[13:36:29] hey sumanah [13:36:35] hi bawolff [13:38:06] So Raylton has been talking to me about the possibility of me and him both co-mentoring the book gsoc project [13:40:19] I just wanted to touch base with you, and ask how that would work if that project got accepted (Like would one of us be the "official" mentor, and the other an unofficial helper?) [13:43:58] bawolff: sorry, got distracted [13:44:42] bawolff: yes, one of you would be the mentor of record, I believe. But it depends on how Google decides to do it in Melange [13:44:47] bawolff: (freaking Melange) [13:45:04] lol [13:45:13] bawolff: last year IIRC Melange let me put down multiple people as mentors, formally, for one student [13:45:35] bawolff: and then the requirement was that at least one of those mentors had to give the midterm evaluation and at least one had to give a final evaluation [13:48:50] bawolff: if we accept that student, then we can juggle it so that you, Raylton, or possibly both are formally the student's mentor(s) [13:48:51] I think [13:49:27] Well personally it doesn't matter that much to me, I was more just curious how that worked [13:53:47] bawolff: ah, ok! [13:53:48] :) [13:54:00] Think of the free t-shirt! [13:54:39] lol, and the bumper sticker [13:54:59] Is that new this year? [13:55:18] I got a bumper sticker when I was a student 2 years ago [13:55:27] hmm [13:56:31] <^demon|away> Mentor's don't get a bumper sticker :( [13:57:40] The joys of being a student [13:58:40] * YuviPanda has his T-Shirts and Stickers safely tucked away [14:00:32] My sticker is still in a box somewhere, but i do wear my shirt from time to time [14:00:56] For halloween i dressed up as a dandelion, and my green gsoc shirt was just the right colour [14:01:39] bawolff: I like this year (2011)'s better. Dark blue and stretchy. [14:01:44] but yeah, doesn't work as a dandelion [14:03:57] * bawolff goes and turns off irc to study for his impending exam [14:04:04] hi ialex [14:04:40] hello sumanah [14:04:52] (you probably saw my email) [14:04:57] yes [14:06:06] Thanks for grading proposals! [14:08:40] hi sumanah [14:09:06] hi Nikerabbit [14:09:30] sumanah: I'm still grading few, hopefully not too late [14:10:10] Nikerabbit: not too late, no. Comments are a nice thing to do [14:10:20] Nikerabbit: so that others can understand why you're grading the way you are [14:10:39] oki [14:27:22] hexmode: guillom ping [14:27:29] hi [15:47:45] hexmode: got a link to the page you mentioned just now in TLDR with 1.20 feedback from yesterday? I'm not finding it... [16:47:36] chrismcmahon: https://commons.wikimedia.org/wiki/Commons:VP#MediaWiki_1.20 [17:15:08] prepping for 20% time checkin.... [17:16:00] it's Tuesday, so I'm talking to Brion, Roan, Trevor, and Rob Moen [17:16:21] none of them are on IRC yet [17:16:45] they really like doing 20% time xD [17:16:52] heh. [17:16:58] It's only 10:15am in San Francisco [17:17:26] wow, that's a 9h offset [17:17:29] is there a shorthand to link to gerrit changesets in bugzilla? [17:17:41] We've not added one.. [17:17:43] We should [17:17:44] I guess you're still on winter time? [17:17:58] +1 [17:18:04] Oh, wait [17:18:14] gwicke: match => qr{\bgerrit(\ change(set)?)?\ ?\#?(\d+)}i, [17:18:29] ahh, thanks! [17:18:50] gerrit shouldn't be required :P [17:19:28] https://bugzilla.wikimedia.org/show_bug.cgi?id=35144 gwicke [17:19:43] added to https://www.mediawiki.org/wiki/Bugzilla#What_syntax_can_I_use.3F gwicke [17:19:44] well, changeset should really be the commit hash [17:20:09] does that work wel with the urls? [17:20:09] which is different from the gerrit changeset [17:20:15] We can add both [17:20:25] * sumanah waits for the 20% cohort [17:20:31] we talked about using the shorthand c[0-9]+ [17:21:07] sumanah: thanks for the pointers! [17:21:35] gwicke: Glad to help [17:21:38] hi TrevorParscal [17:21:41] sumanah, here you have TrevorParscal [17:21:57] so it seems the sequential id should be used [17:21:59] TrevorParscal: hope you're settling into your new home OK? [17:24:09] TrevorParscal: I can write a comment re the progress you made last week in reviewing OnlineStatusBar as part of 20% time - https://bugzilla.wikimedia.org/show_bug.cgi?id=32128 [17:24:43] sumanah: yeah, getting on ok [17:24:54] TrevorParscal: basically, your review is in progress? [17:24:57] still working on the house, that will probably never end [17:25:26] yeah, so i'm reviewing en masse because there are so many deferred revs and things need one good pass [17:25:33] also, i'm doing a design review at the same time [17:25:54] is there an active author I can communicate with, or should I just fix smallish things myself? [17:26:01] it's Petr Bena [17:26:08] right [17:26:29] petan on irc [17:26:48] ah, ok didn't make that connection [17:27:23] TrevorParscal: assigned the bug to you so you can communicate with him there [17:27:28] right on [17:27:53] TrevorParscal: and you can see what a few other folks have said there [17:28:06] sure [17:28:08] TrevorParscal: can you ping Brion, Roan, and rmoen to chat with me re their 20% stuff today? [17:28:24] 3 minutes left during the stated checkin window, and I need to go soon and pack for my plane trip [17:28:49] sure, rob should be on in a sec [17:29:00] sumanah: Hello [17:29:05] Roan was up late, he might be lagging this morning [17:29:13] Roan is just getting online [17:29:28] if brion isn't online, he's off the grid [17:29:47] ok, there you go [17:30:18] :-) [17:30:22] hi brion RoanKattouw_away & rmoen [17:30:25] 20% discussion [17:30:43] rmoen: what were you up to last week? I lost track of you a bit - we were talking about you maybe taking on some extension code review [17:31:12] sumanah: Yes, I was reviewing SubpageSortkey [17:31:20] Good morning [17:31:32] sumanah: I have looked at all of the code, I would like to do a little more testing though [17:31:41] rmoen: aha! [17:31:54] sumanah: then I think I could move onto something else [17:32:12] Rob: ok, I'm assigning https://bugzilla.wikimedia.org/show_bug.cgi?id=22911 to you [17:32:33] sumanah: ok [17:32:49] brion: what are you up to for 20% stuff today? [17:32:59] RoanKattouw: and you? [17:33:14] I am going to do a job runner restart that's needed for AFTv5 [17:33:20] Then I'm going to review some AFTv5 code [17:33:27] RoanKattouw: oh yes, AFTv5 [17:33:34] and the job runner thing - is that still fragile? [17:33:36] Then I'm gonna write documentation for the New Way to Deploy Things in the Magical Git Universe [17:33:41] YYAYAYY [17:33:49] It's no longer fragile, I fixed it [17:34:13] Well sort of -- the fix isn't deployed yet so I have to replicate it ad hoc today, but we do have a generic way to safely restart them [17:35:41] sumanah, some work on mobile Blackberry Playbook port, and some going through gerrit [17:35:52] brion: there are still a bunch of patches in Bugzilla that could use friendly rejection or advice, and I would love help with that - like https://bugzilla.wikimedia.org/show_bug.cgi?id=34819 and https://bugzilla.wikimedia.org/show_bug.cgi?id=33582 [17:35:58] excellent [17:37:29] brion: the backlog in Gerrit I think needs less of my pushing in general, but Gerrit patches you might like: https://gerrit.wikimedia.org/r/#change,4658 https://gerrit.wikimedia.org/r/#change,4550 [17:37:55] csteipp: looking forward to meeting you tomorrow! [17:38:31] ok, sounds like all of you know What's Up. Thanks all [17:38:49] <^demon> RoanKattouw: Writing docs?!? You've gone over to the dark side :( [17:38:56] and RoanKattouw, today is the day my spouse's book comes out in paper! [17:39:14] ^demon: if by "dark" you mean "awesome and Sumana-approved" [17:39:30] sumanah: Ah! I must purchase it some time before my next flight then [17:39:42] sumanah, congratulations for him [17:39:48] <^demon> Buttttt, when you write docs you make yourself replaceable ;-) [17:39:49] Thank you! [17:39:59] ^demon: and how else can you get promoted? :) [17:40:01] ^demon, where are the gerrit groups? [17:40:08] Platonides: in gerrit. Duh [17:40:15] I see references as 4cdcb3a1ef2e19d73bc9a97f1d0f109d2e0209cd [17:40:19] <^demon> sumanah: Some of us like being in the trenches and don't want promotions :) [17:40:23] but I don't know where are such objects [17:40:27] don't seem to appear on any repo [17:40:39] <^demon> They're stored in the database [17:40:46] ^demon: True. More seriously, documentation enables vacations and moving to new areas of the codebase.... but I know you jest [17:41:07] and why are things like 4cdcb3a1ef2e19d73bc9a97f1d0f109d2e0209cd mediawiki ? [17:41:55] RoanKattouw: I think you can buy the paperback by going to http://www.candlemarkandgleam.com/store/delivery/constellation-games-serial/ and signing up for the Plutonium Package [17:42:14] <^demon> Platonides: I have no clue off the top of my head. Ask in #gerrit? [17:42:19] Platonides: in case you want to read the first 2 chapters for free (includes video game programming): http://constellation.crummy.com/Constellation%20Games%20excerpt.html [17:44:55] thanks sumanah [17:45:44] Sure! [17:45:48] ok, now I gotta pack [17:46:12] (it is very strange to be thinking about bringing multiple pairs of shoes on this trip) [17:46:33] * RoanKattouw wonders why sumanah's husband's book order web site has a shipping costs calculator that defaults to, of all places, Vermont [17:46:43] RoanKattouw: because the publisher lives in Vermont [17:47:05] RoanKattouw: Leonard actually got to meet her a few days ago! Evidently she is cool [17:47:15] Aha! [17:47:36] Vermont: Home of Ben & Jerry's Ice Cream and a few friends-of-friends [17:48:05] Anyway [17:48:19] I aim to hike this weekend, so I will be purchasing hiking shoes in San Francisco, as I have none [17:48:40] in addition to that, I have just bought new regular everyday shoes for the first time in like 5 years [17:48:42] Aha, where are you going? [17:48:53] I don't know yet -- my friend will guide me [17:49:19] it'll be Earth Day on Sunday, so we may go someplace that is doing a bit of trash pickup -- evidently used fishing line gets discarded and is hazardous to wildlife [17:50:18] the new everyday walking shoes are not really good for treadmills & other workout-y stuff, so I will probably also bring the end-of-lifed sneakers and toss them at the end of my stay [17:50:36] ok, bringing one pair of shoes in my bag, I can handle that, it's just a big switch from 0 to 1 [17:53:44] TrevorParscal: RoanKattouw: I just realized/remembered -- https://bugzilla.wikimedia.org/show_bug.cgi?id=34885 is more urgent than the OSB review [17:53:47] (onlinestatusbar) [17:54:22] <^demon> Totally misread OSB as LSB. [17:54:25] <^demon> And I freaked. [17:54:26] yeah, i was supposed to work with Timo on that on Friday... [17:54:26] <^demon> :) [17:54:37] I will poke at it now I guess [17:54:54] robla was hoping we could come up with something that works, even if it's mildly evil [17:55:05] Platonides: still struggling to properly amend 5174.. [17:55:52] ^demon: LSB? [17:56:31] <^demon> sumanah: http://php.net/manual/en/language.oop5.late-static-bindings.php - the single most evil feature to make it into PHP in a long long time. [17:57:25] gwicke, why are you having problems? [17:57:47] don't remove the Change-Id line [17:57:56] that should be all that is needed... [17:58:17] Platonides: I think I should have switched back to a fresh master before trying the amend [17:58:46] <^demon> sumanah: Basically, a bunch of people complained that static functions don't inherit like normal functions. PHP devs caved and introduced the feature, and (from what I understand) are regretting it terribly now. [18:00:05] Platonides: likely something to do with doing a soft reset instead of hard [18:00:38] Krinkle also had problems amending [18:00:51] I had no issues when I did it [18:01:01] maybe it's a different git-review version [18:01:10] or perhaps because I amended them just after committing [18:01:26] and further changes in the checkout confuses git [18:01:55] just reset my local state with checkout -f [18:03:59] very weird [18:04:10] I just fetched the changeset I wanted to amend [18:04:21] A few people have been having issues [18:04:22] then changed a line and did a commit -a --amend [18:04:27] then git review -R [18:04:32] <^demon> `git alias ohcrap="checkout --hard origin/master"` [18:04:34] <^demon> :) [18:05:09] I did a checkout -f master [18:06:07] gwicke, see if https://gerrit.wikimedia.org/r/#change,5174 is what you wanted [18:07:03] Platonides: yep, thanks! [18:07:25] it worked flawlessly for me :S [18:07:26] git-review -d 5174; edit includes/Sanitizer.php ; git commit --amend -a ; git-review -R [18:07:50] (actually, I first forgot the -a, but that was a PEBKAC) [18:08:00] hmm, pretty much what I did [18:08:13] except that I used the checkout command from gerrit [18:08:29] instead of git-review -d 5174 [18:08:36] is there an analog to "ssh srv193; sync-common" for test2? [18:08:44] Reedy, RoanKattouw ^ [18:08:54] No [18:08:56] <^demon> test2 is just like any normal wiki [18:09:00] awjr: No, test2 is a real cluster wiki, not special like test [18:09:01] Push to everything [18:09:02] <^demon> Just sync out php-1.20, right? [18:09:06] yeah [18:09:09] oh i see [18:09:11] php-1.20wmf1! [18:09:23] <^demon> Pfft, you had to go and rename that. [18:10:41] well, we'd have started with php-1.20 [18:10:44] so we do not have a staging wiki running 1.20? [18:10:55] no, commons, mw.org and test2 [18:10:58] 1.20wmf1, that is [18:11:02] k [18:11:04] then when we had the 2nd version, we'd need some crappy suffix [18:11:14] so the result of git-review -d seems to be different from running the gerrit-provided checkout command [18:13:57] <^demon> gwicke: Yes, git review -d checks out the changes to a branch new branch for you to work on. [18:14:15] <^demon> The gerrit-provided one just applies it to a detached head, I believe. [18:14:30] * ^demon can't remember, he only does the former. [18:14:40] yeah, it results in a detached head [18:15:36] so we should remove the non-git-review alternative from https://www.mediawiki.org/wiki/Git/Workflow#Amend_your_change [18:15:51] <^demon> Well, not everyone's using git-review :\ [18:16:28] * ^demon goes back to his snack [18:16:58] right now it suggests something that apparently cannot work [18:17:08] How does it not work? [18:17:15] "Detached head" !== "doesn't work" [18:17:36] well, I got the same behavior as the note left there by somebody else [18:18:34] did not test the git-review -d alternative, but the gerrit command at least consistently resulted in a new changeset instead of a proper amend [18:18:54] How the hell does that happen? [18:19:03] When amending, did you change or remove the Change-Id line in the commit message? [18:19:19] not explicitly [18:19:24] Did you fetch without git review but submit with git review? [18:19:30] but I used -m 'Message' [18:19:34] which might be it [18:19:36] Yes [18:19:40] That ist [18:19:43] *That is it [18:19:47] Timo probably did that too [18:19:56] ok, so that part of the documentation is wrong [18:19:56] That's completely unrelated to the fetch stuff [18:20:00] I'll fix the docs to mention that [18:20:08] OH HELL [18:20:11] Who put THAT there? [18:20:19] * RoanKattouw finds out who wrote that [18:21:37] RoanKattouw: while you are at it, you could also remove the fixme note somebody left in that section [18:22:18] Yes, that was timo [18:24:27] lol [18:24:33] First he added the fixme, then he broke the example [18:25:01] sneaky ;) [18:27:59] (with -m) [18:28:02] If you use -m you're removing the Changeset-Id [18:28:18] then gerrit places that at a new changeset [18:28:20] Yes, exactly [18:28:33] So first Timo put a fixme template on the documentation [18:28:41] THEN he rewrote the documentation to use -m xD [18:29:08] maybe he already had the edit window opoen :) [18:29:50] poor gwicke :) [18:30:17] that's what you get for mostly self-reviewing Parsoid stuff ;) [18:47:47] there appear to be broken parser tests in wmf/1.20wmf1 [18:49:48] Are you talking about test#1 failing with a timeout? [18:49:50] That's a known issue [20:39:41] mdale: anything you can add to https://www.mediawiki.org/wiki/API:Upload#Chunked_uploading ? [20:44:16] will do :) [20:46:05] thx! [20:47:10] it was surprising to see you reply after the quit :) [20:47:47] i'm surrounded by computers. one quit don't mean another quit :D [20:48:25] :D [21:20:50] hi kaldari, my first contribution to Wikimedia : https://gerrit.wikimedia.org/r/#change,4778 [21:21:08] \m/ ( Merged code i meant ) [21:21:09] congrats! [21:21:32] thanks ! [21:24:18] sorry to bug you guys, but does anyone know if there is a unary negation in the abuseFilter grammar? [21:24:36] i.e. the first ! in !(lines_added rlike freakyregex) [21:24:55] (or the last, seening there is only one) [21:25:32] ah, yes, there is. [21:25:42] * FooBarMartijn bows and thanks you for your time [21:25:43] ;) [21:26:45] er, now I'm not sure anymore :/ [21:27:18] drecodeam, congrats :-). that was an annoying bug. [21:27:42] Eloquence: thanks. [21:27:59] Eloquence: I deployed the VisualEditor beta warning thingy a few hours back BTW [21:28:06] I saw :) [21:28:32] although I liked it better without the flash of unstyled content introduced by removing the style attribute .. [21:28:58] I was never quite clear on why that was there, but I'll let you fight that out with Trevor [21:29:02] heh [21:29:09] He submitted an improved version which is merged but not deployed yet [22:01:37] Tim-away: ping [22:01:47] Tim-away: is there an issue with the Message Cache [22:01:56] Reedy^^ [22:02:16] preilly: what's the issue? [22:02:50] robla: missing some translations on production [22:03:05] robla: for id.zero.wikipedia.org for example [22:03:44] RoanKattouw: ^ is there possibly a bigger problem that we were tripping into yesterday with the WikiEditor pref? [22:06:23] Possibly [22:06:25] Let me see [22:06:30] preilly: What's the i18n key of the message? [22:09:16] RoanKattouw: zero-rated-mobile-access-banner-text-digi [22:09:35] Is https://id.wikipedia.org/wiki/MediaWiki:Zero-rated-mobile-access-banner-text-digi correct? [22:09:54] RoanKattouw: yes [22:10:07] http://id.zero.wikipedia.org/wiki/MediaWiki:Zero-rated-mobile-access-banner-text-digi displays the same [22:10:12] OK, so the server has the message [22:10:20] Where is it that you're not seeing the message, or seeing the wrong contents? [22:10:21] RoanKattouw: hmm [22:10:30] RoanKattouw: let me ask Dan Foy [22:10:35] RoanKattouw: he reported it to me [22:10:43] OK [22:10:49] Let's drag him in here then [22:11:06] <^demon> What if there's ever a lang code "zero?" [22:11:07] <^demon> [22:11:09] <^demon> :) [22:13:25] RoanKattouw: okay, I think I found the issue [22:24:04] is there anybody here that can grant me translator permissions on http://translatewiki.net/ user account, "preilly" [22:24:12] ? [22:25:04] preilly: Ask #mediawiki-i18n [22:25:51] RoanKattouw: thanks! [22:27:25] wow, http://translatewiki.net/ has Google ad sense ads [22:30:43] it's truly a third party project :) [22:33:42] Eloquence: ha ha [23:45:27] rmoen: thanks for the review [23:46:04] bawolff: np [23:48:50] btw, The reason it doesn't do anything for namespaces that don't have subpages enabled is to be consistent with how core treats namespaces with subpages [23:49:48] bawolff: Ok, I assumed there was probably a reason. It tricked me though at first while testing the new sort keys [23:50:16] yeah, I could see how that could be confusing, I'll add a note to the documentation about that.