Tuesday, 2011-11-22

*** Quits: johnbender ([email protected]) (Ping timeout: 258 seconds)00:46
*** Joins: aaronmcadam ([email protected])00:55
*** Joins: sven_franck ([email protected])02:32
*** Quits: RuAnt ([email protected]) (Ping timeout: 245 seconds)02:47
*** Joins: RuAnt ([email protected])02:49
*** Quits: RuAnt ([email protected]) (Changing host)02:49
*** Joins: RuAnt ([email protected]/ruant)02:49
*** Quits: RuAnt ([email protected]/ruant) (Ping timeout: 244 seconds)02:59
*** Joins: RuAnt ([email protected])03:00
*** Quits: RuAnt ([email protected]) (Changing host)03:00
*** Joins: RuAnt ([email protected]/ruant)03:00
*** Quits: sven_franck ([email protected]) (Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243])03:36
*** Joins: erichynds ([email protected])05:24
*** Joins: DaveMethvin ([email protected])05:36
*** Joins: timmywil ([email protected])05:48
*** Quits: timmywil ([email protected]) (Ping timeout: 260 seconds)05:53
*** Joins: miketaylr ([email protected])05:55
*** Joins: timmywil ([email protected])06:10
*** Joins: rworth ([email protected]et)06:13
*** Joins: sven_franck ([email protected])06:17
*** Joins: danzik17 ([email protected])06:25
*** Joins: rwaldron ([email protected])06:34
*** Joins: kinblas ([email protected])07:26
WiltoHey guys.07:40
*** Quits: sven_franck ([email protected]) (Ping timeout: 240 seconds)07:47
*** Joins: ryanneufeld ([email protected])07:51
*** Joins: sven_franck ([email protected])08:11
gseguin'morning08:38
*** Quits: aaronmcadam ([email protected]) (Quit: Leaving)08:48
*** Quits: ryanneufeld ([email protected]) (Remote host closed the connection)08:51
WiltoSo I have an idea, which is generally kind of a dangerous thing.08:51
Wiltokinblas, you kickin’ around in here this morning?08:51
*** Joins: miketayl_r ([email protected])09:05
*** Quits: miketaylr ([email protected]) (Read error: Connection reset by peer)09:05
*** Joins: ryanneufeld ([email protected])09:07
*** miketayl_r is now known as miketaylr09:11
WiltoAnybody? Bueller?09:16
sven_franckWilto: everyone scared you have an idea?09:17
* miketaylr is09:17
sven_franckforgot09:18
WiltoAs you all should be.09:19
WiltoSo! I am in pretty serious like with jsperf.com, as you may well know.09:19
WiltoWhile refactoring the dialogs, I figured it would be a good idea to keep half an eye on how my changes are impacting performance. I’m thinking this should be a Thing we do, whenever someone is making a semi-major change—put a representative jsperf test together, and include the link in the commit message or something.09:20
Wilto“Show your work,” as toddmparker_ put it.09:21
sven_franck$.extend({playground:true}09:21
*** Joins: johnbender ([email protected])09:23
garannit would be cool to build in perf tests, too09:24
garannso you'd get that info when the project built09:24
garannin some sparkling utopian future, i guess..09:25
Wilto…We can do that?09:26
garannin theory09:26
WiltoBe careful sharing such information with a known data hoarder.09:26
garannheh, noted09:27
garanni mean, we have these unit tests.. do those get run when the project builds right now?09:28
WiltoYep.09:29
kinblasWilto: yeah I'm here :-)09:29
Wiltojqm-jenkins-bot wiles out in here if they break, too.09:29
garannsweet09:29
kinblasWilto: you talkin' performance?09:30
kinblas:-)09:30
WiltoYou know it, brah.09:30
WiltoAnd as The Keeper of the Milliseconds, I thought you might dig the semi-manditory-for-big-changes jsperf idea.09:31
WiltoWell. “Strongly encouraged,” anyway.09:31
kinblasWilto: I like the idea of folks getting in the habit of checking code patterns they write for perf impact09:34
kinblasnot sure how you would do something for dialog though like I thought I read above09:34
johnbendergarann: we have a jenkins server that runs the suite on source change09:35
johnbenderkinblas: Wilto: just beware, it's really hard to take into account all things that affect performance09:35
WiltoOh, I know it.09:36
kinblasjohnbender: +109:36
WiltoStill, lots of cases where it might point out major gains/losses.09:36
johnbenderthat doesn't mean we shouldn't do it, just that you can sometimes get results that will steer you in the wrong direction09:36
WiltoYeah, I wouldn’t want it to become, like “wait drop everything undo that fix—it’s 3.1415% slower.”09:40
*** Joins: eephillip ([email protected]/supporter/student/eephillip)09:43
johnbendertoddmparker_: just making some changes to eddies09:51
johnbenderpr09:51
johnbenderthen the warning should be good to go09:51
*** Joins: aaronmcadam ([email protected])10:04
kinblasjohnbender: I think the parsing of the android UA for the lists-ul-record.php might be off10:05
kinblasjohnbender: it's counting my Android 2.1 or 2.2 device as Android 5.0?10:05
kinblassven_franck: ping10:10
sven_franckkinblas: yo10:10
kinblassven_fanck: in the email what did you mean about providing enhancements for file-uploads?10:11
sven_franckkinblas: 1sec10:11
sven_franckkinblas: <input type="file"> is currently looking pretty lame on a JQM site. Even if it's not supported on all devices, JQM enhancing it would hurt?10:13
sven_franckexample: http://www.franckreich.de/jqm/page3.html10:14
sven_franckkinblas: some make-up woudn't hurt.10:15
kinblassven_franck: oh you're talking about the file form element10:16
sven_franckkinblas: correct, pardon my language10:17
sven_franckdigging through files...10:17
WiltoAndroid 5.0! What are things like in the future, kinblas?10:17
kinblasWilto: shiny ui that still can't animate correctly10:18
WiltoSounds about right.10:18
sven_franckkinblas: back in JQM.a1 I tried myself - http://www.franckreich.de/3/members/b_importb.html10:20
kinblassven_franck: so are you just wanting to dress up the button? Some platforms have both a button and textfield10:22
kinblasnote we also are unable to submit forms with file inputs10:22
sven_franckkinblas: correct.10:22
kinblasvia ajax10:22
sven_franckkinblas: ok. still looks ugly. I was thinking more along UI than AJAX.10:24
*** Joins: eddiemonge ([email protected])10:25
kinblassven_franck: ok, it might take some experimentation to see if that would work. The file input is special and does not respond to things like the other elements due to security10:27
sven_franckkinblas: ok, will look into it, too10:29
johnbendertoddmparker_: just pushed eddie's work with some changes10:37
*** Joins: jqmobilecommit ([email protected])10:38
jqmobilecommit[jquery-mobile] johnbender pushed 6 new commits to master: http://git.io/nI120A10:38
jqmobilecommit[jquery-mobile/master] Allow local browsing of the docs without a web server - Eddie Monge10:38
jqmobilecommit[jquery-mobile/master] Ajax should be disabled instead - Eddie Monge10:38
jqmobilecommit[jquery-mobile/master] Add check to see if ajax should be disabled - Eddie Monge10:38
*** Parts: jqmobilecommit ([email protected]) ()10:38
johnbenderkinblas: that's entirely possible10:38
kinblassven_franck: I just closed issue https://github.com/jquery/jquery-mobile/issues/313510:38
socialhapy★ Issue #3135 on jquery-mobile, reported by frequent (4h, 44m ago): plugins cannot be run on pages pulled into the DOM10:38
kinblaswith comments10:38
eddiemongejohnbender: did you look at the message? the padding seemed to cause overflow-x10:41
johnbendereddiemonge: where?10:41
johnbenderon a phone?10:41
eddiemongein the example todd had at least10:42
johnbendereddiemonge: hmm, are we worried about this being served to a phone?10:42
johnbenderthat is, will the warning ever be served to a phone?10:42
sven_franckkinblas: THANKS! Made my day10:42
eddiemongeprobably not10:43
jzaeffererhey mobile people - I'm looking for an idea or two on how to deal with the sidebar on small screens on my blog http://bassistance.de - for the most part its working good enough already, but completely hiding the sidebar isn't really an option10:43
jzaeffererits basically a trivial challenge in responsive design principles ;)10:43
johnbenderjzaefferer: that counts me out, but that's a lot of content to squish10:43
johnbenderI'll have to deferr to the designers among us10:44
WiltoDo I count?10:44
jzaeffereropen to everyone!10:44
eddiemongei would say the content looks unimportant enough to have it fall below the main listings10:44
WiltoCollapse that sucka, jzaefferer.10:45
jzaeffererbetween main content and footer?10:45
jzaefferercollapse?10:45
eddiemongeyes and yes10:45
WiltoYeah, I’d drop it down there too. Then make the headings toggle their associated content below a certain point.10:45
jzaeffererdo you have some resource on combining media queries with additional JS interactions?10:46
jzaeffererso far its all in CSS10:46
WiltoDO I EVER10:47
WiltoI am your resource, sir.10:47
WiltoBut the key is: when you have weird interactions going on below a certain breakpoint, keep the associated styles in a max-width media query.10:48
WiltoLook quick, ’cause I’m about to break this link on purpose:10:48
Wiltohttp://matmarquis.com/nav/10:48
WiltoSo, shrink your browser down and open the search toggle, then expand your browser again.10:48
WiltoIt drops the open/closed styles above the designated break point, but if you scale it down again: since the class is still in place the style is reapplied.10:49
jzaefferervery nice10:51
jzaeffererthanks10:51
WiltoIt’s also a little weird, but setting a data- attribute for the _scripting_ breakpoint makes things a lot more flexible.10:51
jzaefferercan I forward that link, or will it be gone any minute?10:53
WiltoAny minute, unfortunately—but I’ve got a post in the works for Filament.10:54
*** Joins: jqmobilecommit ([email protected])10:54
jqmobilecommit[jquery-mobile] johnbender pushed 1 new commit to master: http://git.io/QzIN7g10:54
jqmobilecommit[jquery-mobile/master] fix for float value appearing earlier in ua string - John Bender10:54
*** Parts: jqmobilecommit ([email protected]) ()10:54
*** Quits: eddiemonge ([email protected]) (Quit: eddiemonge)11:00
kinblasWilto: that search bar is pretty slick ... especially how you swap the search icon onto the form element11:02
WiltoThanks, man. I’m digging that search pattern, for sure—toddmparker_’s idea, that.11:03
kinblasVisually it looks like the same element but it isn't11:03
kinblas:-)11:03
kinblasmind tricks11:03
kinblasI was trying to figure out how you were killing behaviors with no CSS11:03
WiltoI just gotta learn how to write without three emdashes and two “goddammit”s per sentence, and I’ll be writing up blog posts about some of this stuff.11:03
kinblasthat's when I realized it wasn't the same element11:04
WiltoSlight of hand, sir.11:04
* kinblas looks at Wilto's left hand, while he pick-pockets with the right11:04
Wilto…Who told you about my last career.11:04
*** Joins: jqmobilecommit ([email protected])11:06
jqmobilecommit[jquery-mobile] johnbender pushed 1 new commit to master: http://git.io/roQEKw11:06
jqmobilecommit[jquery-mobile/master] account for optional 'U;' in iPad ua. The fun continues - John Bender11:06
*** Parts: jqmobilecommit ([email protected]) ()11:06
johnbenderkinblas: I've done my UA penance for today11:10
johnbenderkinblas: corrected the android issue and the newer ipad ua string11:11
* kinblas loads visualize11:11
kinblasjohnbender: so out of curiosity11:11
kinblaswhat is this fullboot versus pageload?11:12
Wilto“Optional U?”11:12
WiltoJesus.11:12
johnbenderkinblas: https://github.com/jquery/jquery-mobile/blob/master/tests/speed/stats/startup.js11:13
kinblasyeah ok I was just lookin' at that11:13
johnbenderkinblas: that was my best "guess" really as to when we should be taking those snapshots11:13
johnbenderbased on our conversation11:14
kinblasjohnbender: take a look at /test/tools/page-load-times.js11:15
kinblasjohnbender: so the one thing about those perf numbers11:16
kinblasoh nevermind11:17
kinblasI forgot you are timing the page, but the content is already there11:17
kinblasyou aren't dynamically loading content11:17
johnbendernope11:17
kinblasI was going to mention something about network latency11:17
kinblasbut nevermind11:17
johnbenderyah I didn't want to tackle that bit11:17
*** Joins: bot-t1 ([email protected]/temp01/bot/bot-t)11:17
*** Quits: bot-t ([email protected]/temp01/bot/bot-t) (Disconnected by services)11:17
*** Quits: bot-t1 ([email protected]/temp01/bot/bot-t) (Excess Flood)11:17
kinblasjohnbender: so I was wondering if we should have 3 numbers11:17
johnbenderkinblas: we can have as many numbers as you want11:18
johnbenderalso I can't find that page11:18
johnbendererr11:18
johnbenderthat js you linked11:18
kinblasload (before create) , enhancement(create to init), and then full-load11:18
*** Quits: temp01- ([email protected]/temp01) (Ping timeout: 252 seconds)11:18
johnbenderkinblas: is it tools/page-change-time.js ?11:18
kinblassorry11:18
kinblashttp://jquerymobile.com/test/tools/page-change-time.js11:19
*** Joins: temp01 ([email protected]/temp01)11:19
kinblasjohnbender: the reason, is for example on iPad we see an upward trend11:19
kinblasit would be nice to know if that was enhancement related?11:20
kinblasor was it mostly in the loading/processing before enhancement11:20
*** Joins: bot-t ([email protected]/temp01/bot/bot-t)11:21
kinblasI can't tell if it was due to network latency or what ... actually that can still factor in because the jquerymobile.js is loaded afterwards11:21
kinblasit loads synchronously so we are at the mercy of the network/cache11:22
johnbenderkinblas: https://gist.github.com/33b03cc038c8caf4635011:23
kinblasjohnbender: yeah that looks about right11:24
kinblasjohnbender: is there a way to group the 3 items into a single chart?11:24
johnbenderkinblas: you're right, hmmm11:24
johnbenderkinblas: which items?11:24
*** Joins: fcheslack ([email protected])11:25
kinblasyou know like have a chart that groups full,enhance,pageload for a given day next to each other followed by a gap for the next day with the same 3 things, etc11:25
johnbenderkinblas: we've have to create a seperate query, but it's entirely possible11:25
kinblasI've seen charts like that before, not sure if that plugin you use supports that kind of grouping11:25
johnbenderkinblas: the data is rather raw for that reason11:26
johnbenderI wanted it to be flexible11:26
kinblas+111:26
kinblasI likes my data raw11:26
* johnbender was an oracle dba for a while11:26
johnbenderkinblas: I'm going to do at least one more pull request today and then dig into the deployment stuff11:28
WiltoCue: O.D.B.11:28
johnbenderOracle DBA -> .NET/SQL Server consultant -> Ruby dev -> Javascript dev11:29
johnbenderprogressively getting more hipster11:29
toddmparker_hey all, sorry i was testing devices11:29
toddmparker_forgot to check in11:29
toddmparker_what did i miss?11:30
WiltoHip-hop references, and jzaefferer inadvertently guilt-tripping me into writing that responsive nav blog post.11:30
toddmparker_i see, good11:31
toddmparker_get writing!11:31
WiltoSIR.11:31
toddmparker_so i've been testing some stuff11:31
toddmparker_this article got me fired up again to try using fonts for icons11:31
toddmparker_http://css-tricks.com/15099-using-fonts-for-icons/11:31
toddmparker_demo page: http://css-tricks.com/examples/IconFont/11:31
toddmparker_hover, after testing all our devices, support isn't great11:32
toddmparker_and when if fails, you get typical missing character boxes and odd characters11:32
toddmparker_my results11:32
toddmparker_https://github.com/jquery/jquery-mobile/wiki/Fonts-for-icons---support-on-mobile-browsers11:32
kinblastoddmparker_: heh11:33
kinblastoddmparker_: is there really a standard icon font?11:35
kinblasstandard icon web-font11:35
kinblasthat example just sets a font-family: icons11:35
toddmparker_no, people make their own fonts, then embed them11:36
toddmparker_the font is really just icons mapped to characters11:36
kinblasok so they need @font-face11:36
toddmparker_great idea and i wanted to do this for 1.011:36
toddmparker_but support isn't good enough11:36
kinblastoddmparker_: you got my cc to that google guy right? My email has been flakey lately11:37
kinblasso I can't tell and I'm paranoid11:37
toddmparker_yep11:37
toddmparker_thanks for following up11:37
toddmparker_think my tone was a little defensive :)11:37
kinblastoddmparker_: to roman?11:38
kinblasnah11:38
toddmparker_he just wrote back to me11:38
WiltoI just wish you hadn’t flipped my desk before sending it. _Again._11:38
toddmparker_yeah11:38
toddmparker_sorry Wilto11:38
toddmparker_that's what you get for sitting near me11:39
kinblasI didn't get that impression ... but I think he glossed over the asks in the 1st reply11:39
kinblasyou were just clarifying11:39
kinblasanyways we'll see if Nicolas replies, he didn't the first time11:39
kinblaswe'll see if 2nd time is a charm11:40
kinblasjohnbender: so it's things like the Android 2.2 graphs that got me worried about the network latency11:41
kinblashttp://jquerymobile.com/test/tests/speed/stats/visualize/11:41
johnbendertoddmparker_: have you had a chance to create and try the docs yet?11:41
kinblasI don't think we got that bad :-)11:41
johnbenderkinblas: yah I was wondering about that myself11:42
johnbenderkinblas: we're getting down to the point where it really matters too11:42
toddmparker_sorry johnbender - i can now11:42
johnbenderkinblas: maybe record the start time at mobileinit?11:42
kinblasjohnbender: yeah ... the thing is it *DOES* matter if the device has the jquerymobile.js already in the cache11:42
kinblasit reports lower numbers11:42
kinblasfor sure11:42
johnbenderkinblas: kk I'll make that change and pus11:43
kinblasso that first load on all devices skews the numbers11:43
kinblasjohnbender: yeah mobileinit might be a good place ... but are you trying to measure from doc load to pageready?11:44
kinblasIf so then what you have is about right11:44
kinblasthough it excludes the loading of core11:44
johnbenderkinblas: that was the original idea, but if we want to avoid the network latency issue we have to wait until the js is loaded from the server like you said right?11:45
kinblasya mon11:45
johnbenderkk11:45
johnbenderhmm11:45
johnbenderI'll add it as another data point11:45
johnbenderthat way we can get seperate graphs and it is technically a seperate point11:46
johnbenderhmm11:46
johnbendernevermind11:46
jqm-jenkins-botProject jqm-master build #301: STILL FAILING in 40 min: http://10.59.19.240:8080/job/jqm-master/301/11:50
jqm-jenkins-bot* john.m.bender: Allow local browsing of the docs without a web server11:50
jqm-jenkins-bot* john.m.bender: Ajax should be disabled instead11:50
jqm-jenkins-bot* john.m.bender: Add check to see if ajax should be disabled11:50
jqm-jenkins-bot* john.m.bender: Allow unlimited sub-directory replacements11:50
jqm-jenkins-bot* john.m.bender: Add in more link rewrite test cases, and warning message11:50
jqm-jenkins-bot* john.m.bender: target the page through the event, simplify the href selection, account for ajax page loads11:50
jqm-jenkins-bot* john.m.bender: fix for float value appearing earlier in ua string11:50
jqm-jenkins-bot* john.m.bender: account for optional 'U;' in iPad ua. The fun continues11:50
johnbenderpossible merge issue there, I'll take a look11:51
toddmparker_ok, i just emailed my foxed and overflow testing11:51
toddmparker_fixed11:51
toddmparker_i'll post this to the wiki, but formatting will be rough11:52
toddmparker_on the wiki11:56
toddmparker_https://github.com/jquery/jquery-mobile/wiki/Position:fixed-and-Overflow:-testing11:56
toddmparker_kinblas ^^11:56
kinblastoddmparker_: so with zoom on, most default back to viewport == size of doc?11:57
toddmparker_i think this is just about setting maximum-scale=1.0; user-scalable=0;11:58
toddmparker_yeah, so zoom disabled:11:59
toddmparker_<meta name="viewport" content="width=device-width, initial-scale=1.0, minimum-scale=1.0, maximum-scale=1.0,user-scalable=no" />11:59
toddmparker_(from the test page)11:59
kinblastodddmparker: the 2.2 info is bunched together11:59
toddmparker_not disabled:11:59
toddmparker_<meta name="viewport" content="width=device-width, initial-scale=1.0" />11:59
toddmparker_tough.11:59
toddmparker_ok, fixed12:00
kinblastoddmparker_: heh and you use zoon everywhere :-)12:00
toddmparker_arg12:00
toddmparker_fixed12:01
kinblastoddmparker_: so my one hesitation about ditching the current fixed behavior12:01
kinblasis folks that use navbars today12:01
toddmparker_...is it's so awesome?12:01
kinblasfixed header/footer I meant12:02
toddmparker_sur12:02
kinblasnot fixed positioning12:02
toddmparker_sure12:02
kinblastoddmparker_: that said12:02
toddmparker_we'd need to test this a lot, just seems like a good lead12:02
toddmparker_promising12:02
kinblaswe can always give them a plugin for backwards ease12:02
toddmparker_yeah12:02
toddmparker_it would be worth dropping this in and seeing how good we can get it12:02
toddmparker_and where the gotchas are12:03
toddmparker_i know that form elements in a toolbar bust in android with our dynamic re-positioning12:03
toddmparker_it keeps fighting with the on-screen keyboard12:03
kinblasyeah12:03
toddmparker_so maybe this will help12:03
kinblasyeah because we're not moving anythig12:04
toddmparker_right12:04
kinblasthe browser does ... if it supports it12:04
kinblasdon't get me wrong, I like it ... I just always thought scrolling headers/footers sucked in jqtouch12:04
kinblas:-)12:04
sven_franckkinblas: toddmparker_: sounds like magic?12:04
toddmparker_does jqtouch used position:fixed?12:05
toddmparker_or you just don't like fixed bars in general?12:05
kinblastoddmparker_: I don't think so, headers/footers are inline12:05
toddmparker_so what did that comment mean?12:05
kinblaslol12:05
kinblasI thought if you're going to emulate native12:06
kinblasit should stay fixed12:06
kinblasI now realize it's hard12:06
kinblasbut it's hard because the platform doesn't support fixed the way we are used to on desktops12:06
toddmparker_yeah12:07
toddmparker_but now enough do that it's a good option12:07
toddmparker_i like the idea of still have the option to turn on the tap to hide or an auto-hide after X seconds, just combined with this technique12:07
kinblastoddmparker_: hmm now I'm confused12:11
kinblasI thought what you were talking about was losing the fixedheaderfooter stuff and just relying on good ole CSS fixed?12:11
toddmparker_yep12:11
Wilto+1 to that.12:12
Wilto+2, even.12:12
kinblastoddmparker_: so I guess I'm ocnfused about is what you meant by combining that with tap to hide or auto-hide?12:13
toddmparker_well, you could still fade out the bars, even if using position:fixed12:13
toddmparker_right?12:13
kinblasoh yeah, sure12:13
kinblasbut that would look odd when at the top of the doc12:14
WiltoWe could selectively apply it with a feature test for fixed.12:15
WiltoThere’s no way there isn’t one.12:15
WiltoAaactually.12:16
WiltoWhen we show/hide, we’d just check if the offset from the top of the document is zero. If so, don’t do anything.12:17
toddmparker_that coud work. think it's like that now?12:18
WiltoNo idea. I customized TextMate to point to those files in my project drawer with a little “here there be monsters” flag.12:19
WiltoBut if dialogs don’t completely blow up in my face, that could be next on my hit list.12:28
johnbenderkinblas: so I'm adding two new data points for that page12:40
johnbenderenhancement and pageprocess12:40
johnbenderenhancement = pagecreate -> pageinit12:41
johnbenderpageprocess = mobileinit -> pageinit12:41
johnbenderpageprocess could be pageprocessing12:41
kinblassounds good12:43
johnbenderkewl12:43
johnbenderjust some testing and I'll push12:44
johnbenderbummed I didn't think of that for the original stats :(12:44
kinblasjohnbender: so for the missing data points from previous entries12:44
kinblasnevermind ... are you resetting everything?12:44
johnbendernope12:44
johnbendernew data points12:45
kinblasok cool12:45
johnbenderwe'll keep collecting those too since there is _some_ value there12:45
johnbenderjust not sure what12:45
johnbenderlol12:45
kinblasjohnbender: btw did you see the web selenium driver bustage on jenkins?12:46
johnbenderkinblas: yessir12:47
johnbenderI am aware!12:47
kinblasjohnbender: I guess sometime after Thanksgiving you should show me how to reset that stuff so you aren't the only one having to do it all the time12:47
johnbenderkinblas: job security :D12:48
johnbenderkinblas: I agree though it's always bad to have bottlenecks12:48
kinblasheh ... ok if you're good with always dealing with it12:48
johnbenderand it's not guaranteed that i'm "doing it right" either so its good to get others involved12:48
toddmparker_ok johnbender - do you still need me to test the docs message stuff?12:49
johnbendertoddmparker_: yah I would just take a gander and see if you spot anything wrong12:50
toddmparker_ok. can you remind me what to do? sorry12:50
johnbenderI think we have it sorted fairly well12:50
johnbendercommands from the root:12:50
johnbendermake docs12:50
johnbenderunzip compiled/jquery.mobile.docs.zip12:50
johnbenderopen jquery.mobile/index.html12:50
toddmparker_ok, cool12:51
toddmparker_so this is in master?12:51
johnbenderyessir12:51
toddmparker_ok12:51
johnbenderI pulled in eddie's changes and make some quick tweeks12:51
johnbendertoddmparker_: actually12:51
toddmparker_nice12:51
toddmparker_yeesssss12:51
johnbenderI meant to ask, which browsers had ajax working with the file protocol?12:51
johnbenderI set it up to work with page changes but I need to double check it12:52
johnbenderi think it said in the ticket but none of those worked for me12:52
johnbenderand by none of those I mean "firefox" :P12:54
*** Joins: jqmobilecommit ([email protected])12:55
jqmobilecommit[jquery-mobile] toddparker pushed 2 new commits to master: http://git.io/wd-p3Q12:55
jqmobilecommit[jquery-mobile/master] Update docs/pages/page-dynamic.html - Anne-Gaelle Colom12:55
jqmobilecommit[jquery-mobile/master] Merge pull request #3139 from agcolom/patch-38 - Todd Parker12:55
*** Parts: jqmobilecommit ([email protected]) ()12:55
toddmparker_johnbender: where does the zip go?12:58
johnbendercompiled/12:58
toddmparker_ok, looks good13:02
toddmparker_so you guys got this working in chrome huh?13:02
toddmparker_windows chrome too?13:02
johnbendertoddmparker_: I did not test in windows chrome. I have to be honest with you there13:03
toddmparker_the reason i ask is that message says it may not work in some browsers, such as chrome13:03
toddmparker_is there a browser we know this won't work in?13:03
johnbendertoddmparker_: it doesn't work in chromium, ajax13:03
toddmparker_or we can just keep it vague13:03
johnbender+113:03
johnbenderass covering13:03
johnbenderit's what we do13:04
toddmparker_mind vaguifying?13:04
toddmparker_we need to remember to add some info to that wiki page13:04
toddmparker_looks great.13:04
toddmparker_after that text change, lock and load13:04
toddmparker_big improvement13:05
Wilto“Sometimes in life some things don’t work.” Vague++13:05
toddmparker_like it13:05
johnbendertoddmparker_: I'll just take out the chrome reference13:06
toddmparker_yep13:06
toddmparker_guys - what's your take on this issue? https://github.com/jquery/jquery-mobile/issues/313613:10
socialhapy★ Issue #3136 on jquery-mobile, reported by kirk911 (6h, 37m ago): preventDefault does not prevent history change during pagebeforechange event13:10
kinblastoddmparker: all of that happens *BEFORE* changePage() gets called I think ... the browser has already decremented it's history stack pointer before it triggers the hashchange13:19
sven_franckkinblas: I was sitting on something similar today. Only thing that worked for me was to prevent the hashChange from firing.13:23
kinblassven_franck: you can't do that if they've hit the back button13:23
kinblasif you don't process, the changePage request, our notion of where we are in the stack will be out of sync with the browser13:24
sven_franckkinblas: similar...13:24
*** Quits: rwaldron ([email protected]) (Read error: Connection reset by peer)13:28
johnbenderah which reminds me13:28
johnbenderkinblas: it seems like everyone is finding he hashchange is slow, are they talking about the event that's triggered by the browser?13:29
johnbendersven_franck: ^13:29
johnbendertoddmparker_: I don't think my key is on the cdn server so we'll have to have dan or eddie push the new zip13:30
*** Quits: erichynds ([email protected]) ()13:30
kinblasjohnbender: sven_franck: depends on the browser13:30
toddmparker_doesn't look like either is around, huh?13:30
toddmparker_mind sending a quick email?13:30
johnbendertoddmparker_: negatory :(13:30
johnbendernot at all13:30
kinblasmost browsers fire a hashchange event, other times, it's a polling timer via our hashchange plugin13:30
sven_francktoddmparker_: I'm here13:30
toddmparker_hey sven_franck13:30
toddmparker_we were looking for danheberden or eddiemonge13:31
sven_franckok13:31
johnbenderkinblas: so the polling should be faster than 300ms eh?13:31
kinblasjohnbender: well, the idea is not to tax the cpu too much13:31
kinblastoo much polling wears the battery down13:31
toddmparker_hey gseguin - are you around tomorrow?13:31
gseguinyes13:32
toddmparker_cool13:32
johnbenderkinblas: that would be a fun benchmark13:32
gseguinwhat is it about? did I just volonteer for something?13:32
toddmparker_i might want to see if you can whip up a branch with toolbars using fixed: CSS13:32
kinblasheh13:32
toddmparker_heh13:32
*** Joins: rwaldron ([email protected])13:32
toddmparker_what else is on your plate?13:33
toddmparker_if you're busy, maybe garann would want to tinker with this13:33
gseguinI was looking into making our code AMD modules and use RequireJS as our build system13:33
* johnbender loves that we have so much help13:33
toddmparker_i just need this wired in so i can do more testing13:33
toddmparker_yeah johnbender13:33
gseguindon't know how much of a priority that is though13:34
toddmparker_ok gseguin - that seems more important13:34
toddmparker_paves the way for the download builder, right?13:34
gseguinyes indeed13:34
*** Joins: jqmobilecommit ([email protected])13:34
jqmobilecommit[jquery-mobile] johnbender pushed 2 new commits to master: http://git.io/_NpCqg13:34
jqmobilecommit[jquery-mobile/master] record enhancement time and pageprocessing, both of which ignore the network load time of the jqm js - John Bender13:34
jqmobilecommit[jquery-mobile/master] remove chrome ref - John Bender13:34
*** Parts: jqmobilecommit ([email protected]) ()13:34
toddmparker_think that's importnat13:34
gseguinallows you to declare dependencies13:34
toddmparker_cool13:34
toddmparker_guess we need to get that dependency map worked out too13:35
gseguinI actually found a project on github we could leverage13:35
toddmparker_yeah?13:35
* gseguin looks for it13:35
gseguinthere it is: https://github.com/arian/amd-packager-php13:37
gseguinSomeone can start looking at that13:37
gseguinthe unfortunate thing is that it still has a dependency on PHP :(13:38
gseguinI was dreaming of a NodeJS service that could do that  but haven't wrapped my head around it13:38
gseguintoddmparker_: are you in tomorrow?13:39
toddmparker_yup gseguin13:39
*** Joins: eddiemonge ([email protected])13:39
eddiemongewhere are docs getting pushed to?13:39
toddmparker_hey eddiemonge13:39
gseguinok I'll ping you in the morning to figure out what I should be spending time on13:39
*** Parts: fcheslack ([email protected]) ()13:39
johnbendereddiemonge: it looked to me like the zip gets push to the cdn in the deploy target13:39
garanntoddmparker_: you need someone to do something with fixed CSS?13:39
toddmparker_they should replace the ones up on the cdn13:40
eddiemongeright13:40
toddmparker_yeah that13:40
toddmparker_hey garann13:40
johnbendereddiemonge: without a full deploy though13:40
johnbenderjust scp the zip13:40
toddmparker_what are you looking at right now?13:40
johnbenderof the docs13:40
garanntoddmparker_: not a thing13:40
toddmparker_cool, fun project then13:40
toddmparker_based on my initial research, I'm feeling pretty good about position:fixed support13:40
toddmparker_I'd like to make a branch where we can play around with modding the fixed toolbar script13:41
toddmparker_to use position fixed CSS instead13:41
*** Quits: DaveMethvin ([email protected]) (Quit: Leaving.)13:41
garannah, cool13:41
toddmparker_once that's in, i can do more testing to see how it goes13:41
eddiemongereplace the jqmdocs.js or the whole package?13:42
eddiemongeor should docs get added to the /latest/ folder?13:42
toddmparker_i need to probably make some test pages with different elements in the toolbars - forms, buttons, etc. to make sure those work ok too13:42
johnbendereddiemonge: I was thinking we could just scp a new version of the zip13:42
johnbenderunless people are downloading it some other way13:42
eddiemongeto /1.0/?13:43
toddmparker_garann - if the initial tests look good, we can think about whether a support test could be added for this and maybe even an option to use the current dynamic re-positioning as a polyfill13:44
johnbendereddiemonge: where do people download the zip from now13:44
toddmparker_but the first step is to just wire this this in place of the JS toolbar logic13:44
johnbenderI'm not sure how to answer your questions other than how I did13:44
eddiemongei dont know. i dont even know if we've published a link to them13:44
johnbendertoddmparker_: ^13:44
toddmparker_what's the Q?13:44
toddmparker_whether this should replace 1.0?13:45
johnbendertoddmparker_: where are people downloading the docs from, or are they supposed to build them for themselves from the repo?13:45
garanntoddmparker_: cool, that sounds straight-forward13:45
toddmparker_the zip is linked from the release13:45
toddmparker_http://jquerymobile.com/blog/2011/11/16/announcing-jquery-mobile-1-0/13:45
toddmparker_"download zip" link near the top13:46
toddmparker_we just need to make sure we're not changing the library code at all.13:46
toddmparker_i.e. no changes from release day should be in the JS13:46
eddiemongehave there been any changes to the docs since then that shouldnt be in there?13:47
toddmparker_nah, that's ok anyway13:47
toddmparker_just don't want the actual mobile 1.0 code to change13:47
johnbendertoddmparker_: eddiemonge: we can branch from the tag and just cherrypick the commits we made to address the issues13:48
johnbenderthen rebuild the zip13:48
toddmparker_docs changes are prolly fine to include, they aren't a big deal13:48
eddiemongewerent the only changes in jqmdocs.js?13:48
johnbendereddiemonge: yes13:48
toddmparker_cool13:48
eddiemongeok then i can rebuild using only that file13:48
johnbenderbut doing a `make docs` will include the latest js eh?13:48
johnbendereddiemonge: ah13:48
johnbenderfair enough13:49
eddiemongeim going to download https://github.com/jquery/jquery-mobile/zipball/1.0 , copy in the current jqmdocs.js and run make deploy from that, with only the zip file being generated13:49
johnbendereddiemonge: make sure to give it a good test just in case13:50
jqm-jenkins-botProject jqm-master build #302: STILL FAILING in 40 min: http://10.59.19.240:8080/job/jqm-master/302/13:50
jqm-jenkins-botcoloma: Update docs/pages/page-dynamic.html13:50
johnbenderI need to look into the test problesm13:50
eddiemongehelps to have the correct makefile as well13:59
danheberdenjohnbender eddiemonge did y'all get the zip on the cdn ok?14:02
johnbenderdanheberden: I think we're getting it sorted now14:03
danheberden \o/14:04
toddmparker_eddiemonge: is latest not working? this page is acting busted now: http://jsbin.com/ofowab/214:04
eddiemonge?ftj @ toddmparker_14:05
bot-ttoddmparker_, Fixing These jQuery: A Guide to Debugging - http://fixingthesejquery.com14:05
eddiemongelol jk14:06
toddmparker_heh14:06
eddiemongedanheberden: do you know how to increase openjdk cache?14:06
eddiemongeOpenJDK Server VM warning: CodeCache is full. Compiler has been disabled. OpenJDK Server VM warning: Try increasing the code cache size using -XX:ReservedCodeCacheSize=14:07
danheberdeneddiemonge isn't that upon running it?14:08
eddiemongeactually i was going to ask johnbender about his perf page14:08
danheberdenlike --maxm=1028m or something14:08
danheberdeni know i have to allocate min/max for the ui build process14:08
danheberdenit's in the build script if you want to take a look14:09
eddiemongein this one https://github.com/jquery/jquery-ui/blob/master/build/build.xml ?14:10
*** Quits: miketaylr ([email protected]) (Quit: miketaylr)14:11
eddiemongetoddmparker_: page works for me now14:11
toddmparker_hmm...me too14:14
toddmparker_cdn freakout?14:14
WiltoPretty good band name.14:16
eddiemongeno, it was a compiler error14:17
WiltoLess good band name.14:17
eddiemongehttp://etiger13.com/jquery.mobile-1.0.zip14:20
eddiemonge^^ who wants to do some quick testing?14:20
eddiemongejohnbender: toddmparker_ ^14:22
toddmparker_i'll test14:22
eddiemongejohnbender: https://github.com/jquery/jquery-mobile/pull/3115/files14:24
socialhapy★ Pull request on jquery-mobile by shemigon (3d, 14h ago): Generates right home links14:24
eddiemongethat can be closed14:24
eddiemongetoddmparker_: is this bug satisfied? https://github.com/jquery/jquery-mobile/issues/311314:24
socialhapy★ Issue #3113 on jquery-mobile, reported by eddiemonge (3d, 19h ago): Docs without a Server don't work (Chrome)14:24
toddmparker_think so, just checking now!14:24
johnbendereddiemonge: on the second one I was waiting until we deployed the zip14:25
johnbendereddiemonge: what does 3115 address?14:25
eddiemongethe same thing we fixed in the jqmdocs with the links14:25
johnbendereddiemonge: yah that definitely can be closed14:26
johnbenderit only addresses "../../" links :/14:26
toddmparker_seems to work great guys14:26
eddiemongegood to push then?14:26
toddmparker_the only think i realize is that we shut off ajax nav, right?14:27
toddmparker_guess we can explain all that on the wiki page14:27
eddiemongethere is a difference between that zip and the one that is up there though: there isnt a separate docs.zip in it anymore14:27
eddiemongetoddmparker_: its only shut off if it isnt supported14:27
toddmparker_ok, so if i drop this into a web server, ajax is on and no message?14:27
eddiemongeyup14:28
toddmparker_ok, cool14:28
toddmparker_go for it14:28
toddmparker_looks great. so much more useful14:28
TehShrike|WorkIs it even supported for new elements to be created and styled by jqm after the page is initialized?14:28
toddmparker_sure, you can use create to enhance new markup added14:29
TehShrike|WorkA lot of people are trying to do it, one of them me, but the docs aren't really any help14:29
toddmparker_or call the plugins directly14:29
TehShrike|Workcreate() is the function to be used?14:29
TehShrike|WorkYou're speaking of using .button() or something?14:29
toddmparker_this is a good docs page to point people to: http://jquerymobile.com/test/docs/pages/page-scripting.html14:29
toddmparker_See "Enhancing new markup" halfway down14:30
TehShrike|WorkThat looks highly relevant!14:30
johnbenderso if anyone is interested14:32
johnbenderthe ci server saw an upgrade to firefox 8 that wasn't my doing14:32
johnbenderwhich honestly makes me nervous14:32
WiltoLike, Skynet-nervous?14:33
toddmparker_does FF auto-update now?14:34
toddmparker_soooo...who wants to be the other Bada test device holder? kinblas, johnbender or gseguin14:34
WiltoI kind of hope it does.14:34
gseguinis bada short for bad ass ?14:35
gseguinjoking aside it would make more sense that full time folks get it14:36
eddiemongeok http://code.jquery.com/mobile/1.0/jquery.mobile-1.0.zip will be refreshed shortly. the file is on the cdn14:36
*** Parts: Wilto ([email protected]/web/irccloud.com/x-ikmdtqrtqvqltwiy) ()14:38
toddmparker_thanks eddiemonge and johnbender for gettignthe zip updated. makes me feel a million times better about the 1.0 release.14:39
toddmparker_so kinblas or johnbender - who's the lucky fella?14:39
kinblastoddmparker_: I think bender only has 2 devices14:40
kinblas:-D14:40
kinblasan android and a wp714:40
toddmparker_he wins!14:40
*** Quits: timmywil ([email protected]) (Read error: Operation timed out)14:42
eddiemongeor loses depending on how you look at it14:44
johnbendertoddmparker_: yah I'm "device lite"14:44
toddmparker_you got it buddy14:47
*** Joins: sven_franck_ ([email protected])14:55
*** Quits: sven_franck ([email protected]) (Ping timeout: 245 seconds)14:56
*** sven_franck_ is now known as sven_franck14:56
*** Quits: eddiemonge ([email protected]) (Quit: eddiemonge)14:59
*** Joins: timmywil ([email protected])15:05
jqm-jenkins-botProject jqm-master build #303: STILL FAILING in 4.2 sec: http://10.59.19.240:8080/job/jqm-master/303/15:07
jqm-jenkins-bot* john.m.bender: record enhancement time and pageprocessing, both of which ignore the network load time of the jqm js15:07
jqm-jenkins-bot* john.m.bender: remove chrome ref15:07
toddmparker_ok, heading home15:22
toddmparker_talk to you guys later15:22
*** Quits: danzik17 ([email protected]) (Quit: Leaving)15:22
jqm-jenkins-botYippie, build fixed!15:22
jqm-jenkins-botProject jqm-master build #304: FIXED in 11 min: http://10.59.19.240:8080/job/jqm-master/304/15:22
johnbenderthere we go15:23
*** Quits: eephillip ([email protected]/supporter/student/eephillip) (Remote host closed the connection)15:42
*** Quits: johnbender ([email protected]) (Ping timeout: 276 seconds)16:22
*** Quits: ryanneufeld ([email protected]) (Remote host closed the connection)16:26
*** Quits: sven_franck ([email protected]) (Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243])16:36
*** Quits: rwaldron ([email protected]) (Quit: Leaving...)16:52
*** Quits: aaronmcadam ([email protected]) (Remote host closed the connection)17:10
*** Parts: gseguin ([email protected]/web/irccloud.com/x-oxeapifyxlupabfm) ()17:25
*** Joins: gseguin ([email protected]/web/irccloud.com/x-oxeapifyxlupabfm)17:25
*** Quits: gseguin ([email protected]/web/irccloud.com/x-oxeapifyxlupabfm) ()17:26
*** Joins: gseguin ([email protected]/web/irccloud.com/x-oldtjoegtmpsufgn)17:26
*** Joins: miketaylr ([email protected])17:32
*** Joins: rwaldron ([email protected])18:07
*** Quits: kinblas ([email protected]) (Quit: kinblas)18:29
*** Joins: eephillip ([email protected]/supporter/student/eephillip)20:04
*** Joins: lukekarrys ([email protected])20:08
*** Parts: lukekarrys ([email protected]) ()20:17
*** Quits: eephillip ([email protected]/supporter/student/eephillip) (Remote host closed the connection)20:28
*** Quits: rwaldron ([email protected]) (Quit: Leaving...)20:30
*** Joins: lukekarrys ([email protected])20:56
*** Quits: lukekarrys ([email protected]) (Client Quit)20:57
*** Quits: miketaylr ([email protected]) (Quit: miketaylr)21:08
*** Joins: DaveMethvin ([email protected])21:22
*** Quits: DaveMethvin ([email protected]) (Quit: Leaving.)21:37
*** Quits: timmywil ([email protected]) ()22:10
*** Joins: johnbender ([email protected])22:34
*** Quits: xjjk ([email protected]) (Remote host closed the connection)22:55
*** Joins: xjjk ([email protected])23:03

Generated by irclog2html.py 2.12.1 by Marius Gedminas - find it at mg.pov.lt!