziewoński: Rename RedirectSpecialPage subclasses [core] - 10https://gerrit.wikimedia.org/r/106888 [16:54:38] (03CR) 10Aklapper: "Alright, overwriting is not supported, I just tested this on boogs.wmflabs.org. So we need to stick with a copy of upstream's user-error.h" [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/106883 (owner: 10Tinaj1234) [16:54:40] (03CR) 10Chad: [C: 032] Vector: Delete edit-icon.png [core] - 10https://gerrit.wikimedia.org/r/112301 (owner: 10Bartosz Dziewoński) [16:54:43] (03CR) 10Bartosz Dziewoński: "Rebased again." [core] - 10https://gerrit.wikimedia.org/r/106888 (owner: 10Bartosz Dziewoński) [16:54:54] (03PS2) 10Andrew Bogott: Add createInstance maintenance script for use during migration [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/115185 [16:56:48] (03PS3) 10Andrew Bogott: Add createInstance maintenance script for use during migration [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/115185 [16:57:02] (03Merged) 10jenkins-bot: Vector: Delete edit-icon.png [core] - 10https://gerrit.wikimedia.org/r/112301 (owner: 10Bartosz Dziewoński) [16:57:19] (03CR) 10Aklapper: "This will collide with https://gerrit.wikimedia.org/r/#/c/113685/ so needs rework after." [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/107410 (owner: 10Aklapper) [16:57:30] (03CR) 10Andrew Bogott: [C: 032 V: 032] Add createInstance maintenance script for use during migration [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/115185 (owner: 10Andrew Bogott) [17:06:07] (03CR) 10Bartosz Dziewoński: [C: 04-1] "Thanks for pinging, I forgot about this :) Comments inline." (033 comments) [core] - 10https://gerrit.wikimedia.org/r/107360 (owner: 1001tonythomas) [17:06:11] (03PS1) 10Manybubbles: Setup jenkins configuration [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 [17:06:34] (03CR) 10Manybubbles: "Lets see what Jenkins does with this." [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 (owner: 10Manybubbles) [17:06:34] (03CR) 10jenkins-bot: [V: 04-1] Setup jenkins configuration [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 (owner: 10Manybubbles) [17:27:58] hashar, what's the difference between deployment-staging-cache-mobile02 and deployment-cache-mobile01? [17:28:22] the staging-cache-mobile02 use puppet master self (aka you can hack manifests locally) [17:28:31] it does not serve any traffic, I have created it for yurik to play with [17:28:38] (03PS1) 10Cmcmahon: Attempt to get around Sauce bug where Chrome gets newlines [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115203 [17:28:43] mobile01 serves the m.beta.Wmflabs [17:29:01] (03CR) 10Cmcmahon: [C: 032] Attempt to get around Sauce bug where Chrome gets newlines [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115203 (owner: 10Cmcmahon) [17:29:20] thanks! [17:30:01] (03PS2) 10Hashar: CirrusSearch dependencies for tests [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115199 [17:30:06] hashar, I can't log into deployment-cache-mobile01 - is it something special? [17:30:40] (03CR) 10Hashar: "Dependencies added but the job is still failing https://integration.wikimedia.org/ci/job/mwext-CirrusSearch-testextensions-master/404/cons" [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115199 (owner: 10Hashar) [17:30:49] (03CR) 10Manybubbles: "recheck" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 (owner: 10Manybubbles) [17:31:36] MaxSem: no clue [17:31:52] MaxSem: try to reboot the instance maybe ? [17:32:06] well, it works:] [17:32:10] MaxSem: I can't access it etither [17:32:12] zeljkof: I'm MarkTraceur in Gerrit. :) [17:32:24] rebooting [17:32:34] MaxSem: got to leave, sorry :( [17:32:40] np hashar [17:34:42] (03PS1) 10Tpt: Fix fatal error when usebetatoolbar option was activated and not showtoolbar [extensions/ProofreadPage] - 10https://gerrit.wikimedia.org/r/115205 [17:35:48] (03PS2) 10Gerrit Patch Uploader: Add 'newonly' option to Special:Contributions and related i18n message * Add to Special:Contributions * Add to i18n files * Add 'ucnewonly' (like 'uctoponly') to list=usercontribs * Add to API action=feedcontributions [core] - 10https://gerrit.wikimedia.org/r/115189 [17:35:50] (03CR) 10Gerrit Patch Uploader: "This commit was uploaded using the Gerrit Patch Uploader [1]." [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [17:36:26] (03PS2) 10Manybubbles: Setup jenkins configuration [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 [17:38:32] (03CR) 10PiRSquared17: "Failure because line over 100 characters... will fix." [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [17:38:34] (03CR) 10jenkins-bot: [V: 04-1] Setup jenkins configuration [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 (owner: 10Manybubbles) [17:38:55] hexcode: Hie hexcode, can you tell me how to go about the "catalogue for mediawiki extensions" project? [17:40:06] (03CR) 10PiRSquared17: "Never mind, that seems to not be from my patch. Let another bug or change set handle that." [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [17:40:46] (03PS1) 10Bartosz Dziewoński: DefaultSettings.php: Remove $wgPasswordSenderName [core] - 10https://gerrit.wikimedia.org/r/115206 [17:40:54] (03CR) 10Bartosz Dziewoński: "Definitely. https://gerrit.wikimedia.org/r/115206" [core] - 10https://gerrit.wikimedia.org/r/114111 (owner: 10TTO) [17:41:27] (03CR) 10Bartosz Dziewoński: "Also https://gerrit.wikimedia.org/r/#/c/115206/ for a final followup. :)" [core] - 10https://gerrit.wikimedia.org/r/104707 (owner: 10Theopolisme) [17:42:29] hexmode: Hie hexmode, can you tell me how to go about the "catalogue for mediawiki extensions" project? [17:43:05] rdwrer: Don't forget https://gerrit.wikimedia.org/r/#/c/113951/ I am off to bed now o/ Thanks! [17:43:28] https://gerrit.wikimedia.org/r/#/c/115189/ Does this look okay? [17:44:58] Anyone here? [17:45:39] sure [17:46:07] MatmaRex: https://gerrit.wikimedia.org/r/#/c/115189/ can you review this? [17:46:18] I'd like guidance, as I'm new to this [17:46:51] it doesn't look like frontend-only change, so i dunno if i can, heh [17:46:59] explaining what this does in the commit message would help :) [17:48:04] MatmaRex: I think I did explain in the commit message, no? [17:48:35] Or do you mean a technical explanation like rev_parent_id = 0 ? [17:48:53] i actually mean a layman explanation [17:49:00] you never said what does htis new option do :P [17:49:13] XD I'm an idiot [17:49:30] (i figured it out now, but still) [17:50:07] I thought it was obvious, guess not [17:51:03] (03CR) 10PiRSquared17: Add 'newonly' option to Special:Contributions and related i18n message * Add to Special:Contributions * Add to i18n files * Add 'ucnewonly' (033 comments) [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [17:51:25] huh: sorry, but i have no idea if this is performant enough with the database layout we have [17:51:41] MatmaRex: how could it be any worse than the "toponly" one? [17:51:43] huh: this will potentially result in scanning a lot more rows than now to deliver some results [17:51:53] maybe it's okay [17:52:03] maybe it's not, but i'm not really competent to judge [17:53:06] hexmode: You there? [17:53:58] huh: by the way, add an empty line after the first line of the commit message, otherwise various tools get confused [17:54:01] like grrrit-wm [17:54:15] Oops, sorry [17:54:17] or gerrit itself [17:54:41] MatmaRex: as I said, I'm a n00b [17:54:46] :) [18:01:55] (03CR) 10PiRSquared17: Add 'newonly' option to Special:Contributions and related i18n message * Add to Special:Contributions * Add to i18n files * Add 'ucnewonly' (034 comments) [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [18:02:07] Krenair: did you see it? [18:02:47] see what, huh? [18:03:03] (03CR) 10Legoktm: [C: 032] DefaultSettings.php: Remove $wgPasswordSenderName [core] - 10https://gerrit.wikimedia.org/r/115206 (owner: 10Bartosz Dziewoński) [18:03:09] You're listed as Reviewer [18:04:57] huh, well i see it in my inbox, yeah [18:05:04] (03CR) 10Parent5446: "Thanks! I'll add the entry point today so that tests start passing." [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115092 (owner: 10Parent5446) [18:06:41] (03Merged) 10jenkins-bot: DefaultSettings.php: Remove $wgPasswordSenderName [core] - 10https://gerrit.wikimedia.org/r/115206 (owner: 10Bartosz Dziewoński) [18:08:29] (03CR) 10Technical 13: [C: 031] "Code looks functionally sound and seems to work on http://abusefilter-global-main.instance-proxy.wmflabs.org/w/api.php?action=query&list=u" (033 comments) [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [18:12:50] (03CR) 10PiRSquared17: "Replied to T13." (032 comments) [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [18:12:53] (03PS1) 10Aude: Update Wikidata build [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115208 [18:14:27] andre__, is it clear to you the difference between the Bugzilla keywords "easy" and "need volunteer"? I know how they are different but they are somehow linked, isn't it. [18:15:14] andre__, ... and I had noticed the "testme" keyword. Sweet! [18:15:18] qgil: "need-volunteer" is a Wikidata thing. It basically means "Wikidata team won't work on this" and to me translates to low/lowest priority with the offer to mentor to some extend. [18:15:19] hadn't [18:15:31] greg-g: around? [18:15:54] andre__: hi [18:16:04] qgil: basically we can set "testme" on every single open ticket not updated for two years. I'm not sure how useful it is - could theoretically also be covered by resetting to UNCONFIRMED :-/ [18:16:07] huh: hola [18:16:23] or Reedy [18:16:24] Why did you remove the dependency from https://bugzilla.wikimedia.org/show_bug.cgi?id=58446 ? [18:16:31] (and also the trello bug) [18:16:37] we need https://gerrit.wikimedia.org/r/#/c/115208/ merged / deployed [18:17:47] (03CR) 10Hoo man: [C: 032] Remove another ss_active_users check from SiteStats::isSane [core] - 10https://gerrit.wikimedia.org/r/115198 (owner: 10Umherirrender) [18:18:14] mhm [18:18:24] aude: I tried to ping Reedy earlier today, no luck [18:18:42] huh: because fixing it does not require fixing bug 56040 (cleaning yup our custom patches) first [18:18:49] hoo: btw, I forgot to say yesterday: thanks for the good catch and the fix :) [18:19:10] huh: but you're right that it did require upgrading to MoreBugUrl in Bugzilla 4.4 first. It's my fault as there's two things in one bug report, and the first part is solved... [18:19:24] paravoid: You're welcome ;) [18:20:04] andre__: ok :) [18:20:26] back later but safe to deploy that anytime (so it gets on test.wikidata) [18:20:47] aude: Ok, will probably be around (unless my internet goes down again :/) [18:21:23] (03Merged) 10jenkins-bot: Remove another ss_active_users check from SiteStats::isSane [core] - 10https://gerrit.wikimedia.org/r/115198 (owner: 10Umherirrender) [18:21:40] qgil: I like the displaying of bug report age in the lists at the bottom of http://korma.wmflabs.org/browser/bugzilla_response_time.html - it might motivate me to only have open tickets with less than X days left in there :P [18:21:42] Why can a bot merge things? [18:21:46] I'm configuring git [18:21:55] huh: It does that after a human approved [18:21:57] and in that he said to do cd .ssh [18:22:07] it runs the unit tests one more time, and if they pass, it merges the change ;) [18:22:10] hoo: oh, okay [18:22:12] ? [18:22:13] what if I don't have .ssh folder [18:22:27] Dpp_coder: then mkdir ? [18:22:27] huh: a human can say +2 before the not verifies it and then the bot does and the merge is causes by the combination of human and bot having reviewed.. would be one way [18:22:35] s/not/bot :p [18:22:48] hoo: do non-voting failures affect it too? [18:22:54] What is non-voting exactly? [18:22:54] huh:Its ok right? [18:22:56] andre__, yep, I also find those lists very useful [18:23:11] Dpp_coder, if you have ssh/openssh installed, .ssh folder will already be present I believe.. [18:23:16] Dpp_coder: Honestly, I have never configured git, but the tutorial is simple [18:23:29] huh: nope [18:23:31] BPositive: ok [18:23:43] non-voting just means: Ignore failures silently [18:23:54] Dpp_coder, install openssh package, that would do it for you :) [18:24:13] BPositive: ok [18:25:20] (03Abandoned) 10Umherirrender: Use Language::commaList on Special:ProtectedTitles [core] - 10https://gerrit.wikimedia.org/r/112121 (owner: 10Umherirrender) [18:30:54] (03CR) 10Umherirrender: "Does anybody knows a MediaWiki Core reviewer with (a installed) oracle to test, review and merge this patch set? Thanks." [core] - 10https://gerrit.wikimedia.org/r/110991 (owner: 10Umherirrender) [18:31:17] (03PS1) 10Chad: Remove content index segregation [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115210 [18:32:11] Dpp_coder: did you try cd ~/.ssh [18:32:12] ? [18:32:14] <^d> manybubbles: That ended up being easier than I thought ^ https://gerrit.wikimedia.org/r/#/c/115210/ [18:33:08] (03CR) 10jenkins-bot: [V: 04-1] Remove content index segregation [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115210 (owner: 10Chad) [18:33:11] (03CR) 10Nemo bis: "Do you mean MediaWiki Core the WMF team, or MediaWiki core's reviewers i.e. https://gerrit.wikimedia.org/r/#/admin/groups/11,members ?" [core] - 10https://gerrit.wikimedia.org/r/110991 (owner: 10Umherirrender) [18:33:51] (03CR) 10Chad: "Or even the MediaWiki Core Features team? Too many cores...." [core] - 10https://gerrit.wikimedia.org/r/110991 (owner: 10Umherirrender) [18:34:29] aude: heya [18:35:16] tonythomas: I have instaled ssh and tried :) [18:35:29] tonythomas: thanx :) [18:35:42] Dpp_coder: its working right ? [18:35:53] yeah [18:35:59] Dpp_coder: welcome ;) [18:36:27] (03CR) 10Manybubbles: [C: 04-1] "OK! So two things to think about:" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115210 (owner: 10Chad) [18:42:21] (03CR) 10Bartosz Dziewoński: "I'm pretty sure no one with +2 rights actually cares about this :(, but I'm also pretty sure that you can find someone to take it on faith" [core] - 10https://gerrit.wikimedia.org/r/110991 (owner: 10Umherirrender) [18:46:27] " connect to host gerrit.wikimedia.org port 29418: Network is unreachable" [18:47:00] This is what I'm seeing for ssh Dharmana.prudhvi@gerrit.wikimedia.org -p 29418 [18:47:35] Dpp_coder: If you run 'host gerrit.wikimedia.org' , what do you get? [18:48:05] On my system, I get: [18:48:07] RoanKattouw: gerrit.wikimedia.org has IPv6 address 2620:0:861:3:208:80:154:81 [18:48:11] gerrit.wikimedia.org has address 208.80.154.81 [18:48:13] gerrit.wikimedia.org has IPv6 address 2620:0:861:3:208:80:154:81 [18:48:19] <^d> RoanKattouw: I'm seeing a VE bug on ptwikibooks. I'm clicking "Edit" on a user JS page (trying to copy+paste the source for testing). It sends me to veaction=edit but still looking like action=view. [18:48:32] ^d: Let me check that out [18:48:34] RoanKattouw: same thingg [18:48:53] Dpp_coder, are you behind a proxy? [18:48:56] ^d: Can we trade? You help Dpp_coder with his accessing Gerrit problem and I look at ptwikibooks? :) [18:49:01] BPositive: yes [18:49:05] (or her) [18:49:23] <^d> port 29418 network unreachable means your network is blocking port 29418 99% of the time. [18:49:35] You can try setting the http_proxy and https_proxy environment variables [18:49:36] <^d> Workarounds include: gerrit patch uploader, using https. [18:49:51] Still if it does not work, it means the port is firewalled [18:50:07] https://www.irccloud.com/pastebin/RgLqTNni [18:50:09] ^d: Link to the page you mentioned? [18:50:12] (03PS3) 1001tonythomas: Splitted date and time in message 'rclistfrom' [core] - 10https://gerrit.wikimedia.org/r/107360 [18:50:21] <^d> RoanKattouw: https://pt.wikibooks.org/wiki/Utilizador:Helder.wiki/vector.js [18:50:30] Thanks [18:50:32] <^d> Clicked "Editar" and got the symptoms above. [18:50:35] <^d> :) [18:51:21] ^d: WTF, that page is protected (from non-owners at least), there shouldn't be a VE link [18:51:39] It says "*Ver* código-fonte", not "*Editar* código-fonte" [18:52:14] <^d> Are we seeing different things? Wanna look at my screen? [18:52:24] Hmm, I'm anon [18:52:26] I'll log in [18:52:29] <^d> I'm logged in. [18:52:30] (03CR) 10Parent5446: "What would be a good test case for testing this? Trying to think of where upsert() is used in core." [core] - 10https://gerrit.wikimedia.org/r/110991 (owner: 10Umherirrender) [18:52:51] Hah [18:52:58] Yeah now I saw "Edit" being replaced with "Editar" [18:53:04] Which is weird [18:53:36] Hmm [18:53:44] There's a ca-edit in the page source but no ca-ve-edit [18:54:04] And the former is replaced with the latter at run time [18:54:09] BPositive: what should I do? [18:56:14] export http_proxy="http://:" and if you have authentication as well then export http_proxy="http://:@:" [18:56:31] repeat the same for https_proxy [19:00:28] BPositive: the same result [19:00:28] I have used proxy before a lot of times from terminal [19:00:50] BPositive: this is the onlt thing saying netwrok unreachable [19:01:37] Dpp_coder, Sadly, the port is blocked then :( [19:02:05] Soo what can I do?? [19:02:05] :( [19:03:00] BPositive: Is tor the only option? [19:04:08] I haven't used tor buddy! Sorry. But you can google for some ways to do that. [19:04:49] you should be able to use proxychains w/ ssh if it's not statically linked ...although I haven't tried it [19:05:03] ^d: I found the cause :S [19:05:09] /* MediaWiki:Gadget-VisualEditorByDefault.js *//** * Ativa Visual editor por padrão * TODO: Disable this once [[bugzilla:61762]] is fixed */ [19:05:11] is editing erroring out for anyone? [19:05:20] Where https://bugzilla.wikimedia.org/show_bug.cgi?id=61762 os [19:05:27] *is "please actually enable VE" [19:06:40] (03CR) 10MarkTraceur: [C: 04-1] "I'm having trouble getting it working, standby." (032 comments) [extensions/Popups] - 10https://gerrit.wikimedia.org/r/113951 (owner: 10Prtksxna) [19:07:36] ^d, RoanKattouw: Done in https://gerrit.wikimedia.org/r/115213 – feel free to merge and deploy. ;-) [19:07:54] James_F: Feel free to book a deployment window for me :) [19:09:59] greg-g: ^^ ;-) [19:10:27] que es? /me is in meeting [19:12:26] greg-g: LD request for today, will sling in calendar. [19:12:48] what is the change though? [19:13:07] (03CR) 10Anomie: Add 'newonly' option to Special:Contributions and related i18n message * Add to Special:Contributions * Add to i18n files * Add 'ucnewonly' (032 comments) [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [19:13:15] 14:06 < grrrit-wm> (PS1) Jforrester: Make VisualEditor opt-out on Portuguese Wikibooks [operations/mediawiki-config] - https://gerrit.wikimedia.org/r/115213 [19:13:18] ? [19:13:23] greg-g: Enabling VE by default on ptwikibooks. [19:13:24] greg-g: Yes. [19:13:28] cool [19:13:43] I trust all relevant non-tech steps have been taken :) [19:14:33] (03PS1) 10Chad: Don't use parsed wikitext when dealing with CSS/JS pages [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 [19:14:39] YuviPanda: Do you know if there's anything special one has to do to get Popups running locally? [19:14:42] I'm having zero luck. [19:14:52] rdwrer: vagrant enable-role popups? [19:15:00] rdwrer: if not [19:15:05] (03PS2) 10Chad: Don't use parsed wikitext when dealing with CSS/JS pages [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 [19:15:06] No, won't work [19:15:14] And he doesn't have the latest code on chicken.wmflabs [19:15:15] rdwrer: you need the following extensions: TextExtracts, PageImages, BetaFeatures [19:15:17] last is optional [19:15:31] rdwrer: I've chicken access, what do you want me to put on chicken? [19:15:34] Oh. [19:15:40] See, documentation is fucking helpful [19:15:41] greg-g: Well, it's a community request with a VP post without dissent, so… :-) [19:15:45] * rdwrer grumbles [19:15:47] (03CR) 10Chad: "Untested (having local dev problems with elastic 1.0), but should basically deal with my theory from the linked bug." [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [19:15:53] James_F: sounds good ;) [19:16:40] !e TestExtracts | YuviPanda, this doesn't exist [19:16:40] YuviPanda, this doesn't exist: https://www.mediawiki.org/wiki/Extension:TestExtracts [19:16:46] rdwrer: Text, sorry [19:16:49] TextExtracts [19:16:51] wait [19:16:52] Oh, my bad, misread [19:17:12] (03CR) 10jenkins-bot: [V: 04-1] Don't use parsed wikitext when dealing with CSS/JS pages [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [19:17:41] Lol white page [19:17:58] rdwrer: added dependency info to https://www.mediawiki.org/wiki/Extension:Popups [19:18:06] rdwrer: I can put things on chicken if you want [19:18:18] YuviPanda: If I can't get it working locally then something's still wrong [19:18:36] <^d> y u mad jenkins? [19:18:41] rdwrer: true. did you get both textextracts and pageimages? [19:18:52] ^d: cuz u no fix its bugz [19:18:58] I will make sure [19:19:05] ok [19:19:12] <^d> MatmaRex: Solution: ignore jenkins [19:19:25] Yes. [19:19:40] ^d: grep the logs for me saying that, i'm advocating it for months [19:19:41] Both are in my LS and in the directory [19:19:48] Fatal error: Class 'ResourceLoaderSchemaModule' not found in /home/marktraceur/projects/wikimedia/mediawiki/core/includes/resourceloader/ResourceLoader.php on line 409 [19:19:53] rdwrer: aaaaaaaah [19:19:56] rdwrer: EventLogging too [19:19:59] Hahaha [19:20:06] most jenkins failures i saw in the last few weeks were false positives [19:20:23] it's beginning to stop being useful [19:20:28] (03PS13) 10EBernhardson: No need to load workflow in formatters [extensions/Flow] - 10https://gerrit.wikimedia.org/r/112846 [19:20:33] Huzzah [19:20:37] OK, thanks YuviPanda [19:20:37] rdwrer: updated it again [19:20:42] the doc page, that is [19:20:43] (03CR) 10jenkins-bot: [V: 04-1] No need to load workflow in formatters [extensions/Flow] - 10https://gerrit.wikimedia.org/r/112846 (owner: 10EBernhardson) [19:20:53] rdwrer: I've been updating the vagrant role appropriately, but yeah, should keep others in mind too [19:21:02] YuviPanda: Is PageImages something Infobox-specific? [19:21:25] !e PageImages [19:21:25] https://www.mediawiki.org/wiki/Extension:PageImages [19:21:37] rdwrer: not *infobox* specific, but it usually picks up things from infoboxen [19:21:45] rdwrer: it picks up the first image in the page that's not 'too small' [19:21:54] Hm. [19:22:19] greg-g: In fact, this community has already created an on-by-default Gadget that tries to enable VE by default and is almost successful but subtly broken [19:22:32] YuviPanda: OK, so the default thumb size is "too small". :P [19:22:50] rdwrer: heh, probably :D [19:22:55] rdwrer: MaxSem wrote it! [19:22:55] RoanKattouw: niiice [19:22:56] (03CR) 10Jackmcbarn: [C: 04-1] "Mostly minor nitpicks." (036 comments) [core] - 10https://gerrit.wikimedia.org/r/114966 (owner: 10Mjnovice) [19:23:03] Wonderful stuff though [19:23:03] rdwrer: it used to be part of MF but got split out [19:23:07] (03PS1) 10Manybubbles: 1.0 updates [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115215 [19:23:19] *nod* [19:23:23] Good on ya [19:23:25] (03CR) 10Manybubbles: [C: 04-1] 1.0 updates [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115215 (owner: 10Manybubbles) [19:23:26] Right [19:23:31] Gonna merge, I guess [19:24:02] (03CR) 10MarkTraceur: [C: 032] "Yuvi updated the docs and I'm now able to test locally!" [extensions/Popups] - 10https://gerrit.wikimedia.org/r/113951 (owner: 10Prtksxna) [19:24:41] (03CR) 10jenkins-bot: [V: 04-1] 1.0 updates [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115215 (owner: 10Manybubbles) [19:27:35] MatmaRex: Reviewing your delayedBind patch - do you want to note specifically the extensions that need updates? I have them here [19:27:56] I guess it's just WikiEditor and MassMessage actually [19:29:01] (03PS14) 10EBernhardson: No need to load workflow in formatters [extensions/Flow] - 10https://gerrit.wikimedia.org/r/112846 [19:29:13] (03CR) 10jenkins-bot: [V: 04-1] No need to load workflow in formatters [extensions/Flow] - 10https://gerrit.wikimedia.org/r/112846 (owner: 10EBernhardson) [19:30:34] rdwrer: huh, yeah, if anything actually used that, i definitely want to update it [19:30:42] (03PS15) 10EBernhardson: No need to load workflow in formatters [extensions/Flow] - 10https://gerrit.wikimedia.org/r/112846 [19:31:10] (03PS2) 10Merlijn van Deen: +sidebar userscript [wikimedia/bugzilla/triagescripts] - 10https://gerrit.wikimedia.org/r/105618 [19:31:54] MatmaRex: Like I said, WikiEditor and MassMessage [19:32:04] Seems trivial-ish [19:32:50] MatmaRex: Do you know of anything using expandableField? [19:33:05] rdwrer: i'm positive nothing uses it [19:33:20] rdwrer: it was used by the Vector extension, off by default, until we killed it [19:33:32] This was the expandable search box thing, right? [19:33:34] (it was also horribly broken) [19:33:36] yes [19:33:56] Oh, huh. [19:34:05] rdwrer: i tested that with the box on Special:Search in browser console :) [19:34:07] Oh, rdwrer. [19:34:12] Hah. [19:34:13] (03CR) 10Merlijn van Deen: "Strangely enough Chrome had no issues with it. Could also be a jQuery vs built-in-$-operator issue, but I'm not 100% sure." [wikimedia/bugzilla/triagescripts] - 10https://gerrit.wikimedia.org/r/105618 (owner: 10Merlijn van Deen) [19:34:25] MatmaRex: 'kay [19:34:27] (it doesn't work with vector's search box these days) [19:34:30] rdwrer: no kenkins for popups :) [19:34:34] jenkins [19:34:42] rdwrer: you need to hit submit [19:35:07] * Nemo_bis is surprised anything works in Vector's search box [19:35:20] * YuviPanda takes away monobook from Nemo_bis [19:35:43] YuviPanda: FFS [19:35:48] It's like the gorram stone age [19:35:55] bwaaaaaaaaaaaaaaa [19:36:02] * Nemo_bis goes back to UseMod [19:36:11] (03CR) 10MarkTraceur: [V: 032] "Grumble grumble grumble" [extensions/Popups] - 10https://gerrit.wikimedia.org/r/113951 (owner: 10Prtksxna) [19:36:48] * YuviPanda makes mediawiki a purely facebook app, just for Nemo_bis [19:36:53] rdwrer: i'll check out the two and look [19:37:01] Seems to work, thanks [19:37:26] I guess next up is deprecating expandableField? :) [19:37:30] (03CR) 10MarkTraceur: [C: 032] "Awesome, MatmaRex, thanks a bunch." [core] - 10https://gerrit.wikimedia.org/r/114953 (owner: 10Bartosz Dziewoński) [19:37:32] (03CR) 10Chad: "Bleh, I forgot about (1). That's annoying." [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115210 (owner: 10Chad) [19:37:53] greg-g: can we get https://gerrit.wikimedia.org/r/#/c/115208/ deployed today? [19:38:16] * Nemo_bis laughs at YuviPanda from his renovated UseModWiki dock while the panda flies away [19:38:17] rdwrer: i'm not sure if there's a point in deprecating it, if we want to kill it then let's just do it [19:38:25] MatmaRex: Fair enough [19:38:39] Yeah, no matches in any deployed extensions, may as well [19:40:51] (03PS1) 10Bartosz Dziewoński: jquery.delayedBind → jquery.throttle-debounce [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/115222 [19:40:59] rdwrer: ^ not tested [19:41:06] (03Merged) 10jenkins-bot: Stop using and deprecate jquery.delayedBind [core] - 10https://gerrit.wikimedia.org/r/114953 (owner: 10Bartosz Dziewoński) [19:41:43] Fun times [19:41:54] I'll let legoktm deal with that, MatmaRex [19:42:29] i'm cloning wikieditor repo too [19:42:51] it takes a while because of how slow gerrit is [19:42:57] it's sad that no one cares [19:44:19] (03PS4) 10Addshore: Implement ApiQuerySites class [core] - 10https://gerrit.wikimedia.org/r/110427 [19:44:26] (03PS5) 10Addshore: Implement ApiQuerySites class [core] - 10https://gerrit.wikimedia.org/r/110427 [19:45:15] (03CR) 10Umherirrender: "Upsert is used when a category is added to a wikipage." [core] - 10https://gerrit.wikimedia.org/r/110991 (owner: 10Umherirrender) [19:45:36] <^d> MatmaRex: I care when it's slow. But it't not. [19:45:40] <^d> http://p.defau.lt/?IlFCrm_abNkhRQTvn1Vcuw [19:46:18] ^d: 15 KiB/s is not slow? :/ [19:46:23] <^d> That's your connection. [19:46:23] (03PS2) 10Matthias Mullie: Non-hacky way to feed title to Parsoid [extensions/Flow] - 10https://gerrit.wikimedia.org/r/115112 [19:46:26] ^d: ooooh, it went up to 40 now [19:46:34] <^d> I cloned at 2.51 MiB/s [19:46:45] well i cloned at 15 KiB/s [19:47:02] ^d: at the office? (is greg-g around) [19:47:08] fetches are just as slow [19:47:18] <^d> Makes no difference. [19:47:24] <^d> THey're both fetching objects. [19:47:38] <^d> aude: Yes, and not this second. [19:47:44] ok [19:47:45] clones are fetching more, that might matter, so just saying [19:47:50] <^d> He was a bit ago, but he's not at his desk. [19:48:09] <^d> MatmaRex: Well yeah, more objects, but you can't tell me 7.14MiB of objects is a lot :) [19:48:16] i'll poke him again later [19:48:23] <^d> So if clone is ~7M, and fetch is some number less. [19:48:23] is there no clean way to get the redirect target of a given page from wikitext? i just found https://en.wikipedia.org/wiki/Module:Redirect and it seems like a terrible hack [19:48:34] <^d> And it can be cloned at > 2 M/s [19:48:35] (03CR) 10jenkins-bot: [V: 04-1] Implement ApiQuerySites class [core] - 10https://gerrit.wikimedia.org/r/110427 (owner: 10Addshore) [19:48:38] <^d> I blame connections :( [19:49:30] rdwrer: heh, the wikieditor thing is in the iframe code [19:49:41] rdwrer: that code is *so* dead [19:49:54] it probably doesn't work anyway already [19:50:17] MatmaRex: Did you yet remove it? [19:50:38] hoo|away: are you really away? [19:51:09] huh: Not really, but I'm doing sorting some paper works and whatever... [19:51:48] (03PS1) 10Bartosz Dziewoński: jquery.delayedBind → jquery.throttle-debounce [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/115229 [19:55:59] (03CR) 10Bartosz Dziewoński: "I'm not sure if anyone cares about this though…" [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/115229 (owner: 10Bartosz Dziewoński) [19:58:27] (03PS6) 10Addshore: Implement ApiQuerySites class [core] - 10https://gerrit.wikimedia.org/r/110427 [19:59:47] ssh Dharmana.prudhvi@gerrit.wikimedia.org -p 29418 [19:59:47] ssh: connect to host gerrit.wikimedia.org port 29418: Network is unreachable [19:59:47] plz Help me with this error [20:00:23] I'm behind proxy [20:00:58] I have used al roxy settings but only this command doesnt work [20:01:02] Dpp_coder: What happens if you don't use -p 29418 ? [20:01:16] Do you get a connection failure, or Permission denied (publickey)? [20:01:22] ssh: connect to host gerrit.wikimedia.org port 22: Network is unreachable [20:01:26] (03CR) 10Freakolowsky: "Tested on 11.2.0.2.0" [core] - 10https://gerrit.wikimedia.org/r/110991 (owner: 10Umherirrender) [20:03:49] (03PS6) 10Mjnovice: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 [20:03:55] RoanKattouw : any idea? [20:05:44] (03Abandoned) 10Hashar: Added unit testing. [extensions/Git2Pages] - 10https://gerrit.wikimedia.org/r/53268 (owner: 10Terrrydactyl) [20:08:51] (03CR) 10Aklapper: "Works for me in Firefox, thanks." [wikimedia/bugzilla/triagescripts] - 10https://gerrit.wikimedia.org/r/105618 (owner: 10Merlijn van Deen) [20:08:51] (03CR) 10Technical 13: Add 'newonly' option to Special:Contributions and related i18n message * Add to Special:Contributions * Add to i18n files * Add 'ucnewonly' (031 comment) [core] - 10https://gerrit.wikimedia.org/r/115189 (owner: 10Gerrit Patch Uploader) [20:09:31] (03CR) 10Jackmcbarn: [C: 04-1] Automatically copy the LocalSettings.php file to the desired location (033 comments) [core] - 10https://gerrit.wikimedia.org/r/114966 (owner: 10Mjnovice) [20:11:32] (03PS1) 10Aizhuli90: added simple schema for db [extensions/OpenBadges] - 10https://gerrit.wikimedia.org/r/115238 [20:11:34] (03CR) 10Merlijn van Deen: "The three listed projects are just examples (and the ones I use ;-)) See the comment above for the 'why':" [wikimedia/bugzilla/triagescripts] - 10https://gerrit.wikimedia.org/r/105618 (owner: 10Merlijn van Deen) [20:11:45] (03PS3) 10Manybubbles: Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [20:11:53] Can someone here please check https://gerrit.wikimedia.org/r/115189 ? [20:12:12] (https://gerrit.wikimedia.org/r/115189 just adds option to Special:Contributions, API) [20:12:14] qgil: Hi [20:12:50] (03CR) 10Manybubbles: "Added test and fixed the case where the index was already corrupted and needed to be fixed. This had to happen because we update all page" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [20:12:52] (03CR) 10jenkins-bot: [V: 04-1] Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [20:12:58] qgil: Would you mind expanding a bit on your message to the Tech News talk page? [20:13:02] hi odder ! [20:13:14] qgil: You wrote "With a bit of organization of the current Tech News content, you could isolate the MediaWiki-centric news (e.g. feature X is coming) separating it from the Wikimedia centric news (e.g. something broke our servers)." [20:13:23] I'm interested in hearing what you have in mind [20:13:57] odder, see the News block at https://www.mediawiki.org/wiki/File:MediaWiki_Homepage_Mockup_Graphic_With_Help_Links.png [20:14:02] aude: hey, sorry, meeting morning [20:14:18] aude: is there a corresponding change for wikidata for that for easy review? [20:14:25] odder, Imagine a World in which every week that block would be updated with news coming from Tech News,, automagically [20:14:55] aude: /me goes to heat up lunch, brb [20:14:56] odder, but only MediaWiki centric news aka Feature X is coming! Not Wikimedia-centric news aka Our servers went down! [20:15:10] odder, that's ti [20:15:12] Yes, you wrote that already, so let me rephrase. [20:15:12] it [20:15:18] How do you imagine we do that? [20:15:22] Technically [20:15:29] Two parts [20:15:59] the first part is editorial: make sure that there is a tiny separation between both types of news. [20:16:40] The second part... take that MediaWiki section and transclude it somehow (tm) [20:17:18] I don't think this is possible at the moment; including sections from one wiki in pages in another one. [20:17:56] Yep. I spoke with guillom about this today. He didn't promise anything but basically he said that if there is a will, there is a way. :) [20:18:19] I only pinged him now, but he is basile, so :-) [20:18:54] * odder wondering [20:19:05] (03Draft2) 10Addshore: Split Exception.php [core] - 10https://gerrit.wikimedia.org/r/115240 [20:19:09] I agree it is not straightforward to automate this, but if we agree on the concept, first steps could be done in the editorial side [20:19:11] We already go through core and extension patches in Gerrit and Bugzilla bugs [20:19:29] (03CR) 10Nemo bis: "I assume this is the point when log_params for newusers log got serialized? so it's not like we don't see this kind of information [20:19:50] "we" being who? [20:20:03] Guillaume and I; Tech News is basically us two [20:20:22] with infrequent involvement from about 5 people [20:20:24] The target for this would be basically newcomers and casual visitors landing at the mediawiki.org homepage and having no clue about our day to day, tech news, etc. [20:20:37] Why not just add
? [20:20:45] and transclude it? [20:20:54] 21:17 odder: I don't think this is possible at the moment; including sections from one wiki in pages in another one. [20:20:56] interwiki transclude? :) [20:21:04] huh, because Tech News is published from Meta, and we are talking about mediawiki.org [20:21:16] qgil: it's possible to deliver the news to a certain page on MW.org [20:21:18] well, we could just test interwiki transclusion [20:21:33] huh: We are talking about brand new enterprise [20:21:39] ? [20:21:42] (03PS1) 10Addshore: Add scope to Exception class methods [core] - 10https://gerrit.wikimedia.org/r/115242 [20:21:43] Tech News MediaWiki-centric [20:21:45] ? :) [20:21:49] noooo [20:22:00] huh: indeed, we already do that e.g. on https://it.wikipedia.org/wiki/Wikipedia:Wikipediano [20:22:11] odder, you keep doing Tech News as you do, just paying a bit of attention in how you group the items [20:22:27] I understand [20:23:07] Who is best to code-review changes to core? [20:24:18] Nemo_bis: so how is this achieved? [20:24:50] huh: depends on what kind of changes. a few people probably understand to some level the entirety of mediawiki, but i'm sure no one is proficient with every part of it [20:24:51] Because I see it's just a MassMessage delivered to a local page, and transcluded there [20:25:22] But anyway qgil, I guess we can figure this out; even if we'd have to start by hand for a short period [20:25:41] odder, I'm happy copy/pasting if the separation is clear [20:25:56] huh: this is mostly up-to-date, i think: https://www.mediawiki.org/wiki/Developers/Maintainers [20:25:59] MatmaRex: in this specific case (remember?) [20:26:08] Looking at the past Tech News, it could suffice with a separation between "software changes" and "deployments" [20:26:11] qgil: as huh said, we could just include it in a section [20:26:14] huh: but definitely not comprehensive [20:26:15] (03PS2) 10Aude: Update Wikidata build [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115208 [20:26:19] Now "software changes" include both [20:26:23] huh: i'd ask brian or tyler [20:26:29] greg-g: https://gerrit.wikimedia.org/r/#/c/115195/ [20:26:31] Either around? [20:26:32] huh: aka, bawolff and parent5667 or something [20:26:49] qgil: the separation we are currently using is between past and future changes [20:27:07] though there is some overlap in VisualEditor items [20:27:40] odder: just automatic delivery to a local page with suitable tags + LST [20:27:40] Think also that one thing is a visual separation in sections and another is a semantic separation using
[20:27:51] (see the source, it's very trivial) [20:27:51] odder: Yeah. :-( [20:27:54] I have installed git review but there is no git-review folder in .config [20:28:06] You could keep Tech News looking the way they look now, but identifying the bits ready for the mediawiki.org homepage [20:28:17] I agree,
will work [20:28:26] locally, yes. [20:29:04] anyway, as I said, we can figure it out at a later point, no need to put non-existent obstacles so early in our path [20:29:13] Why reinvent the wheel with a separate Tech News? [20:29:38] nobody wants to do this [20:30:03] qgil: so that would be core+extension commits/bugs/announcements? [20:30:20] Which extensions? Any? [20:30:26] (03PS1) 10Hashar: Fix path in UnitTestsList hook [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115244 [20:30:39] MediaWiki software. Any extension deployed in Wikimedia servers is good. [20:31:19] commits/bugs is not a clear division, because some commits/bugs refer specifically to Wikimedia deployments, not to the software itself [20:31:30] (03CR) 10jenkins-bot: [V: 04-1] Fix path in UnitTestsList hook [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115244 (owner: 10Hashar) [20:31:52] I'm a bit confused by this focus though. Few people will consume such stuff, in the meanwhile we don't have usable release notes and release overviews like [[MediaWiki 1.22]] are always very depressing (like, why in earth would I want to upgrade??? did they really do nothing in 6 months?). [20:32:05] By the way, this discussion should be in #mediawiki [20:32:24] As soon as a non-Wikimedia person will get involved, yes. [20:32:50] Well anyway, I'm sure we can find five or so items worth noting every week. [20:32:51] (03CR) 10Hashar: "On my setup:" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115244 (owner: 10Hashar) [20:33:00] qgil: deployed? I assume you meant hosted? [20:33:22] I'm worried about long-term manpower, but I expect we can deal with the additional workload for a while, and then see what happens. [20:34:15] (03PS3) 10Aude: Update Wikidata build [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115208 [20:34:34] !e [20:34:35] https://www.mediawiki.org/wiki/Extension:$wiki_encoded_* [20:34:42] !e Translate [20:34:42] https://www.mediawiki.org/wiki/Extension:e [20:34:42] How many of the changes are in non-Wikimedia extensions? [20:34:47] aude: wow [20:34:50] wtf is that shit. [20:35:10] greg-g: hope it's clear enough :) [20:35:11] Reedy: can you look at aude's wikidata update? [20:35:45] we're moving away form javascript formatters (to php/api ones), hence issues with our formatters [20:36:01] odder, Nemo_bis I quickly took the items of the last Tech News and organize them at https://www.mediawiki.org/wiki/User:Qgil/Sandbox#Tech_News_for_mediawiki_org [20:36:01] Nemo_bis: none that we currently report [20:36:05] (03PS7) 10Mjnovice: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 [20:36:16] think it's good now with the update [20:37:18] I mean "The new search tool (CirrusSearch) was added to the Italian Wikiquote and all Wikiversity projects." wuldn't go to the mw.o homepage because it is a Wikimedia-only thing [20:37:30] (03CR) 10jenkins-bot: [V: 04-1] Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 (owner: 10Mjnovice) [20:38:07] "You will soon be able to use the tag again." is also relevant only to users of Wikimedia projects. Etc. [20:38:26] (03PS8) 10Mjnovice: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 [20:39:17] Is it? [20:39:26] AFAIK it was broken in the extension itself [20:39:46] (03CR) 10jenkins-bot: [V: 04-1] Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 (owner: 10Mjnovice) [20:40:01] Ok, then my mistake, I thought it was one of those cases of "disabled in Wikipedia because of performance issues" [20:40:17] The problem with tech news is that non-Wikimedia users don't do weekly updates, so they're not interested in most of teh cycles of break-unbreak-rebreak-reunbreak that we have [20:40:58] Tech News contains the closest we have to "MediaWiki news" [20:41:05] https://www.ohloh.net/p/mediawiki-extensions-wmf-hosted have about half of the activity of WMF-deployed exts https://www.ohloh.net/p/mediawiki-extensions-wmf [20:41:11] Perhaps we should have some competition. [20:41:16] It only helps the end user. [20:41:42] On the other hand WMF extensions are getting less commits and the others more, compared to last year [20:42:11] Once we are having regular updates of Core and extensions featured in Tech News, it will be easy to allow ohers to chime in in the mw.o homepage only, leaving alone Tech News [20:42:16] odder: I don't understand what's in your opinion the higher workload, checking the non-WMF extensions? [20:42:57] I get odder and guillom, they care about maintaining a good resources for Wikimedia tech readers. This is hard enough (as seen with the many failed attempts) and I don't want to distort their plan [20:43:25] I want to help them bringing more readers while they help us bringing more news to mediawiki.org [20:44:03] What I would like is something replacing the summaries of the wmfX release notes, that also helps making the MediaWiki 1.N pages. Separation of MediaWiki-only stuff is good even without adding anything else. [20:44:31] (03PS1) 10Hashar: TimedMediaHandler depends on MwEmbedSupport [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115248 [20:44:39] Nemo_bis: No; the necessity to write down, check, re-write, simplify items we wouldn't do if it wasn't for MW.org [20:44:43] (03CR) 10Hashar: [C: 032] TimedMediaHandler depends on MwEmbedSupport [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115248 (owner: 10Hashar) [20:45:08] (03Merged) 10jenkins-bot: TimedMediaHandler depends on MwEmbedSupport [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/115248 (owner: 10Hashar) [20:45:47] odder: mediawiki.org has a more technical audience than Tech News, I don't see such a need. It's mostly devs/sysadmins vs. editors [20:45:48] So qgil is offering us "More weekly work for uncertain gain in readership." Hmm. [20:46:02] odder, am I proposing more weekly work? [20:46:17] Nemo_bis: How do you mean, you don't see such a need? [20:46:21] I'm proposing to sort items differently [20:46:33] For a News section on MW.org, or what exactly? [20:47:15] qgil: If that's all you suggest, then sure. I thought you wanted us to give more focus to MW-related items [20:47:37] Let me repeat again: "noooo" :) [20:47:38] Because right now, we just don't feature some important technical stuff if it doesn't directly influence Wikimedia users [20:47:40] It all depends on what "more focus" means. :) [20:48:48] odder, now that we have discussed a bit, go back to https://meta.wikimedia.org/wiki/Talk:Tech/News#Isolating_MediaWiki-centric_news [20:49:08] IMHO you can just separate the MW stuff, don't pretend you represent the whole MW, distribute it further. If someone cares about non-WM MW stuff being in it, they'll add it. Otherwise, nobody will (or not systematically) [20:49:30] odder, still, all I'm asking is to "isolate MediaWiki-centric news" within the news that anyway you publish [20:49:34] qgil: My point precisely. [20:49:42] We don't do "feature X is coming" things [20:49:52] you do [20:49:57] like we had a change in how i18n messages are kept, I think [20:50:07] there was a change so now they are kept as JSON files? [20:50:15] or whatever that was about a couple of weeks ago [20:50:51] (03PS1) 10Hashar: Unit tests for TimedMediaHandler (non voting) [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115249 [20:50:54] We didn't do that because no Wikipedian cares about it; I expect some devs would've cared. [20:50:54] (03PS9) 10Mjnovice: Automatically copy the LocalSettings.php file to the desired location [core] - 10https://gerrit.wikimedia.org/r/114966 [20:51:18] odder, as said, I'd take the Yes + Yes, polishing content at https://www.mediawiki.org/wiki/User:Qgil/Sandbox#Tech_News_for_mediawiki_org and transclude it in the mediawiki.org homepage right away [20:51:30] (03CR) 10Manybubbles: "This is trying to run unit tests that aren't in cirrus. I don't think this change is what did it, but something is funky." [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115244 (owner: 10Hashar) [20:51:33] This is a bit off-topic, but is it possible to exclude a specific file from grep ? [20:52:18] (03CR) 10Hashar: [C: 032] Unit tests for TimedMediaHandler (non voting) [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115249 (owner: 10Hashar) [20:52:25] (03Merged) 10jenkins-bot: Unit tests for TimedMediaHandler (non voting) [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115249 (owner: 10Hashar) [20:57:13] odder, "The latest version of MediaWiki..." piece that all tech News have are relevant for mediawiki.org because they announce a new release that anybody can go and test. Even if they are "wmf" releases, in practice they become "MediaWiki unstable". [20:57:35] (03Restored) 10Hashar: validate jenkins jobs [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/36974 (owner: 10Hashar) [20:57:40] (03PS2) 10Hashar: validate jenkins jobs [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/36974 [20:57:45] qgil: I know; I reverted that edit [20:57:55] just for clarity :) [21:00:13] (03CR) 10Chad: [C: 032 V: 032] Fix path in UnitTestsList hook [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115244 (owner: 10Hashar) [21:00:21] (03CR) 10jenkins-bot: [V: 04-1] Fix path in UnitTestsList hook [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115244 (owner: 10Hashar) [21:02:06] (03CR) 10UltrasonicNXT: "Hmm, in that case it sounds like we need a better solution... That code is a bit yucky really IMO anyway" [extensions/Comments] - 10https://gerrit.wikimedia.org/r/114972 (owner: 10UltrasonicNXT) [21:07:22] (03PS1) 10Hashar: flag API tests with @medium [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115272 [21:07:25] (03PS1) 10Hashar: Typo TesVideoTranscode -> TestVideoTranscode [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115273 [21:09:45] (03PS2) 10Hashar: flag API tests with @medium [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115272 [21:17:02] (03CR) 10Aude: "@anomie sites are in core, so think sites (via api) belongs there and is generally not wikibase-specific." [core] - 10https://gerrit.wikimedia.org/r/110427 (owner: 10Addshore) [21:20:30] (03Abandoned) 10Hashar: validate jenkins jobs [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/36974 (owner: 10Hashar) [21:24:22] (03PS3) 10Mglaser: Make MySQLi work with non-standard port [core] - 10https://gerrit.wikimedia.org/r/109634 [21:25:02] (03PS1) 10Aaron Schulz: Avoid key fragmentation in FileBackend [core] - 10https://gerrit.wikimedia.org/r/115300 [21:25:20] (03PS1) 10Aaron Schulz: Avoid key fragmentation in FileBackend [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115301 [21:25:32] (03PS1) 10Aaron Schulz: Avoid key fragmentation in FileBackend [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/115302 [21:25:50] ^d: ^ [21:28:25] Hi, is someone around, who would like to do a a review on some new phpunit tests https://gerrit.wikimedia.org/r/#/c/112353/ [21:28:52] (03PS14) 10Bsitu: Add diff view of changes to board header [extensions/Flow] - 10https://gerrit.wikimedia.org/r/109431 [21:29:17] (03CR) 10Mglaser: "Know your framework ;) There's actually a method in MediaWiki that parses out the port." [core] - 10https://gerrit.wikimedia.org/r/109634 (owner: 10Mglaser) [21:36:32] (03PS1) 10Hashar: Ignore TestVideoThumbnail::testApiThumbnails [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115303 [21:36:38] (03PS1) 10Hashar: Ignore TestVideoTranscode::testTranscodeJobs [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115304 [21:37:40] (03PS2) 10Hashar: Typo TesVideoTranscode -> TestVideoTranscode [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115273 [21:37:54] (03PS2) 10Hashar: Ignore TestVideoThumbnail::testApiThumbnails [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115303 [21:37:59] (03PS2) 10Hashar: Ignore TestVideoTranscode::testTranscodeJobs [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115304 [21:39:41] (03CR) 10MarkTraceur: [C: 032] flag API tests with @medium [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115272 (owner: 10Hashar) [21:39:50] (03Merged) 10jenkins-bot: flag API tests with @medium [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115272 (owner: 10Hashar) [21:40:08] (03CR) 10MarkTraceur: [C: 032] Typo TesVideoTranscode -> TestVideoTranscode [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115273 (owner: 10Hashar) [21:40:17] (03Merged) 10jenkins-bot: Typo TesVideoTranscode -> TestVideoTranscode [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115273 (owner: 10Hashar) [21:41:00] (03CR) 10MarkTraceur: [C: 032] Ignore TestVideoThumbnail::testApiThumbnails [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115303 (owner: 10Hashar) [21:41:08] (03Merged) 10jenkins-bot: Ignore TestVideoThumbnail::testApiThumbnails [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115303 (owner: 10Hashar) [21:41:29] (03CR) 10MarkTraceur: [C: 04-1] Ignore TestVideoTranscode::testTranscodeJobs (031 comment) [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115304 (owner: 10Hashar) [21:42:23] (03PS3) 10MarkTraceur: Ignore TestVideoTranscode::testTranscodeJobs [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115304 (owner: 10Hashar) [21:42:53] (03CR) 10MarkTraceur: [C: 032] "\o/ speed" [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115304 (owner: 10Hashar) [21:42:58] (03Merged) 10jenkins-bot: Ignore TestVideoTranscode::testTranscodeJobs [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/115304 (owner: 10Hashar) [21:44:29] (03CR) 10Chad: [C: 032] Avoid key fragmentation in FileBackend [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/115302 (owner: 10Aaron Schulz) [21:45:43] (03CR) 10Chad: [C: 032] Update Wikidata build [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115208 (owner: 10Aude) [21:45:51] (03PS1) 10Hashar: mwext-TimedMediaHandler-testextensions-master is now voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115305 [21:46:05] (03CR) 10Hashar: [C: 032] mwext-TimedMediaHandler-testextensions-master is now voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115305 (owner: 10Hashar) [21:46:31] (03Merged) 10jenkins-bot: mwext-TimedMediaHandler-testextensions-master is now voting [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/115305 (owner: 10Hashar) [21:47:27] (03PS8) 10Hashar: Jenkins job validation (DO NOT SUBMIT).. [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/70205 [21:47:40] (03CR) 10Aaron Schulz: [C: 032] API: Add prop=redirects and list=allredirects [core] - 10https://gerrit.wikimedia.org/r/104764 (owner: 10Anomie) [21:48:21] (03Merged) 10jenkins-bot: Avoid key fragmentation in FileBackend [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/115302 (owner: 10Aaron Schulz) [21:48:24] (03CR) 10Aaron Schulz: [C: 032] Avoid key fragmentation in FileBackend [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115301 (owner: 10Aaron Schulz) [21:48:40] ^d: did you mean to do the master one? [21:48:50] <^d> Blahh, yes [21:49:05] ^d: FYI the CirrusSearch job got broken since I added as dependencies TimedMediaHandler which had broken tests :/ Fixed now! [21:49:10] (03CR) 10Chad: [C: 032] Avoid key fragmentation in FileBackend [core] - 10https://gerrit.wikimedia.org/r/115300 (owner: 10Aaron Schulz) [21:49:12] (03Merged) 10jenkins-bot: Update Wikidata build [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115208 (owner: 10Aude) [21:49:25] <^d> AaronSchulz: master +2'd now, sorry about that [21:49:31] <^d> hashar: Thanks! I was wondering what broke :) [21:49:53] ^d: I have been Nik guinea pig earlier today :-] [21:49:59] (03PS3) 10Manybubbles: Setup jenkins configuration [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 [21:50:09] manybubbles: rebasing some CirrusSearch chanages [21:52:04] (03Merged) 10jenkins-bot: API: Add prop=redirects and list=allredirects [core] - 10https://gerrit.wikimedia.org/r/104764 (owner: 10Anomie) [21:53:02] (03CR) 10jenkins-bot: [V: 04-1] Setup jenkins configuration [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 (owner: 10Manybubbles) [21:53:04] (03Abandoned) 10Chad: Minor updates for settings with Elasticsearch 1.0 [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/112940 (owner: 10Chad) [21:53:16] (03Merged) 10jenkins-bot: Avoid key fragmentation in FileBackend [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/115301 (owner: 10Aaron Schulz) [21:54:02] (03Abandoned) 10Hashar: Jenkins job validation (DO NOT SUBMIT).. [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/70205 (owner: 10Hashar) [21:54:27] (03PS4) 10Chad: Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 [21:54:31] (03CR) 10Manybubbles: "This is failing because we try to reset Elasticsearch, I think, but we don't have an Elasticsearch instance for unit testing. I'm happy t" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115193 (owner: 10Manybubbles) [21:55:08] (03Merged) 10jenkins-bot: Avoid key fragmentation in FileBackend [core] - 10https://gerrit.wikimedia.org/r/115300 (owner: 10Aaron Schulz) [21:56:09] (03CR) 10jenkins-bot: [V: 04-1] Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [21:58:50] ori, can you review the XDebug PECL change (https://gerrit.wikimedia.org/r/#/c/113519/) when you have a chance? [22:06:20] (03PS5) 10Manybubbles: Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [22:06:22] (03CR) 10jenkins-bot: [V: 04-1] Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [22:06:57] (03CR) 10Manybubbles: "recheck" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [22:08:34] (03CR) 10Hashar: "You would need to add a dummy .jshintrc file containing {} . This is due to bug 52456, e Jenkins jshint jobs fallback to /srv/.jshintrc f" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [22:08:46] (03CR) 10Hashar: [C: 04-1] Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [22:25:37] ^d, yt? [22:25:55] <^d> meeting. sup? [22:26:46] ^d, wanted a bit of your Elastic wisdom [22:26:58] <^d> mmmm, stretchy wisdom [22:27:54] <^d> I saw your response to my e-mail over the weekend. Was going to play with your gist this afternoon.\ [22:31:08] (03PS1) 10Gergő Tisza: Fix ForeignDBViaLBRepo favicon [core] - 10https://gerrit.wikimedia.org/r/115314 [22:48:29] (03CR) 10Bartosz Dziewoński: [C: 032] Split Exception.php [core] - 10https://gerrit.wikimedia.org/r/115240 (owner: 10Addshore) [22:49:22] (03CR) 10Bartosz Dziewoński: [C: 032] Add scope to Exception class methods [core] - 10https://gerrit.wikimedia.org/r/115242 (owner: 10Addshore) [22:52:27] (03Merged) 10jenkins-bot: Split Exception.php [core] - 10https://gerrit.wikimedia.org/r/115240 (owner: 10Addshore) [22:52:40] (03Merged) 10jenkins-bot: Add scope to Exception class methods [core] - 10https://gerrit.wikimedia.org/r/115242 (owner: 10Addshore) [22:56:42] mediawiki vagrant - can I log stuff from the apache process? the "vagrant" user owns the debug log file [22:58:08] (03PS1) 10QChris: Set uploader to current user in "patchset-created" event upon rebasing [gerrit] (wmf) - 10https://gerrit.wikimedia.org/r/115316 [22:59:21] manybubbles: ^d: CirrusSearch unit tests should be fine now. Failures you can see should be related to the patch you uploaded. If there is any trouble, do fill bug / mail me and I will look tomorrow :] [22:59:32] hashar: cool [22:59:47] manybubbles: wanna pair again tomorrow? [23:00:02] will be at home so we can hangout :-] [23:00:08] sure [23:00:48] \O/ [23:03:30] (03CR) 10QChris: "Corresponding upstream change is at" [gerrit] (wmf) - 10https://gerrit.wikimedia.org/r/115316 (owner: 10QChris) [23:13:13] FYI: beta-code-update builds are failing since 4 hours: logs says something like: error: Unable to append to /data/project/apache/common-local/php-master/extensions/.git/modules/MobileFrontend/logs/refs/remotes/origin/master: Permission denied [23:14:38] also: fatal: failed to read object b4ac8c5cf408da55c8b36519df2f35d460400c6d: Operation not permitted [23:17:25] wtf, they belong to root [23:18:27] se4598: Should be fixed [23:20:02] (03PS1) 10Chad: Multiply by 100 for percentages, not 200 [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115319 [23:22:57] (03CR) 10Chad: "recheck" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [23:23:38] hoo: ok now its new: 23:23:04 error: unable to unlink old 'MobileFrontend.i18n.php' (Permission denied) [23:23:38] 23:23:04 error: unable to unlink old 'javascripts/modules/editor/EditorApi.js' (Permission denied) [23:23:38] 23:23:04 error: unable to unlink old 'tests/javascripts/modules/editor/test_EditorApi.js' (Permission denied) [23:23:47] sigh [23:23:52] will have a look [23:25:02] spagewmf, am I missing something at MediaWiki:Common.js ? "Get MediaWiki" at https://www.mediawiki.org/wiki/MediaWiki/Homepage_redesign/Preview still doesn't get the mediawiki.ui button class [23:25:36] <^d> hoo, se4598: chown'ing everything in /data/project/apache/common/ to mwdeploy as it should be. [23:25:51] (also, I'm not sure whether editing MediaWiki:Common.js is ok since the page was deleted...) [23:26:25] ^d: There are no more instances left [23:26:30] AFAIS [23:26:45] <^d> Can't hurt :p [23:26:48] not sure how long a whole chown would have taken... depends on no. of files :P [23:27:09] qgil: maybe mw.loader.getState( 'mediawiki.ui.button' ) isn't "registered" at the time common.js is loaded? [23:27:29] se4598, it's my experimentation [23:27:44] hoo^ [23:27:45] MaxSem: you should not experiment on beta using live hacks :D [23:28:16] se4598, I'm trying to follow spagewmf 's instructions at https://www.mediawiki.org/wiki/Talk:MediaWiki/Homepage_redesign/Preview#mediawiki.ui [23:28:24] if I knew how t repro it anywhere else:P [23:28:36] MaxSem: :-D [23:28:41] qgil that seems legit. I didn't write the enwiki common.js, but I believe it works, so ? [23:29:00] hm [23:29:02] anyway really off now. have a good day! [23:29:06] <^d> MaxSem: protip: don't sudo to root, sudo to mwdeploy. [23:29:14] <^d> :) [23:29:33] mh... is Chris Steipp on holiday or otherwise not at work? [23:29:42] he mentioned something like that, but I forgot... [23:29:46] ^d, I intentionally chowned MobileFrontend to prevent jenkins from messing with my files;) [23:30:14] you could have at very least logged that :P [23:30:17] <^d> hoo: At a conferene this week. [23:30:19] <^d> *conference [23:30:25] hoo, I did;) [23:30:26] mh, ok [23:30:45] 17:47 MaxSem: Investigating a mobile bug, might cause intermittent problems [23:30:53] well, that's IMO not clear enough [23:30:57] and it was hours ago [23:31:52] sorry for messing with your stuff [23:32:40] At best just leave something obvious like a file "IM-HACKING-IN-HERE-DONT-TOUCH--MaxSem" in the folder :P [23:33:50] se4598, spagewmf do you see a "Get MediaWiki" big blue button at https://www.mediawiki.org/wiki/MediaWiki/Homepage_redesign/Preview ? [23:33:52] qgil: the snipped on enwiki in common.js is capsualted in a "mw.loader.using( ['mediawiki.util', 'mediawiki.notify', 'jquery.client'], function () {" so it's likly executed later, so there will be then mw.loader.getState( 'mediawiki.ui.button' ) === 'registered' [23:34:04] qgil: no [23:34:40] greg-g: https://gerrit.wikimedia.org/r/#/c/114100/ is officially a major change ("Enable Parsoid's edit caching on all public wikis"), so we should schedule it properly, I guess? The Parsoid team are happy for it to go out, but it probably warrants a proper slot. [23:34:54] qgil: for sure some race condition or side effect [23:35:10] se4598, I do, but only after reloading [23:35:49] James_F, in terms of request volume it should hardly be noticeable [23:36:00] !log Rolled back [23:36:12] !log deployment-prep Rolled back [23:36:18] gwicke: Yeah, but better safe than sorry. [23:36:41] gwicke: Alternatively I could ask Roan to do it in 24 minutes' time in the LD, but it seems like a bad idea. [23:36:42] qgil: or it was again some cache issue. I'm remembering such cache issue with MediaWiki:Common.js I think [23:36:55] se4598, I will leave Common.js as it is, and see what happens [23:37:07] thank you for your help se4598 [23:37:20] (03CR) 10Chad: [C: 032] Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [23:37:26] qgil: works with chrome (has nothing cached b/c not used) :) [23:37:26] (03Merged) 10jenkins-bot: Don't use parsed wikitext when dealing with CSS/JS [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/115214 (owner: 10Chad) [23:37:34] James_F: good call, next week... say Monday afternoon, after parsoid's deploy? [23:37:45] greg-g: WFM. [23:38:06] cool, can you add a line to next month? I'll add it later to the week when I have it [23:38:07] alright [23:38:15] greg-g: Sure. [23:38:26] James_F: ty [23:40:34] MaxSem: where do your !log end up? [23:40:56] ...repeaded in the right channel [23:41:06] in Labs Server Admin Log? [23:41:13] qgil the button's blue now on shift-Reload [23:41:32] ok [23:43:33] hoo, MaxSem: beta-code-update build is succeded again [23:43:52] :)