**** BEGIN LOGGING AT Tue Nov 13 02:59:57 2012 Nov 13 03:00:02 nobody is gonna help with #php Nov 13 03:00:27 i'm trying to get coarse location via the LocationManager, using getLastKnownLocation() and requestLocationUpdates() passing 10 minutes as the minTime and 1km as the minDistance. However in practice when the phone is not moving I don't get any updates at all… Should I simply assume that the last location I've received is still good? Nov 13 03:00:28 php is the shits Nov 13 03:05:18 spifff: yes. Nov 13 03:06:55 ctate: thanks Nov 13 03:10:05 someone suggested a good subreddit on reddit to learn programming Nov 13 03:10:27 someone suggested a good subreddit on reddit to learn programming Nov 13 03:14:13 what Nov 13 03:16:54 ya Nov 13 03:21:37 * offby1 idly wonders if factory80 is a non-native English speaker, and actually meant to say "Could someone please suggest a good subreddit ..." Nov 13 03:27:19 i just like to speak in jibbeish Nov 13 03:27:21 jibberish Nov 13 03:27:23 ;) Nov 13 03:28:40 "gibberish", and I'm standing by my "non-native speaker" hypothesis :) Nov 13 03:31:29 if you wanrt help for php go to fucking #php jackass Nov 13 03:32:12 I have two Intents, and am using those two intents to launch two services. Yet when I call getIntent().getComponent.getClassName(), they both return the same value - of the first Service. Nov 13 03:33:04 Sinofsky leaving MS :O Nov 13 03:33:15 ? Nov 13 03:33:18 who dat Nov 13 03:33:26 is that the guy that authored the book Code Nov 13 03:33:31 What's weirder is that I put a logging statement in the ServiceConnection when the service connects - and when I log componentName.getClassName(), it prints the name for the correct class. Nov 13 03:33:52 Ballmer tossed him because he wanted his job :) Nov 13 03:34:11 Ballmer will be tossed soon and Sinofsky will have his job after all :D Nov 13 03:34:43 go to #nobodycares :-p Nov 13 03:34:57 maslen Nov 13 03:35:10 where are you callin getIntent().getComponent... from? Nov 13 03:35:14 the two different services? Nov 13 03:35:57 yes. Nov 13 03:37:27 When receiving an intent, this is the component that was found to best handle it (that is, yourself) Nov 13 03:37:31 speaking of intents.. who's using their own app intent schema here? ie. myappname://dostuff/123 Nov 13 03:38:00 I know Dianne Hackborne(sp?) would murder me in my sleep for talking of such things Nov 13 03:38:05 speakingcode-wor: I don't think I understand what you're saying Nov 13 03:38:15 but I hear the counterpoint being that it's a requirement in the iOS world so why not android Nov 13 03:38:20 curious why you would need to call getintent.getComponent from a service when it's just going to return the name of that service.. Nov 13 03:38:48 speakingcode-wor: Logging purposes Nov 13 03:38:51 erm, does Service even have getIntent? Nov 13 03:38:57 not sure what you're doing but seems like this.class or somethin Nov 13 03:39:01 Leeds: pretty sure Nov 13 03:39:25 Leeds: It's possible that's the issue :) Nov 13 03:39:32 im not sure, if the code #jdjjjmdj its ok Nov 13 03:39:34 so bye Nov 13 03:39:43 oh, no it doesn't Nov 13 03:39:45 Yup. Nov 13 03:39:47 if anyone can show me where it's referenced on http://developer.android.com/reference/android/app/Service.html ... Nov 13 03:39:48 you're right. Nov 13 03:40:02 that's the problem then Nov 13 03:40:04 you just get passed the inent during onStartCommand Nov 13 03:40:21 so i odn't know where he is calling it from but it's not from his service Nov 13 03:40:44 thanks Leeds, speakingcode-wor Nov 13 03:41:10 what are you trying to log? Nov 13 03:42:18 Trying to find the cause of a NullPointerException, when the data should be populated. Nov 13 03:42:42 why do you need the class name? Nov 13 03:43:26 I was using it for logging Nov 13 03:43:47 we've successfully gone in a circle Nov 13 03:44:47 maslen, have you tried just passing the intent to Log() ... Nov 13 03:44:54 i think it prints all that shit out Nov 13 03:45:11 Log.i(TAG, myIntent) Nov 13 03:47:00 who puts the "sleep" function key on f4 T_T Nov 13 03:47:00 i tried to close a tab, hit fn instead of ctrl... Nov 13 03:47:16 haha Nov 13 03:47:18 lenovo? Nov 13 03:47:23 it's on my laptop too Nov 13 03:47:34 yea a lenovo thinkpad Nov 13 03:47:45 acer aspire one Nov 13 03:47:49 if iwant to go to sleep, i will go to the ubuntu menu and do it that way Nov 13 03:47:58 ubuntu x_x Nov 13 03:48:13 i know, i'd prefer windows, but it's hard to build android on it Nov 13 03:48:23 g00s: I'll try that, thanks Nov 13 03:48:23 windows?!?!?! xx_xx Nov 13 03:48:30 i <3 windows Nov 13 03:48:33 speakingcode-wor: No idea why you were doing that either :) Nov 13 03:48:36 canadiancow: what model did you get? I was going to replace my aging macbook with a thinkpad Nov 13 03:48:49 T430s Nov 13 03:48:52 but i didnt choose it Nov 13 03:48:54 it's for work Nov 13 03:49:07 i had a better one at zynga Nov 13 03:49:12 but at zynga, it was my only computer Nov 13 03:49:13 canadiancow: and I used to respect you... Nov 13 03:49:27 I have a Z580 - $400 for this one - I'm cheap :) Nov 13 03:49:33 heh Nov 13 03:49:40 i have a macbook pro at home for personal use Nov 13 03:49:47 hmm, the 8GB n7 is going for 160 at staples if available Nov 13 03:49:55 yea Nov 13 03:50:01 finding laptops that run ubuntu nicely with an nvidia card is difficult Nov 13 03:50:16 i got a aspire one netbook, got it for $40 and needed a screen replacement, cost me $60 for new screen Nov 13 03:50:38 The only problem is, it's a netbook :( Nov 13 03:50:43 I genuinely don't understand developers working on a laptop Nov 13 03:50:49 what's wrong with that? Nov 13 03:50:57 hmm? i use a laptop too Nov 13 03:51:05 i can take it with me anywhere Nov 13 03:51:09 Leeds: I go to and from an office, there are external monitors at both.. Nov 13 03:51:09 unless you're just docking it Nov 13 03:51:21 Leeds, i just docked it at zynga Nov 13 03:51:28 but it was a very nice portable dev machine Nov 13 03:51:30 code in my bed.. code in a lounge, code in a coffee shop, code at a friends house, Nov 13 03:51:34 my computer lives on my desk - I can remote access anything from anywhere if I need to, but otherwise I like having a couple of screens, a decent mouse and keyboard Nov 13 03:51:37 http://www.lenovo.com/products/us/laptop/thinkpad/t-series/t430s/ <-- lol Nov 13 03:51:58 realistically, I've never felt the need to code while visiting friends - and if I did, I'd use vim-over-ssh on their computer Nov 13 03:52:01 wat, amazon is selling the 8gb n7 for $227? wtf Nov 13 03:52:05 preds, why is that lol? Nov 13 03:52:12 i have a dual monitor setup, but i've found it just causes a lot more eye and head movement Nov 13 03:52:20 g00s: is it Amazon, or a seller on Amazon? Nov 13 03:52:27 canadiancow - Internal Server Error - Read (500) Nov 13 03:52:35 it works for me Nov 13 03:52:35 Leeds: oops, good call didn't see that Nov 13 03:52:43 using a tiling window manager n trying to stay mouseless i'm more productive, big monitor is just more screen to not look at Nov 13 03:53:17 *not* a big monitor, a couple of 19" 4:3 ones - again, widescreen is not for work, it's for watching movies Nov 13 03:53:33 i have a tv for watching movies Nov 13 03:53:56 i have two 21 led's on my desktop, but it mostly sits there Nov 13 03:54:04 oh yeah duh, i guess amazon probably wouldn't be selling that >.> Nov 13 03:54:08 i have two 24" at home and at work Nov 13 03:54:12 when i need to run emulators and vm's and heavy stuff i use it, but most of the time, meh. bels n whistles get in the way Nov 13 03:54:27 i prefer a big ass single monitor than 2 smaller ones :) Nov 13 03:54:46 i think you can get 27" IPS monitors reasonably cheap now Nov 13 03:55:08 a w aouldn't mind a two layer panoramic display tho... like 2x5 Nov 13 03:55:14 hmm.. so swiping doesn't work on HTC live wallpapers Nov 13 03:55:37 I wonder what the reasoning behind that was Nov 13 03:58:56 doens't work how? Nov 13 03:59:28 my guess is there was none (reasoning behind that) Nov 13 03:59:28 i find working on 1 screen really hard Nov 13 03:59:48 how come alex_PP ? Nov 13 04:00:03 i feel squashed Nov 13 04:00:09 i like to have a lot of space Nov 13 04:00:20 differnnt things on each is nice Nov 13 04:00:23 Don't fence me in! Nov 13 04:00:31 one for a terminal or 2, one for eclipse Nov 13 04:00:43 another desktop with email and irc Nov 13 04:00:59 hemls me with mental context switching Nov 13 04:01:01 when i have multiple em's, it's nice to have debugger windows, logcat, console, emulator windows, code editors etc all spread out Nov 13 04:01:03 speakingcode-wor: onOffsetsChanged not getting called on newer devices, onTouchEvent not getting called on old ones Nov 13 04:01:05 *helps Nov 13 04:02:00 I like having a second monitor for a separate task - for example, referencing docs, or running an emulator. Basically, anything I'd have to minimize to do. Nov 13 04:02:22 right now i've got console, two files in vim, and irc on the screen together on my netbook, no issue Nov 13 04:02:53 and i have 9 virtual desktops i can switch between with win+1..9 faster than i can tur Nov 13 04:02:53 * offby1 uses Eclipse on a laptop with a 1024x768 scren Nov 13 04:02:54 screen Nov 13 04:02:56 n my Nov 13 04:02:58 he Nov 13 04:03:01 turn my head* Nov 13 04:03:07 like living in a Tokyo apartment :) Nov 13 04:03:44 offby1: Why would you do that to yourself? Nov 13 04:03:56 maslen: it's the best machine I've got! Nov 13 04:04:07 i'm about to post a screenshot of my desktop Nov 13 04:04:23 widescreens are great for having 2 documents side by side too Nov 13 04:04:36 or an editor and the layout preview side by side Nov 13 04:04:47 Is it possible to put a breakpoint on accessing a variable? Nov 13 04:05:48 http://imgur.com/DIcpk Nov 13 04:07:30 I haven't figured out how to debug in Eclipse, at all. Nov 13 04:07:33 i've been meaning to try a tilling WM Nov 13 04:07:42 look like fun Nov 13 04:07:49 best thing i ever did Nov 13 04:10:16 hi Nov 13 04:10:38 hi Nov 13 04:10:56 what page are using here for paste code? Nov 13 04:11:12 whatever u want. pastebin or gist work good Nov 13 04:11:29 k Nov 13 04:11:30 hi-res projector is great for coding till the bulb burns out ;) Nov 13 04:11:47 anyone use c9.io? Nov 13 04:13:43 ok guys i am starting with android using a book , but i have a error eclipse say me i need set a width and height but is seted : http://pastie.org/5369822 (sorry for my bad english :) ) Nov 13 04:14:19 error: No resource identifier found for attribute 'layout_height' in package ' android' Nov 13 04:15:04 Ugh, I hate stupid mistakes. I seem to have an abundance of them tonight Nov 13 04:16:32 bambanx: did you try saving the file or cleaning the project? Nov 13 04:16:44 sometimes eclipse will get "stuck" with the error after its corrected Nov 13 04:17:05 saving speakingcode-wor Nov 13 04:17:15 Hey guys, anybody have any experience "s-off"ing newer LG phone, I'm working with a locked bootloader and help would be appreciated Nov 13 04:18:54 Calcdsatr1: you might want to try #android-root i think Nov 13 04:19:06 I need to implement a custom URL scheme in my app and I'm just trying to figure out how to best integrate this with the whole task/activity/backstack lifecycle. A single MainActivity is responsible for most of my app, but in a few cases I call other activities (with startActivityForResult, generally). How can I get any additional intents routed to the existing MainActivity instance, without necessarily tearing down activities in the sta Nov 13 04:19:18 Alright, Thanks Nov 13 04:21:20 spiff i'm not sure i'm following but have you looked at launchMode? Nov 13 04:21:44 http://developer.android.com/guide/topics/manifest/activity-element.html#lmode Nov 13 04:23:29 seems like that may be what ur looking for. if you wnat to handle intents _without_ the activity coming into view, you probably nood to look into a service Nov 13 04:24:09 speakingcode-wor: yes i've been looking at that… the problem seems to be that when an Intent comes in, the system decides what's going to happen to any activities that are on top of the stack before my code gets a chance to look at it Nov 13 04:25:19 hi all, been playing with the ndk native activity example. it's working. now i want to try adding a lib dependency by adding LOCAL_LDLIBS += /path/to/libtest.so in Android.mk, it builds but it won't run because it doesn't know where to find libtest.so, what's the proper way to do this? Nov 13 04:25:32 i am exited is my first app on android :) Nov 13 04:25:55 e.g. if i have MainActivity | ViewDetailActivity(123) i.e. ViewDetailActivity being on top of the stack. Now I'd like to have a URL like myapp://returnToDetail/123 but if I register MainActivity for that scheme it seems the system will either (depending on launchmode) create another instance of MainActivity, or tear down the current ViewDetailActivity and then deliver the intent to my existing MainActivity. Nov 13 04:26:18 bambanx: awesome Nov 13 04:26:23 hehe Nov 13 04:26:45 I'd like to be able to look at the intent, and decide that for example the current top activity is already the right one, and just pass the data into the existing instance Nov 13 04:27:27 i cannot understand what i do wrong dude i cannot run the project cos eclipse say me that: error: Error: String types not allowed (at 'layout_width' with value 'wrap parent'). Nov 13 04:27:39 i changed for fill_parent and dont work too Nov 13 04:28:38 yeah wrap parent is wrong Nov 13 04:28:45 do match_parent or wrap_content Nov 13 04:29:06 the diff doesn't matter at this stage of learning, but either of those are valid Nov 13 04:29:47 uh lemme check Nov 13 04:29:54 if you're targetting older version you'll have to do fill_parent instead of match_parent, but that's now deprecated Nov 13 04:32:16 thanks men speakingcode-wor i change fill for match and now is all fine Nov 13 04:33:26 well not at all but now i have only one error Nov 13 04:33:28 :P Nov 13 04:34:12 in this line android:layout_width="fill parent" Nov 13 04:34:20 error: Error: String types not allowed (at 'layout_width' with value 'fill parent'). Nov 13 04:34:26 do fill_parent Nov 13 04:34:29 not fill parent Nov 13 04:34:33 underscore Nov 13 04:35:51 oh hell me eclipse highlight my string so i cannot see my _ lol Nov 13 04:35:53 thanks men Nov 13 04:36:07 eclipse le doo doo Nov 13 04:36:10 np Nov 13 04:36:15 lol Nov 13 04:36:29 ron_frown: guess u use intellij? Nov 13 04:36:39 i hate eclipse Nov 13 04:36:52 i like sublime text but it dont have tools for android Nov 13 04:36:58 intellij Nov 13 04:37:01 but only 12 Nov 13 04:37:13 i tried intellij once, thought it sucked Nov 13 04:39:43 eclipse burn my pc (is old ) Nov 13 04:45:57 it seems I really want launchMode singleTask, but I've had issues with that in the past… this looks interesting as well: http://stackoverflow.com/questions/6575730/notification-to-restore-a-task-rather-than-a-specific-activity Nov 13 04:47:05 effectively the URL scheme handler could be a separate activity that just brings the task into the foreground and communicates with whatever other activities are in it Nov 13 04:47:32 bahaha just started watching this show called Catfish on MTV where they go n expose online pretenders who pretend to be people they arne't Nov 13 04:55:28 what are you trying to do exactly? Nov 13 05:03:29 ron_frown: andy rubin : android :: sinofsky : windows. its like saying Andy Rubin is leaving google Nov 13 05:04:41 g00s I know you are an optomist and optimistic about the future of windows rt and such Nov 13 05:05:04 ron_frown: i'm watching :) Nov 13 05:05:10 but frankly I think the writing is on the wall Nov 13 05:05:15 possibly Nov 13 05:05:53 ms is stereotypically too little too late Nov 13 05:06:04 trying to catch up to a market that has like 4-5yr ahead of time Nov 13 05:06:36 eh Nov 13 05:06:57 google did it to apple, but ms has none of the competing aspects Nov 13 05:07:03 could work out to their advantage. let the space mature n settle a bit, address the common pain points for users n devs, see what works n what doesn't Nov 13 05:07:20 now, BB on the other hand... Nov 13 05:07:30 frankly this is what gonna take Nov 13 05:08:00 open, free to license, major incentives Nov 13 05:08:15 and fantastic dev tools Nov 13 05:08:31 ms actually has incentives and dev tools Nov 13 05:08:35 yeah Nov 13 05:08:45 but but I've heard licensing was pretty rediculous Nov 13 05:10:07 I think nokia builds pretty decent hardware Nov 13 05:10:54 but I'd much much rather deal with meego vs windows rt Nov 13 05:11:25 heh, i remember when i was going to look at meego - and to run it, you had to have an nvidia card Nov 13 05:11:30 (the sim) Nov 13 05:12:39 Heh. Meego. Nov 13 05:12:45 I mean Moblin. Nov 13 05:12:47 I mean Tizen. Nov 13 05:13:44 g00s in the emu maybe Nov 13 05:13:50 but it was just clutter and c++ Nov 13 05:14:17 shrug. in the mobile space, i want android to be the dominant player, i could care less for the others Nov 13 05:14:45 tizen is more misdirected folks trying yet again to make an "OS" out of fucking html Nov 13 05:15:30 i want something more open than android to be the dominant player Nov 13 05:15:35 ms had to do _something_ to modernize windows but i don't see it being dom in mobile in any way, Nov 13 05:15:45 more open than android? Nov 13 05:16:12 sure. meego / tizen were more open. firefoxOS will be more open Nov 13 05:16:27 define more open Nov 13 05:16:36 developed in the open Nov 13 05:17:04 firefox os is html+js Nov 13 05:17:13 I think html+js has come a long long ways Nov 13 05:17:16 dont get me wrong Nov 13 05:17:27 but as a replacement for regular gui applications, I dont agree with that Nov 13 05:17:29 also, fireFoxOS … slim chance on it … but yeah, built by a nonprofit on open standards … at least philosophically, i like that Nov 13 05:17:32 firefox may do xul Nov 13 05:17:33 * jfpoole mehs. Nov 13 05:17:39 firefox browser sucks shit these days Nov 13 05:17:48 I am skeptical they could pull off even a remotely capable os Nov 13 05:17:51 look at chrome Nov 13 05:17:54 Mobile phones are getting faster and faster. A UI written in HTML + JS might be usable on current hardware. Nov 13 05:17:58 google has tons and tons of UI experts... etc Nov 13 05:18:03 I'd say chrome is mostly a failure Nov 13 05:18:06 chromeos Nov 13 05:18:26 conceptually speaking, the web & web stqndards & browser can be seen as an os Nov 13 05:18:37 there's tons of applications there Nov 13 05:19:25 and there's no reason the capabilities can't be extended to a more concrete os. javascript is a language, the limit is in the platform upon which it runs Nov 13 05:20:11 javascript is complete shit Nov 13 05:20:22 I really dont get the recent boom in javascripts reputation Nov 13 05:20:26 vs a REAL programming language Nov 13 05:20:30 lol wtf Nov 13 05:20:37 do u use the internet? he Nov 13 05:20:39 heh Nov 13 05:20:43 I write c/c++/java/c# etc Nov 13 05:20:53 as well as building webapps with the latest html5 and js standards etc Nov 13 05:20:57 javascript is shit. Nov 13 05:21:05 so how is it not a real language? Nov 13 05:21:05 ron_frown: What? JS is turing complete. :-) Nov 13 05:21:32 everybody seems to want to replace it though; dart; coffeescript; compiling to js, etc Nov 13 05:21:42 yeah clearly there is a problem wiht js Nov 13 05:21:44 its the lowest common denominator Nov 13 05:21:50 but imo compiling somethign decent INTO js seems... dumb? Nov 13 05:22:01 js is extremely powerful Nov 13 05:22:03 ron_frown: you have seen that though, right ? Nov 13 05:22:12 functional, prototyping, metaprogramming Nov 13 05:22:15 sadly I think ms had web technolgy 100% right with silverlight... except that it was portrayed as a flash replacement Nov 13 05:22:24 gaaaah Nov 13 05:22:24 vs browser hosted dlr with dom access Nov 13 05:22:27 * g00s hits ron_frown Nov 13 05:22:49 execution of dlr code was like 100x faster Nov 13 05:22:52 which makes sense Nov 13 05:22:56 closures, objects.. Nov 13 05:23:08 speakingcode-wor I can do that in c# Nov 13 05:23:10 next? Nov 13 05:23:14 ron_frown: curious on your thoughts on say, lisp, or scala Nov 13 05:23:26 yeah, and you consider c# real, so what's wrong with js? Nov 13 05:23:30 I like scala... though honestly I dont have much use for it Nov 13 05:23:49 so why is scala good/real but not js? Nov 13 05:24:03 I dont know enough abotu scala to answer that Nov 13 05:24:13 you know enough to say you like it Nov 13 05:24:21 so? Nov 13 05:24:31 I cant even pinpoint what pisses me off so much about js =) Nov 13 05:24:54 just seems like an ungrounded hatred, almsot naive, no offense Nov 13 05:24:57 other than slow, obfuscated vs compiled Nov 13 05:25:09 its not naivety Nov 13 05:25:28 I'm no language scholar Nov 13 05:25:31 slow is a metric of the execution environment, not the language Nov 13 05:25:56 slow vs native or dlr/jvm runtimes Nov 13 05:26:04 sure you can keep throwing more and more and more hardware Nov 13 05:26:09 and it'll speed up the js Nov 13 05:26:12 buuuuut Nov 13 05:26:19 no i'm not talking about hardware Nov 13 05:27:35 talkin about the exectuion engine Nov 13 05:27:36 I spoke to a very intelligent spanish gentleman who was a very big advocate of js... very knowlegeable Nov 13 05:27:39 open minded Nov 13 05:27:50 and he didnt change my mind Nov 13 05:28:38 I tried to find all the shit I disliked with js, and have him explain if it was just a knowlege shortcoming or if there were better ways to achieve x Nov 13 05:28:53 sure, being dynamic and loosely type and interpreted, it can't really be compiled to get the same performance optimizations, but the trade off is the power of a dynamic, loose langauge Nov 13 05:29:12 and that power lies.... Nov 13 05:29:12 ? Nov 13 05:29:42 nope it tells the truth Nov 13 05:29:51 ? Nov 13 05:30:02 playing on your words ;-) Nov 13 05:31:00 look I didnt mean to start an argument about this, and honestly I am all for an open platform for apps etc Nov 13 05:31:08 its all good Nov 13 05:31:09 I like html5, I like css (mostly) Nov 13 05:31:28 I just very much dislike js... and honestly I think that part should be negligible Nov 13 05:31:35 I think language really shouldnt matter Nov 13 05:32:51 have you ever programmed in c? Nov 13 05:32:55 yep Nov 13 05:33:11 several times a week? Nov 13 05:33:20 ok. so can you do OO in c? Nov 13 05:33:40 Yes, but it's a pain Nov 13 05:33:43 haha you 'can', but it's not pretty Nov 13 05:33:48 see glib, gtk Nov 13 05:33:54 exactly. you can, but the langauge doesn't lend itself to it Nov 13 05:34:01 exactly Nov 13 05:34:19 I honestly just use c++ (and not half assed c++ as much as I can) Nov 13 05:34:22 js lends itself to many things; i'll just put it like that Nov 13 05:34:40 not that I am condoning it, but so does java for that matter Nov 13 05:34:44 and its proved that Nov 13 05:34:59 you can do most things in any language, more or less Nov 13 05:35:06 PYTHON! /endthread Nov 13 05:35:20 but how concise and straight forward it will be Nov 13 05:35:45 hahah Nov 13 05:35:48 definitely not concise Nov 13 05:36:06 java is the engineers language Nov 13 05:36:12 or rather the overengineers Nov 13 05:37:07 java can be verbose but not necessarily Nov 13 05:37:18 I know its more the actual implementation Nov 13 05:37:45 the explicitness of types and such can actually save a lot of syntax Nov 13 05:40:06 compile-time checking, etc. no need to check a bunch of shit in runtime Nov 13 05:41:30 Are there any resources for android phone hardware development? Like say, building a device that would be capable of running android? Nov 13 05:41:49 there's a platform develpment kit Nov 13 05:42:03 android pdk Nov 13 05:42:12 does anyone know whether the app encryption feature added in JellyBean has the server send down an encrypted APK? Or, does the device encrypt it once it is sent down unencrypted? Nov 13 05:49:23 jbermudes: another reason why we need something more open; we need an open phone : hw and sw Nov 13 05:50:25 not sure something will get the market success n commercial support being totally open, i dunno, maybe so Nov 13 05:53:12 I was thinking of trying to figure out the feasibility of transplanting another phone's guts into an HTC Dream/G1, so I was wondering how compatible the various parts might be... the keyboard, screen, etc. Nov 13 05:53:44 sounds like a lot of work Nov 13 05:54:24 pdk would probably be ur start but i suppose mfrs can deviate n add on their own stuff as much as they want to Nov 13 05:54:32 oh, but it would be so worth it! The G1 has the best physical keyboard! heh Nov 13 05:54:57 g00s: do u use a fully open hardware platform for ur pc? Nov 13 05:55:26 n do use only open software n drivers? Nov 13 05:55:43 no. its an ideal Nov 13 05:56:27 even if the ideal is unachievable, its worthy of consideration as a benchmark to what could be but isnt Nov 13 05:56:45 i agree, just curious Nov 13 06:38:10 is there a difference between oncreate and onstart? i know that onstart is called after oncreate but for example if I had variable++; in the last line of oncreate wouldn't it be the same as putting it first in onstart? or is there difference between the super.onstart? Nov 13 06:39:02 nvmd found my answer Nov 13 07:25:01 Hey guys Nov 13 07:27:24 Hm Nov 13 07:28:29 . Nov 13 07:31:31 :) Nov 13 07:35:19 What cool projects are you guys working on? Nov 13 07:36:09 cleaning up the refrigerator Nov 13 07:36:13 cool stuff Nov 13 07:43:21 Tezt Nov 13 08:00:49 hey guys... can android apps listen on sockets? Nov 13 08:05:43 socket() call is failing :| Nov 13 08:06:32 socket(AF_INET,SOCK_STREAM,0); :P Nov 13 08:07:11 =) ah Nov 13 08:29:28 How to translate a view and its click area? Is there any proper docs on how to do this? I could apply translate animation to a view, but it does not move the click area. Nov 13 08:29:58 napster: If you use the android.animation package, it does Nov 13 08:31:47 cyrilmottier: what is it? Nov 13 08:32:07 http://developer.android.com/reference/android/animation/package-summary.html Nov 13 08:32:57 oh got it, I'm using view.animation Nov 13 08:33:12 let me try animation package. ty cyrilmottier Nov 13 08:33:25 napster: it will work, I'm sure of it ;) Nov 13 08:33:45 ok Nov 13 08:38:58 cyrilmottier: for your new map library, did you ever consider osm instead ? Nov 13 08:39:16 g00s: nope :( Nov 13 08:39:37 did they say during i/o the map library was going to be updated ? Nov 13 08:40:16 g00s: Yes they said "we are working on it" Nov 13 08:40:23 hehe Nov 13 08:43:00 * p_l|work ponders the evil idea of ZFS on Android Nov 13 08:44:34 p_l|work: i like the idea of a 20tb cellphone Nov 13 08:45:33 lets write an ipad emulator and copy all apple content to the device! Nov 13 08:45:33 g00s: :) Nov 13 08:45:40 zfs will make a lot of sense Nov 13 08:51:38 When there's an incoming call, the class com.android.phone.PhoneApp is called, right? Anyone here knows the specific part of the code where the caller info (phone number and contact name) is send to a Bluetooth handsfree device display/LCD? Nov 13 09:14:47 is the aapt tool a java thing or a linux binary? Nov 13 09:16:11 mitch0: ever used the 'file' command? Nov 13 09:17:09 I don't have it installed, I'm a freebsd, and if it's a binary, I won't bother with it... Nov 13 09:17:36 you're a freebsd, and I'm a linux? Nov 13 09:17:49 erm Nov 13 09:17:54 I'm on freebsd ;) Nov 13 09:19:04 bash-4.2$ file aapt Nov 13 09:19:05 aapt: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), stripped Nov 13 09:19:23 thanks. too bad. will do this at home then Nov 13 09:20:03 just wanted to change the long-press-back-to-kill-app timeout :( Nov 13 09:20:25 should be configurable from some advanced settings tbh... Nov 13 09:31:23 Hi, how can my dump my mmcblk partitions so that could be recovered by fastboot flash? Nov 13 09:31:33 s/my/I/ Nov 13 09:32:31 hi all Nov 13 10:05:37 why is my mutex dead locking :S Nov 13 10:07:49 it's like, so dead :S Nov 13 10:08:13 I *know* no other thread locked that mutex... yet it just keeps waiting there at a pthread_lock :S Nov 13 10:08:27 *pthread_mutex_lock Nov 13 10:13:25 wow Nov 13 10:13:38 so many quits (.split) Nov 13 10:13:46 what just happened? Nov 13 10:14:42 -mrmist- [GLOBAL NOTICE] - In a few minutes there will be some network disruption whilst we sort out some internal routing issues. It shouldn't take too long, but could be a bit noisy. Thank you for your patience. Nov 13 10:17:18 hello. I am creating a dialog that is set to an xml I created. Inside the xml I have some buttons and a tablelayout that grows dynamically (by adding more rows to it). is there any way to set the maximum height of the dialog? I can't find that anywhere Nov 13 10:23:41 hello. I am creating a dialog that is set to an xml I created. Inside the xml I have some buttons and a tablelayout that grows dynamically (by adding more rows to it). is there any way to set the maximum height of the dialog? I can't find that anywhere Nov 13 10:24:51 hi Nov 13 10:24:55 hi Nov 13 10:25:20 i am i on the right place - questions about android development ? Nov 13 10:25:28 yes Nov 13 10:25:36 cool Nov 13 10:26:39 if i define a single button under the merge root, can i append additional layout parameters not defined inside that merge layout? for instance, all buttons in my layouts have width 5. if i use include on a merge which contains a single element button with the attribute layout_width 5, can i then set the text with the include tag? Nov 13 10:26:56 is it better to have an asynctask for each screen of my app or only one seperate asynctask class that handles the processing of all the screens on my application ? Nov 13 10:28:12 hello Nov 13 10:28:16 hi Nov 13 10:28:19 hi Nov 13 10:28:20 hexadecimal_: that's a wierd question Nov 13 10:28:26 what does that asynctask do? Nov 13 10:28:45 @hexadecimal_, the first one, since all screens should be activities and it makes more sense to keep all activity related code together as much as possible Nov 13 10:28:48 for sanity's sake. Nov 13 10:28:48 i would like some help may i ask here? its about the samsung s5830i not connecting Nov 13 10:29:22 it does inserts/updates and select sql statements on the sql lite db Nov 13 10:30:41 @hexadecimal_: if its specific to the activity, i would keep it local to it for readability. avoid extra longitudinal checks you have to make to ensure correctness Nov 13 10:31:04 hello. I am creating a dialog that is set to an xml I created. Inside the xml I have some buttons and a tablelayout that grows dynamically (by adding more rows to it). is there any way to set the maximum height of the dialog? I can't find that anywhere Nov 13 10:31:08 unless its the same thing over and over Nov 13 10:31:16 then just extend asynctask and use your custom Nov 13 10:31:40 er disregard that. Nov 13 10:31:52 thx! Nov 13 10:32:44 if i define a single button under the merge root, can i append additional layout parameters not defined inside that merge layout? for instance, all buttons in my layouts have width 5. if i use include on a merge which contains a single element button with the attribute layout_width 5, can i then set the text with the include tag? Nov 13 10:32:51 i was thinking it should also be good for performance not havnig one looong asynctask class... Nov 13 10:33:39 Guys, why isn't pthread_mutex_lock ever coming out? the mutex is 'not' locked by any other thread :( (NDK) Nov 13 10:33:54 gdb it? Nov 13 10:36:22 ctate> re: appt performance. K9 mail ( https://github.com/k9mail/k-9/wiki/BuildingK9 ) takes 2 minutes 6 seconds to clean+export on my system with original AAPT, it takes 23 seconds on my system with modified AAPT. Not the same performance increase as I have on some of my own projects, but clearly noticable. (just a clean goes down from 32 to 8 seconds) Nov 13 10:36:23 I'm not really set up with gdb for Android yet :| Nov 13 10:36:33 but I'm LOGV'ing the heck out of it Nov 13 10:36:40 ctate> you're probably asleep now... ping me later if you get this or I'll just post it again :P Nov 13 10:41:16 is there a way to remove all these extra messages ? Nov 13 10:46:36 so for the record, the answer to my question is no Nov 13 10:46:40 the question was: Nov 13 10:46:44 if i define a single button under the merge root, can i append additional layout parameters not defined inside that merge layout? for instance, all buttons in my layouts have width 5. if i use include on a merge which contains a single element button with the attribute layout_width 5, can i then set the text with the include tag? Nov 13 10:47:14 it doesnt work because the layout in which the include tag is featured doesnt know anything about the included layout Nov 13 10:47:58 My message got lost in the chat. I am creating a dialog that is set to an xml I created. Inside the xml I have some buttons and a tablelayout that grows dynamically (by adding more rows to it). is there any way to set the maximum height of the dialog? I can't find that anywhere Nov 13 10:49:46 dioHere: not familiar with anything like that. what happens if it adds more rows than can be seen on screen? are you going to use scroll? why not simply limit the number of rows so it is independent of screen size/density Nov 13 10:51:35 greatjustice I cannot limit the rows, because the tablelayout shows the bidding for a multiplayer game. So the tablelayout can grow a lot.. Yes I want it to be scrolled, but in order to be scrolled I have to set a maximum height in order for the scrolling to work Nov 13 10:51:57 when adding more rows, the dialog grows up too Nov 13 10:52:56 I don't want the dialog to grow bigger than a certain point in the screen. That's where I want it to start scrolling Nov 13 10:53:25 maxHeight? Nov 13 10:53:40 monsti I cannot find a maxHeight property for the dialog Nov 13 10:53:53 yeah but for the layout inside of the dialog Nov 13 10:54:00 dioHere: dont know anything about tablelayout or scrollview (lol) but can't you just set the dialog parent view to a size defined in dimens and use that value as the maximum height for scrolling to work Nov 13 10:54:00 that's your code ;) Nov 13 10:54:15 greatjustice: yeah that works great, too Nov 13 10:54:44 i wrote some helper to determine the tablet size and now i can have a dialog e.g. "30% of the physical height/width" Nov 13 10:54:54 in dimens.xml, choose a dp maxHeight. Limit the inflated view for the Dialog to that maxHeight and use it whever you need it Nov 13 10:55:10 yeah user /dev/brain Nov 13 10:55:12 -r Nov 13 10:55:44 monsti greatjustice I will try the things you said. thank you very much Nov 13 10:56:10 yw take care Nov 13 10:56:59 why does PDF suck so hard? Nov 13 10:57:17 wasnt designed for mobile devices Nov 13 10:57:29 Does anybody know when there are new n4 available? Had no luck... Nov 13 10:57:52 PDF was designed for faithful representation. Coupled with sometimes shitty source material for said PDF... Nov 13 10:58:01 lasserix: yeah that's the point Nov 13 10:58:28 guys what do you suggest for creating simple online reports on a tablet? Nov 13 10:58:29 is there someone here who know why Intent Action.SEND do not let you share image+url+text on facebook? Nov 13 10:58:37 currently i am using a text template engine to create html Nov 13 10:59:06 Hello folks, when using a dialog fragment with a list. is it possible to insert a Html.fromHtml() object into the list? Nov 13 10:59:14 I'm looking for a reason behind this choise: is it a google decision? is it a facebook decision? etc... Nov 13 10:59:34 mastro probably fb decision, don't they have some api you are supposed to use? Nov 13 11:01:01 eish - this android emulator loads forever Nov 13 11:01:03 Hi there :) Nov 13 11:01:38 hexadecimal_, funny you should say that, i´ve been waiting for 20 minutes till my emulator boots up.. i just see ´Android´ .. using 4.1.2 Nov 13 11:01:50 was wondering if i´m doing something wrong -.- Nov 13 11:01:54 lol - me too Nov 13 11:02:22 WouterNL: i am using virtual box with an x86 version of android Nov 13 11:02:29 hold - i'll send you a link Nov 13 11:02:46 oh, thats clever actually Nov 13 11:04:10 androvm.org Nov 13 11:04:26 my emulator boots in 20 seconds... Nov 13 11:04:30 ready to use including libhudini a arm emulator for x86 machines Nov 13 11:04:55 greatjustice: mine in 7 seconds :) Nov 13 11:04:59 just tested Nov 13 11:05:05 o.O Nov 13 11:05:33 hm, awsome Nov 13 11:05:43 wich realease should i get / Nov 13 11:05:45 WouterNL: you need to connect via adb tcp connect - are you familiar with that? Nov 13 11:06:01 androvm.org/blog/download Nov 13 11:06:24 the "t" is the tablet, the "p" is the phone, the "tp" is tablet+phone Nov 13 11:06:37 gapps & Houdini = ? Nov 13 11:06:50 and the vmwplayer archives is the bundled version with the vmbox included? Nov 13 11:06:50 tp googleapps + houdini Nov 13 11:07:03 you need to install virtualbox and import the appliance Nov 13 11:07:14 what are the AndroVMplayer archives then? Nov 13 11:07:15 that's the "ova" file Nov 13 11:07:26 something for vmware i don't use it Nov 13 11:07:49 hm i have vmware player installed Nov 13 11:07:57 then have fun trying :) Nov 13 11:08:08 can´t hurt =p Nov 13 11:08:14 currently there are different vmware bundles depending on your host os for the opengl acceleration Nov 13 11:08:27 that's pre-beta-testing-eats-your-hamster Nov 13 11:08:42 'ello Nov 13 11:08:43 right, virtualbox it is =p Nov 13 11:09:19 WouterNL: we use androvm for development and for presentations for custumers via laptop/beamer Nov 13 11:09:37 very stable then i assume ^^ Nov 13 11:09:43 adb tcp is to connect eclipse to the vm ? Nov 13 11:09:49 Hello folks, when using a dialog fragment with a list. is it possible to insert a Spanned (Html.fromHtml) object into the list? Nov 13 11:09:51 much better then the google emulator :) Nov 13 11:10:00 WouterNL: one second Nov 13 11:10:05 I have to implement a pattern where I need to show a number of images (2 of them in each row). But I have to do this inside a parent scrollview. But as per the guidlines and best practices I see that its not good to insert a gridView inside a scrollview. Is there any other suggested way of doing this? Nov 13 11:10:41 WouterNL: boot the virtual machine, detect the IP adress of the machine Nov 13 11:10:41 napster - if you have to nest a gridview inside of a scrollview then you've done something wrong Nov 13 11:10:45 as a gridview IS a scrollview Nov 13 11:10:51 WouterNL: then do at the cmd line a - C:\Android\android-sdk-windows\platform-tools\adb.exe connect 10.2.3.101:5555 Nov 13 11:11:08 if you really HAVE to have a scrollview and can't just use a gridview, then I'd suggest just using two side-by-side imageviews Nov 13 11:11:18 set their layout size dynamically Nov 13 11:11:29 Tee_Pee: Yes I understand. But in my requirement, I need to scroll this set of images and a lot of other widgets along. Nov 13 11:11:32 WouterNL: in my case 10.2.3.101 is the IP of the virtual machine Nov 13 11:11:53 monsti and then the sdk manager will see it as a usable test-environment ? Nov 13 11:12:07 WouterNL: yeah ;) and the device is "rooted" as the emulator Nov 13 11:12:17 I do something similar with my gallery app - I have a minimum 200px width for the image. I then calculate how many of those can fit onto the screen (screenWidth/minWidth), and see how much spare pixels I got Nov 13 11:12:21 Tee_Pee: Using two side by side images are fine, but there are unknown number of images, so I would like to use some kind of adpaters. Nov 13 11:12:24 sweet, let me try all that :) Nov 13 11:12:50 I then scale the imageviews to minwidth + (screenWidth - count*minWidth) / count Nov 13 11:13:10 or "the remaining space divided by the number of imageviews" Nov 13 11:13:20 so I get the imageviews to always fill up the screen Nov 13 11:13:55 Tee_Pee: no no, I mean there are X number of images, and I need to show them in two columns which can be scrolled if there are more images that would fit into the screen vertically. Nov 13 11:14:00 yay another network connection, i now have wifi, lan, 3g, tunnel vmware wifi, tunnel vmware lan, tunnel 3g, tunnel virtualbox lan, wifi and 3g -_-¨ Nov 13 11:14:01 no hirzontal scroll is required Nov 13 11:14:04 well you can use an adapter to fill up the scrollview as normal, it's just how you display the content that's the issue Nov 13 11:14:29 ok Nov 13 11:15:14 Here's an interesting issue I am trying to work around right now - I have a slideshow gallery (essentially a viewpager with a huge number of fragments in it) Nov 13 11:15:24 I only keep 3 (previous, current, next) fragments in memory, that's fine. Nov 13 11:15:42 However, as each fragment gets created, it querries the server for the image. Nov 13 11:15:48 via an asynctask Nov 13 11:16:05 The problem arises if the user is rapidly swiping from one to the other Nov 13 11:16:10 use loaders Nov 13 11:16:15 :) Nov 13 11:16:42 loaders aren't the issue.... the issue is getting 10-ish async tasks downloading quite big bitmaps at the same time Nov 13 11:16:45 flooding the memory Nov 13 11:17:48 I believed that simply canceling the task in the adapter's destroyItem() method would suffice Nov 13 11:17:54 monsti, well my hats are off for you, this emulator works like a freaking lightning bolt Nov 13 11:18:03 but while that did solve the memory overload, it also made the async tasks go haywire Nov 13 11:18:39 what I think about doing is adding a 200ms delay on the start of the async task, so if the user rapidly swipes through they images don't load at all... Nov 13 11:20:29 you shouldn't be using an AsyncTask for things like that anyways Nov 13 11:20:43 a ThreadPoolExecutor with reasonable limits is probably much better Nov 13 11:22:08 guys any clue about my pthread_mutex_lock issue? lock() doesn't return, trylock says EBUSY but it's not locked anywhere else :S Nov 13 11:22:10 back it by a priority queue you can reorder downloads as the user swipes Nov 13 11:22:40 how do i mess with onClick() events of buttons in a layout XML from a Service Nov 13 11:22:56 when I printout the internal 'value' of the pthread_mutex_t structure I get different values between the unlock and the lock... Is that normal? Nov 13 11:23:09 Can the value for the same mutex actually change before it's destroyed :S? Nov 13 11:24:01 JakeWharton - I run the AsyncTasks like task.executeOnExecutor(ImageTask.THREAD_POOL_EXECUTOR); Nov 13 11:24:05 lasserix, as you probably know android can choose which applications are able to handle an intent. And that's the way to PROPERLY develop on Android. So that if I say "send this" Android will let the user choose which application... Facebook, Twitter, Google+, Pinterest, Whatever_app_support_that_intent_would_come_up_in_the_future --- starting to directly use an API means you drop that feature of the Android system Nov 13 11:24:11 I assume you meant that when you said thread pool executor? Nov 13 11:24:22 lasserix, furthermore there's no Twitter API for Android (for example) Nov 13 11:25:12 has anyone worked with custom Notifications? how to act upon clicks of Buttons in the Notification (layout XML + RemoteViews) from a Service Nov 13 11:26:55 no, I meant ThreadPoolExecutor the class Nov 13 11:26:57 which make's the facebook decision to not properly support that intent a very poor and stupid decision. I was looking for an official declaration about this or some unofficial-but-from-facebook statement about this stuff Nov 13 11:27:36 AsyncTasks are useful for hitting the SD card or maybe making a one-off HTTP request but for things like downloading images and/or making multiple HTTP requests they break down really quickly Nov 13 11:28:15 how the heck do pthread_mutex_t 's value work :S Nov 13 11:28:22 I get the same value for all mutexes created Nov 13 11:29:36 "for things like downloading images and/or making multiple HTTP requests they break down really quickly" and what is you advice? Nov 13 11:29:48 guava? Nov 13 11:30:08 Oh I guess that's the attributes :S Nov 13 11:31:39 mastro: oh yeah I know Nov 13 11:31:54 ah ha! finally have a legitimate use of the sliding drawer: app widget configuration optiosn Nov 13 11:32:08 lasserix, :) if you happen to find some statement from facebook about this I'll be glad if you share it with me Nov 13 11:32:55 no I just see so posts about problems with the fb api, so I am assuming they want devs to use their api since their whole strategy is intractable homogenization of the internet Nov 13 11:33:25 JakeWharton: I d/l'd the new ABS but after importing I get a whole host of @Ovveride annotation errors, is there some setting I missed to fix that do you know? Nov 13 11:33:36 change your JDK to 1.6 Nov 13 11:33:44 under java compiler settings Nov 13 11:33:48 JDK compliance Nov 13 11:34:13 Ahh thanks! Nov 13 11:34:56 how can i make so that if my Notification is tapped, something happens in the background & the Notification is updated Nov 13 11:35:05 pthread_mutex_lock ? not working? Nov 13 11:35:12 broadcast receiver + service Nov 13 11:35:17 regedit: ^^^ Nov 13 11:36:17 JakeWharton: yes i have a foreground service (that's why i need my ongoing Notification) but how is the Notification rigged to act upon interactions in such a way Nov 13 11:37:15 I just connected my androvm to adb.. but now how do i get eclipse to use it as a test platform ? Nov 13 11:37:38 you specify a pending intent when you create it Nov 13 11:37:47 when the user clicks on the notification, the pending intent is fired Nov 13 11:37:57 JakeWharton: if the Notification had a button or 2 for example, how can those be manipulated/responded to from my Service Nov 13 11:38:21 they're all remote views which means they can only fire pending intents when clicked Nov 13 11:38:38 I have one more qustion. I have one-column gridview. With Imageviews. I have how i can stretch imageview (w=100px h=50px)to fill all column width with saving bounds. When i set AdjustViewBounds for imageView + android:stretchMode="columnWidth" - no result. If i set only android:stretchMode="columnWidth" = image streched by width and fill all column but height not stretched and View looks... Nov 13 11:38:40 ...bad. How can I force stretch image in both ways? Nov 13 11:39:09 JakeWharton: so how do i make different buttons fire different pendingintents as you describe Nov 13 11:39:21 Yauhen90: subclass ImageView and override onMeasure to force the height to be an aspect ratio of the measured width Nov 13 11:39:51 Can i set marque efect to text in notification (cistom view) Nov 13 11:40:30 WouterNLD: you have your vbox up and running? Nov 13 11:40:39 monsti yes, but when i try adb connect Nov 13 11:40:41 Will inJustDecodeBounds work with decodeStream(MyBufferedInputStream)? Nov 13 11:40:44 JakeWharton: my remoteviews layout is XML, how can my service set up the XML-defined buttons to act upon interaction (send pendingintents) Nov 13 11:40:45 i.e. an image from the internet Nov 13 11:40:48 it says unable to connect to 10.0.3.15:5555 Nov 13 11:40:53 set the pending intents in code Nov 13 11:41:08 WouterNLDwo yeah you might have the 2nd adapter set to "NAT" Nov 13 11:41:09 JakeWharton: thanks! Nov 13 11:41:12 when you're creating the notification Nov 13 11:41:20 you have to set it to internal or hostonly and then reboot the machine Nov 13 11:41:26 okay Nov 13 11:41:31 or even bridged Nov 13 11:41:55 JakeWharton: setOnClickPendingIntent ? Nov 13 11:41:56 anyone ever have their app download a zip and unzip? if so do u unzip to sdcard or some other place? Nov 13 11:41:56 Yauhen90: this might help get you started: https://gist.github.com/2856179 Nov 13 11:42:17 yours can be much simpler, though Nov 13 11:42:46 Hello everybody. Anyone knows if I can get the phone to supply 5v over USB to my Android Open Accesory device? From my understanding the AOA-device is host in the begining and some handshaking going on between the phone and the AOA-device. After that the AOA-device is set to client and the phone to host. When the phone is host can I gnd the OTG pin (middle usb pin) and get 5v from the phone? Or do I loose the previous connectio Nov 13 11:43:40 when should i create an asynctask - why can't i just always use SQLiteOpenHelper directly ? Nov 13 11:43:53 monsti bridged works :) Nov 13 11:44:09 for accessing the local sql lite db. Nov 13 11:45:18 anyone ever have their app download a zip and unzip? if so do u unzip to sdcard or some other place? Nov 13 11:45:39 hexadecimal: usually you want to push intensive operations to the background so the ui thread doesn't idle while you are doing your query Nov 13 11:46:32 hexadecimal: ie loading a long listview but you can use async loader or cursor loader (requires content provider) Nov 13 11:47:34 guys? How could I debug this pthread mutex on Android? :S Nov 13 11:49:02 thx, i'm just doing normal selects,inserts and updates and displaying them to the user - thats not intensive right? Nov 13 11:50:12 anyone know how big Internal Storage usually is? Nov 13 11:51:06 hexadecimal ive done without offloading same, but i wonder if should wrap it in a thread just in case ;p Nov 13 11:51:59 Gaz`: like... between 1G and 64G :P Nov 13 11:52:37 ahh, cool, thanks, this app has to get zips containing json :-{ Nov 13 11:52:44 hm, seems like my gf's x10 mini only has 128 MB internal storage Nov 13 11:52:57 and i get zips of jpegs.. Nov 13 11:53:01 prob under 100 meg tho Nov 13 11:53:15 even if it's under 100 MB, said 128 MB are already used by other apps Nov 13 11:53:29 good point, i guess then i can divert to SD Nov 13 11:53:41 I mean, that's really a low-end phone, but if you want to include it in your target group, you better save on external Nov 13 11:54:02 yeh i wonder if i should just go for external straight away Nov 13 11:56:33 how do i set a custom layout (subclassed relativelayout) to be the root view for an xml layout? Nov 13 11:56:49 omfg - there is no 17" macbook pro? Nov 13 11:58:46 JakeWharton: do i need PendingIntent.getService() or .getBroadcast() ? i want the button to cause some execution in my running Service (as well as update the Notification) Nov 13 12:05:47 JakeWharton: found this http://stackoverflow.com/questions/7308298/should-i-use-pendingintent-getservice-or-getbroadcast-with-alarmmanager thanks Nov 13 12:10:22 What's the slide-from-the-side drawer thing? Nov 13 12:12:10 Why is pthread_mutex_lock broken? :( Nov 13 12:12:11 Anyone have the link to the side-panel slider on github? Nov 13 12:17:12 lasserix: https://github.com/SimonVT/android-menudrawer Nov 13 12:27:54 JakeWharton thanks, I have the perfect use: holding the list of app widget configuration options so I can put the preview in the content Nov 13 12:55:36 JackWharton, thanks again. Could you give me a hint on how to make the drawer always stick out some amount? Nov 13 13:15:39 If I'm loading images into a listview from my app package, how best to do this asyncronously and keep smooth scrolling Nov 13 13:17:53 Estel, unfurtunately there's no build in solution nor library that does that I know about Nov 13 13:18:16 Estel, it's not an easy task either because you need handle view recycle and memory issues Nov 13 13:18:28 without having much control on when an image is recycled Nov 13 13:19:53 the "official" example you find around is broken: it use AsyncTask, which is a very bad idea, and only cache in memory... it should cache on the filesystem instead Nov 13 13:20:07 this might be a good place where to start: http://developer.android.com/training/displaying-bitmaps/cache-bitmap.html Nov 13 13:20:15 (and in memory) Nov 13 13:20:28 Devastator-, Estel yest, THAT's the official example I'm talking about Nov 13 13:20:47 mastro: I think you can use the LRUCache over a file based cache. Nov 13 13:21:57 DeviantPeer, sure you can, but that doesn't solve the issue of getting an image in background and attaching it to a view that could be recycled and, at the same time, avoiding getting decode errors due to memory exception hidden by the underling decoding mechanism Nov 13 13:22:03 the link I sent actually has a section about it. Nov 13 13:22:40 mastro: well.. that's a vyper nest you dealing with... ;) Nov 13 13:22:46 DeviantPeer, that's not a practical solution, retain fragment Nov 13 13:23:05 hey DeviantPeer got the ellipsizing on the textview app widget to work, thanks again for the help Nov 13 13:23:09 Isn't caching on the filesystem redundant if my files are already on the file system? Nov 13 13:23:12 DeviantPeer, I mean: it's only suited to a small set of needs Nov 13 13:23:25 lasserix: u welcome. Nov 13 13:23:36 Estel, if your files are on the filesystem why don't you put them as reasources instead? Nov 13 13:23:56 mastro: every problem has it's own particular solution. Nov 13 13:24:05 Estel, and if they are from "external user activities" then you obiously do not need a caching :) Nov 13 13:24:12 not the cache on filesystem at least Nov 13 13:24:26 Estel: depends. If you decode the images and resize them for the dimensions you'll use in the app then caching in the storage makes sense. Nov 13 13:24:36 DeviantPeer, and the Android platform do not really help in that matter :) Nov 13 13:25:17 mastro: well... I don't see a lib for that being created. What are you waiting for? ;) Nov 13 13:25:22 just kiding. Nov 13 13:25:33 mastro, http://negativeprobability.blogspot.fr/2011/08/lazy-loading-of-images-in-listview.html Nov 13 13:25:40 DeviantPeer, I have a library for that Nov 13 13:25:57 open? like apache license or something? Nov 13 13:25:58 I adapted something from there, which also cache in the filesystem for more persistence Nov 13 13:26:17 DeviantPeer, but it's wasn't easy writing it, and it still have many issues that simply can't be solved because you do not have access to some of the Android stuff that handle images Nov 13 13:26:18 DeviantPeer: oh yeh, forgot about the resizing, thanks Nov 13 13:26:48 the only problem i hit is sometimes i reach max handlers Nov 13 13:26:54 (around 128, iirc) Nov 13 13:27:09 but i'm sure there are ways (did not look yet into) to avoid this issue Nov 13 13:27:19 adq: nice link. Nov 13 13:27:31 it happens when i scroll fast and where there is lot of images to load Nov 13 13:27:48 I think the key is to use two sets of images, low and full detail (resolution), since detail doesn't matter when quickly scrolled, so use low res and then when the listview comes to a stop or slows to a threshold level, use the cache of full detail/high resolution bitmaps and invalidate Nov 13 13:27:55 adq: maybe use a pool of handlers? (is it possible?) Nov 13 13:28:08 yup for sure there are many ways Nov 13 13:28:31 even a predefined list of handlers (so number < 128), used as a FIFO Nov 13 13:28:50 would work too Nov 13 13:29:00 true Nov 13 13:29:12 but... adaptive is so much more kool :) Nov 13 13:29:19 DeviantPeer, nope, it's not open, I'm discussing it with my company. It's an option they didn't refused. There's a possibility I'll release it open in the next couple of month, not sure tough Nov 13 13:29:24 the main problem is that you don't know the size in advance so if you have a listview, it becomes problematic Nov 13 13:29:38 if you want to act as a webview (loading html) Nov 13 13:29:50 setellipsis works on pre-ics Nov 13 13:30:10 adq, handlers are not a good idea for the very reason you experienced :) Nov 13 13:30:10 but i never found a solution for later OS (but it was few monthes ago when i played with it, so things may have change) Nov 13 13:30:15 mastro: good. Getting a company to release code (some companies at least) is like taking a piece of meat from a hungry wolf.. so good luck. Nov 13 13:30:20 mastro, agreed :) Nov 13 13:31:37 adq, the other two main issues are that you do not know the size of a view until it has the image in it, so if you want to use wrap content you need to set up a fake Bitmap of the size of the actual bitmap (decode without sampling) then register an observer to know when the view size has been set and properly decode / set the image inside it Nov 13 13:32:33 adq, and you can't control when the system will release those images you loaded from memory, if you are loading in background you may be loading an image that is not needed anymore and it's not easy to know when you will not need it anymore Nov 13 13:33:20 mastro: yeah... vyper nest. Nov 13 13:33:22 adq, decode of an image may just fail because when you ask for it there's not enough memory to decode it and you have no way to know this will happen or to distinguish between this case and a corrupted image (they both return null) Nov 13 13:33:53 yup Nov 13 13:33:56 adq, the only think you see is a warning in the log Nov 13 13:34:51 Android developer did a very bad job defining that API :) -- I also discovered android do not support jpeg with CMYK color space.. not found it written anywhere in the doc nor the android log error helped in finding it out Nov 13 13:35:10 Would be nice to dig in the webview code, since I never experienced the problem(s) you described above Nov 13 13:35:18 I had to look at the C code that does the decode to discover it, no kidding Nov 13 13:35:20 but i'm sure it's like taking a caterpillar for nailing :/ Nov 13 13:35:45 * DeviantPeer likes C Nov 13 13:35:47 :) Nov 13 13:35:47 adq, that would be cheating :D -- the webview can use internal API .. Nov 13 13:36:36 mastro: android has its own implementatiof of jpeg decoding? Nov 13 13:36:39 adq, I looked at the code of the Android Gallery: it handle huge images, scroll, scale, etc without ever any memory issue... Nov 13 13:37:35 adq, the code that does that is a custom view that apply video-streaming techniques to images: it decode small portion of the images keeping what's needed "what you see" and a little "around the border" and "stream" those image portions when you scroll around.... Nov 13 13:37:37 mastro, try to ask again in few hours (because most of ppl awake right now are in europe), i'm sure you will find the light later on with ppl "on the other side" Nov 13 13:38:21 adq, "cool" I though.. I can do a similar thing... but nope, I can't because to do so they directly control the scrollX / Y attributes of the view, which are private for the *common developer*, and there's no direct set either Nov 13 13:38:50 adq, do not need help now :) I was just telling you my experience Nov 13 13:39:31 thx for sharing :) Nov 13 13:39:40 i'm feeling kinda noob :3 but it's ok ^^ Nov 13 13:39:52 adq, actually it was Estel who asked for help in lazy loading images from a listview, I was explaining how bad is android in helping out there Nov 13 13:40:20 for sure this issue is redundant. Nov 13 13:40:23 mastro: the app QuickPic does a very good job at dealing with images and caches. but I think it closed source. Nov 13 13:41:11 DeviantPeer, never heard of it Nov 13 13:41:20 mastro: but it would be interesting to know how they deal with those problems. Cause it really felt faster than the "native" gallery. Nov 13 13:41:27 adq, everyone is a noob on matters that they didn't worked / spent time on Nov 13 13:41:35 mastro: at least in the time of android 2.3 Nov 13 13:42:04 mastro: well said. hands on is the only way to learn. Nov 13 13:42:29 DeviantPeer, agree Nov 13 13:42:49 talking about hands on :p let's get back to code! Nov 13 13:43:02 touché Nov 13 13:43:09 but actually going for lunch now. Nov 13 13:43:10 brb Nov 13 13:43:12 adq, waiting for the SDK upgrade here :) Nov 13 13:45:05 I think the main issue with android development is how Android handle libraries: they did a so bad job that every single developer almost re-invent the wheel every time they write a new application because it's less of a pain then trying to write and maintain a library :) (not 100% true but not far from it) Nov 13 13:47:17 hello Nov 13 13:51:03 Hello ppl. Is there a tool or API that allows a developer to read from gralloc ? I've seen some commits at nvidia gerrit that include read method but I can't find them on aosp. Nov 13 13:51:22 mastro: very true. on the other hand libraries also bring their own set of problems (linking, versioning, etc). Nov 13 13:53:13 p.s why does nvidia choose don't provide access via /dev/graphics/fb0 like they should ? that gralloc method is very complicated /undocumented. Nov 13 14:03:37 Guys, please help. I need some example of reading sqllite db file. What I need to read it? Example of existing and working example if somebody can. Please. Nov 13 14:19:52 im new to this so any help would be greatly appreciated. im working on the multiplayer app "collaborative writing", where players join a game with several players, and input sentences one at a time. after every player has submitted a sentence they collectively vote on which one is the best and that sentence gets selected to be a part of the story. They do the process over and over again Nov 13 14:19:52 so that a story develops. i have a webserver with a sql database, and i have to make use of GCM. i've been bending over backwards to organize the php and the tables in the simplest way possible. can anyone explain how they can help me? Nov 13 14:20:39 I'd suggest hitting stackoverflow for database suggestions, maybe their db-oriented stack too Nov 13 14:20:54 db-oriented stack? Nov 13 14:21:01 hmm ok Nov 13 14:21:06 in your case, I'm not sure you need RDBMS at all Nov 13 14:21:23 is db-oriented stack not a RDBMS? Nov 13 14:21:26 http://www.blackbox.com/Store/Results.aspx/Cabinets-Racks/n-4294965579/p-0 Nov 13 14:21:29 sorry Nov 13 14:21:32 wrong link Nov 13 14:21:37 http://dba.stackexchange.com/ Nov 13 14:21:56 ^ that's what I meant - a "stack" for DBAs ;) Nov 13 14:22:22 DBAs? Nov 13 14:22:26 administrators? Nov 13 14:22:43 im using mysql i have to Nov 13 14:22:52 but i have no idea how to work the backend Nov 13 14:23:20 i have to wrtie php scripts to create and read info off the db right? Nov 13 14:25:47 profligacy: I really have no idea what you'd use an RDBMS for Nov 13 14:25:59 you also may want to rethink PHP as your choice of language for this. Nov 13 14:26:03 when ppl end a room Nov 13 14:26:10 or wehen ppl come in a room Nov 13 14:26:18 why lov? Nov 13 14:26:33 php should be fineee Nov 13 14:26:38 aside from the fact that PHP is a fractal of bad design Nov 13 14:26:46 what?? Nov 13 14:26:51 uh Nov 13 14:26:55 seriously? You missed that one? Nov 13 14:27:07 http://me.veekun.com/blog/2012/04/09/php-a-fractal-of-bad-design/ Nov 13 14:27:16 the webbertubes were all abuzz when it came out Nov 13 14:27:29 because the post pretty much distilled all of the problems that php has into one concise blag post Nov 13 14:27:37 DeviantPeer, sure, but if ios has some advantages over Android they can be narrowed down to this two simple thing: 1) fixed set of device screen size = easier to develop an UI, 2) easier to include any library -- more software available (aka: you do not have to start *almost* from scratch) Nov 13 14:27:39 short version is "PHP considered harmful" Nov 13 14:27:43 anyhoo Nov 13 14:28:02 I guess you could use a database, but it really seems like something better suited to living in-memory Nov 13 14:28:03 lov what should i use then? Nov 13 14:28:17 i have to as a requirement to my prject Nov 13 14:28:20 use a db Nov 13 14:28:22 idk. servlets? appengine? ror? Nov 13 14:28:26 ugh Nov 13 14:28:28 ehh Nov 13 14:28:29 is this a university project? Nov 13 14:28:32 im using GCM Nov 13 14:28:34 yeah Nov 13 14:28:37 aha. Nov 13 14:28:39 welp. Nov 13 14:28:41 mastro: maybe so Nov 13 14:28:42 have fun. Nov 13 14:28:47 ? Nov 13 14:28:54 mastro: but you can also include libs with your app. Nov 13 14:29:00 profligacy, why you are talking about php? isn't a little too much out of topic??? -- anyway if you want to know why php is bad: http://me.veekun.com/blog/2012/04/09/php-a-fractal-of-bad-design/ Nov 13 14:29:04 it just that they are unique to each app Nov 13 14:29:10 just be sure to forget php the moment you're done with it, preferably with strong liquor. Nov 13 14:29:23 i dont have to use php lov Nov 13 14:29:27 oh, good. Nov 13 14:29:29 but i have to use the backend db Nov 13 14:29:30 then for the love of god please don't. Nov 13 14:29:32 mastro: good link. lol Nov 13 14:29:40 lov, what do i use instead? Nov 13 14:29:50 09:28:27 < lov> idk. servlets? appengine? ror? Nov 13 14:29:55 is php really that bad? Nov 13 14:29:57 yes. Nov 13 14:29:59 yes it is. Nov 13 14:30:01 why is this the first time i m hearing about it Nov 13 14:30:02 it is actually that bad. Nov 13 14:30:07 I honestly don't know. Nov 13 14:30:13 profligacy, python, java, actually ANY other language is better than that... even cobol Nov 13 14:30:16 PHP is "good" in that it's an extremely forgiving language when you're a beginner. Nov 13 14:30:17 i thought it was the most ppopular server side scripting language Nov 13 14:30:22 oh, it is. Nov 13 14:30:24 that doesn't mean that it's good. Nov 13 14:30:29 it just means that it's popular. Nov 13 14:30:39 go read that blog post, and you'll understand. Nov 13 14:30:40 i wanted something popular so that it's easier to find help resources Nov 13 14:30:46 do you really think ill ahve issues Nov 13 14:30:49 It's the Visual Basic 6.0 of serverside development Nov 13 14:30:52 with the simplicity of my project Nov 13 14:30:59 anybody know any good Java <> Android pitfalls to avoid? Nov 13 14:30:59 really I have no idea. Nov 13 14:31:04 profligacy, milions of flies love shit too :) that doesn't mean you should love it too Nov 13 14:31:05 why is vb6.0 bad? Nov 13 14:31:08 (literature) Nov 13 14:31:16 I literally don't know how to respond to that one. Nov 13 14:31:38 lets just say that as you go out into the world, you'll learn some things about software engineering. Nov 13 14:31:40 why not..is this common knowledge? Nov 13 14:31:41 profligacy: go ahead. Nov 13 14:31:56 profligacy: leave this two old guys complaining about php.. Nov 13 14:31:58 belgianguy, most common one is memory leak when rotating the device: most programmer do not even know what a memory leak is :) Nov 13 14:32:07 :) Nov 13 14:32:35 mastro: memory leak: is that when the oil from the memories start to run? Nov 13 14:32:35 DeviantPeer: do you think ill have issues (beig that my project requires simple php tasks)? Nov 13 14:32:45 profligacy: just do it in php, go knock your socks off. Nov 13 14:32:48 mastro: ah, I see, and is there anything else than handling the lifecycle events that matters here? Nov 13 14:32:57 profligacy: there are lot's of "big" projects done in php. for better or worse it's doable. Nov 13 14:33:10 as the Activity gets destroyed, but Fragments can be retained (setRetainInstance(true)) Nov 13 14:33:25 belgianguy: yeah, the lifecycle in general is really something you have to look out for. Nov 13 14:33:30 DeviantPeer, profligacy it's doable in php, it's the best choice? nope Nov 13 14:33:34 profligacy: it's just that... you need nlots of discipline to make it so that it doesn't become a mess of entangled code and markup Nov 13 14:33:47 mastro: whats the best choice? Nov 13 14:33:47 mastro: true... Nov 13 14:33:57 go Nov 13 14:33:59 :) Nov 13 14:34:12 golang.org Nov 13 14:34:51 mastro: whats the best choice? Nov 13 14:35:04 mastro: php has one big advantadge over other choices: there's hosting everywhere that supports it and it's fast to get something quickly. Nov 13 14:35:09 lov: yeah, I've studied on that one, and have had enough FC's to know that by now :) Nov 13 14:35:18 does an imageview have something equivalent to an activity's onCreate, or a fragment's onCreateView method? Nov 13 14:35:19 mastro: would I use php? no... :D Nov 13 14:35:44 Tee_Pee: draw ? Nov 13 14:36:40 let's see.... Nov 13 14:36:50 profligacy, there's no such thing as "best choice" expecially in programming, the best choice depends on what you have to do, your knowledge of the tools and the time you can spend learning. The best choice having infinite time to spend is to learn very well every tool that could allow you to do what you need and then choose the best fitted. Since time is not unlimited you have to go down to compromise :) And, from my experience, try to avoid PHP if you Nov 13 14:36:50 can, or to drift out from it as soon as you can. Reason is in the link I gave you before Nov 13 14:37:26 so python is ok then? Nov 13 14:37:50 do you guys hava any links to learn pyhton with json Nov 13 14:38:04 Tee_Pee, what you want to do? Nov 13 14:38:11 profligacy: google.com Nov 13 14:38:12 python is ok as a language, but then you must choose the web framework you want to use... Nov 13 14:38:25 DeviantPeer: the web framework?! Nov 13 14:38:33 profligacy: I'd recommend that you look at whoever is providing your webhosting and figure out what they actually support, then choose from that list. Nov 13 14:38:37 profligacy: http://docs.python.org/2/library/json.html Nov 13 14:38:45 love they support python php and perl Nov 13 14:38:56 then for the love of god learn and use python. Nov 13 14:39:07 profligacy: and remember... channel is android-dev, not web-dev ;) Nov 13 14:39:07 DeviantPeer, profligacy if you know nothing about python starti with django or choose one from here: http://wiki.python.org/moin/WebFrameworks Nov 13 14:39:20 Yes, that too. Nov 13 14:39:27 You're highly encouraged to ask python questions in #python Nov 13 14:39:28 I have a gridview with a bunch of imageviews (custom class extending imageview that is)... for some reason, if the views are recycled, the images sometimes do not show up... however they do still hold a String object containing their intended image uri... so I just want to check if that uri is non-null on the "onCreate" method and set the drawable to that URI Nov 13 14:39:37 also, mysql questions in #mysql, etc. Nov 13 14:40:45 Tee_Pee, you are doinging something wrong, if the image is recicled you shouldn't even use that layout anymore Nov 13 14:40:52 recycled Nov 13 14:41:18 Tee_Pee, and if you are manually recycling the image then you are doing it at the wrong time Nov 13 14:42:35 mastro - Let me just fetch the relevant code, easier that way than to try and elaborate Nov 13 14:42:37 on the engineering front, are there groundrules about how to divide a problem into different Activities? Nov 13 14:44:39 belgianguy: generally, for every different screen you'd show the user, use a new activity. Nov 13 14:44:44 also, look into Fragments. Nov 13 14:45:32 lov: ah thanks, yeah that was my ground rule as well, I've looked into Fragments, they're pretty powerful Nov 13 14:45:42 esp wrt UI phone and UI tablet Nov 13 14:45:46 yeah Nov 13 14:46:02 Think of activities as being logical UI activities Nov 13 14:47:29 mastro - http://pastebin.com/VFmryiF7 These are the relevant code snippets from the adapter and imageview respectively. Nov 13 14:47:55 lov: yeah, with the simple UI elements and layouts it's all pretty straightforward, but when I get to a 'game', it has to be updated in many more ways Nov 13 14:48:36 The bug that occurs is that, if I start scrolling and just jerk it up and down (recycling a few items down and up respectively), those items occasionally show the default image. Nov 13 14:48:55 I imagine the problem relies with an incorrectly canceled BackgroundDecode task Nov 13 14:49:05 lies with* Nov 13 14:49:11 belgianguy: what usually happens for games is that you'll have an activity that handles displaying the game engine Nov 13 14:49:22 and the game engine does whatever Nov 13 14:49:38 it's totally cromulent to have just 1 activity if you're making a game which has its own engine that handles rendering Nov 13 14:50:55 lov: yeah, it's that I'm struggling with, but haven't investigated just yet how to do it Nov 13 14:52:15 is it easy to make an app to get android notifications on desktop? Nov 13 14:52:32 well, there are several game engines out there that you can use. If you feel like getting dirty with the NDK, there are some more options too. Nov 13 14:52:49 Tee_Pee, what is mPicsInfo ? where do you modify it? Nov 13 14:52:59 gbs: if you mean computer desktop, not really. Nov 13 14:53:22 lov, yeah, i only saw like 2 apps that do this Nov 13 14:53:43 lov: yeah, NDK is totally fine Nov 13 14:54:00 I've experimented with a physics engine, seemed to work OK Nov 13 14:54:16 It's just a list containing the uri of the image, id and some other variables about it I need, it's in the adapter, and it is populated in the adapter's constructor from a JSONArray I get from the server. Nov 13 14:54:19 from my server* Nov 13 14:54:23 perhaps I should revisit that Nov 13 14:54:26 so Nov 13 14:54:30 Tee_Pee, use my name or I may miss your reply Nov 13 14:54:31 private List mPicsInfo; Nov 13 14:55:25 that being said, I tried commenting out the background task cancel command just now, and it didn't help, so that isn't the culprint. Nov 13 14:55:59 actually scratch that, 1s Nov 13 14:56:14 belgianguy: go hog wild then. Just bear in mind that at the NDK layer, issues will be somewhat more opaque. You will, of course, have the benefit of the raw speed that running C code has. Nov 13 14:56:31 lov: yeah, I know, it's a bit of a minefield Nov 13 14:57:05 Tee_Pee, do you see something in logs? Nov 13 14:57:22 but while I know loads of stuff already, I'm building this game to iron those unknown regions out Nov 13 14:57:40 Tee_Pee, also, using asynctask is not the best choice for this, you are limited to 20 async task at the same time If I remember correctly Nov 13 14:58:17 Tee_Pee, oh.. that's why you used the Executor I see Nov 13 14:58:44 mastro - Nothing that would help me identify this... that being said, I just changed if(bdecode == null || bdecode.cancelPotentialWork(imagepath)), to if(true), so that it does the task every time, and that solved the missing image issue, but I hit the task cap rather quickly and crash... so the issue is in the cancelPotentialWork() method Nov 13 14:59:31 I gotta change that so that it doesn't kill tasks it's not supposed to kill. Nov 13 15:01:06 Tee_Pee, I would say it's more likely that Bitmap b = BitmapFactory.decodeFile(imagePath); return null Nov 13 15:01:38 I would add a warning log there if cancelled != true && b is null Nov 13 15:01:44 Tee_Pee, ^^ Nov 13 15:01:58 Tee_Pee, and you'll probably notice a weird error in the log like.. let me re-find it Nov 13 15:02:32 let's see.... Nov 13 15:03:08 Tee_Pee, something like this: DEBUG/skia(xxxx): --- decoder->decode returned false Nov 13 15:03:55 in the postExecute I had if(bitmap!=null), added an else{log info...} Nov 13 15:03:58 let's see if that triggers... Nov 13 15:05:08 Is the Nexus4 and Nexus10 launch a joke with..like 100 devices available per country ? Nov 13 15:05:18 Tee_Pee, if that doesn't trigger then it may simply be a synchronization issue that you can solve by synchronizing the check for cancelled with the setImageBitmap((Bitmap) bitmap); Nov 13 15:05:32 bubbleguuum, that's called marketing :) Nov 13 15:05:43 that's non-launch Nov 13 15:06:07 DeviantPeer: *ping* Nov 13 15:06:27 it's silly: if you do not order it within 5min when thesold outy are available, poof there are Nov 13 15:06:27 mastro - 'fraid it doesn't trigger. Nov 13 15:06:29 monsti: pong Nov 13 15:06:34 *sold out Nov 13 15:06:41 mastro - it's something in the cancelation check... let's try and smarten that one up a bit. Nov 13 15:06:46 DeviantPeer: i got the composed ID stuff working, interessted in the solution? Nov 13 15:06:58 monsti: yes Nov 13 15:07:15 monsti: always willing to learn new tricks. :) Nov 13 15:12:11 Tee_Pee, add logs to see what it's happening Nov 13 15:13:51 belgianguy: anyway, best of luck. Nov 13 15:14:29 * monsti got a S3 today Nov 13 15:14:33 lov: thanks, I'm looking for an Android related job, and thought such a personal project would help :) Nov 13 15:14:38 thanks for the pointers Nov 13 15:16:20 hey Nov 13 15:16:20 im new to this so any help would be greatly appreciated. im working on the multiplayer app "collaborative writing", where players join a game with several players, and input sentences one at a time. after every player has submitted a sentence they collectively vote on which one is the best and that sentence gets selected to be a part of the story. They do the process over Nov 13 15:16:20 and over again so that a story develops. i have a webserver with a sql database, and i have to make use of GCM. i've been bending over backwards to script this in a language that is convenient. can anyone explain how they can help me? Nov 13 15:16:20 id like to make some powerpoint slides for my presentation Nov 13 15:16:20 talking briefly about django, mysqlAlchemy and briefly how it ties to my project (that i wrote about above)...its only 10 simple slides by today...can anyone help me? Nov 13 15:16:34 Tee_Pee, make sure you put the Thread ID in the log Nov 13 15:16:35 /ignore profligacy Nov 13 15:16:42 Tee_Pee, (Thread.currentThread().getId()) Nov 13 15:18:12 profligacy: I don't get it, are you looking for someone to make your powerpoint presentation? Nov 13 15:18:27 yes evancharlton Nov 13 15:18:35 or at the very least help a lot Nov 13 15:18:50 profligacy, what makes you think #android-dev is suited for that question? Nov 13 15:19:07 its an android app Nov 13 15:19:48 mastro - I know WHAT is happening - the imagePath within the BackgroundDecode task is equal as the new one being sent to it (i.e. even though it's a new image, the old pre-recycle uri gets forwarded to it) Nov 13 15:19:54 mastro - what I don't know is WHY :p Nov 13 15:21:16 profligacy: what on earth? Why would anyone in here help you make a presentation about your own app? Nov 13 15:21:37 its for school evancharlton Nov 13 15:21:43 Tee_Pee, because Android GridView / ListView recycle views -- and when android think an ImageView is going to be recycled it also recycle it's Bitmap, since the ImageView is the same one the path is the same one but the bitmap is being recycled Nov 13 15:21:52 profligacy: so what? Nov 13 15:22:04 so... Nov 13 15:22:45 Tee_Pee, so when that happen the only thing you can do is re-decode. But you have no way to know if an image is being recycled or not from Android... I would guess this is what happening Nov 13 15:23:49 Tee_Pee, I have a completely different way of handling the same issue and I do not know enough of yours to think of a solution Nov 13 15:24:36 aye, it's a bitch really... I'll hopefully hit my head against a wall on my way home and have an epiphany Nov 13 15:25:46 thanks though :p Nov 13 15:28:43 Tee_Pee, no problem :) I know very well how hard it is to play with android images :) Nov 13 15:30:28 I'm getting "trouble processing "javax/xml/parsers/DocumentBuilder.class" " while trying to compile my project.. anyone ran into this problem? Nov 13 15:31:43 where is that file? in the jdk or something Nov 13 15:32:16 or ant maybe Nov 13 15:35:17 are you using the newDocumentBuilder() method? Nov 13 15:35:27 mastro - solved it :D Nov 13 15:35:58 I don't got time to explain, here are the code dumps, if you're interested you can take a look Nov 13 15:38:31 Tee_Pee, ^^ where's the core dump? Nov 13 15:38:43 mastro - http://pastebin.com/RkMXLs8F Nov 13 15:38:43 there Nov 13 15:38:58 anyway, gotta rush now :D Nov 13 15:41:17 Hello! I have layout of fragment. RelativeLayout android:layout_height="match_parent". In RelativeLayout - GridView (android:layout_height="fill_parent"). In hierarchyviewer i see that RelativeLayout height always const. When and how i can obtain RelativeLayout height??? Nov 13 15:42:32 getMeasuredHeight() or another method? and when? i call getMeasuredHeight() in onCreateView () and receive 0. Nov 13 15:49:12 Now that Adobe is putting the kibosh on Flash on Android, will it still be possible to develop apps for Android using Flash? I mean, are they just killing the browser plug-in or the whole ecosystem? Nov 13 15:50:03 jedrek: it's never been wise, possible or not Nov 13 15:51:53 Yeah, but I need to knock out a simple puzzle game to be run during an event Nov 13 15:56:28 Can't you still use Air? Nov 13 15:56:48 I dunno, that's why I'm asking :) Nov 13 15:57:58 It was my understanding that they still supported Air on Android, but I've never had occasion to try it Nov 13 16:00:18 Yauhen90: after onMeasure has been called you can get its height Nov 13 16:01:38 ok, cool Nov 13 16:01:41 thanks Estel Nov 13 16:09:25 JakeWharton_: i mean following: onCreateView of Fragment, after inflate i call getMeasuredHeight of main RelativeLayout and i receive 0. when i call getMeasuredHeight of ViewGroup container - i receive not null. Nov 13 16:09:34 can't send multiple attachments with default email client Nov 13 16:10:18 hmm, ACTION_SEND_MULTIPLE seems doesn't work for me Nov 13 16:21:02 ctate> as a developer, I could really do with a method to purchase Nexus phones to test my stuff on. Being in a country where you all refuse to release Nexus's (Nexii?!) to the general public ... Nov 13 16:23:44 Chainfire: not that I'm disagreeing with you, but it stands to reason that if they had the means to sell you one phone, they would have the means to sell to the country as a whole Nov 13 16:23:53 right Nov 13 16:23:58 cause my country is one person Nov 13 16:24:05 ello Nov 13 16:24:10 is there a minimum size for the play store badge? Nov 13 16:24:34 dang`r`us: which "badge" is this? Nov 13 16:24:43 Chainfire: that's not what I said at all Nov 13 16:25:37 evancharlton most likely the reason they don't sell here is because the new generation of Nexus devices (incl 7) is heavily betting on the Play ecosystem, which is not properly supported here Nov 13 16:25:39 Chainfire: selling 50,000 units is exactly as difficult (for non-technical reasons) as selling just 1 Nov 13 16:26:07 evancharlton as such, such a device isn't much use here. But as a developer, I'd still like to play and test with the hardware Nov 13 16:26:28 I suspect that's an incomplete view of the situation Nov 13 16:27:33 Chainfire: I was essentially told that by a local dev advocate when I asked about N7 availability Nov 13 16:27:43 no media store=no hardware store Nov 13 16:28:19 evancharlton purchasing it in the USA and getting it shipped here is fine... and certainly possible without any legal mumbojumbo... if they can ship to England, they can ship to the Netherlands Nov 13 16:28:38 I don't care about triple price, nor do I care about warranty, or any such consumer considerations Nov 13 16:28:46 Chainfire: I genuinely wonder if there are any EU issues about selling to only some countries Nov 13 16:28:51 what makes you think they're shipping to england as opposed to shipping from within england? Nov 13 16:29:06 didn't they use to have a program where every developer account could purchase a Nexus ? Nov 13 16:29:30 that was back in the ADP/ION days, before Nexus Nov 13 16:29:42 right, before my time Nov 13 16:29:43 yeah, and wasn't that through some weird third-party thing? Nov 13 16:29:48 brightstarcorp or something? Nov 13 16:30:02 evancharlton the point is they can get the hardware inside europe Nov 13 16:30:06 which would of course be a perfect method to sell to developers worldwide still Nov 13 16:30:17 meh Nov 13 16:30:21 which means there isn't any non-artificial limitation to get it shipped to my door Nov 13 16:30:34 Chainfire: I'm so glad you're an expert on all European regulatory issues Nov 13 16:30:47 there are political limitations for sure, depending some country Nov 13 16:30:53 you should call up Google and offer assistance Nov 13 16:32:41 evancharlton: snark is fine, but it is causing genuine negativity towards Google around the world... Nov 13 16:33:10 Leeds: I doubt they're aware of that Nov 13 16:33:12 even if fixing it is too hard/expensive/complicated, it could be dealt with better Nov 13 16:33:16 evancharlton really, if they can sell it in England they can ship it to the Netherlands. Doesn't mean they need to sell the stuff here. But I guess you know better as always. Nov 13 16:33:43 Is this the right place to ask about frame buffer stuff on android ? I need help with *gralloc* Nov 13 16:34:13 XORi: yes, this is the right place, but that's pretty niche; you might have better luck on StackOverflow or the android-developers google group Nov 13 16:34:34 XORi: at the OS/platform level, or NDK? Nov 13 16:35:14 evancharlton: Stackoverflow didn't help for some days . Any way to read the buffer to stdout ? both NDK or os level is fine Nov 13 16:35:56 I wish that there was /dev/fb0 to read gralloc but those tegra chips doesn't provide /dev/fb0 Nov 13 16:36:09 i have a layout i've made via xml, and one i've made by subclassing relativelayout… how do i set my custom layout to be the root view of the xml layout? Nov 13 16:36:40 this is for a dialogfragment, if that matters Nov 13 16:37:24 XORi> which device Nov 13 16:37:38 Chainfire: nexus 7 / xoom Nov 13 16:37:45 evancharlton, sorry, colleague dragged me away. I'm referring to http://www.webpronews.com/google-play-badges-available-for-android-developers-2012-03 Nov 13 16:38:00 anyway, I guess this page answers it anyway. Nov 13 16:42:42 hm, crappy page. but http://developer.android.com/distribute/googleplay/promote/brand.html has some sensible info. Nov 13 16:42:45 thanks, and bye Nov 13 16:49:37 how do you guys get around low precision gpu problems? Nov 13 16:55:24 android-root is down it seems, so I came here as a last resort... anyone have experience with D2G or recovery failing to boot? Nov 13 16:55:50 I bootstrapped my D2G and everything, but even when I SBF to stock ROM the recovery is screwed. Nov 13 16:56:52 Takes me to esclamation symbol, then I get to menu to factory restore, etc... wipe cache, etc... nothing. Nov 13 16:58:14 "Oops, we had trouble processing your request. We're working on fixing the issue. Please try again later." Nov 13 16:58:17 SNORE Nov 13 16:59:49 Why does IRC have to be filled with people like you... Nov 13 17:00:02 cus im adorable o/ Nov 13 17:00:14 lol Nov 13 17:00:18 irc has to be filled with ppl. period. Nov 13 17:01:26 I've been trying to fix this damn issue for 5 days now, and there's no help on any forums I've checked :/ Nov 13 17:02:01 You're in the wrong channel Nov 13 17:02:11 I know I am, I stated that Nov 13 17:02:18 It seems that android-root is down Nov 13 17:02:23 I can't connect. Nov 13 17:02:33 Tough luck Nov 13 17:02:40 it's not down. Nov 13 17:02:49 you don't connect to a chan, but you "join" a chan Nov 13 17:02:53 /join #android-root Nov 13 17:03:12 I tried that Nov 13 17:03:24 are you banned? Nov 13 17:03:28 Nah Nov 13 17:03:37 Never been to the channel yet Nov 13 17:04:05 type the command I wrote for you just above Nov 13 17:04:30 nodda Nov 13 17:04:53 XORi I'm sure I have some gralloc code that works at least on Tegra2 (haven't tested on Tegra3), trying to remember where I got it from ... Nov 13 17:05:09 It won't join #android or #android-root Nov 13 17:05:43 anyone who knows his/her stuff about RemoteViewsService, factory and parcels and has time to perhaps help me out? i seem to have hit a wall i cant get past >.< Nov 13 17:07:02 Chris6647: just ask your question Nov 13 17:07:48 Chainfire: wow thanks man. I've found some code in droid-vnc-server, but I can't compile it with my aosp sources , the sources has to be patched Nov 13 17:08:24 and the patch doesn't work on any aosp version I tried. Nov 13 17:08:37 in brief, i cant pass my parcelable objects to the remoteviewsservice. ive explained my issue in more detailon http://stackoverflow.com/questions/13363046/passing-custom-parcelable-object-extra-or-in-arraylist-to-remoteviewsservice-bre Nov 13 17:08:58 XORi> what exactly are you trying to do ? Nov 13 17:09:10 a noob is in the house *.* Nov 13 17:09:37 Chris6647: how much data are you trying to send? Nov 13 17:09:38 emulate frame buffer so I can feed ffmpeg and produce a video Nov 13 17:09:42 either the remoteviewservice just never goes on, ie. the log flags that ive got in the remoteviewsservice and factory are never actually reached Nov 13 17:09:50 Chris6647: also, I'm super skeptical that there's _nothing_ in logcat Nov 13 17:10:20 yeah i saw what i wrote a few moements ago and was about to try and generate a log with no filters to post Nov 13 17:11:09 im not actually sure how i'd go about measuring the amount of data im trying to send. on a normal use case id say a decent size, but the issue is there even with just an object containing a few strings. Nov 13 17:12:38 yeah, run logcat without any filters, post the log from around that time Nov 13 17:17:37 XORi> ok I remember where I got the code... from the screenshot code on a tegra Nov 13 17:17:45 source was somewhere on some nvidia server Nov 13 17:18:40 Chainfire: somewhere here? http://goo.gl/rMpjM Nov 13 17:19:31 something like that, yeah Nov 13 17:20:07 there a binary in /system/bin on some Tegra systems that takes screenshots Nov 13 17:20:17 not sure but I think it's actually built into toolbox Nov 13 17:20:28 hope that helps you find it... should've saved the link somewhere, heh Nov 13 17:21:57 Chainfire: there is /system/xbin/screenshot and /system/bin/screencap. I see that screencap but it uses android screenshot client. is this what you mean Nov 13 17:22:38 updated my question with a log which has clues :) http://stackoverflow.com/questions/13363046/passing-custom-parcelable-object-extra-or-in-arraylist-to-remoteviewsservice-bre Nov 13 17:23:08 XORi> look into older, this code was from Tegra2, before Android had a centralized screenshot function Nov 13 17:24:38 XORi> https://github.com/gojimi/gtablet_platform_system_core/commit/fe488b64a532993a77e9df0c9667f641bf108864 Nov 13 17:24:55 search page for "readfb_gralloc" Nov 13 17:26:03 is there a straightforward way to use a different background for a RadioButton than the little circle, short of writing a custom RadioButton? Nov 13 17:26:23 so people howdy Nov 13 17:26:51 SimonVT: I did some troubleshooting on the problem I discussed with you yesterday Nov 13 17:27:22 the arraylist values being null in onClickListener Nov 13 17:27:28 good m00rning Nov 13 17:27:46 my friend got charged for 3 devices Nov 13 17:27:55 i have a feeling there will be a lot of cancellations Nov 13 17:28:01 the problem is that add method is appending null in the list :-( Nov 13 17:29:11 Chainfire: Thanks man.. btw, are you the developer of supersu? Nov 13 17:30:11 yw, and yes Nov 13 17:30:34 Chris6647: it looks like your log explains what's going on, more or less Nov 13 17:31:09 Chainfire: than make it double !! tanks again. Nov 13 17:31:26 iNoob: (notice) Hi there, welcome to #android-dev !! Nov 13 17:31:32 autojoin message Nov 13 17:31:40 thanks Nov 13 17:32:31 heya turb1ne247 Nov 13 17:34:48 Chris6647: Your service is running in another process, so you probably have to tell it which classloader to use.. Bundle has a setClassLoader method, you could try that.. Nov 13 17:34:55 Tho I'm wondering why you're doing this at all Nov 13 17:35:23 I wouldn't count on your service running at all times, so you would probably lose any data you sent to it every once in a while Nov 13 17:38:47 evencharlton, SimonVT: yeah im researching all that now, and actually looking at the setClassLoader method. Well i supose the backbone of my program is the intentservice which is triggered from events and alarms, this service then figures out what to do, perhaps fetches new information from a website, then passes it along to the remoteviewsserivce (im actually doing setRemoteAdapter Nov 13 17:38:48 everytime this is called, not sure if this is wrong and if i should somehow invalidate the data instead) Nov 13 17:41:41 SimonVT: and in case i dont fetch new info from the website, and i need certain info, i am actually deserializing my objects elsewhere. i just thought it was stupid to deserialize it several times in one go, which is why i figured it would be way better to pass it as a parcel, and only deserialize it once Nov 13 17:42:02 is there frameworks that make it easier to do cool android ui design ? Nov 13 17:43:43 Hello! I am trying to figure out the best way to update my UI. Of course I tried the wrong way, and got an exception. I've been googling a little bit, and I found something about using a handler, but it doesnt seem right for my use Nov 13 17:44:40 AndreasBr: the question is unanswerable from a description like that. Nov 13 17:45:46 I see. One issue is this: I am communicating with a payment terminal. The events coming from the terminal are sent to a listener called PayListener. (Its a class of its own). I need to be able to update the UI from that class Nov 13 17:46:37 alankila: From the examples I've seen, I really dont understand how you choose what to update, do you need to update everything everytime? :S Nov 13 17:47:09 provide a context for the paylistener to use, such as applicationcontext, and then broadcast intents from it. This is one way. Another would be to pass a handler which communicates with UI thread and sendm essages to it Nov 13 17:48:03 Is there any method where I can just do something like updatethisbutton(String withthistext) ? Nov 13 17:48:07 you update what you need to update. It's not like android actually updates anything at all by default. It might inflate a layout tree from call such as setContentView, but even that just sets the values as they are defined in the resource XML files. Nov 13 17:48:24 most likely button.setText("foo"). Nov 13 17:48:33 Yes, something like that Nov 13 17:48:46 But that needs to be done by the UI thread? Nov 13 17:48:51 indeed. Nov 13 17:48:59 I saw this example: http://crodrigues.com/updating-the-ui-from-a-background-thread-on-android/ Nov 13 17:49:13 But it didnt look right for what I want to do, or atleast what I think I want to do. Nov 13 17:50:21 well, it shows how to use a handler, I guess. Nov 13 17:50:34 I see, but where do I decide what to update? Nov 13 17:50:45 All I see is myHandler.post(updateRunnable); Nov 13 17:50:52 well android does nothing by default. The updateUI method in that example must do the changes. Nov 13 17:50:57 In my head that is where I would want to choose what to update Nov 13 17:51:17 hey together Nov 13 17:51:20 alankila: yes, that I kinda understood. But where do you pass what to change, and what to? Nov 13 17:51:44 I got a problem with android encryption - and I'd like to disable it in the ramdisk Nov 13 17:51:52 does anyone know how to do so? Nov 13 17:52:16 AndreasBr: well you need to decide a way to represent changes you wish to perform as objects. Nov 13 17:52:42 *scratches head* Nov 13 17:53:15 here's a way that is fairly elite: Nov 13 17:53:21 final String newButtonText = ...; Nov 13 17:53:23 It has to be easier than it seems in my head Nov 13 17:53:46 handler.post(new Runnable() { public void run() { updateUI(newButtonText); } }); Nov 13 17:53:56 Ok Nov 13 17:54:16 So I could also do like updateUI(mybutton, myText) ? Nov 13 17:54:20 this captures the value of the newButtonText in the anonymous inner class that inherits/implements Runnable Nov 13 17:54:35 and use a switch Nov 13 17:54:40 AndreasBr: if you have the button object in your hands Nov 13 17:56:10 So I could make a class of its own, called updateUI ? and do like handler.post(new updateUI(mybuttonobject,newstring)) ? Nov 13 17:56:36 if you like. *Shrug* Nov 13 17:56:50 Is that a messy way to do it in your opinion? Nov 13 17:57:01 I dont want my main file becoming way too big =P Nov 13 17:57:11 I would probably simply generate broadcasts in your paylistener that correspond to events that the paylistener hears Nov 13 17:57:27 and then make activities listen to these events so that their state would update as they receive the broadcasts Nov 13 17:57:28 ok, you see, now you confused me again Nov 13 17:57:43 Is there a way to pad a ListView so when you scroll to the bottom there's some blank space? Nov 13 17:57:48 Sorry for being stupid, I blame it on being Norwegian Nov 13 17:58:17 Do you mean making my Paylistener throw an event to the activity? Nov 13 17:59:03 eh, more or less Nov 13 17:59:41 it's at least making a clean separation between the UI and the backend, if nothing else Nov 13 17:59:56 Since I'm new at this shit: Each layout does not have its own activity?= Nov 13 18:00:07 having the button objects and texts you want to appear in some backend class seems like it isn't very well layered Nov 13 18:00:11 Can an activity be compared to a form in windows programming? Nov 13 18:00:23 well no, thats wrong. Nov 13 18:00:30 I don't do windows programming. And you must have missed the 'setContentView' call in activity Nov 13 18:00:53 No, I havent.. I was just thinking wrong for a second there Nov 13 18:00:55 the layouts are just data, they know nothing and understand nothing. The activity selects a layout for itself with the setContentView call, which actually causes the objects described in the XML to be created. Nov 13 18:02:23 I need to google more on what an activity is Nov 13 18:02:24 and you can put layouts in layouts Nov 13 18:02:29 An activity is a single, focused thing that the user can do. Nov 13 18:02:31 similarly, another common beginner mistake is attempt to call findViewbyId before the setContentView call. Nov 13 18:03:06 Right now I think I need to try to understand what an activity is Nov 13 18:03:09 brb :) Nov 13 18:03:11 *google* Nov 13 18:03:37 For example, an email application might have one activity that shows a list of new emails, another activity to compose an email, and another activity for reading emails. Nov 13 18:03:50 Why wouldnt you do this in one activity, and just change the layout? Nov 13 18:04:10 (Im sorry if these questions are a little TOO basic for you) Nov 13 18:04:15 AndreasBr: to avoid making a very large single class where you have multiple objects that are valid depending on some internal modal state Nov 13 18:04:44 I think I just read the answer actually Nov 13 18:04:51 . For example, a camera application can start the activity in the email application that composes new mail, in order for the user to share a picture. Nov 13 18:05:20 I think I only need one activity atm for my use Nov 13 18:05:24 well that could always be achieved even if you crammed everything into a single activity Nov 13 18:05:41 you just receive the broadcast that tells to compose email and switch the internal views appropriately Nov 13 18:06:02 hm ok Nov 13 18:06:21 This stuff is hard when you dont know how to do it, haha Nov 13 18:06:32 Most things are Nov 13 18:06:39 I dont understand why updating UI has to be such a .. fubar process Nov 13 18:06:48 Its so simple in a normal windows app Nov 13 18:06:50 xD Nov 13 18:07:02 I actually once saw a viewflipper based design that had something like 3 different major views crammed into a single activity Nov 13 18:07:11 that was ... unreadable. Very large class. Nov 13 18:07:17 from the sound of it your "normal windows program" is VB Nov 13 18:07:25 c# Nov 13 18:07:28 and this is a lot easier than real win32 Nov 13 18:07:31 AndreasBr: close enough Nov 13 18:07:33 haha Nov 13 18:08:29 Say that I have a button in my layout Nov 13 18:08:31 SimonVT: you wouldnt by any chance have any other ideas in regards to my issue, would you? doesnt seem like i can get it to work still Nov 13 18:08:36 AndreasBr: it honestly isn't very difficult but you sort of need to pick a model you want to use to achieve the update. Handlers, Intents and AsyncTasks make a showing when you do multithreaded android programming. Nov 13 18:08:47 you wouldn't think to write java by considering the syntax in C# and then finding hte nearest translation; likewise don't write an android app by considering how a windows app works then finding a literal android api equivalent Nov 13 18:09:28 Chris6647: Put the data somewhere the widget can access it, tell it where via settings Nov 13 18:09:29 In order to get that as a object I would do something like Button myButton = (Button)Find...somethingByID(R.id.my_button); ? Nov 13 18:09:51 (..something because I forgot what it was, haha) Nov 13 18:09:59 AndreasBr: findViewById, and yes, that is a way to acquire the Button object after setContentView has inflated it Nov 13 18:10:26 See, im not that stupid, haha =P Nov 13 18:10:39 and then I could pass myButton to the handler? Nov 13 18:10:57 That is such a vague description of the process that it makes me cringe. Nov 13 18:11:00 ehm. I'm just thinking here Nov 13 18:11:07 SimvonVT: i'm not quire sure exactly what you mean. "putting it somewhere reachable" kinda sounds like what im doinog now with deserializing the files from internal storage. but seeings how everybody says serialization is slow, im thinking this isnt optimal? Nov 13 18:11:13 Because I could imagine gazillions of other things to update in UI Nov 13 18:11:20 that isnt text Nov 13 18:11:31 add a new button, disable a button, etc Nov 13 18:11:44 Thats why this look so difficult to me. Nov 13 18:11:47 AndreasBr: you can't pass the button to a handler, it has no method that accepts a button. You give it a Runnable, which is kind of like a task you want to execute later, which will be executed by the counterpart thread of the Handler Nov 13 18:12:20 Yes, my bad. Nov 13 18:12:55 and if you have complicated update needs, you need to model the UI mutations somehow so that you can describe the changes you want to occur in some nice way Nov 13 18:13:19 I do not really recommend giving the other thread access to your UI objects, it's going to tie them together fairly strongly. Nov 13 18:13:20 At this point I just need to: update UI with text Nov 13 18:14:01 I have a menu called settings, a layout called activity_settings, and in those two I have a textbox, and a button Nov 13 18:14:07 it might be better to try to design a data structure that contains the information the UI needs to display, and you update that from the other thread, and then the UI thread builds the views to match it Nov 13 18:14:24 I imagine getting text from something is fairly simple, while setting text is "harder" Nov 13 18:15:23 On another note, does android have anything like a listbox? Nov 13 18:15:26 SDK is up, folks. Get it while it's hot! Nov 13 18:15:33 AndreasBr: what's a listbox? Nov 13 18:15:34 single column, several rows Nov 13 18:15:41 * alankila doesn't know what a listbox is, but it sounds like spinner Nov 13 18:15:43 AndreasBr: ListView? Nov 13 18:15:53 spinner is a stupid name, but it's basically a drop down list Nov 13 18:16:25 Let me try to explain what I need. The payment terminal sends events to my program. "Insert card" "Waiting for PIN" Nov 13 18:16:51 I just want to display them in a orderly fashion underneath eachother Nov 13 18:17:14 perhaps a job for a listview then. Nov 13 18:17:24 if you only have a few items, just use a LinearLayout and ScrollView Nov 13 18:17:41 ListView is for datasets whose size can approach infinity, pretty much Nov 13 18:18:03 Aha, and add text to the LinearLayout? Nov 13 18:18:06 listview is my go-to for repeating structures, even if they were known to be finite Nov 13 18:18:29 Which brings me back to my updateui "challenge" Nov 13 18:19:15 well, challenge for me, haha Nov 13 18:20:06 4.2 notes up! http://developer.android.com/about/versions/jelly-bean.html Nov 13 18:20:39 In my head I want several methods, like .. "addbutton(LinearLayout, "nameofbutton",onclickthingy)" Is that possible? Nov 13 18:21:13 AndreasBr: it's most likely something like linearLayout.add(new Button(this)); and you manipulate the button beforehands maybe to add an onclicklistener Nov 13 18:21:16 API17 is available to download too Nov 13 18:21:19 or is it .addView Nov 13 18:21:40 So I need an onclicklistener? Nov 13 18:21:47 well you said 'onclickthingy' Nov 13 18:21:53 I was looking into that earlier, looked complicated as f*** Nov 13 18:21:54 so I assume you wnated one Nov 13 18:22:27 I guess theres also an easy way to determing what button was pressed? Nov 13 18:22:34 It's not complicated, but honestly you don't seem to think in a very object oriented fashion Nov 13 18:22:44 alankila: Hmm, maybe not' Nov 13 18:22:50 AndreasBr: well unless the button has an onclick listener, I doubt anybody cares if you press it Nov 13 18:22:53 so you want one on it Nov 13 18:22:59 maybe my basic thinking is a littlebit off?= Nov 13 18:23:03 Maybe thats the problem Nov 13 18:23:14 /issue/challenge Nov 13 18:23:58 alankila: I like honesty :) Nov 13 18:25:11 Perhaps thats why it seems so hard for me, because my basic thinking is off? Nov 13 18:25:24 I'm not thinking in an object oriented way Nov 13 18:25:52 well, maybe you are; but in any case you are lacking familiarity with android which is a serious difficulty for now. Nov 13 18:25:58 Hi all, I'm trying to set a background for my activity with an image. I've set the android:background attribution of my linearlayout which is at the root of my activity and the bacgkround shows, but now hte color of the background blends with my buttons on the acitivy. How can I prevent this? Nov 13 18:26:12 alankila: Yes, but I do have to start somewhere, right? Nov 13 18:27:19 seanjean999/seanjean718: got a screenshot of the problem? Nov 13 18:27:26 As far as I've understood there is an UI thread, and only that thread is allowed to touch the UI. Nov 13 18:27:26 AndreasBr: in OO, you should solve a problem at a time. The way you defer a problem to be solved later is by defining an interface that will solve the problem, and then just leaving it unimplemented, or just mock-implemented for now. Compartmentalize and conquer. Nov 13 18:27:35 Ikee, sure how do I upload it? Nov 13 18:27:41 The Support library has also had a version bump Nov 13 18:27:54 AndreasBr: yes, only UI thread is allowed to manipulate the views Nov 13 18:28:06 Ikee, the buttons by default theme is grey but after I set the background blue, it turns blue as well Nov 13 18:28:10 AndreasBr: that reason alone is why I suggest NOT leaking the objects anywhere where other threads might touch them, even by mistake Nov 13 18:28:20 http://www.pasteall.org/pic/ Nov 13 18:28:21 Ikee, the button is a child of the linearlayout for which I've change the background Nov 13 18:28:30 Ok, may I ask how you suggest that I update the ui? Nov 13 18:28:32 one of the key things in keeping threading in check is to carefully control the environment available to the threads so that they can only touch stuff that is safe to touch Nov 13 18:29:10 Somehow I have to tell the UI thread to change something. Which could either be done by handler, asynctask, or.. a last option I dont remember? Nov 13 18:29:10 Ikee, Nov 13 18:29:13 Ikee, http://www.pasteall.org/pic/40368 Nov 13 18:29:20 AndreasBr: well you can broadcast intents from the keylistener, every time you receive an event you will just send it as broadcast and let the activity listen to those broadcasts and manipulate its state accordingly. This keeps the keylistener and activity completely separate from each other, they don't have to even be aware that the other exists Nov 13 18:30:07 Ok, does that mean that I take the event that comes from the terminal, containing what text etc, and create a new event? Nov 13 18:30:15 or do I just pass it along? Nov 13 18:30:19 seanjean718: holo buttons are semi transparent. Nov 13 18:30:33 Ikee, ohhh? how do I make it non-transparent? Nov 13 18:30:39 or is that up to me to decide? Nov 13 18:30:44 AndreasBr: another option is to define methods in an activity that do stuff like 'displayKeyEvent(String text)' or whatever, something like that, and give the KeyListener the Activity instance, and a Handler, and post runnables that call methods on the activity inside the run() Nov 13 18:31:16 AndreasBr: whatever you do with broadcast intents is up to you. You do what is convenient and what allows the system to work Nov 13 18:31:23 guys, what's a good way of storing data in common storage between multiple applications? Nov 13 18:31:25 *read what you wrote* Nov 13 18:31:31 seanjean718: http://droidapp.co.uk/2011/07/11/android-dev-custom-button-with-no-pngs/ dont know if it works but worth a try Nov 13 18:31:35 Ikee, another follow up question. After I apply a background to the buttons, the gravity attribute appears to stop working for the buttons Nov 13 18:31:41 Ikee, any reason that is? Nov 13 18:32:52 AndreasBr: yet another way would be to define a datastructure that is manipulated by KeyListener, and read by Activity, in such a way that as KeyListener adds stuff into it, the Activity gets some kind of 'update' method fired that causes the activity to rebuild its view tree from data of the datastructure. This might be slightly less code than working with broadcasts, I hate the buggers Nov 13 18:33:30 Whatever you do, it's really a question of personal style and modelling. With some experience, you will be able to choose the methods that seem to make most sense for you. Nov 13 18:33:34 alankila: But that would re"draw" the entire view every time? Nov 13 18:33:57 Maybe. It's probably a trivial cost, in any case. Nov 13 18:34:01 I am trying to repeat my question so that if someone who came after the time I asked can also have the opportunity to help me :-P Nov 13 18:34:20 seanjean718: Yeah seem to replicate it... Don't know why really Nov 13 18:35:33 seanjean718: could you link you button xml? http://pastebin.org Nov 13 18:35:57 One of the real problems that occur with android is the need to retain the state when stuff happens like activity gets destroyed and recreated because of orientation change. If the state of the view tree is only in activity, it is lost unless work done to keep it around is done. With broadcasts, they are lost if the activity is not listening to them -- unless sticky broadcast is used, but they suck Nov 13 18:36:10 alankila: So can I have a method in my main activity thats like displayEvent(textEvent); ? Nov 13 18:36:25 I am iterating (while) over and xmlpullparser instance and using getAttributeValue(). The problem is that getAttributeValue gets value in Log.d but adds a null to ArrayList when I use the add function of the list Nov 13 18:36:33 and have that put something into my LinearLayout?= Nov 13 18:36:34 "For additional security on SurfaceView objects, your app can set a secure flag to indicate that the contents should never appear in screenshots or on a non-secure display output, even when mirrored." Nov 13 18:36:39 wonderful Nov 13 18:36:47 so really a lot of this stuff is imho pretty crappy and the sophistication of code required for "just work, damnit" is actually pretty high in android. Nov 13 18:37:18 AndreasBr: sure, you can add any methods you think are useful in classes you define Nov 13 18:37:22 ok Nov 13 18:37:29 But then they need to be run by the ui thread, right? Nov 13 18:37:44 because any other thread wont have access. And to do that I would need a handler? Nov 13 18:38:13 handler is a way to pass chunks of work between threads, yes. Nov 13 18:38:27 who creates the Runnable is not always the one who executes it. Nov 13 18:38:51 *scratches his head again* Nov 13 18:39:31 I think I would like to have something like updateUI.displayEvent(String eventFromTerminal); Nov 13 18:39:55 Is it possible to get it that simple? That no matter what thread you are in, you can do that Nov 13 18:39:58 and the UI is updated? Nov 13 18:40:14 Or am I missing the object-orientated thing again now? Nov 13 18:40:44 well... I do not really see a lot of value in this 'UpdateUI' class. Nov 13 18:40:50 Ok Nov 13 18:40:55 it still needs to defer the actual work somehow to an Activity or bunch of Views Nov 13 18:41:02 You would just have the methods in the main activity? Nov 13 18:41:08 probably, yes. Nov 13 18:41:18 just use the handler, and a bg Thread Nov 13 18:41:30 belgianguy: yes, we are getting there... Nov 13 18:41:35 "just use" easy for you to say, you arent the noob here, I am Nov 13 18:41:36 :P Nov 13 18:41:39 Ikee, one second Nov 13 18:41:54 Ok alankila, so say I have displayEvent in my activity Nov 13 18:41:55 his other thread will do something like 'handler.post(new Runnable() { public void run() { activity.someEventOccurredBlahblah(text); } });' in the end Nov 13 18:42:04 AndreasBr: well I'm not much further than you, don't worry ;) Nov 13 18:42:13 alankila: Ahaa Nov 13 18:42:20 So inside my payListener Nov 13 18:42:26 I would do what you just typed there? Nov 13 18:42:45 for instance. You need to pass it both the handler, and the activity, for this to be possible, of course. Nov 13 18:42:48 Would the handler need to be inside my payListener, or in mainactivity? Nov 13 18:43:06 well the handler is defined in the activity, and instantiated there, but passed to the keylistener Nov 13 18:43:16 what is "keylistener" ? Nov 13 18:43:24 who instantiates the handler is important because it establishes the counterpart who runs the 'post()' events Nov 13 18:43:27 is that my paylistener where all the events end up? Nov 13 18:43:33 oh yeah sorry Nov 13 18:43:36 I got that mixed up Nov 13 18:43:45 In top of my activity I have "public final Handler myHandler = new Handler();" Nov 13 18:44:09 you should make it private, or at best protected, but that's fine otherwise. Nov 13 18:44:26 no direct access to class members is encouraged in java. Nov 13 18:44:26 But if its private, can I still use it from another class? o.O Nov 13 18:44:40 AndreasBr: that just controls visibility to the name, not the object. Nov 13 18:44:47 aha Nov 13 18:44:56 ok, private now Nov 13 18:45:03 the name 'myHandler' can not be accesses by anybody who has instance of your activity if it's not public Nov 13 18:45:26 Question: If I have done an animation on a View and set fillAfter="true" with the animation setting the alpha to zero, how to i get it back to 1? Nov 13 18:45:45 But would that mean Nov 13 18:45:52 that when I err Nov 13 18:46:16 I need to pass the instance on to my payListener? Nov 13 18:46:40 yes, or define a method called 'post' in activity that can be called safely by paylistener, and which just pases the work on to the myhandler Nov 13 18:46:53 alankila: I just want to say that you have the patience of a saint Nov 13 18:46:56 it'll look a bit weird for sure. Nov 13 18:47:04 evancharlton: And I sure do appreciate that! Nov 13 18:47:53 there was actually a guy here earlier today for which I didn't really even have patience for. Nov 13 18:48:22 alankila: only one? Nov 13 18:48:42 I mean, hypothetical code like: 'activity.post(new Runnable() { public void run() { activity.doSomething(); } })' :-p Nov 13 18:48:56 perfectly logical, but still. Nov 13 18:49:13 *scratches head again* Nov 13 18:49:14 haha Nov 13 18:49:32 but we are going to be in trouble with this design before long. You'll discover you don't control activity lifecycle Nov 13 18:49:46 so both the handler and the activity sometimes go stale underneath you Nov 13 18:49:53 So that last part is a bad idea? Nov 13 18:50:12 using activity.post Nov 13 18:50:25 well, passing activity anywhere really is fundamentally probelmatic Nov 13 18:50:28 See! UI updating is hard -.- Nov 13 18:50:59 AndreasBr: what exactly happens when the UI has to be updated? Nov 13 18:51:27 Well, one of the things I want is a settings screen (to set/show the IP of the terminal being connected to) Nov 13 18:51:38 and to be honest I'm really sorry that android is like this. It isn't my fault, it's just a lot of poor decisions over time. Nov 13 18:51:38 kroot: You forgot http://imgur.com/jacoj Nov 13 18:51:39 But atm thats just a textfield Nov 13 18:51:58 Thanks (y) Nov 13 18:52:09 happened* Nov 13 18:52:09 incoming signal/event? Nov 13 18:52:10 Another thing is that I want a LinearLayout and a Scrollthingy.. And whenever I get an event from the terminal, the UI needs to be updated Nov 13 18:52:47 the event comes in the PayPointListener Nov 13 18:52:51 public synchronized void getPayPointEvent(PayPointEvent event) { Nov 13 18:53:21 that is really odd name for a method Nov 13 18:53:32 I didnt write it :D Nov 13 18:54:11 what does this actually mean? It's obviously not "getting" anything because it's a void method. Unless it fills in data to such an instance. Nov 13 18:54:23 hmm, Nov 13 18:54:23 but then again, instance reuse like that makes anybody frown Nov 13 18:54:24 and the networking stuff, is that happening in a Service? Nov 13 18:54:31 *scratches head* Nov 13 18:54:52 guys any clue about my pthread_mutex_lock issue? lock() doesn't return, trylock says EBUSY but it's not locked anywhere else :S Nov 13 18:54:55 I "dont" have access to the sourcecode of the java library I am using. Nov 13 18:55:26 (The library is to make it easy to integrate an application with Point payment terminals Nov 13 18:56:22 AndreasBr: how many threads do you plan on having? Nov 13 18:56:38 Ok, now you are way above my level Nov 13 18:56:51 (I'm sorry to say, hehe ^^) Nov 13 18:57:31 At the moment this really just needs to be a BASIC app Nov 13 18:57:40 AndreasBr: who creates PayListener? How do you make sure that when activities come and go, the paylistener is still the same? For instance, if user leaves activity, android might free it when it's not visible Nov 13 18:57:59 alankila: My app is not that far yet Nov 13 18:58:09 AndreasBr: so you just make an instance in the activity? Nov 13 18:58:16 Ive modified the PIM (Point Integration Module) to work on android Nov 13 18:58:28 try { Nov 13 18:58:28 payPoint.setPayPointListener(this); Nov 13 18:58:36 or new PayListener() Nov 13 18:58:59 The best thing for ME i think is to have its own class Nov 13 18:59:09 AndreasBr: make sure no networking is done on the main thread, as it'll always cause exceptions (Android does not allow it) Nov 13 18:59:21 I ran into this one and was befuddled, it's just a heads up Nov 13 18:59:32 belgianguy: I believe the PIM creates its own thread Nov 13 18:59:36 The network shit is working Nov 13 18:59:43 I can send an amount to the terminal Nov 13 18:59:48 that works ok Nov 13 18:59:53 AndreasBr: ah, ok Nov 13 19:00:03 The PIM does that Nov 13 19:02:56 Ok Nov 13 19:03:05 Nvm, google before asking Nov 13 19:04:22 is there an "easy" (built-in) way to make a multichoice alertdialog that only allows X>1 choices? Nov 13 19:06:24 AndreasBr: to skip every possible false start, here's what I think you really need to do: you need a service that holds the paylistener, and which is a foreground service -- don't allow android task killer to ever murder it as long as it's communicating or listening to events. Nov 13 19:06:24 EditText test = (EditText)findViewById(R.id.txtAmount); Nov 13 19:06:24 int amount = Integer.parseInt(test.getText().toString()); Nov 13 19:06:27 Thats correct right? Nov 13 19:06:46 AndreasBr: new events arrive via the paylistener, and are put into an event list that is maintained by the service -- the service will remember ALL of them. Nov 13 19:07:34 AndreasBr: when activity is created, it starts thie service, or contacts a running instance, and asks it for the list of all events, and displays them by making the required views. It also registers itself as an interested party to the service using some API you define, so when new events occur, the activity will update. Nov 13 19:08:02 alankila: sounds cool Nov 13 19:08:13 your service needs to contain a thread or some such, it might be AsyncTask or whatever, which chats with the actual device/network/what-have-you. Nov 13 19:08:14 anybody here having an account for droidrzr.com? Nov 13 19:09:12 alankila: *sigh* Nov 13 19:09:14 AndreasBr: yes, that code looks correct to me, in sense, though usually you only call findViewById once and store the EditText instance to a member Nov 13 19:09:37 Ok, From that line alone I understand that I'm a _NOOB_ Nov 13 19:09:50 how do I force R.java generation? Nov 13 19:09:57 run aapt Nov 13 19:10:24 aapt? Nov 13 19:10:27 Delete it and build your app Nov 13 19:10:30 just kidding, run clean Nov 13 19:11:47 Guys? how to debug this mutex thing? :( Nov 13 19:11:59 alankila: It sounds kinda.. hmm Nov 13 19:12:02 Complicated Nov 13 19:12:28 I need to go locate some food here, brb Nov 13 19:12:29 that didn't help Nov 13 19:12:41 hey guys anybody? Nov 13 19:13:32 didnt find any, im back Nov 13 19:14:01 hey guys, opinion on an app name? Nov 13 19:14:10 LottoTrackr. Yay or nay? Nov 13 19:14:11 AndreasBr: every android application that needs to chat with some external service long term and accumulates state inside the application process is complicated Nov 13 19:14:13 alankila: Im grateful for your idea on using a service to keep things alive, but atm this is so basic that I only need to display stuff in UI Nov 13 19:14:24 ideal android applications are just fart soundboards that have no state worth keeping Nov 13 19:15:03 * Hodapp looks at alankila Nov 13 19:15:39 first you must defeat the risk of task killing, so you need Service -- something with no visible UI that nevertheless must be kept alive -- then you need to establish communication between activity and service, and you will encounter the marvel that is Binder Nov 13 19:16:06 I wonder how much research has gone into the generation of synthetic, non-sampled fart sounds. Nov 13 19:16:20 So what you are saying is that the system may decide to kill a thread, just for the heck of it? Nov 13 19:16:48 then you will grapple with the fact that Service connection establishment is asynchronous in nature, so you sort of can't call any service methods until the connection is established ... it's another little cringe-worthy affair of complexity, and all the Binder does is hand you a fucking object -- it's a travesty -- then you can finally start programming sort of like normal person would Nov 13 19:17:06 AndreasBr: not a thread, but the entire process Nov 13 19:17:12 So it will kill my app? o.O Nov 13 19:17:22 when its active Nov 13 19:17:33 in my hand, for the 5-10 minutes I will have it up without closing it? Nov 13 19:17:42 AndreasBr: yes, android programs are designed to only maintain illusion that they run. They can be killed any time their UI is not visible. Objects that aren't visible are also risk at being destroyed. Nov 13 19:17:58 my UI will be visible at all times Nov 13 19:18:06 and when its not, I'd like it not running Nov 13 19:18:10 atleast at this point =P Nov 13 19:18:13 so you think, but user presses home screen and your activity is fair game for android. Nov 13 19:18:25 Well, this wont be an app that is distributed Nov 13 19:18:48 Most likely it will just be showed off to the guys developing at work (I work at Point) Nov 13 19:18:49 you can avoid a lot of complexity if you can rely on the fact that multitasking is not required. Nov 13 19:19:01 Multitasking is NOT required at this point Nov 13 19:19:16 you can disable the screensaving feature, making the screen stay alive, and you can disable orientation changes from destroying the activity by forcing it in portrait, say. Nov 13 19:19:21 s/NOT/not/ :D Nov 13 19:19:39 then android is never at liberty of killing your activity as long as it's in screen Nov 13 19:19:52 Thus I dont need a service? (yet) Nov 13 19:20:38 Hodapp: I don't know about fart sounds, but I know that I had to read a few math papers for something of similar cultural value Nov 13 19:20:43 My basic and inital need was to be able to update the UI Nov 13 19:21:09 You have explained alot, but it still seems like rocket science just to ... change the name of a button Nov 13 19:21:23 Which in my head should be a fairly simple thing to do Nov 13 19:21:31 it is trivial, sure. Nov 13 19:21:43 not any one of us finds 'button.setText("foo")' especially complicated Nov 13 19:21:54 haha Nov 13 19:22:04 But that has to be done by the ui thread? right Nov 13 19:22:22 but you need it tied to some external system and invoked from another thread. Even then, it's a matter of modeling the relevant inter-thread communication somehow. Nov 13 19:22:46 and people are using AsyncTasks which are especially suited for this kind of work, though android makes them have certain corner cases as well Nov 13 19:22:53 isnt that the definition of rocket science? :D Nov 13 19:23:08 So what I really want is an asynctask? Nov 13 19:23:23 it is just a possibility. Nov 13 19:23:32 asynctasks hide the Handler from you Nov 13 19:23:33 ok Nov 13 19:24:33 you have a method called doInBackground that runs in non-UI thread from some singlethreaded executor pool or multithreaded one depending on android version, and it can do stuff like 'publishProgress(T)' which allows it to report something, and that causes UI thread to run onProgressUpdate(T). Nov 13 19:25:13 in the background, there is probably a Handler involved that passes the object given to publishProgress to a Runnable that executes onProgressUpdate Nov 13 19:25:24 ok Nov 13 19:25:31 but afterwards you can model the class T any way you want, to have it encode the UI updates you want somehow Nov 13 19:25:56 "use onProgressUpdate() method of AsyncTask. It is performed on UI thread" Nov 13 19:26:53 hmm … any new sdk? when i go to d.android.com the file is just called blah-bundle.zip; i dont see any version info Nov 13 19:27:49 ok, new compat lib and sdk :) Nov 13 19:28:00 does it have the actionbar, that is the question, bwahaha Nov 13 19:28:17 I hate being a noob at something, haha Nov 13 19:28:46 It feels like you are standing at the bottom of mnt everest, and looking uo Nov 13 19:28:47 up Nov 13 19:29:02 When even the "simplest" task is at the peak Nov 13 19:30:32 Seems to be atleast Nov 13 19:30:38 that is why one has to make studies, or just resign to solving problems as they occur, or look through code that does stuff with some patience Nov 13 19:31:02 in case of android, one has to have roughly the right model of the framework because it's unusual and huge part of android applications, also huge part of the pain Nov 13 19:31:37 I bet the stuff I am looking for is in one of the example sources asweell? Nov 13 19:31:53 the key to android is that it manages object lifecycles for you, for the most part. It creates your activities, views and fragments (though you usually partake in the creation of the latter objects somehow). The objects can in theory vanish underneath you, and be replaced by new copies, the process itself can die and be recreated. Nov 13 19:33:23 for these reasons android applications need to be unusually careful about how and where to keep their state. This makes android applications far harder to write than desktop applications (single process which never dies abruptly, as many threads as you want, you control them pretty much fully), web applications (many sophisticated server-side frameworks exist that nearly reach convenience of desktop application writing). Nov 13 19:33:51 I have experience writing both desktop and web applications in java, and they aren't even halfway as difficult to write than android applications Nov 13 19:34:17 it's all relative Nov 13 19:34:40 once you 'get' the lifecycle, who lives, who dies, who persists Nov 13 19:34:48 it's not so bad Nov 13 19:35:22 * kbs is in a constant state of android nirvana. my process is always extinguished. Nov 13 19:35:42 Im still lost when it comes to UI Nov 13 19:35:43 haha Nov 13 19:36:40 the lifecycle is easy; the consequences of all the callbacks and reactive programming are not Nov 13 19:37:43 http://creating-industrial-solutions.com/2011/06/threads-in-android-part-2-updating-the-ui/ Nov 13 19:37:43 ooh Nov 13 19:37:47 I think I found what I need Nov 13 19:38:24 but now I need to go to the store and buy a frozen pizza..! dang i wish this was easier Nov 13 19:38:36 fuking awesome; Nokia is going to make their new maps API available for android devs Nov 13 19:38:59 who the f*** uses Nokia these days? Nov 13 19:38:59 g00s: is that a good thing? Nov 13 19:39:03 at least somebody takes ownereship of maps apis on android :) Nov 13 19:39:08 : p Nov 13 19:41:17 has anyone been able to compile aosp for toro yet? :D Nov 13 19:43:46 g00s: API 17 is out Nov 13 19:44:01 \o/ ;P Nov 13 19:44:39 so in 4.2, does anybody know how to use pm/am/whatever to install an already installed APK for user 0 to a secondary user ? Nov 13 19:46:22 does anyone know how the encrypted volume (/mnt/asec) on Android prevents people from forwarding apps to other phones? Nov 13 19:47:07 I am trying to understand it, but if it is like any other encrypted volume, then while the device is running, the volume is decrypted and the APKs are able to be read Nov 13 19:47:46 anyone managed to update to api 17? sdk manager seems broken Nov 13 19:48:44 xorgate: i always just nuked and reinstalled; its had issues for years Nov 13 19:48:51 working fine... thanks for the reminder :) Nov 13 19:49:22 meh Nov 13 19:49:23 g00s: I see nothing for BTLE Nov 13 19:49:48 dragorn: FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFUUUUUUUUUUUUUUUUUUUUUUUUUUUUU Nov 13 19:51:10 dragorn: thanks for looking into that already Nov 13 19:51:24 g00s: inquiring minds and all that Nov 13 19:51:41 g00s: nothing in the release highlights and nothing in a quick search of the BT API stuff Nov 13 19:57:45 looks like new NDK has clang Nov 13 19:57:52 goodie, I can drop gcc compat in some source I got Nov 13 20:00:00 4.2 has nested fragments ? Nov 13 20:00:23 cool, support lib has that also Nov 13 20:01:18 that makes lots of things with viewpagers having fragments easier :D Nov 13 20:01:38 yeah I welcome this Nov 13 20:01:38 and low latency audio I heard :-) Nov 13 20:01:43 o// Nov 13 20:02:09 Drakonite ^^^ Nov 13 20:02:43 wut Nov 13 20:02:52 are you claiming Fragments are now usable ? Nov 13 20:03:25 lets see how many bugs there are first :P Nov 13 20:03:44 for a start firefox crashes on 4.7 N7 Nov 13 20:04:02 MX Player doesn't run (unsupported version) Nov 13 20:04:07 bubbleguuum: running android 4.7 eh Nov 13 20:04:19 hi all Nov 13 20:04:22 yup 4.7 :p Nov 13 20:04:41 i have several questions regarding using the Camera class Nov 13 20:06:08 my app has an activity that feeds the rear camera to the screen (like in API Demos) and can take pictures Nov 13 20:06:45 Im back Nov 13 20:06:52 is there a way to have continuous focus of the camera? Nov 13 20:07:16 i call camera.autoFocus() before taking a picture Nov 13 20:07:45 that sets the focus, but then it stays at the focus level until the next time i take a picture Nov 13 20:08:02 and before the first picture the preview is a bit fuzy Nov 13 20:10:22 is there a way to dynamically load code from for example sd card? say i want to download a precompiled lib of some sorts Nov 13 20:11:11 can't wait for Android 6.0 with fragment API deprecated for something much simplier... Nov 13 20:12:41 Chainfire: "most likely the reason they don't sell here is because the new generation of Nexus devices (incl 7) is heavily betting on the Play ecosystem, which is not properly supported here" Nov 13 20:12:54 most likely they don't sell there because your country's laws make it hard for them to do so. Nov 13 20:12:56 which eclipse version is included in the adt bundle? Nov 13 20:12:57 FWIW. Nov 13 20:13:13 ctate> never made a difference for the other Nexus devices ... :) Nov 13 20:13:15 it's not like google doesn't *want* to sell worldwide; i mean sheesh Nov 13 20:13:17 eclipse included in the adt bundle ? Nov 13 20:13:22 ctate> more importantly, did you see my comments re: AAPT ? Nov 13 20:13:26 i did not! Nov 13 20:13:39 and uh i have days of scrollback to sort through, ugh Nov 13 20:13:42 g00s: yes, this one: http://developer.android.com/sdk/index.html Nov 13 20:13:52 alankila: Do you happen to still be here? :) Nov 13 20:13:56 foudn it! Nov 13 20:13:59 yes. Nov 13 20:14:02 Chainfire: many many thanks for the repro Nov 13 20:14:17 Im reading about asynctask, and it says that it can only be executed once Nov 13 20:14:27 Does that make it a bad solution for ui updating? Nov 13 20:14:43 AndreasBr: not really. Nov 13 20:14:51 Normally one would want something one could use multiple times? Nov 13 20:14:52 ah, found the release notes, it's 3.8 Nov 13 20:14:53 what the fuck … i already have an eclipse why would they bundle another one Nov 13 20:15:08 g00s: To see your reaction Nov 13 20:15:08 AndreasBr: you don't have to ever let the asynctask finish Nov 13 20:15:20 Chainfire: are you building with Eclipse, or with ant directly, or what/ Nov 13 20:15:21 g00s: you don't have to install the bundle if you have eclipse, there's "Setting Up and Existing IDE" in the menu Nov 13 20:15:25 or you can let it finish when there's nothing more to do Nov 13 20:15:34 eh, afaik the main thing is done in a worker thread Nov 13 20:15:35 I'm gonna delete my eclipse and use the adt bundle though because I like it's splash screen :P Nov 13 20:15:47 but the onPostExecute is ran using ui thread, or main thread Nov 13 20:15:49 ctate> do you happen to know on 4.2 how to make an app thats installed for the main user, appear for a different user ? can't seem to find a PM command that's relevant Nov 13 20:15:51 s/it's/its/ Nov 13 20:15:57 ctate> compiled with standard Eclipse and ADT Nov 13 20:16:09 AndreasBr: again, see onProgressUpdate / publishProgress Nov 13 20:16:18 you can use that while your other thread is still executing in doInBackground Nov 13 20:16:19 when my activity is launched by an intent it does a lot of initialization based on that intent, when the screen orientation changes oncreate is called again but the intent isn't the same anymore? is this normal? do i have to save the intent manually? getIntent() doesn't result in the same one? Nov 13 20:16:20 Ahaa Nov 13 20:16:33 You mean somehing like an asynctask that "lookforsomethingtoupdate, Nov 13 20:16:34 Chainfire: after it's already installed for the owner? Nov 13 20:16:40 yes Nov 13 20:16:43 if it finds it, publishprogress ? Nov 13 20:17:03 not really. Nov 13 20:17:07 oh, ok Nov 13 20:17:45 angelos: menu ? i we used to just unzip it; wtf have they done Nov 13 20:18:00 or well, I don't know if we agree. I mean that you'd listen to the PayListener events in the doInBackground and when you get them, publishProgress() those suckers Nov 13 20:18:08 my camera preview turns black and stays black when i turn my screen off and on... Nov 13 20:18:15 then handle the events in UI thread context in the onPostUpdate method Nov 13 20:18:30 alankila: Yes, but I'd want to update ui from other places aswell Nov 13 20:18:50 so, is it recommended to switch to the new bundle pack (eclipe + sdk + etc..) or we can keep our current environment and update components?? Nov 13 20:18:50 nothing stops you. Nov 13 20:18:51 g00s: menu on that page, which leads to http://developer.android.com/sdk/installing/index.html Nov 13 20:19:25 What about Activity.runOnUiThread(Runnable) ? Nov 13 20:19:28 or rather http://developer.android.com/sdk/installing/installing-adt.html which includes the "old" instructions Nov 13 20:19:33 Chainfire: i don't think there's a way to do that, actually Nov 13 20:20:05 without redoing the 'adb install' operation for the target user Nov 13 20:20:16 but .. there is not way to dl the sdk w/o downloading eclipse in the bundle ? Nov 13 20:20:19 ctate> ok, that's what I was afraid of Nov 13 20:20:36 ctate> is there any way to determine if the app is installed for a certain user ? Nov 13 20:20:43 AndreasBr: what about it? you can use it if you like. Nov 13 20:20:48 ctate> because that's the second thing I couldn't figure out Nov 13 20:20:56 g00s: there is, on the page I just linked Nov 13 20:20:58 Is it a bad way to do it? Nov 13 20:21:44 AndreasBr: I really have no opinion on the matter. Nov 13 20:21:49 Ok Nov 13 20:22:01 i will just have to try then :D Nov 13 20:23:04 Chainfire: adb shell pm list packages --user 10 Nov 13 20:23:09 or whatever the user handle is Nov 13 20:23:38 the owner is always user 0 Nov 13 20:24:32 ctate> ah right! Nov 13 20:24:34 thanks Nov 13 20:25:00 is it ok to call stopSelf() from onStartCommand() ? Nov 13 20:25:30 Chainfire: also, of course you can just look at Pm.java in the source ;) Nov 13 20:29:24 regedit: of course Nov 13 20:29:34 just like it's valid to call finish() from an activity's onCreate(). Nov 13 20:29:43 ok thanks Nov 13 20:29:47 why on earth would my mutex not work :| Nov 13 20:31:23 Chainfire: pm ok? Nov 13 20:32:34 ESphynx: they dont work on eC :P Nov 13 20:32:43 g00s :P Nov 13 20:32:49 g00s seriously man... http://pastebin.ca/2251329 Nov 13 20:33:01 I *obviously* unlock the mutex in the other thread Nov 13 20:33:08 yet pthread_mutex_lock never returns... wtf? Nov 13 20:34:23 is it ok to call startForeground() from within a Service's onCreate() method? Nov 13 20:35:14 regedit: of course :) Nov 13 20:35:36 ctate: why thank you kind sir :) Nov 13 20:35:59 g00s it doesn't make any sense, right? Nov 13 20:36:15 i havent used pthreads in a while Nov 13 20:36:42 but I mean .. pthread_mutex_lock / pthread_mutex_unlock ... Works for me all the time. Nov 13 20:36:47 here it's just being silly? Nov 13 20:36:57 is it an emulator bug? Nov 13 20:37:28 have you verified some first principles, like it being the same mutex instance? and the mutex was properly initialized? Nov 13 20:37:39 ctate: yes I've verified both of these things Nov 13 20:37:54 the pthread_mutex_t struct only as a volatile int 'value' member... which changes values :S Nov 13 20:38:01 but i don't know how the implementatino works Nov 13 20:38:28 http://pastebin.ca/2251329 -- in my pasteBin the lockMytex thing is the struct containing my pthread_mutex_t object Nov 13 20:38:54 Locking/Unlocking is that implemetnation int value 49233925 / 49233926 Nov 13 20:39:06 it changes values but looks like it's a count or something? Nov 13 20:39:14 how can i manipulate a RemoteViews/layout-XML from within a Service? in this case I want to edit the text of a Button in my foreground-Service's Notification (which is a RemoteViews layout XML) Nov 13 20:39:16 implementation thing :| Nov 13 20:40:25 also pthread_mutex_unlock returned 0 Nov 13 20:40:32 dunno why the damn thing won't lock! Nov 13 20:43:26 ctate: any ideas? ^ Nov 13 20:44:04 this is totally messed up. I'm going to write a simple test :| Nov 13 20:44:35 regedit: not a UI guy, sorry :) Nov 13 20:45:02 i have to wonder; if nokia is partnering w/firefox as plan B in case wp8 doesnt work well for them; kinda makes sense … maps & location = killer apps for mobile; wp8 seems like a platform for higher end phones but nokia used to have tons of lower end phones; firefoxOS works on low end phones. etc Nov 13 20:45:14 seems like there's no findViewById() from a Service.... Nov 13 20:45:22 regedit: :| Nov 13 20:46:21 one thing nokia does smartly with their maps is employ fedex & ups for data collection Nov 13 20:46:38 dunno why DHL; heh … maybe they were always in the weeds or something :P Nov 13 20:46:48 g00s: wha? might you know anything about manipulating UI from a Service? Nov 13 20:47:01 regedit: no i dont :) Nov 13 20:47:15 unless you mean UI as 'notifications' Nov 13 20:47:43 anyone know where I could ask hw related q's Nov 13 20:47:46 I do: don't Nov 13 20:47:56 at XDA they don't know Nov 13 20:48:07 g00s: indeed, i'm dealing with a foreground-Service's mandatory Notification Nov 13 20:48:20 which i'm guessing you want to animate or something Nov 13 20:48:33 ctate: who me? Nov 13 20:48:38 yes :) Nov 13 20:48:47 SimonVT: was that directed at me? Nov 13 20:48:54 <`teh1> having crazy problem with android sdk O_O Nov 13 20:48:58 no Nov 13 20:49:02 ah, k Nov 13 20:49:05 the notification has a button, when pressed i want the service to and update the button Nov 13 20:49:21 <`teh1> when it pulls op the FBDialog it is completely invisible and doesnt draw things below it??? Nov 13 20:49:30 <`teh1> any ideas? Nov 13 20:50:49 g00s: ^ you mean that? Nov 13 20:50:56 <`teh1> http://cl.ly/image/0N051z0C2B3Z Nov 13 20:51:10 <`teh1> see it doesnt draw things below it as i should if it was really there, but it doesnt appear :[ Nov 13 20:51:25 <`teh1> also clicking anywhere on it hides it Nov 13 20:51:26 <`teh1> so confused Nov 13 20:51:55 <`teh1> any ideas? Nov 13 20:52:24 access denied on your image Nov 13 20:52:43 will the android xmlns ever support an attribute inheritance model? Nov 13 20:52:55 Could someone recommend an ssl cert provider that would be in Android's trusted store? Nov 13 20:53:16 duplicate code keeps me awake at night Nov 13 20:58:17 Hi guys. I'm new to android development and I was wondering: How should I run network sockets in a separate thread? I need to be monitoring for UDP broadcasts, which will need to be handled, and I need to establish long-running connections to servers. Nov 13 20:59:40 <`teh1> AndreasL: AsyncTask? Nov 13 21:00:35 `teh1: I read in the docs that using AsyncTask is discouraged for long-running operations.. Nov 13 21:00:41 <`teh1> oh :P Nov 13 21:00:52 new Thread(Runnable)? Nov 13 21:00:57 <`teh1> ^was about to say Nov 13 21:00:59 Omg! Another Andreas! Nov 13 21:01:06 NORSK OGSÃ… Nov 13 21:01:09 Sorry for caps :D Nov 13 21:01:15 i'm at a loss here.. i have an xml layout, and custom layout that subclasses relativelayout… how do i connect them so when i inflate the xml layout, it uses my custom class? this is for dialogfragment if that matters Nov 13 21:01:17 must be SAN Nov 13 21:01:29 <`teh1> AndreasBr: shows how unoriginal your name is Nov 13 21:01:33 pft Nov 13 21:01:42 is this an english only channel ? :D Nov 13 21:01:43 originality is overrated Nov 13 21:01:48 yes please :P Nov 13 21:01:54 AndreasBr: hehe, morsomt Nov 13 21:02:01 damn ivdorelian Nov 13 21:02:05 Oh, right. I guess 'morsomt' isn't a very established English word Nov 13 21:02:54 is there something equivalent to Camera.setDisplayOrientation(angle) for API10? Nov 13 21:03:45 alankila: So i'm doing something in a wrong way.. Which obv didnt work. I tried using runOnUiThread in my payListener class, how would I do this the right way? Nov 13 21:03:53 I would need to use a handler somehow, yes? Nov 13 21:04:07 I suspect the static method already has a valid handler for ui thread Nov 13 21:05:04 http://pastebin.com/05yKesmc Nov 13 21:05:21 ew Nov 13 21:05:29 Indeed Nov 13 21:05:34 why do you think that would show anything? Nov 13 21:05:42 I dont know, because im stupid? Nov 13 21:05:44 *grins* Nov 13 21:05:49 whoa nested fragments Nov 13 21:05:51 yo dawg Nov 13 21:06:09 There must be some easier way to create a new textview? Nov 13 21:06:10 we heard you like fragments Nov 13 21:06:15 well the textview isn't just magically going to appear into the screen. It needs to be part of the activity's view hierarchy Nov 13 21:06:40 rght, in my layout I have a scrollview inside a linearlayout Nov 13 21:06:59 Or should it be the other way around? :P Nov 13 21:07:15 its correct the way you're stating it in English, who knows what's happening in the code Nov 13 21:07:24 the textview now goes inside the LinearLayout Nov 13 21:07:26 I believe evancharlton suggested a scrollview with linearlayout inside Nov 13 21:07:46 Sicp: say what? Nov 13 21:07:58 what what? Nov 13 21:08:02 So both scrollview and textview are children of linearlayout? Nov 13 21:08:11 anyone using jake whartons viewpagerindicator? Nov 13 21:08:22 i imagine JakeWharton might be Nov 13 21:08:26 or Scrollview -> LinearLayout -> TextView ?= Nov 13 21:08:36 ^ that way Nov 13 21:08:38 ok nothing makes sense here!! Nov 13 21:08:51 Sicp: *changing* Nov 13 21:09:21 ehm Nov 13 21:09:32 In layout editor, I add the scrollview first right? Nov 13 21:09:44 aren't you XML'ing? Nov 13 21:09:47 so much easier Nov 13 21:09:48 umm Nov 13 21:09:49 no¨ Nov 13 21:09:59 I like graphical stuff :D Nov 13 21:10:18 never worked there personally, but I guess..yea, you add the scrollview first and drag the linear layout to there Nov 13 21:10:24 and inside the linear you add a textview Nov 13 21:10:24 http://pastebin.com/PdQBNZXa Nov 13 21:10:58 ctate> yes you can PM me Nov 13 21:10:59 uno secondo Nov 13 21:11:06 (I was afk) Nov 13 21:11:19 Chainfire: People ask about that? o.O Nov 13 21:11:54 AndreasBr: yes, that's standard etiquette here: no unexpected PMs please Nov 13 21:11:58 o.O Nov 13 21:12:06 Weird, but OK Nov 13 21:12:13 people are busy here Nov 13 21:12:19 can't just go into a pm and start talking Nov 13 21:12:26 <-- lol Nov 13 21:12:42 pthread_mutex_unlock is completely b0rked. Nov 13 21:12:46 how can i tell an xml file to use a custom layout i've written? Nov 13 21:12:47 True, but whats the difference between answering here, or answering on pm ? (The question that might be "got 2 seconds?" Nov 13 21:12:56 xml layout, rather Nov 13 21:12:58 PM is a more personal engagement Nov 13 21:13:04 we don't agree to this just by hanging in here Nov 13 21:13:10 True Nov 13 21:13:34 questions are better answered publicly Nov 13 21:13:36 My respond to any question such as that would be "of course" Nov 13 21:13:39 most of them anyway Nov 13 21:13:57 jeppy`: yeah, then others in the channel may see useful information, or even others may find the irc log googling Nov 13 21:14:44 i don't believe I've ever ran into an irc log while googling Nov 13 21:14:53 yeah Nov 13 21:14:55 AndreasBr: if you randomly PM me a question, i'll almost certainly tell you "ask on the public channel please" Nov 13 21:15:00 then you haven't googled enough jeppy` Nov 13 21:15:07 I just ignore pm's :F Nov 13 21:15:12 ctate: I agree on that Nov 13 21:15:12 and if random people PM me with other kinds of stuff, i'll probably just ignore them Nov 13 21:15:16 hmm, anyone tried the 4.2 emulator yet? Nov 13 21:15:23 I have a "problem" Nov 13 21:15:24 belgianguy no Intel image yet? :| Nov 13 21:15:24 ESphynx, I have to doubt that Nov 13 21:15:34 I can't get it to start (works when I run it in 4.1.2) Nov 13 21:15:49 ESphynx: nope, not yet, but even the regular one won't work Nov 13 21:15:52 well several really, but this one is: I have a String called test, in a class called payListener Nov 13 21:15:56 first time ever it does that Nov 13 21:16:02 this works here, AndreasBr http://pastebin.com/UWjCFmD8 -- adds a button Nov 13 21:16:07 I would like to update the ui, and add a textview Nov 13 21:16:09 *looks* Nov 13 21:16:25 The TV needs to be dynamic -.- Nov 13 21:16:37 But thanks Nov 13 21:16:51 how much have you read about this topic so far? Nov 13 21:17:10 tutorials/etc.. Nov 13 21:17:14 I mean, all the topics I see either .. Seems overly complicated for accomplishing such a small task Nov 13 21:17:24 Hi, Can I ask a question related with memory leaks here? Nov 13 21:17:25 maybe you started at the wrong place Nov 13 21:17:44 check the Docs first, and if you are more visual, try looking for Marakana Android tutorial on YouTube Nov 13 21:17:49 Like the lower part here Nov 13 21:17:49 http://creating-industrial-solutions.com/2011/06/threads-in-android-part-2-updating-the-ui/ Nov 13 21:17:55 dude.. Nov 13 21:18:04 that's not what you're looking for at all Nov 13 21:18:17 charlypu: ask Nov 13 21:18:19 this is asynctask stuff Nov 13 21:18:20 I'm getting "emulator: ERROR: This AVD's configuration is missing a kernel file!!" Nov 13 21:18:29 and I don't know which one :s Nov 13 21:18:38 as I just updated the SDK Nov 13 21:18:43 well it should be in asynctask at least Nov 13 21:18:45 I'm having problems with memory leak from GLThread, of the SurfaceView Nov 13 21:18:52 or thbis Nov 13 21:18:53 http://crodrigues.com/updating-the-ui-from-a-background-thread-on-android/ Nov 13 21:19:17 you want to add/remove the textview dynamically Nov 13 21:19:18 ? Nov 13 21:19:19 It has reference to the Activity Nov 13 21:19:32 So when the activity tries to get destroyed it is not erased Nov 13 21:19:36 does 'set dex.force.jumbo=true' influence the maximum number of methods that can be in a dex file? or is this just 'strings' ? Nov 13 21:19:46 i think the method max was 65k Nov 13 21:19:58 Sicp: I have a String called test in a Listner. Every time it gets a .. "ping", I want it to add a new textview to the linearlayout with the text contained in 'test' Nov 13 21:20:11 Listener* Nov 13 21:20:19 I'm coming from Qt and C++, and being used to its Model/View system, I can't seem to find anything equivalent in Android. What's the android way of keeping, say, a List view in sync with underlying data? Nov 13 21:20:23 that isn't "simple".. Nov 13 21:20:29 but what you linked to will do the job Nov 13 21:20:40 well it isn't simple for me at least, maybe other people here do that everyday Nov 13 21:20:56 To me that should be really simple, All I want to do is add a textthingy, and set the text o.O Nov 13 21:21:09 atleast in my theory :P haha Nov 13 21:21:19 threads complicate stuff though Nov 13 21:21:42 you're doing it dynamically man, while the application is running Nov 13 21:22:00 How would you go about changing something static then? Nov 13 21:22:05 nm, seems there's some stuff I didn't install Nov 13 21:22:12 but it didn't prompt me about it :s Nov 13 21:22:16 not sure I understand what you're asking; what's "something static"? Nov 13 21:22:30 Say... Change the text on a button Nov 13 21:22:41 .setText() Nov 13 21:22:46 From another thread. Nov 13 21:22:56 as in: Not main thread Nov 13 21:23:12 <`teh1> getActivity().runOnUiThread() Nov 13 21:23:19 *tries* Nov 13 21:23:34 <`teh1> you may not need getActivity() if you are not using Fragments, for instance Nov 13 21:23:46 Seems like I didnt Nov 13 21:23:50 Well Nov 13 21:23:54 Didnt even have it Nov 13 21:24:00 he method getActivity() is undefined for the type PayListener Nov 13 21:24:05 AndreasBr: http://developer.android.com/guide/components/processes-and-threads.html#WorkerThreads may be of interest Nov 13 21:24:17 andreas: *checks out* Nov 13 21:25:07 eh, that one starts out in main thread Nov 13 21:25:11 I start out in a worker thread Nov 13 21:25:28 also asynctask can only be ran once Nov 13 21:25:36 I need it to run every time there is an event Nov 13 21:25:47 AndreasBr: you should still be able to use the post principle though, I reckon Nov 13 21:27:34 ivdorelian: any idea? Nov 13 21:28:09 f*** ui, who needs to know what happens anyway Nov 13 21:30:01 uhmm Nov 13 21:30:03 AndreasBr: why not System.out.print() until the UI part clears up? :p Nov 13 21:30:07 null the reference in onDestroy? Nov 13 21:30:19 im not sure without seeing code Nov 13 21:30:19 belgianguy: And where would that be shown? :P Nov 13 21:30:30 in your debugger Nov 13 21:30:32 belgianguy: I'm ready to ditch android, haha =P Nov 13 21:30:44 Well how does that help me with showing without an usb cable? Nov 13 21:30:45 =P Nov 13 21:30:58 AndreasBr: not at all, didn't know that Nov 13 21:31:03 :DS Nov 13 21:31:03 what is *wrong* with Github! Nov 13 21:31:08 pebkac Nov 13 21:32:11 AndreasBr, you can still debug over WiFi, if you don't have usb cable Nov 13 21:32:35 Greetings. I'm writing a simulator & set of tools, and would want to run it on android. Did not yet do much beyond getting SDK and looking through it Nov 13 21:32:45 I want to show the messages from the payment terminal in the app Nov 13 21:32:46 heh Nov 13 21:33:10 My question is... I don't want to bother with android development too much, and I need to port my simulator as it is written in C, plus android specific things. Should I look for anything specific, is that possible? Nov 13 21:33:34 I know you can write in C, but I don't know if that is possible without invoking some complex processes or handling a lot of things Nov 13 21:34:25 frack this, putting my listener back in main class Nov 13 21:35:02 ivdorelian: it is impossible to have access to the GLThread... >( Nov 13 21:35:31 ivdorelian: so its impossible to null it Nov 13 21:36:14 hm, i guess they didnt get around to adding bt support to the emu yet :/ Nov 13 21:37:03 AndreasBr: Professional Android 4 Development has a pretty detailed example Nov 13 21:37:27 I think you mind find that worth taking a look at, instead of trial/error mode :) Nov 13 21:37:47 charlypu well, i'm not familiar with GL stuff, but just a thought: why do you care? it can't waste a lot of memory can't it? Nov 13 21:38:00 and after the GLThread is gone so is the reference, right? Nov 13 21:38:05 is it really an issue? Nov 13 21:38:28 it also explains the Handler, SQLite, Services, UI handling etc Nov 13 21:38:47 just make sure you have the latest version, there are books that are outdated, which are called the same Nov 13 21:39:06 ivdorelian: The GLThread never dies, so it stays there. The Activity consumes about 10MB. several activities... Nov 13 21:39:51 charlypu can you show some code? it will help others help you too Nov 13 21:41:49 ok, anyway to show the code here? Nov 13 21:41:58 i have a DialogFragment that uses a custom view i've written… if i set the background color to transparent, it ends up default dark gray… how can i actually make it transparent so i can redraw everything on my own? Nov 13 21:42:21 charlypu use pastebin Nov 13 21:42:53 HAHA Nov 13 21:42:55 I DID IT Nov 13 21:43:00 I frikkin did it Nov 13 21:43:08 you did? Nov 13 21:43:11 OH yeha! Nov 13 21:43:13 congrats! Nov 13 21:43:18 i don't know what you did Nov 13 21:43:26 I did something :-/ Nov 13 21:43:32 Greetings all Nov 13 21:43:33 but i love those 'ah ha' moments Nov 13 21:43:51 i just had one, but it quickly led to more questions Nov 13 21:44:50 ultra-: small steps Nov 13 21:45:09 ivdorelian: http://stackoverflow.com/questions/13367800/android-game-leaking-memory-because-of-glthread-reference I just posted it there, just because I dont have access to the GLThread or SurfaceView code (is from android source), instead I think MAT images are more "usefull" Nov 13 21:45:18 yeah there's been millions of those… i enjoy them tho :) Nov 13 21:45:47 how far i've come in a few months… from not much understanding of this to a 100% custom (necessary) interface Nov 13 21:46:34 I have a new problem, that probs has an easy fix Nov 13 21:46:40 I have a textbox in my layout Nov 13 21:46:50 But the damn keyboard stays open :@ Nov 13 21:46:54 i'm surprised sdk 3 and 4 still show up in the sdk updater :P Nov 13 21:47:13 Im running API10 Nov 13 21:47:17 Android 2.3.3 Nov 13 21:47:53 AndreasBr: 05 21:32:17< tophyr> whomp_: (InputMethodManager)getSystemService(Context.INPUT_METHOD_SERVICE).showSoftInput(...) Nov 13 21:48:27 charlypu ok, personally i'm not sure what the solution could be. maybe someone else will jump in. Nov 13 21:48:30 my bad. Nov 13 21:48:38 clever: For some reason I cant remove focus from the textbox Nov 13 21:48:41 will that fix it? Nov 13 21:48:55 AndreasBr: no clue, ive yet to actualy do any text input in my apps Nov 13 21:49:06 ok Nov 13 21:49:12 i've never used a built in view, but maybe request focus from something else Nov 13 21:49:13 Any examples / suggestions on making a large bitmap be drawn, unscaled, into an ImageView that has scrollbars? Nov 13 21:50:48 nevermind, found the link I was looking for Nov 13 21:52:10 dang, almost 2300. Nov 13 21:52:43 it seems changing which activity in my app responds to the launcher intent invalidates existing icons that have been dragged onto the home screen? Nov 13 21:53:25 correct Nov 13 21:53:30 Hey. Nov 13 21:53:40 What happend when I try to compile, Nov 13 21:53:45 xml file. Nov 13 21:53:49 ctate: ah, i better not do that then ;) Nov 13 21:53:55 Is SharedPreferences an OK method to use to store an IP-address? Nov 13 21:53:58 spifff: :) Nov 13 21:54:06 so I dont have to hardcode it Nov 13 21:54:07 I mean with "eclipse" and android plugin. Nov 13 21:54:22 SharedPreferences is a fine way to store a set of key/value pairs Nov 13 21:54:29 and modify them occasionally Nov 13 21:54:44 if you're just storing one thing, why bother putting it in SharedPreferences instead of just writing the damn thing out to disk directly? Nov 13 21:54:53 humm Nov 13 21:54:57 idk Nov 13 21:54:59 ? Nov 13 21:55:18 you mean like an ini file? Nov 13 21:55:46 no, i mean like a file that has four bytes in it Nov 13 21:55:49 which are the IP address :) Nov 13 21:55:59 is there a simple way to run a pice of code in a new thread ?? Nov 13 21:56:19 ctate: doesnt that take code? Nov 13 21:56:28 sharedprefs looked pretty simply Nov 13 21:56:34 without working working with asynctask.. Nov 13 21:56:58 AndreasBr: i'd go for SharedPrefs even for a single thing… invariably you'll want to store a second thing in the future, and then you'd have to deal with migrating the data from the flat file into shared prefs when users upgrade etc Nov 13 21:58:02 why .. Nov 13 21:58:02 The method getActivity() is undefined for the type MainActivity Nov 13 21:58:06 is my API too old? =P Nov 13 22:16:32 Hey guys, you alive? Nov 13 22:16:38 Any idea why TextView txtIp = (TextView)findViewById(R.id.editText2); gave me null? Nov 13 22:16:41 nope. all dead Nov 13 22:16:43 Probs a rookie mistake Nov 13 22:16:49 invalid ID? Nov 13 22:17:19 Ah Nov 13 22:17:22 mighht be it Nov 13 22:17:23 or maybe you missed out a setContentView Nov 13 22:17:40 check the import of the R class Nov 13 22:17:41 ? Nov 13 22:18:16 the ID may be valid elsewhere, otherwise you'd get an unidentified identifier error Nov 13 22:18:21 but maybe it's not in the current layout Nov 13 22:18:46 I think I have it now Nov 13 22:18:52 nexus s people stuck on 4.1.2 Nov 13 22:18:54 good job Nov 13 22:21:42 A last thing before I go to bed Nov 13 22:21:50 same with Xoom, apparently, g00s Nov 13 22:21:51 I want to be able to unfocus the textedit Nov 13 22:22:02 I only have 1 textbox thingy Nov 13 22:22:38 i think google should really look at tightening the belt on the os so it runs on lower end hw Nov 13 22:23:02 AndreasBr, why do you want to do that? Nov 13 22:25:09 google could also tighten product launch days were stuff is sold out in 10 mins :p Nov 13 22:25:15 leslie, nvm, i did it Nov 13 22:25:24 Because I dont want it up hiding half my app? Nov 13 22:25:24 ok Nov 13 22:25:30 g00s: yea it was nice that the nexus s had factory images for 2.3, 4.0 and 4.1 Nov 13 22:26:00 I got all the way up to the payment screen. pressed accept… and nothing. it just sat there for 10 minutes Nov 13 22:26:07 dunno if its true, but i thought 4.x unofficially required a dual core; i think the nexus s had a single core Nov 13 22:26:16 hmm, VZ including the amazon store on Droid DNA Nov 13 22:26:28 AndreasBr, there's an API to hide the virtual keyboard Nov 13 22:27:05 leslie: i got to press the "Accept and Buy" button and got the "Your purchase will complete momentarily …" message and then nothing. :( Nov 13 22:27:07 InputMethodManager imm = (InputMethodManager)getSys Nov 13 22:27:08 hblaba Nov 13 22:27:16 louielouie, yup, same here Nov 13 22:27:26 After all Google is small company inexperienced in the web... Nov 13 22:27:36 Anyways, my app works kinda nice now, going to bed, and showing off to our development team tomorrow Nov 13 22:27:38 :) Nov 13 22:27:39 gnight Nov 13 22:27:43 night Nov 13 22:27:56 Where should I put stuff that runs in the background of all the activities of my app? I mean, where do I instantiate it? Nov 13 22:27:59 one last thign Nov 13 22:28:04 there are no number-of-cores requirements, nor have there ever been Nov 13 22:28:05 cn we use ORMs with GCM? Nov 13 22:28:06 how do you shut down an app with a button? :D Nov 13 22:28:13 AndreasL, create a service Nov 13 22:28:23 I want to shut it down completely Nov 13 22:28:28 terminate the app Nov 13 22:28:33 why/ Nov 13 22:28:36 err, why/ Nov 13 22:28:38 AndreasBr, don't bother Nov 13 22:28:40 sldkfjalksfdhl stupid keyboard. Nov 13 22:28:41 leslie: I see, thanks Nov 13 22:28:50 Can I use ORMs with Google Cloud Messaging? Nov 13 22:28:52 Because I dont want to go into applications and do it manually? Nov 13 22:28:53 it is bad UX design to put an explicit 'quit' or 'exit' action into your activity Nov 13 22:28:57 Because the moon aint green Nov 13 22:29:05 Can I use Object-RElational Mapping with Google Cloud Messaging? Nov 13 22:29:08 well yes, but in this case I need it. Nov 13 22:29:08 ctate: thats the official word Nov 13 22:29:14 AndreasBr: why? Nov 13 22:29:16 AndreasBr, do you mean as a user? or as a developer? in both cases, the answer is probably don't bother. Nov 13 22:29:27 ah Nov 13 22:29:30 finish(); Nov 13 22:29:31 i mean, the user already has two different ways to get out of your app Nov 13 22:29:44 why do you need to give them additional ones that are custom to your activity? Nov 13 22:29:56 anyone running tests with robolectric? after much trial and error I have it basically working with ant, abs, and fragments. now i'm trying to make it work with json. Nov 13 22:30:23 I tried manually adding the jar as many threads suggested, however I still get a RuntimeException: Stub! from robolectric when it hits JSONObject in my code. Nov 13 22:30:51 retomeier on exit buttons: http://twitter.com/retomeier/status/266623572519288833 Nov 13 22:31:11 ctate: The "user" at this point is ONLY me Nov 13 22:31:22 Its just an app to show integration of our payment terminals on android Nov 13 22:31:24 so hit the 'back' or 'home' key Nov 13 22:31:29 boom! out of the activity :) Nov 13 22:31:37 And when I pull the thingy down and choose BrisPOS Nov 13 22:31:45 will it initiate a NEW one, or resume? Nov 13 22:31:52 Can I use Object-RElational Mapping with Google Cloud Messaging? Nov 13 22:32:05 from back, a new one; from home, resume. Nov 13 22:32:10 same as it ever was Nov 13 22:32:10 Aha Nov 13 22:32:16 So if I use back it terminates it? Nov 13 22:32:29 if you press it enough times Nov 13 22:32:32 ok Nov 13 22:32:35 the default action of the back button, if your activity does not intercept it, is to finish() the foreground activity Nov 13 22:32:38 I have a finish button anyways atm Nov 13 22:32:43 can somebody tell me if i can use ORM with GCM??? Nov 13 22:32:44 ok, and if the app does? Nov 13 22:32:45 :D Nov 13 22:32:49 Nvm, going to bed Nov 13 22:32:55 the app is kinda the way I want it now Nov 13 22:33:04 apps don't interact with the UI Nov 13 22:33:05 activities do Nov 13 22:33:09 profligacy: thats like asking if you can use a chainsaw with a leafblower Nov 13 22:33:26 both are completely unrelated. but you might use them both in your app. Nov 13 22:33:32 gnight! Nov 13 22:33:37 ok Nov 13 22:33:47 probs be back tomorrow :D Nov 13 22:34:03 tdignan: so can i use an ORM to mapp data so that i can send it to GCM servers? Nov 13 22:34:13 g00s: ? Nov 13 22:34:15 becuase i think they only take text/json Nov 13 22:34:25 profligacy: no -- they are separate tools. I suggest you learn what they both do before you try to combine them. Nov 13 22:34:34 Drakonite: somebody said 4.2 has better low latency audio Nov 13 22:34:35 im not cominbing them Nov 13 22:34:38 tdignan Nov 13 22:34:46 Well sure, you can use them in the same app! Nov 13 22:35:06 Drakonite: you'll have you scroll back quite a bit Nov 13 22:35:24 yeah, I didn't want to scroll back that much :P Nov 13 22:39:16 and yeah... JB is supposed to have some better audio latency after some changed that were experimental in ICS. It's not so much by spec as it just happens that the systems were improved a bit in the main branch Nov 13 22:39:57 in other words, it's not a requirement to be better, it's just that the code in vanilla android results in lower software based latency Nov 13 22:40:06 does anyone know what would make a call to Camera.startPreview() throw a RuntimeException? Nov 13 22:40:18 its a native method Nov 13 22:40:25 karakuri hard to say without a stacktrace Nov 13 22:40:48 its happening in surfaceChanged callback Nov 13 22:40:54 using API demos Nov 13 22:42:15 check logcat Nov 13 22:43:22 the only part of code that i have access to is the surfaceChanged callback Nov 13 22:43:29 * ctate points to the imgur link in the channel topic Nov 13 22:43:33 everything else is either system calls or native method calls Nov 13 22:43:52 i'm looking at log cat stack trace now Nov 13 22:45:21 RuntimeException: startPreview failed Nov 13 22:45:38 not enough information to help Nov 13 22:45:49 at android.hardware.Camera.startPreview(Native Method) Nov 13 22:46:07 at .CameraPreview.surfaceChanged() Nov 13 22:46:18 don't paste here, use something like pastie.org Nov 13 22:46:24 Can I pass wheel events to the emulator? :P Nov 13 22:46:33 at android.view.SurfaceView.updateWindow Nov 13 22:46:37 after that is all system stuff Nov 13 22:46:51 more surfaceViews and windows and things i do not control Nov 13 22:47:09 facepalm.jpg Nov 13 22:48:03 http://pastebin.com/Zjgfu2pa Nov 13 22:49:18 were there no other log messages related to it? Nov 13 22:49:29 not just the exception, any log messages too Nov 13 22:50:07 like what? Nov 13 22:50:37 something that might actually describe why the exception was thrown Nov 13 22:52:46 how does updating a custom Notification work; is it with NotificationManager.notify() as with simple non-custom Notifications? Or do i manipulate the notification's layout with a LayouInflater? Nov 13 22:53:23 leslie: i have this much, http://pastebin.com/7CMXdpgB Nov 13 22:54:32 and yes, i have camera permission in manifest Nov 13 22:56:18 what device are you using? Nov 13 22:56:31 I'm guessing some kind of Samsung device Nov 13 22:56:50 i think that one was a samsung Nov 13 22:57:15 I don't know why it'd fail tbh, can't really see anything useful in that log Nov 13 22:57:38 the only thing it says is Nov 13 22:57:39 11-09 15:19:44.718: E/SecCamera(81): ERR(int android::SecCamera::initCamera(int)):Cannot open /dev/video0 (error : Device or resource busy) Nov 13 22:57:43 stuff marked in Android sources as @hide are not available for 3rd party apps, correct ? Nov 13 22:57:54 !seen d4rken Nov 13 22:58:07 no !seen Nov 13 22:59:00 maybe the camera is somehow locked? Nov 13 22:59:37 but then I would expect an exception from Camer.open() Nov 13 23:00:21 another weird thing just happened... Nov 13 23:00:48 my camera preview activity crashed, but the app didn't completely close Nov 13 23:01:15 normally the process is terminated, no? Nov 13 23:02:31 not sure, I haven't really looked at the process model much Nov 13 23:10:03 g00s: fwiw, i suspec the N10 will take over from the Nexus S as the best-supported device for AOSP/platform work. Nov 13 23:11:15 Is there a resource path? Or all teh resource expected to be in the root 'res/' directory? Nov 13 23:13:25 ... anyone here can tell me how much the N4 package weights? :D Nov 13 23:15:42 roughly the same as the galaxy nexus package Nov 13 23:15:46 but i dont' know any specifics Nov 13 23:16:47 well, I don't know heathen units like pounds, never had GN to compare, and the smuggling service wants dimensions and weight ;) Nov 13 23:17:42 Different question: I'm including a third party library that has its own 'res/' folder. My project is built with this library, but my Android code can't find the resources within that library Nov 13 23:18:21 ctate > stuff marked @hide in java source is should-be-off-limits to 3rd party apps, correct ? Nov 13 23:18:43 *is* off-limits to 3rd party apps :) Nov 13 23:18:44 yes Nov 13 23:18:54 if it isn't in the official sdk, dont' use it Nov 13 23:19:24 right, I was torn between it being like that, or maybe d.android.com wasn't fully updated on 4.2 docs yet ;) Nov 13 23:19:38 (if you're writing a bundled package that you know a priori will always be rebuilt for a given product rom, then situations are different) Nov 13 23:19:52 heh. you've got the 4.2 sources, i assume? Nov 13 23:20:12 even if d.android.com hasn't been updated you can build a local offline sdk package from those :) Nov 13 23:21:42 yeah I have the sources, some stuff that might be useful ( ACTION_USER_SWITCHED, specifically ) was marked @hide. Not sure what to make of that. But the Intent didn't exist on d.android.com, so I figured @hide meant it wasn't for 3rd party apps. Just making sure that is indeed the case. Nov 13 23:22:21 correct; very little of that is public API yet Nov 13 23:22:22 seems this may be related to the strange behavior i observed: http://stackoverflow.com/questions/4389301/why-does-android-revives-crashing-apps Nov 13 23:23:20 Well hello again! Nov 13 23:25:37 Tell me, I want to send a message from my BroadcastReceiver to one of 2 'active' Activities. If they aren't up, I don't need it. Essentailly, I'm trying to run the function Update() when my app receives an SMS so it can re-populate the list. Suggestions? Nov 13 23:26:01 register the receiver in onResume, unregister in onpause Nov 13 23:26:59 I have 2 receivers at this point. 1 created upon SMS receive via the manifest, and 1 created onResume Nov 13 23:27:42 I cant find the intent execution code for use within onReceive in the BroadcastReceiver Nov 13 23:28:58 What's the best Sliding Menu library at th emoment? Nov 13 23:30:05 Different question: I'm including a third party library that has its own 'res/' folder. My project is built with this library, but my Android code can't find the resources within that library Nov 13 23:30:53 "intent execution code"? Nov 13 23:31:34 well, there's context.startActivity(intent), but it crashes if I don't flag it as a new activity. Nov 13 23:31:35 frewsxcv: the "R." files have of the libraries are not in the default namespace Nov 13 23:31:40 you might have to import them Nov 13 23:32:26 monsti: What do you mean they're not in the default namespace? Nov 13 23:33:51 If it starts a new activity, then even when I'm not running the application, it will boot. I would like to keep it quiet? Nov 13 23:34:37 Komak57: so... you want the receiver to do something when the activity is already running, but not when the activity is *not* already running? Nov 13 23:34:42 this.sendBroadcast(i); is the preferred method, but only works in a Service, not a BroadcastReceiver. Nov 13 23:35:48 ehhh, close enough. I want to "refresh" the list when the BroadcastReceiver gets onReceive AND the activity is running? Nov 13 23:36:16 creating the intent and running it as a new task will currupt the navigation Nov 13 23:36:17 so just downloaded the eclipse plugin, and giving an error 'This version of ADT requires Android SDK Tools Revision 21.0.0 rc9 or above. Current revision is 20.0.0" but inside the Android SDK Manager, i only see a revision 20 listed for Android SDK Tools, with no option to upgrade. thoughts? Nov 13 23:36:20 so don't declare a in your manifest; instead, register one at runtime when your activity is started, and unregistere it when it's stopped Nov 13 23:36:32 that way you know for certain that the activity is running whenever the receiver is invoked Nov 13 23:36:56 ctate, the BroadcastReceiver is required for the SMS... weather it is up or not? Nov 13 23:37:13 ACTION Hey guys anyone know how to fix this problem --> E/webcoreglue(618): Should not happen: no rect-based-test nodes found Nov 13 23:37:39 then you need to do some other bookkeeping within your app about whether the activity is already running Nov 13 23:37:49 the OS is not going to do that for you Nov 13 23:40:17 whats the difference between Context and Activity? Nov 13 23:40:49 Can jars not provide resources to an android project? Nov 13 23:41:06 a Context is an interface object through which your code gets access to resources and system services Nov 13 23:41:22 ctate: so if I have an activity Nov 13 23:41:26 an Activity object acts as a Context, but is not the only kind of Context Nov 13 23:41:35 Service objects also act as Contexts Nov 13 23:41:38 etc Nov 13 23:41:39 called MyActivity and pass this to a class Nov 13 23:41:59 "this"*, I can make Instance variables of Context and Activity correct? Nov 13 23:42:00 then it can do Context operations on that instance, sure Nov 13 23:42:05 yes Nov 13 23:42:14 How do I include a LibraryProject? Nov 13 23:42:18 would it better for me to do, "this" or "MyActivity.this" Nov 13 23:42:21 or is there no difference? Nov 13 23:42:21 anyone know how to fix this error ? -- >E/webcoreglue(618): Should not happen: no rect-based-test nodes found Nov 13 23:42:33 there is no difference in most cases Nov 13 23:42:42 but one is static correct? Nov 13 23:42:55 but it also sounds like you are not super familiar with Java Nov 13 23:43:06 re static you might be thinking of MyActivity.class Nov 13 23:43:46 when the system looks at an intent to decide if there is already a task active for that intent, which parts of the intent get compared? Nov 13 23:43:46 am i deVoiced :/ Nov 13 23:44:02 EVX: err, i just saw that fine...? Nov 13 23:44:38 kk was just woundering i asked so many questions and did not even get a google is ur friend response thats why i ask thx ctate Nov 13 23:44:49 typically it means that nobody knows :) Nov 13 23:44:56 True; Nov 13 23:45:04 people *assume* you already know about google! Nov 13 23:45:11 otherwise you get lmgtfy links Nov 13 23:47:15 maybe if i explain the error more in detail i get this error -> 11-13 23:34:08.715: E/webcoreglue(618): Should not happen: no rect-based-test nodes found when running a fragment with a webview that runs a embed of a video from youtube and when click on the screen to start the video i get that error hope that explains it better and hopin to get some pointers cause i google it and did not find a help on this error.. or atle Nov 13 23:47:15 ast one i understand thats why i was hoping anyone here could help out thx in advance ;) ((((sorry for a lot of txt)... Nov 13 23:47:50 EVX: video inside WebView is a can of worms Nov 13 23:48:04 aka Cthulhu Nov 13 23:48:55 may i ask the best way to run a video embed in a fragment do u have any pointers? Nov 13 23:50:15 "In nearly all cases, the activity stack just works as expected." <- probably one of the most misleading phrases in the android documentation Nov 13 23:50:52 working as expected is subjective Nov 13 23:50:59 "all" and "no" are close together on the keyboard ;-) Nov 13 23:51:05 are there any benefits in using android.app.Activity over android.support.v4.app.FragmentActivity if you already require the support lib in your app? Nov 13 23:51:25 assume 14 <= minSdkVersion < 17 Nov 13 23:51:47 IMHO, the back button only very rarely does what I expected it to do (as a user). It's like that maximize-ish button in Mac OS X that isn't really maximize. Something happens, you just never know what ;) Nov 13 23:51:57 spifff: got any pointers? Nov 13 23:52:05 Chainfire, lol Nov 13 23:52:08 it goes back! Nov 13 23:52:15 except when it doesn't Nov 13 23:52:16 :) Nov 13 23:53:18 heh. I've had this macbook pro for 3 years, and I hit the green + button for the first time! Nov 13 23:53:42 how about u ctate? Nov 13 23:54:11 EVX: i know zilch about either webview or about video embedding Nov 13 23:54:43 ctate: yea it shouldve been .class Nov 13 23:54:52 not .this my bad haha, shouldn't have any difference at ll then Nov 13 23:55:26 canadiancow: as far as I'm aware, FragmentActivity is just for platform versions without native fragments support Nov 13 23:56:09 so if the minimum is 14, no benefit. unless the new "embedded fragments" api changes things. haven't tried it out yet Nov 13 23:56:11 EVX: it depends on lot which SDK version you're targeting, and whether or not the video is meant to play in-line or pop up a full screen player, whether or not hardware acceleration is enabled, etc etc. In my app I ended up displaying videos via a separate video player activity and only triggering it from within a webview Nov 13 23:58:12 EVX: there's quite a few threads on stackoverflow around that sort of stuff.. use teh google Nov 14 00:01:07 spifff: i would like to not be worried if someone with a very old SDK gets problem so my target would be everyone from old to new.. Nov 14 00:01:13 but will google on it thx :P Nov 14 00:01:39 leslie, thats not quite true, because as you said, 17 has features 14 doesnt Nov 14 00:01:55 there are definitely benefits in using the support version Nov 14 00:01:59 im just curious if there are any drawbacks Nov 14 00:02:04 yeah. but you said < 17, implying you didn't care about 17 Nov 14 00:02:19 no, i said minSdkVersion, implying it will run on devices that do not have native nested fragments Nov 14 00:02:26 ahh ok Nov 14 00:02:46 hm, the docs FLAG_ACTIVITY_RESET_TASK_IF_NEEDED don't seem to nearly cover what's actually going on with that flag... Nov 14 00:03:00 I imagine it'll work fine unless you use nested fragments. Android is pretty good when it comes to backward compatibility Nov 14 00:03:54 e.g. why am I getting FLAG_ACTIVITY_BROUGHT_TO_FRONT even though the docs say it only applies in launchMode=singleTask (and I'm using standard) Nov 14 00:04:44 yea but again, not what im asking Nov 14 00:04:45 :P Nov 14 00:05:26 well, bleh Nov 14 00:11:15 hi all .. can anyone point me in the right direction i have an ArrayList i want to sort by a value in ProjectLanguage do i have to do this manually or are there some premade functions if so what do i need to search for ? Nov 14 00:12:04 you can use a custom comparator, fanno Nov 14 00:12:16 I think Nov 14 00:12:25 maybe I'm thinking of C# Nov 14 00:13:03 yeah, make a comparator and use Collections.sort(), fanno Nov 14 00:13:32 is there any way in 1 or 2 lines of code to minimize (not disable) the soft keyboard? Nov 14 00:13:51 leslie: i'll try and look look for that Thanks Nov 14 00:15:00 fizyplankton http://stackoverflow.com/questions/1109022/close-hide-the-android-soft-keyboard Nov 14 00:17:15 I am attempting to modify AOSP (specifically the portion underneath the Android Platform) to add special enhancements. This is for a research project. One really important requirement is that I securely implement a communication channel between my linux system service and an Android App. Does anyone have any suggestions for this? Nov 14 00:18:41 leslie: thanks alot that was easy =) much more simple than making a costum function everytime i need to sort something =) Nov 14 00:19:04 np :) Nov 14 00:19:34 this is the one i used as a base =) http://www.vogella.com/blog/2009/08/04/collections-sort-java/ Nov 14 00:19:51 Has 4.2 started rolling out to existing nexus? Nov 14 00:20:24 I think so Nov 14 00:20:32 definitely to Nexus 7 devices Nov 14 00:20:41 I have seen happy GN users Nov 14 00:23:14 oh dear god nooooo Nov 14 00:23:25 * Chainfire is reading Jeff Sharkeys preliminary writeup Nov 14 00:23:37 of the new segmented multi-user external storage Nov 14 00:23:47 secondary external storage being read-only for apps is going to be standard Nov 14 00:23:50 ffs Nov 14 00:23:58 does anyone know if it's possible to clip a canvas by text? Nov 14 00:25:00 Anyone finding fastboot not detecting a device after flashing 4.2 (yes, usb debugging enabled) Nov 14 00:27:47 Chainfire: ... elaborate? Nov 14 00:29:04 grantland: I haven't done it myself, but Paint has a getTextPath() method that purports to extract a Path from a piece of text. Nov 14 00:29:26 [which, if successful, can be used to clip the canvas via clipPath()] Nov 14 00:29:39 p_l https://android.googlesource.com/platform/docs/source.android.com/+/master/src/tech/storage/index.md Nov 14 00:30:33 Chainfire: can you also elaborate on your issue with it? Nov 14 00:31:19 kbs thanks, just found that! Nov 14 00:31:40 SO led me to look into TextView where it didn't exist D: Nov 14 00:31:43 p_l I have shitloads of customers that need the (real) sdcard to be writable by my app Nov 14 00:32:26 in other words Nov 14 00:32:28 on tablets, I assume? Nov 14 00:32:28 cold hard money Nov 14 00:32:29 :) Nov 14 00:32:34 both tablets and phones Nov 14 00:32:38 Chainfire: "real" sdcard? Nov 14 00:33:16 Chainfire: well, it's only enabled for tablets so far... and I suspect you can disable multi-user mode Nov 14 00:33:41 jsharkey people have taken to callign internal storage internal sdcard, just to clarify I mean a real removable thing here Nov 14 00:33:51 wait what?? Nov 14 00:33:55 no RW on SD? Nov 14 00:33:56 Chainfire: you mean secondary external storage provided through physical media? there isn't any public API for secondary external storage Nov 14 00:33:57 why does my "Building Workspace" window seem to pop up so often now and take soooo long to finish Nov 14 00:34:08 so ppl not only are limited to skiny internal storage Nov 14 00:34:15 jsharkey> I know, and I've been nagging people about that for ages Nov 14 00:34:21 but when a device actually allows 64GBs SD it's RO ? Nov 14 00:34:24 but that doesn't mean we can't use it Nov 14 00:34:30 it's actually writable on most devices Nov 14 00:34:31 Chainfire: its on my todo list :) Nov 14 00:34:45 even with that crappy ass new permission that was introduced some time ago Nov 14 00:34:52 It just meant you couldn't rely on it remaining I guess :( Nov 14 00:34:55 that made secondary external storage provided through physical media read-only Nov 14 00:35:02 the media_rw permission should be system|signature Nov 14 00:35:07 luckily, Samsung had brains enough to fix that Nov 14 00:35:27 the media_rw permission should go away. Nov 14 00:36:09 well, yeah, I agree, it should be something settable per-device (maybe with extra permission?) Nov 14 00:36:36 jsharkey maybe you can explain to me the reasoning for media_rw Nov 14 00:36:38 also, possibly make it possible to select whether external sdcard is to be user-partitioned Nov 14 00:36:50 I've cornered Reto a couple of times, he always tells me he'll get back to me on that and never does (surprise!) Nov 14 00:37:03 aahhaha Nov 14 00:37:10 Chainfire: hey, google never answering is a tradition Nov 14 00:37:38 p_l: i'm here answering :P Nov 14 00:38:22 Chainfire: media_rw is being used to lock down secondary external storage until we have a solid public API story in place Nov 14 00:38:22 I'm here asking :) Nov 14 00:38:38 right Nov 14 00:38:47 and you would lock it down because ... Nov 14 00:38:48 jsharkey: how about "mark device as shared"? Nov 14 00:39:05 you're trying to solve a problem nobody is having Nov 14 00:39:10 and thereby creating problems nobody wants Nov 14 00:39:57 if it's not defined by the public API, developers can't rely on its behavior; writable or otherwise Nov 14 00:40:17 developers not being able to rely on its behavior by itself is not a reason to break it Nov 14 00:40:20 and yeah, when all tablets use beefy internal storage, making media_rw for *external* is... bullshit Nov 14 00:40:32 its really really simple: developers will make what their customers want Nov 14 00:40:36 with your blessing or not Nov 14 00:40:46 why dont we just run the entire system on an sd card Nov 14 00:40:48 external sd card is in wide use Nov 14 00:40:53 then theres only one location, and it solves everything Nov 14 00:41:18 Chainfire: so, "application advisory" about not upgrading to 4.2 on tablets? ;P Nov 14 00:41:19 will be too dangerous in corporate environment Nov 14 00:41:20 canadiancow: the quality of physical sdcard media can vary wildly Nov 14 00:41:26 and as Google hasn't made or supported any real external sd card device in quite some time, I find it doubtful such an API will *ever* be supported Nov 14 00:41:27 canadiancow: external storage controlers are even slower than internal ones Nov 14 00:41:30 jsharkey, i was joking ;) Nov 14 00:41:36 and it's only the beginning of how they handle mobile in companies Nov 14 00:41:41 adq: corporate environments will actually want media_rw for external storage Nov 14 00:41:42 just one example: http://www.bunniestudios.com/blog/?p=918 Nov 14 00:41:42 (i think it's a valid reason :p) Nov 14 00:42:08 or rather, want their applications to *have* media_rw for external storage Nov 14 00:42:08 "ghost shifts" in factories Nov 14 00:42:12 cause... wtf Nov 14 00:42:15 The new ADT is so slooowwww D: Nov 14 00:42:28 i bet many policies will not allow sdcards Nov 14 00:42:32 (as usb previously) Nov 14 00:42:41 then let it be done through a policy Nov 14 00:42:46 or device admin or whatnot Nov 14 00:42:49 acutally yea i had a bad microsd card, and the company sent me a new one that was made in a different country, and it was awesome Nov 14 00:43:11 :-) Nov 14 00:43:33 adq: the places that truly want that, also ban BYOD practices Nov 14 00:43:53 or ban them from being used with company data directly (i.e. outside of vetted channels) Nov 14 00:43:59 if they don't, they are already stupid Nov 14 00:44:08 yes, penetration is slow Nov 14 00:44:16 there are various solutions to this problem Nov 14 00:44:58 the currently chosen path is the one most likely to piss people off, and therefore, naturally, the one taken Nov 14 00:45:45 cyanogenmod had a decent solution back in the day Nov 14 00:45:48 * p_l sometimes wishes that instead of MTP we had plain old network protocol Nov 14 00:46:06 there's always adb push/pull Nov 14 00:47:21 p_l: just for that i'm tempted to make it use Gopher Nov 14 00:48:26 ctate: better than MTP! :P Nov 14 00:48:57 not for 90+% of Android users :p Nov 14 00:49:15 ctate: provide MTP as another endpoint, and RNDIS as another Nov 14 00:50:04 the amount of crazy hacks related to trying to get android connected to network over USB astonished me, all because I couldn't configure it as client, not router (or previously, as network device at all) Nov 14 00:50:07 sud how about emulate adb internally and push from there? :P Nov 14 00:50:42 p_l: because host-side usb is insane, basically. Nov 14 00:52:16 ctate: still, it could be selectable action Nov 14 00:52:36 oh well, once I get a bit more time (and money for another dev device) I might actually try something Nov 14 00:59:18 Hey guys, could someone help me with this pice of code i want it to run in a new thread -> http://pastebin.com/vvc77h46 Nov 14 01:01:08 EVX, fragment creation should probably be on your main thread Nov 14 01:01:26 it is Nov 14 01:01:37 all it's doing is allocating a Fragment? why on earth would you want to offload that to a different thread? Nov 14 01:01:44 +1 Nov 14 01:01:46 ^ Nov 14 01:01:46 but i get a notice that i might be working to hard :/ Nov 14 01:01:48 you waste more in overhead than the cost of just doing it inline Nov 14 01:01:49 +100 Nov 14 01:02:04 EVX: the warning is not about the line you quoted Nov 14 01:02:05 EVX: is that in the emulator? Nov 14 01:02:06 unless your constructor is retarded, it is not causing you to skip 37 frames Nov 14 01:02:11 unless its the emu...yeah Nov 14 01:02:13 s/line/section of code/ Nov 14 01:02:26 hey, anyone using robolectric? :) Nov 14 01:02:30 profile it and see where you're spending time Nov 14 01:02:37 yeah it's in emu Nov 14 01:02:48 EVX: dont worry about it Nov 14 01:03:05 its probably not even complaining about your code Nov 14 01:03:11 lol Nov 14 01:03:30 ok thats good Nov 14 01:03:35 thanks guys! Nov 14 01:03:54 Do I need to update the NDK to work with this 4.1.2 stuff? :S Nov 14 01:04:27 tdignan: not what you're looking for, but fwiw - after playing with a few different test frameworks, I've gone to just separating my code, so all the non-ui stuff has no android references and can be tested independently. Nov 14 01:05:17 btw, 37 frames is more than half a second Nov 14 01:05:36 i do the same as kbs. but for actually testing the ui code i use robotium. jenkins can spawn a headless emu to run the integration tests Nov 14 01:06:02 ive not heard good things about roboelectric Nov 14 01:06:15 ESphynx, you mean 4.2? Nov 14 01:06:55 canadiancow: no I mean 4.1.2 Nov 14 01:07:09 I set up my emulator to be 4.1.2 since there's no Intel on 4.2 Nov 14 01:07:27 you should probably always be using the latest ndk anyway Nov 14 01:19:00 tried creating a static handle and a static update function, but the android sends leak errors =.= Nov 14 01:20:12 wat Nov 14 01:21:02 still trying to work on getting my SMS BroadcastReceiver to let my app know when to update. Nov 14 01:21:08 My native glue message pipes are getting messed up =( Nov 14 01:21:55 Maybe i'm missing something somewhere.. Nov 14 01:22:40 pfft Nov 14 01:24:39 reading the msgread pipe is jamming :| Nov 14 01:26:04 JakeWharton_, remember the tip you mentioned a couple of days ago about taking a snapshot of a certain UI element (or just any chosen area)? about how it should be in a canvas and that canvas shou be in a Bitmap and that bitmap would be drawn somewhere? Nov 14 01:26:14 yes Nov 14 01:26:35 well Galaxy Note 2 can do that when you highlight any portion of the screen; Note 2 has JB; think the SDK has that inside? Nov 14 01:27:18 highlight as in, you draw an area on the screen with the pen Nov 14 01:27:31 no, the Android SDK does not have that Nov 14 01:27:59 so it's in JB's capabilities as shown in Note 2 but it won't be in the JB SDK? Nov 14 01:28:03 so not-so-open? Nov 14 01:28:24 JakeWharton_: have you seen the issue where a tab in ABS has the blue selected-state indicator both at the bottom of the tab and below the tab text? Nov 14 01:28:33 (this is on a 2.3 device) Nov 14 01:31:23 ugh, too sleepy Nov 14 01:31:27 good night Nov 14 01:32:50 I asked this ealier but got sidetrack with other work so .... I'm getting "trouble processing "javax/xml/parsers/DocumentBuilder.class" " while trying to compile my project.. anyone ran into this problem? Nov 14 01:35:09 tnzr: screenshot? Nov 14 01:37:01 nvm.. had to remove com.google.android:android dependency Nov 14 01:37:55 JakeWharton_: one sec, I'll try to snag one for you Nov 14 01:38:13 xastey: with Maven or Gradle? Nov 14 01:38:22 xastey: it should be marked as provided Nov 14 01:38:33 maven Nov 14 01:38:51 do you have provided? Nov 14 01:38:55 it was marked provided Nov 14 01:38:59 in parent and module pom's Nov 14 01:39:13 o sorry it wasn't in parent Nov 14 01:39:22 how can i know, wath boot uses my tablet? Nov 14 01:39:29 JakeWharton_: no go, sorry.. I'm at home and my coworker (who has the test devices) isn't online so I won't be able to get one :( basically the light-blue strip indicating the tab is selected shows up both at the bottom of the tab (normal) and right under the tab text, wtih the same length as the text (abnormal) .. If this info is too vague to consider, don't worry about it, maybe I'll catch you next time I'm at work :) Nov 14 01:39:38 like uboot or somethink Nov 14 01:39:55 xastey: it only needs to be marked as provided in the actual app's pom Nov 14 01:40:23 umm.. in parent pom it wasn't marked provided.. but in the actual app it was Nov 14 01:40:41 tnzr: I believe I understand what you are describing. Would love a screenshot and perhaps a sampling of any styles that are in use. Nov 14 01:41:09 xastey: that should be enough. the parent pom usually defines simply the version number and the child pom (where it's actually being used) merely defines how it is used Nov 14 01:41:10 i swear i spent almost all day reimplementing something that is "almost" in the framework Nov 14 01:41:49 JakeWharton_ : I'll give it another try.. trying to track down a bug using RoboGuice Nov 14 01:42:12 well RoboSherlockListActivity.. saying something about "Could not load Finalizer in its own class loader. " Nov 14 01:42:22 i think you just described your bug right there :) Nov 14 01:42:38 kidding, of course... RoboGuice is perfectly... fine Nov 14 01:42:48 hehe Nov 14 01:43:01 May just remove it.. not like I'm using it a lot anyways Nov 14 01:43:44 i have nothing against dependency injection itself, just question a lot of RoboGuice's "helpers" Nov 14 01:43:58 JakeWharton_: cool, maybe I'll catch you on at a later date and if you aren't busy then I can get you taht stuff..thanks for trying! Nov 14 01:44:14 absolutely. hit me up on here or PM or email me Nov 14 01:44:28 wow thanks! Nov 14 01:44:30 finally get to work on something fun tomorrwo Nov 14 01:44:31 or email ABS' mailing list Nov 14 01:44:33 yeah JakeWharton_ its causing a lot of problems right now.. my app is pretty simple so I'm just going to drop it Nov 14 01:44:44 no more of this balls with serial port support =( Nov 14 01:44:44 we are using ABS in all of our android projects at work btw, everyone considers you a legend Nov 14 01:45:18 parallel port was always superior anyways Nov 14 01:45:29 you forget the industry I work in Nov 14 01:45:32 industrial automation Nov 14 01:45:41 lots of serial port / rs485 Nov 14 01:45:51 can/etc Nov 14 01:46:28 How do we provide an icon? :) Nov 14 01:46:56 JakeWharton_: hey, I've got some robolectric stuff working with ant. But I don't think I'm going to get it to work with intellij. Intellij can run my ant task fine with it's ant-build tab. But, since I don't have the intellij side configured, the static analysis highlights the 'problems' in my code (which, when I build with ant, don't happen, since it's properly configured there) Nov 14 01:47:07 Any idea how to shut off the static analysis for one module? Nov 14 01:47:40 man jetbrains has significantly improved my life Nov 14 01:47:58 Yeah, it's a nice IDE especially now that they fixed the bugs I filed over the summer :) Nov 14 01:48:10 well I use intellij, I use youtrack, and teamcity Nov 14 01:48:17 they all integrate very very well Nov 14 01:49:48 wasn't there a chan for intellij? Nov 14 01:50:25 man, if anyone knows how to turn off static code checking just for one module, please.. :) Nov 14 01:53:16 well, I tricked it! Nov 14 01:53:35 I just added it as an android project and I'll never build it with intellij and just use the custom ant build Nov 14 01:56:54 ugh Nov 14 01:57:01 my workspace seems to take so long to build lately, its really slowing me down Nov 14 01:57:15 I end up queueing up things for it to do so that dumb window pops up to tell me to hold on while it finishes crap Nov 14 01:57:29 are you using eclipse? Nov 14 01:57:35 yeah Nov 14 01:58:15 tnzr: what was the last message you got from me :) Nov 14 01:58:22 man, that was a fast auto reconnect. Nov 14 01:58:30 07:53PM are you using eclipse? Nov 14 01:58:41 answer: yeah Nov 14 01:58:50 my other question was, whether you're actually using all the projects currently in your wrkspace Nov 14 01:59:44 tnzr - use faapt Nov 14 01:59:48 yeah, only 3 - zxing, ABS and mine Nov 14 01:59:49 makes HUGE HUGE improvements Nov 14 01:59:52 the problem with eclipse is that workspace literally is your workspace for everything.using multiple workspaces means losing your settings. I was shown workspace mechanic by mre_nouf|away, but it seemed like a lot of work just for one guy Nov 14 02:00:03 ron_frown: I'll google it, sounds interesting Nov 14 02:00:03 especially on windows Nov 14 02:00:14 tnzr: well you dont actually have to rebuild abs and zxing each time Nov 14 02:00:16 its aapt for building xml/resources but they integrated caching layer inside Nov 14 02:00:42 the default in eclipse is that it builds the whole workspace, but you should be able to change that Nov 14 02:00:46 ewww ok Nov 14 02:01:21 i'd tell you how but I retired eclipse for the time being as the latest version has tendency to segfault the oracle jvm, 1.6 and 1.7 versions. Nov 14 02:01:40 its ok, I'll google that too :) Nov 14 02:01:42 and thanks Nov 14 02:20:21 Do anyone tried to get the streetview enabled in your apps? I'm wondering if it is possible to do it. Nov 14 02:22:20 hi Nov 14 02:23:32 is possible do startActivity and back to previous activity without indicating which was explicitly? Nov 14 02:23:43 can anyone see why once stop is pressed i can not play the sound file again code -> http://pastebin.com/JXCTU3qp (error Error (-38,0) Nov 14 02:23:43 11-14 Nov 14 02:23:43 E/MediaPlayer(1323): start called in state 0 Nov 14 02:23:43 ) Nov 14 02:29:07 hello interwebs Nov 14 02:29:23 hellooo Nov 14 02:29:30 I've got an app that works with no problems on 4.0 but is throwing a NoClassDefFoundError on a 2.2 VM Nov 14 02:29:43 awe lool Nov 14 02:29:45 solved it Nov 14 02:29:45 :p Nov 14 02:30:18 @evana try running under 2.3.5 API10 Nov 14 02:30:40 turb1ne247: I'd like to get it working on 2.2 Nov 14 02:31:15 evana, thats because you're using a class that doesnt exist on api 8... Nov 14 02:31:22 what class is it Nov 14 02:31:25 exactly.. Nov 14 02:31:25 canadiancow: that is not the case Nov 14 02:31:42 the class not found error is for one of my activities Nov 14 02:32:06 is it the main activity? Nov 14 02:32:12 turb1ne247: nope Nov 14 02:32:37 did you add the activity to the manifest xml? Nov 14 02:32:42 ok can you show some code? Nov 14 02:32:47 how are you starting the activity? Nov 14 02:32:47 etc Nov 14 02:32:56 it is in the manifest Nov 14 02:32:57 ya..pastebin the code for us Nov 14 02:33:50 Take a look at stack overflow: Nov 14 02:33:51 http://stackoverflow.com/questions/8678630/noclassdeffounderror-android Nov 14 02:34:27 I need some pointers.. say Have a list of items I want to display(ListActivity) . this will should display the Section title plus its children (check layout layout) http://pastebin.com/TaAih0af Nov 14 02:34:58 I know I need to create a custom ArrayAdapter for this.. but whats the proper way to display its children while still having the list scrollable (not its children) Nov 14 02:35:00 bass_ox: Call finish()? Nov 14 02:35:31 I was thinking of adding a nested ListView to the Section.. but wouldn't that intercept the main/parent ListView scrolling? Nov 14 02:35:52 do not put a listview in a listview Nov 14 02:36:06 I have acquired an IRL Canadian to help with this Nov 14 02:36:09 thought someone would say that.. thus my question Nov 14 02:36:20 what the proper way I should handle this canadiancow ? Nov 14 02:37:49 xastey: Expandable listview may be. Nov 14 02:38:07 turb1ne247: https://gist.github.com/4069936 Nov 14 02:38:18 turb1ne247: here is my gist ^^ Nov 14 02:38:26 well i want it expandable all the time Nov 14 02:38:39 for instance http://pastebin.com/8eHJB9sc Nov 14 02:39:36 I was just thinking to add an LinearLayout for the Section parts.. then subviews.. but then I run into the problem with the Children being clickable Nov 14 02:40:09 so I would have to attach a listener to each child view... then propagate it up.. was looking to see if there was a better way Nov 14 02:40:11 onclick=null? Nov 14 02:41:41 I think you can achieve what you are looking for using an expandable listview xastey Nov 14 02:41:54 herp a derp Nov 14 02:42:02 I'll check that out napster Nov 14 02:43:18 looks like something napster.. now to see if I can get it to work with com.github.chrisbanes.pulltorefresh Nov 14 02:53:09 napster.. this will work perfect.. thanks a lot **** ENDING LOGGING AT Wed Nov 14 02:59:59 2012