[00:01:04] thanks [00:02:30] maplebed: yup, i can replicate it [00:03:33] it only affects articles that were loaded from near by [00:03:44] any subsequent page load doesn't show an extra search bar [00:04:08] preilly: $wgMobileDomain simply defines a URL to indicate and force the MobileFrontend skin, right? There is no need to have anything in a /m subdirectory, right? I'm sure the question sounds stupid to a sysadmin :) [00:05:30] $wgMobileDomain is the the subdomain [00:05:57] it can be replaced with nothing if it's not on that domain [00:05:59] preilly: do we see the double search bars because the browser is caching the page ? [00:06:09] tfinc: yes, I believe so [00:08:07] what I don't understand is why a specific subdomain must be defined. Isn't enough to use skin Monobook or MobileFrontend based on the browser detected and call it a day, without changing the URL? [00:08:26] yuvipanda: 33013 has the new icon [00:10:41] philinje: let me check. trying to get search behavior to behave... [00:11:56] preilly: while we should fix that . i don't think its the end of the world to ship without it for 2.2.1 [00:12:08] it only happens on previously viewed near by links [00:17:39] tfinc: no progress on the search bug :( [00:17:49] focus happens, all necessary events are actually being triggered [00:17:51] just.. no focus [00:18:28] hi [00:19:09] [wikipedia-iphone] preillyme pushed 1 new commit to master: http://git.io/tN_gMg [00:19:09] [wikipedia-iphone/master] In the case of nearby me pages that are cached — the data for the URL load should be loaded from the originating source. No existing cache data should be used to satisfy a URL load request. - Patrick Reilly [00:19:32] yuvipanda: would getting brion help this out ? [00:19:35] tfinc: okay, the fix is pushed [00:19:46] k [00:20:02] tfinc: I've got the highlight working. Typign chars is fine too. Just no blinking cursor [00:20:13] maplebed: ---^ near by me search box fixed [00:20:44] possibly, he might've encountered this before. [00:21:05] brion was dealing with this before [00:21:26] i think it;s ok without the cursor [00:22:07] philinje: sent a pull req with focus but no blinking [00:32:30] [WikipediaMobile] preillyme pushed 2 new commits to master: http://git.io/mqzOhw [00:32:30] [WikipediaMobile/master] Slightly better search-button experience - YuviPanda [00:32:30] [WikipediaMobile/master] Merge pull request #99 from yuvipanda/search-button - Patrick Reilly [00:32:43] Project WikipediaMobile - Nightly builds build #57: SUCCESS in 6.7 sec: http://integration.mediawiki.org/ci/job/WikipediaMobile%20-%20Nightly%20builds/57/ [00:32:44] yuvipanda: Slightly better search-button experience [00:42:01] preilly: what ip did we stage zero with? [00:44:58] tfinc: none yet [00:45:12] tfinc: I was busy on some other changes [00:45:19] tfinc: but, it will be the office ip [00:47:22] preilly: how aggressive do you think the native safari cache is in flushing out old pages in the app ? [00:47:51] tfinc: not sure — I'll research [00:48:34] i'm curious as i hear ben saying how he'd like to keep as much cache in as possible. but i don't have a good view of how much is really kept in if we just use the browser cache [00:48:57] * tfinc goes to do a new build of the android app [00:48:58] tfinc: it seems like it flushes on reloads of the app [00:49:27] reload as in when the app resumes ? [00:49:29] or restart [00:49:42] on resume [00:49:48] so thats pretty often [00:51:10] iPhone browser was able to cache a maximum of 19 external 25KB components. The maximum cache limit for multiple components is found to be 475K – 500 KB. [00:54:15] that's pitifully small, considering the massive amount of unused storage on my iPhone. [00:54:26] yuvipanda: still there? [00:54:34] philinje: yes. [00:54:46] someone in #phonegap responded, suggested a few things for the cursor [00:54:51] Heather posted graphics to the About bug [00:54:53] tried 'em and got nowhere [00:55:02] lame [00:55:51] the About screen could be a graphic, or it could be text with a couple of graphics inserted [00:56:06] if text, it could be translated [00:56:07] if we're localizing it obviously has to be text... [00:56:49] so let's give Siebrand a heads-up whenever you get it done [00:57:11] sure, will do. [00:57:19] also will need to scour the app for other untranslated strings [00:57:50] yuvipanda: i have a new bug for you if your up for it [00:58:21] tfinc: sure! [00:58:24] tell me? [00:58:38] https://bugzilla.wikimedia.org/show_bug.cgi?id=33574 [00:58:46] 1) type anything into the search box [00:58:50] 2) tap clear [00:59:10] assigned [00:59:20] the cursor will jump to the beginning of the line but nothing clears [01:01:03] philinje: are we having a go/no go meeting next week for a 1/13 android app launch ? [01:01:55] another non-obviosu bug [01:02:02] that textbox is intent on messing with us isn't it? [01:02:06] yes it is [01:02:18] all the hours chasing browser rendering bugs, not sure if phonegap is worth it :( [01:02:19] who would have thought it wold be the most annoying piece of this [01:02:40] its been interesting to see all this to say the least [01:03:25] not a fan [01:06:05] how much is nit obi vs phonegap? [01:06:20] meaning, the way they implemented things vs the framework? [01:06:30] all of these have been android browser issues [01:06:45] yes, the biggest issues have been browser issues [01:06:50] not nitobi's fault [01:06:59] and those would happen regardless of whatever framework you sue [01:07:00] though tbh they too added to the confusion a bit. [01:07:04] agreed [01:08:07] and their the same kinds of issues that we'll hit with the mobile web [01:08:12] so its good that were tackling them now [01:08:18] even if its tough [01:08:36] making of these changes on the mobile web will be much easier [01:08:38] so we have a win there [01:08:53] not sure, problem with .val('') not actually clearing the textbox... [01:08:58] doubt we'll hit that in the mobile web [01:09:09] *this* particular problem seems to be an intersection of phonegap and the web [01:09:12] *webview [01:10:37] * tfinc is having fun looking at what pl articles are around him [01:10:40] thats working nicely [01:11:43] maplebed: see patricks note above about the browser cache? it looks to be flushed on each app resume [01:12:01] which leads to exactly the behavior that annoys me. [01:12:04] *load a page [01:12:06] * switch apps [01:12:11] * go into no-service zone [01:12:15] * go back to read the page [01:12:17] * page blanks ... [01:12:20] oh fuck. [01:12:25] oh well, guess I'm screwed. [01:12:28] * go play angry birds. [01:12:35] ::sigh:: [01:14:22] yup, the only way to really have a consistent offline experience is to control the whole process [01:14:24] and not let the browser doit [01:14:28] we might get lucky [01:14:33] but chances are it'll suck [01:15:59] tfinc: browsers have offline capabilities too... [01:16:18] our saved pages could easily fit directly into the browser [01:16:47] sure but what i'm saying is that we shouldn't depend on the browser to do it for us as part of its own caching. we should tell it what to do [01:18:13] ah, control 'that' whole process [01:18:14] gotcha [01:32:12] tfinc: https://bugzilla.wikimedia.org/show_bug.cgi?id=33527 [01:34:27] zindex issues. [01:34:38] tfinc: do you want to cut an 'RC' now and get feedback? [01:34:45] or wait for about + the clear button to be fixed? [01:35:33] i really want this to be the last RC [01:36:00] in that case. [01:36:10] how about we get rid of the 'clear' button? :D [01:36:19] long press on backspace does the same thing (almost0 [01:36:21] *almost0 [01:36:22] err [01:36:24] *almost) [01:37:02] thats an option. but id love to know if we can make it work [01:37:11] as we can make it work just fine on the mobile web [01:39:59] tfinc: http://code.google.com/p/android/issues/detail?id=6721 :( [01:40:47] tfinc: problem is that 'clear' is on top of the input box, rather than to the side [01:41:39] and hence it never gets any touch events due to that bug [01:43:07] okay, there's a workaround of sorts there.. [01:55:11] tfinc: i'm removing it for now. [01:58:52] * yuvipanda goes to grab coffee [02:02:15] hmm [02:06:06] no coffee :( [02:06:47] tfinc: any idea what the 'go/no go android' meet request from phil is for? [02:06:58] wether we can push to the store on 1/13 [02:07:08] where we are, whats left to do [02:07:10] etc [02:07:16] ah, right. [02:07:58] tfinc: just thinking a bit ahead - what after 1/13? [02:14:25] hehe [02:14:27] so [02:14:35] we have a couple of options [02:14:49] phil and i ware talking about them yesterday to brainstorm [02:15:03] we can a) start tackling v1.1 features as we have tons of those [02:15:13] b) we can refactor and get the iOS app [02:15:20] out [02:15:29] c) we port to another platform [02:15:37] 'platform' as in? [02:15:38] d) we re-align you and do mobile web work [02:15:45] symbian s60 [02:15:49] ah! [02:16:02] now all of those are good ideas [02:16:11] but we only have so many of us [02:16:14] i like b) [02:16:21] but a) would be nice too [02:16:25] and c) is important [02:16:28] choices! [02:16:38] I like b) then d) with a side of a) and c) [02:16:43] multitasking :D [02:17:02] tfinc: not to mention that there's a good chance i'll be around in SF for 2 weeks. [02:18:37] tfinc: btw, b) should always come before a) [02:18:44] code needs cleanup. [02:19:02] i've a branch that separates out code into UI, network, 'fixups', etc. [02:19:43] would be much easier maintining iOS, bb, WP7, etc [02:22:38] that makes A LOT of sense [02:22:47] and will make a) go much fastere [02:22:50] faster* [02:24:00] true [02:24:17] also will help us keep the platform specific kludges safely separated [02:26:41] tfinc: going to crash. About page will have to wait :( [02:26:54] sent a pull req a while ago for hiding the clear button, so if you want to cut an RC... [02:26:59] and next time we call them 'beta' [02:28:57] yuvipanda: go sleep. lets chat tomorrow [02:29:01] yes [02:29:06] beta1234 [02:29:09] and then rcs [02:29:21] i think the RC'1 worked really well for the iOS 2.2.1 release [02:29:22] it's 8000 pm, I think I should just stay up some more, hit the doctor and then sleep [02:29:29] are you sick again ? [02:29:32] no [02:29:37] just some chronic shoulder issues [02:29:41] ahh ok [02:30:02] be 20, have all the issues of people who are 40 [02:30:05] mmm geek life :) [02:30:13] go to bed [02:30:14] ;) [02:30:17] Hitting the doctor won't help [02:30:23] are you going to be on tomorrow ? [02:30:28] were hosting a meet up here so i'll be around [02:30:29] I find it's best to be nice to people drugging you up [02:30:38] tfinc: i'd probably finish up the about page [02:30:42] k [02:30:43] thanks [02:31:15] s/hit/hit up/ [02:31:22] Reedy: am too young to do any of the other hits :) [02:31:47] tfinc: btw, everything is 'set' for my visa interview on tuesday. [02:50:58] yuvipanda: excellent [22:17:24] yuvipanda: how goes it [22:17:25] i've been busy organizing here in the office [22:17:26] and looking for new designers :)