**** BEGIN LOGGING AT Wed Sep 17 02:59:58 2014 Sep 17 03:00:03 Interesting, I did not even know about the silver program, directly attacking Apple tends to be a mistake. Sep 17 03:00:47 it almost sounds like something MS would come up with. Sep 17 03:01:36 gotta be creative when companies like dell are butchering your product :) Sep 17 03:02:16 true, I just notice people get in trouble when they focus on what Apple is doing, rather than what they are doing ;) Sep 17 03:02:18 g00s: Eclipse is still the most popular IDE, I think. Sep 17 03:07:48 this looks good (if you're into startup stuff) http://www.amazon.com/Zero-One-Notes-Startups-Future/dp/0804139296 Sep 17 03:19:43 hmm, does anyone know how long it takes for your app to get indexed on the play store search? Sep 17 03:19:56 I mean, it works when using an explicit link but not the search, for some reason Sep 17 03:19:58 a few hours Sep 17 03:20:43 it has to propagate for a while. Sep 17 03:20:57 hmm, fair enough Sep 17 03:21:01 it's been half a day though Sep 17 03:21:32 hm...not sure. Sep 17 03:22:17 I don't remember how long after unloading a searched for my app, but it was visible the next day iirc Sep 17 03:22:23 I* Sep 17 03:22:33 I guess I just have to be a bit more patient Sep 17 03:22:56 yup, I think that is about your only option to be honest :) Sep 17 03:23:15 also, any tips on which blogs to send your apps to? Sep 17 03:23:23 besides androidpolice Sep 17 03:23:30 being an obvious one Sep 17 03:24:21 I don't really promote my app, so I don't know personally. Sep 17 03:25:53 you don't promote your app? Sep 17 03:26:05 do you try to make a living out of your android apps? Sep 17 03:27:06 no, I just wanted to go thru the process right now, also my app is dev-art so its free etc. Sep 17 03:27:23 in the future I might try to make some money I guess Sep 17 03:27:42 dev-art? Sep 17 03:27:48 I have a lot of apps planned Sep 17 03:27:59 ya, meaning: using programming to make art. Sep 17 03:28:05 ah fair enough Sep 17 03:28:37 EoN ok, stop right there. there are no "note apps" that haven't been done. there are no "new twists" in that area. it's funny you say that. I'm after a notes app which doenst have ANY 'todo' list or 'extra bits' on it. just notes. simple notes. when i click on the list i want it to open the note. then be able to edit it and save it via clear 'edit' and 'save' or 'done' buttons. no weirdness in switching between edit/view mode. also the font Sep 17 03:28:37 should be good/solid, and design be nice Sep 17 03:28:38 its kind of a movement right now, artists trying to figure out what the "native" media of computer art is and so on Sep 17 03:28:43 i'm yet to find it. please tell me what has that! Sep 17 03:29:10 EoN google Keep is like that :| Sep 17 03:29:19 i dont want any server side component Sep 17 03:29:31 google keep has a weird interface IMO Sep 17 03:30:18 the best notes app i've ever used was the iphone default notes app. i have an android copy but its a bit crap Sep 17 03:30:46 basically i'm after 'less', not more. and everything seems to have extras Sep 17 03:40:30 i'd use less over more any day Sep 17 03:43:04 it doesn't matter how beautiful a note taking app is, at this point you'll have to cut through the noise of 100K other beautiful note taking apps Sep 17 03:44:39 why can’t you save a parcelable to shared prefs -.- Sep 17 03:45:00 that would be too convenient Sep 17 03:45:06 also prefs are stored in XML Sep 17 03:45:19 i need a dingleberry to do that for me Sep 17 03:45:28 mvstore is your dingleberry Sep 17 03:45:56 well, that won't woth parcelable directly, hmm Sep 17 03:45:58 o yea? Sep 17 03:46:29 right now im just doing gson.toJson() Sep 17 03:46:31 but its a bit annoying Sep 17 03:46:42 because i need to get a type token to deserialize it Sep 17 03:46:42 g00s: true, too many notes app these days huh Sep 17 03:47:38 shingshang seems like everyone say, "i'm going to make a note taking app... but this one is going to be different: its going to be /beautiful/ Sep 17 03:48:22 g00s: that's what I did for the iPhone Sep 17 03:48:27 it worked, but not because of the note app Sep 17 03:49:12 one can substitute note taking app for 'messaging app that deletes messages after xxx' Sep 17 03:49:18 etc etc Sep 17 03:49:27 that's true too Sep 17 03:49:37 but the latter is pretty hot for investors I think Sep 17 03:49:50 i just get ppl to pay me to make their apps Sep 17 03:50:02 takes less effort than having your own idea Sep 17 03:50:34 how come adb shell pm clear is just ‘failed’ ? Sep 17 03:55:36 maybe you've already removed that app before? That's what I get if I try and uninstall an app that isn't there Sep 17 03:55:51 i wish :\ i can’t uninstall or clear from shell right now Sep 17 03:56:01 have to do it manually Sep 17 03:56:18 that's strange. how come? Sep 17 03:56:27 oh nm Sep 17 03:56:28 my bad Sep 17 03:56:42 debug { applicationIdSuffix '.dev' } Sep 17 03:56:45 :D Sep 17 03:56:57 forgot the .dev at the end of the app id Sep 17 03:57:02 that ol' chestnut! Sep 17 04:19:38 recyclerview is pretty cool Sep 17 04:20:07 yeah i've been meaning to play around with that Sep 17 04:20:26 been too lazy to copy it out of the 21-rc support lib Sep 17 04:20:27 like in the past i’d extend framelayout or something to make a custom view Sep 17 04:20:53 g00s: you can do this hack http://www.reddit.com/r/androiddev/comments/297xli/howto_use_the_v21_support_libs_on_older_versions/ Sep 17 04:22:04 thanks i'll check that out Sep 17 04:26:17 huh, I thought recycler view is packaged separately already... Sep 17 04:26:32 can’t compile it with so unless you have l preview installed on a device, or want to use the stock android emu, it’s not really viable Sep 17 04:27:00 no compatibility with genymotion, etc Sep 17 04:27:16 Haven't noticed Sep 17 04:28:25 what do you mean havent noticed? Sep 17 04:32:20 the manifest in recycler Sep 17 04:33:15 Haven't even tried it yet, doesn't help that the new support libs cause my compiler to crash Sep 17 04:33:26 at least appcompat Sep 17 04:33:51 pfn javac ? Sep 17 04:34:02 no Sep 17 04:34:24 but the new annotation bullshit they implemented in the new appcompat Sep 17 04:34:26 ... Sep 17 04:36:29 JacobTabak keep mvstore tucked away in the toolbelt, its pretty awesome. if you need k/v its very fast Sep 17 04:36:48 i do Sep 17 04:36:51 need ti right now Sep 17 04:37:18 can you summarize in 1 sentence why its awesome Sep 17 04:37:22 how is it backed Sep 17 04:37:59 its backed by memory or log structured storage Sep 17 04:38:27 its the new engine of h2; just like sqlite is being refactored into a k/v store and sql layer ontop Sep 17 04:38:51 so if i have objects i want to store by id Sep 17 04:38:56 numeric id Sep 17 04:39:06 that i never need to modify Sep 17 04:39:12 just save and read Sep 17 04:39:25 you can modify all you want, but yeah Sep 17 04:39:35 the id would be the key of course Sep 17 04:40:35 why is this faster than shared prefs? Sep 17 04:40:40 doesnt shared prefs cache after the first retrieval Sep 17 04:40:58 JacobTabak i would only use mv store if you need performance, like you have billions of kets Sep 17 04:41:01 *keys Sep 17 04:41:11 or you need range queries, like between x and y Sep 17 04:41:13 ok i’ll keep that in mind Sep 17 04:41:19 probably not good for what im doing now Sep 17 04:41:25 yeah, otherwise its a dingleberry ;) you can use sqlite of prefs Sep 17 04:41:26 which is just caching like 100 objects with 4 fields Sep 17 04:42:33 i try to just use what the android framework gives unless its a big mismatch or performance problem Sep 17 04:42:49 * g00s doesn't like unnecessary dependencies Sep 17 04:42:56 i have so many dingleberries Sep 17 04:43:23 24 gradle deps Sep 17 04:43:27 problem is, all the digleberries get caught in each other and emergen bugginess ensues Sep 17 04:43:44 no problesm for me really Sep 17 04:43:52 :) Sep 17 04:54:09 hey dudes Sep 17 04:54:13 weird thing here Sep 17 04:56:03 http://cl.ly/image/2X3O0d3t3b11 this is my layout preview, and this is the app http://cl.ly/image/393j0F0j440u Sep 17 04:56:09 the line spacing isn’t previewed correctly Sep 17 04:56:11 or something Sep 17 04:56:14 is this always? Sep 17 04:57:35 how are you positioning it ? Sep 17 04:57:58 linearlayout Sep 17 04:58:00 on top of eachother Sep 17 04:58:21 pastebin, and you not setting any special formatting on the text? Sep 17 04:58:55 http://pastebin.com/sL6HYqEW Sep 17 04:59:02 oh sorry Sep 17 04:59:06 i just put some padding in there to make up for it Sep 17 04:59:11 let me repaste quick Sep 17 04:59:26 http://pastebin.com/jLErGaDH Sep 17 05:00:39 man i have never spent this much time on a layout Sep 17 05:00:55 huh, very odd, I know things like kerning are off in the preview, but not soo far out and not at the origin :) Sep 17 05:01:07 heres what it looks like in reality Sep 17 05:01:08 http://cl.ly/image/0F2b1u2o2G1O Sep 17 05:01:36 very annoying -.- Sep 17 05:02:01 well 2 things to do to visually and practically debug this Sep 17 05:02:23 1. turn on view bounds , confirm it's the bounds or the inner text print that the culprit Sep 17 05:03:00 2. to diagnose the same thing, use relative layout with topBottom and leftLeft rules Sep 17 05:03:48 3. Make the text size and style (and no all caps) identical to also add to this mystery :) Sep 17 05:03:55 http://cl.ly/image/0Q3c2Y0l2q00 Sep 17 05:04:05 view bounds are ok Sep 17 05:04:09 its just… weird Sep 17 05:04:44 in the layout preview, the clip box ends before the letters do http://cl.ly/image/172M2p2g1A0d Sep 17 05:04:55 like the y goes over the clip box Sep 17 05:05:00 and the g Sep 17 05:05:13 same amount it's offset at the beginning Sep 17 05:06:01 maybe the view is a little drunk ? given the text content context :) Sep 17 05:06:17 haha Sep 17 05:11:34 another build system http://pantsbuild.github.io/ Sep 17 05:11:56 from twitter, foursquare, and square Sep 17 05:12:02 false Sep 17 05:12:11 it's from twitter Sep 17 05:12:14 my friend always tells me about his crappy ideas for android apps that I should make Sep 17 05:12:21 and it's nowhere near new Sep 17 05:12:24 this time he's actually got a great idea Sep 17 05:13:08 an app that runs in the background so that if you lose it, you can say something like "where's my phone?" and it would starting ringing Sep 17 05:13:31 so it’s going to be listening to your mic 24/7? Sep 17 05:13:37 sounds legit Sep 17 05:13:59 ok apparently you can’t put radiobuttons in a tablerow Sep 17 05:14:41 this may actually be a problem Sep 17 05:14:58 JakeWharton pedantic Sep 17 05:14:59 so don't use them Sep 17 05:15:10 they're just compound buttons with a different background Sep 17 05:15:15 JacobTabak, yeah but thats a problem as well Sep 17 05:15:21 it will eat battery life most likely Sep 17 05:15:31 JakeWharton: i mean you need to have a radiogroup as the direct parent Sep 17 05:15:42 i’m going to have to handle the only-one-selected-at-a-time manually Sep 17 05:15:52 i believe that with API 19 and above, you can do voice recognition without internet Sep 17 05:15:53 just annoying Sep 17 05:16:42 moto x let's you say where's my phone Sep 17 05:16:43 g00s: pedantic is concerned with minor details (such as correcting your use of pedant), you had two glaring inaccuracies Sep 17 05:16:44 if I could figure out an ingenious way to save battery life, it might work Sep 17 05:17:00 pfn, oh really? I didn't know that Sep 17 05:17:22 pfn was it a HW dongle or something? i was looking for it here http://www.motorola.com/us/accessory-family-page-1/ Sep 17 05:18:09 oh i thought they had some new keychain fob too Sep 17 05:20:13 meh, i was thinking of the Motorola Skip Sep 17 05:20:18 i have to make a custom tablelayout that adds checked changed listeners to all its children Sep 17 05:20:39 or just do it manually… either way, lame Sep 17 05:21:07 or could modify radiogroup to look 2 lvls deep for compound buttons Sep 17 05:23:12 eh i’ll just use butterknife to get an array of the checkboxes i need to manage… should be easiest Sep 17 05:24:16 RadioGroup shouldn't be a ViewGroup Sep 17 05:24:27 yeah its a weird linearlayout Sep 17 05:24:29 it should just be an object that manages multiple views Sep 17 06:19:57 pretty interesting theory http://blog.sephiroth.it/2014/09/16/why-its-hard-to-find-android-developers/ Sep 17 06:27:12 It's hard because most of then want to hire supertechs with all the answers who don't require time, investment, and long term commitments Sep 17 06:28:55 capella-s3 thats why i've been doing my best (on the side) to learn about startups, marketing, hw design, and getting stuff manufactured ... Sep 17 06:31:03 :) I like IT as much as the business functional side... but I'd rather not own one Sep 17 06:31:47 * capella-s3 would rather gave my ass caned in Singapore Sep 17 06:32:37 capella-s3 ... Sep 17 06:32:46 heh wat Sep 17 06:33:31 "owning a business is hard" Sep 17 06:34:18 finance profits without producing anything, it can't be that hard :P Sep 17 06:35:27 yah... it's all "profit"... XD Sep 17 06:42:32 Companies make interviewing as unpleasant as possible...I find the asking for references most unpleasant Sep 17 06:45:03 The author of the article seems to have a case of Apple envy... faults it for crowding out interest in Android Sep 17 06:45:08 The one thing the HN crowd talks about is how cheap it is nowadays to launch a startup that can potentially have massive success Sep 17 06:46:17 Most Android developers I know, the designer thinks it's all like iOS, your job is basically to get the designers layout onto an Android Sep 17 06:46:20 I thing it's more a matter of market reality... there's no profit motive in open source software dev Sep 17 06:46:52 MS and Apple earn profits Sep 17 06:47:21 One billion Android devices are being sold this year, there is money to be made there... Sep 17 06:47:32 Hardware Sep 17 06:48:10 My apps made over $3000 last December, and that's one person working part-time on them Sep 17 06:48:46 And I used to make 50$hr for IBM Sep 17 06:49:01 That sells stuff Sep 17 06:49:30 People don't focus on earning money Sep 17 06:49:52 not in Android world, i agree Sep 17 06:49:53 I started programming a photo collage app for teen girls or whatever and I'm already bored with it Sep 17 06:50:35 I know it would be worth the development time if I finished it...but I'm currently futzing with cool live wallpapers that probably won't do much Sep 17 06:53:16 In the early days of the web people said there was no money to be made on the web Sep 17 07:04:54 Ologn, also google doesn't allow all countries to participate so i cant charge money or make money from the play store what so ever which is very very sad Sep 17 07:15:40 Ologn capella-s3 noticed this on HN http://startupclass.samaltman.com/ Sep 17 07:17:05 g00s, yes I'll watch the videos when they come out Sep 17 07:20:12 sounds sketchy... in that... I wonder how much rah-rah you'll get vs. useful detail Sep 17 07:20:21 anyone know if proguard can specifically strip classes, regardless of whether they're used or not? Sep 17 07:29:41 so why can't you use custom fonts with android? Sep 17 07:29:53 ...never, ever, ever, ever, ever works Sep 17 07:34:08 MartialLaw: you can't? Sep 17 07:34:49 I can't add AndroidStaggeredGrid to my project. Gradle says: Failed to find: com.etsy.android.grid:library:1.0.5 . Sep 17 07:37:45 barbs, g00s: "Error:Execution failed for task ':app:processDebugManifest'. > Manifest merger failed : uses-sdk:minSdkVersion 11 cannot be smaller than version L declared in library com.android.support:support-v4:21.0.0-rc1" this message is from a migrated project on Android Studio Sep 17 07:38:01 transition ain't that smooth anyway Sep 17 07:38:59 change your version of support library to 4:20.0.0 Sep 17 07:39:28 barbs: Thanks and ya! Did that, same Sep 17 07:40:19 barbs: I think I need support-v4-20.0.0.jar Sep 17 07:40:41 if you're using gradle, no Sep 17 07:40:56 put compile 'com.android.support:support-v4:20.0.0' Sep 17 07:41:05 in your dependencies in your build.gradle Sep 17 07:42:11 f2prateek: yeah Google FAILED to mention .ttf doesn't work anymore with Android...so you have to use .otf Sep 17 07:42:14 barbs: uses-sdk:minSdkVersion 11 cannot be smaller than version L declared in library com.android.support:support-v4:21.0.0-rc1 Sep 17 07:42:22 what an extremely frustrating thing Sep 17 07:42:23 eeinteresting https://github.com/davidmoten/rxjava-jdbc Sep 17 07:44:46 did you put it in the right gradle file? Sep 17 07:45:06 I changed the build.gradle Sep 17 07:45:14 which one? Sep 17 07:45:17 compile 'com.android.support:support-v4:20.0.0' Sep 17 07:45:27 inside my /app Sep 17 07:45:31 ok Sep 17 07:45:48 and you did a sync? Sep 17 07:46:14 sure Sep 17 07:47:45 I stackoverflow-ed a bit before asking on here Sep 17 07:48:09 do a clean? Sep 17 07:49:22 seriously, I need to draw outside the screen..why are the drawings being clipped off to whatever portion of the screen that's visible! Sep 17 07:49:32 okay, cleaned the project and the message is still there Sep 17 07:49:36 zero google results on this! Sep 17 07:50:13 maybe I can increase the borders of Canvas/ Sep 17 07:51:15 yea I think that's my issue, it sees that canvas spans the width/length of the screen, so that's where it draws. anything outside of that is clipped off Sep 17 07:53:10 barbs: I think I should update my AndroidManifest.xml as well Sep 17 07:53:35 In my AndroidManifest.xml, the uses-sdk is android:targetSdkVersion="21" Sep 17 07:54:02 you should be changing the target in your gradle Sep 17 07:56:25 barbs: changing it on my gradle fixed something. But the Error: "Manifest merger failed : uses-sdk:minSdkVersion 11 cannot be smaller than version L declared in library com.android.support:support-v4:21.0.0-rc1" is still there Sep 17 07:58:43 barbs: Ahh, great! I've changed the com.android.support:appcompat-v7:+ to com.android.support:appcompat-v7:20.0.0. Works fine now. Sep 17 07:59:10 barbs: Thanks Sep 17 07:59:24 ahh, that lil bugger Sep 17 07:59:29 knew it was one of those!! Sep 17 08:02:05 hi, i am an iOS developer that is thinking about jumping ship. Sep 17 08:02:10 i was wondering what phone that is either currently out or is soon for release would be good to switch to? Sep 17 08:02:13 preferably im looking for one with an unlocked bootloader Sep 17 08:02:16 JUMP! DO IT!! Sep 17 08:02:35 i also hate java, so what can i do to avoid it? Sep 17 08:03:09 im more of a C/C++/ command line person and java is one of those languages that seems to push you to an IDE Sep 17 08:03:29 oh... you hate Java Sep 17 08:03:32 it is possible to make android apps without an IDE, though generally not ideal Sep 17 08:03:47 it's also possible to make apps without java - using scala or kotlin Sep 17 08:03:52 or Qt Sep 17 08:03:55 Eclim solves the IDE problem for command line people Sep 17 08:03:59 you don't use IDE, but I think you are still coding in Java. Sep 17 08:04:00 so according to this, http://developer.android.com/guide/topics/graphics/2d-graphics.html I do not have to create a bitmap to initialize the canvas with, if Im placing my calls to Draw inside onDraw() Sep 17 08:04:05 I'm using it with Emacs; works flawlessly Sep 17 08:04:09 so that's great, no need for bitmap, one is already provided Sep 17 08:04:10 or using other frameworks (like Xamarin) Sep 17 08:04:13 rweichler: Not much, sadly. There are Scala, Clojure and all the other JVM languages, but an IDE helps these too, even Clojure (a Lisp), because of the Java ffi. Sep 17 08:05:01 barbs: xamarin is worse lol Sep 17 08:05:05 if you want to stick with c++ you can deploy Qt applications on Android Sep 17 08:05:05 dhamidi: thanks ill check it out Sep 17 08:05:06 i'd start with java and then try other stuff as you feel more adventurous Sep 17 08:05:24 g00s: ok Sep 17 08:05:25 you can't access all of the Java stuff from C Sep 17 08:05:37 Yeah, what g00s said. Also, Qt apps feel... weird. :) Sep 17 08:05:46 and the NDK is a .. pita ;) Sep 17 08:05:48 For games that's probably okay though. Sep 17 08:06:50 flan3002: they are adding Android look and feel in Qt 5.4 flan3002 Sep 17 08:07:04 oh another question Sep 17 08:07:08 GrecKo: Oh, I didn't know about that. :/ Sep 17 08:07:14 uh so iOS has this framework called mobilesubstrate Sep 17 08:07:17 that allows for jailbreak tweaks Sep 17 08:07:20 it's not yet out Sep 17 08:07:25 only the alfa Sep 17 08:07:26 is there anything like that for android? Sep 17 08:08:06 xposed ? Sep 17 08:08:06 there are no jails on android Sep 17 08:08:11 rweichler: If I understood that correctly, it allows to alter the behaviour of applications? Xposed maybe? Sep 17 08:09:02 yeah Sep 17 08:09:02 thanks Sep 17 08:09:17 hm, really there are no books on renderscript O.o Sep 17 08:10:41 Yeah, RS doesn't seem to be that popular... :/ Sep 17 08:11:20 ugh, the docs are terrible Sep 17 08:11:42 if i didn't know any better, i'd think most of android was the result of some highschool science project Sep 17 08:12:21 maybe this really is the best google can do. perhaps - not as awesome as i though Sep 17 08:12:23 *thought Sep 17 08:13:07 the renderscrit reference seems like a doxygen generated dump Sep 17 08:14:01 hi! is that possible to know that java class was already loaded? Sep 17 08:14:17 Why do you need that? Sounds like a design mistake... Sep 17 08:14:54 come on man, how is this pawsable Sep 17 08:15:22 let me show Sep 17 08:16:27 checking to see if a class was loaded isn't too shady ... logging frameworks do that to see if the provider is there, for example Sep 17 08:16:42 I tried drawBitmap too, but I really dont think its needed cause the docs say you dont do it unless you're not using Draw calls in onDraw callback where canvas is already provided Sep 17 08:20:03 god damn it, screenshots! come on! Sep 17 08:20:50 hi guys, not really a android question, but is it possible in eclipse to just add the value-XXX string.xml files to a git repo, like github? Sep 17 08:21:09 i want to make it easier for the people that help me with the translations to work on the files Sep 17 08:21:21 you can add absolutely any file to github Sep 17 08:22:13 oles__: http://developer.android.com/reference/java/lang/ClassLoader.html#findLoadedClass%28java.lang.String%29 Sep 17 08:22:33 alright, here we go! Sep 17 08:22:39 here's how the drawings begin https://dl.dropboxusercontent.com/u/19390574/Screenshot_2014-09-17-11-15-46.png Sep 17 08:22:43 dhamidi, thx Sep 17 08:22:45 Sicp, any tips on how to do that, the whole project is stored in a svn repo today, so preferably it should work together with that Sep 17 08:22:49 and here's when I move them! https://dl.dropboxusercontent.com/u/19390574/Screenshot_2014-09-17-11-15-52.png Sep 17 08:22:59 no, see #git Sep 17 08:23:10 you see the drawings clipped off? why! Sep 17 08:24:59 I tried to draw them at the same X of the middle circle, and then craftily moving them back to where I wanted them to be but this produced the same result Sep 17 08:25:31 *craftily* Sep 17 08:26:17 dhamidi, it is not a public method Sep 17 08:30:17 Hi Sep 17 08:30:26 How can I make a backup of stock recovery? Sep 17 08:30:37 Any ideas? Sep 17 08:34:46 oles__: https://gist.github.com/dhamidi/b611baeb1cc860e08880 Sep 17 08:40:19 How do you reprepare the options menu on fragment transaction? Sep 17 08:44:42 dhamidi, have you tested that code? Sep 17 08:47:30 no Sep 17 08:52:03 is SparseBooleanArray an Android specific thing but not java ? Sep 17 08:57:11 Yes. Sep 17 08:57:17 God studio is such a memory hog. Sep 17 08:57:41 Starts ok, but as I work it'll grow to ridiculous proportions Sep 17 08:58:05 God studio is still in development Sep 17 08:58:12 use Android Studio or Eclipse Sep 17 08:58:28 huh? I uses only like 1-1,5GB of RAM Sep 17 08:58:35 I meant android studio. There's a studio by itself? Sep 17 08:58:42 Was 4gigs here Sep 17 08:58:42 there's God studio Sep 17 08:59:04 Aha I meant God, android-studio Sep 17 08:59:14 oh Sep 17 08:59:26 Yeah ponctuation counts! Sep 17 08:59:37 o/ Sep 17 08:59:38 well I just revealed a secret project Sep 17 09:00:36 4GB is still ok Sep 17 09:00:38 Don't know if it's really secret, but never heard of it and can't seem to google it. Sep 17 09:00:51 hi, doessomebody happen to know wifi socs used in xperia P and xperia U, or else a way I could find out? Sep 17 09:01:26 ive been workign with 4gb ram with android studio and its absolute rubish Sep 17 09:01:30 basically Sep 17 09:01:35 Yes well I am starting to realize how fast 8 gigs can go. Sep 17 09:01:52 if i use chrome or rythmbox or whatever compile takes over 5 minutes Sep 17 09:01:54 and freezes Sep 17 09:01:58 haha Sep 17 09:02:01 I am doing something wrong with EventBus: http://pastie.org/9561234 onEvent methond calls several times, when I reopen activity, but should calls only once. Any reason? Sep 17 09:02:01 that is awful! Sep 17 09:02:05 ya Sep 17 09:02:13 my swap is usually at 2gb Sep 17 09:02:27 it helped when i moved to firefox Sep 17 09:02:34 and installed openjdk 7 Sep 17 09:02:39 its lighter than oracles Sep 17 09:02:45 Luckily for me I have permanent info on RAM available, so I can react in time if not too concentrated! Sep 17 09:03:49 mhm, I'm running 16G on all my machines now, the cost really outweights all the frustration Sep 17 09:04:50 I was planning on it for next rig, might not wait after all. Sep 17 09:05:41 Which is ridicolous by itself... People have been able to work on machines with a lot less RAM decades ago. Sure, they did less, but then it's also people choosing Node.js instead of something that compiles to native code... Sep 17 09:06:17 *ridiculous ^^ Sep 17 09:09:00 I was going to say those words Sep 17 09:09:05 but then I remembered who cares Sep 17 09:09:20 Sure and we didn't have phones then either, but you know, progress etc. Sep 17 09:09:44 not what he's saying Sep 17 09:09:54 Hey, I have an activity that contains a list of about 50 buttons(each with a background), but loading that activity takes about 1.5 seconds , I have seen apps with lots of buttons but faster , so how can I improve the performance/load time ? Sep 17 09:09:59 he's saying that devs nowadays disregard memory issue because "there's so much of it, who cares" Sep 17 09:10:01 I think Sep 17 09:10:14 at least on the desktop Sep 17 09:10:39 Oh ok. Indeed. Sep 17 09:11:26 Someone on HN compared it to old US cars, which were designed with 'fuel is cheap, engineer time is expensive' or something along the lines. Anything that is supposed to be used by a large userbase is worth quite some time on the engineering side. Not saying Jetbrains does a bad job, their products are amazing, it's just the general notion, see Atom. Sep 17 09:11:28 but then you get those users who do not have 145,0000,000GB of ram Sep 17 09:12:12 yea totally valid Sep 17 09:12:35 I mean... I have 8G of RAM on my two-year-old laptop, so I only run into problems when simultaneously compiling stuff like LLVM and some Android app, but... Sep 17 09:14:49 little proof is that everybody touts C for being so memory efficient, and C is so old Sep 17 09:16:00 more languages would do that if the devs cared about being memory efficient, just like any other feature they implement in the language Sep 17 09:16:19 its not like a black art that only C knew how to do Sep 17 09:17:28 Well, in some places it is. Try optimizing for cache performance/data locality in Python... Sep 17 09:17:56 Or in Java for that matter. Basically, and all-heap language. Sep 17 09:17:58 *any Sep 17 09:18:15 (Yeah, primitives, but... no.) Sep 17 09:19:15 Not saying the productivity gains aren't worth it in many occasions. :) Sep 17 09:19:24 I'm talking about the design phase, before Java was created, to have efficiency on the list of things it needs to be Sep 17 09:19:36 they could've worked to make it do what C does Sep 17 09:19:40 but didnt Sep 17 09:20:28 Hey there. I wrote an activity which registeres itself as a Launcher. This activity starts a new activity with startActivity(..); in the onCreate() - but this somehow blocks the BOOT_COMPLETED intent from being sent, system wide. Any idea why? Sep 17 09:20:36 Well, that simplifies the mental and computational model in favor of productivity (as long as you don't have to optimize stuff), which is a valid thing to do imo. ^^ Sep 17 09:21:19 Java (Hotspot) is still insanely fast... so they did a good job. Sep 17 09:22:11 anyone knows the answer to this https://github.com/greenrobot/EventBus/issues/116 ? Sep 17 09:29:36 I come from a C background and am astonished by how much memory other languages use. My school made it very important that we control memory and have no leaks what so ever. Sep 17 09:30:05 memory leaks and memory usage are two different things Sep 17 09:30:17 using 8GB of RAM so you can cache stuff you would have to load form a slow external storage is NOT a memory leak Sep 17 09:30:36 flan3002, you're just forgetting how SLOW old computers were back then. Sep 17 09:30:45 The slightest leak would get us an F on the project. And of course it was important that memory usage be in accordance with what the prog was doing Sep 17 09:30:53 because they had to go to utterly slow floppies and HDDs to get data instead of keeping them around in RAM Sep 17 09:31:15 Mavrik: To be honest, my first computer was an Asus eeepc. So... yeah, no real experience with the really old stuff. Sep 17 09:31:50 yeah, booting a PC or PC XT took up to 5 minutes easily Sep 17 09:32:07 and no, I do not miss disk scratching and waiting for EVERYTHING in Win3.1 either :P Sep 17 09:33:16 Mavrik: Sure they are different, but e.g. an ArrayList is something that doesn't have to/should exist. C++/Rust like specialisation, ie. /real/ generics, would have prevented this awfulness. Sep 17 09:33:26 um Sep 17 09:33:28 *shouldn't. Damn. Sep 17 09:33:30 why? :) Sep 17 09:34:03 I mean, type erasure is annoying as heck in Java... but ArrayList is just like C++ vector or any other languages list the basic workhorse of a data structure :) Sep 17 09:34:44 Mavrik: Because on a 64bit JVM, a boxed integer takes 24 bytes, instead of the supposed 4 bytes. Yes, that is awful. Not good for the cache, not good for anyone... Sep 17 09:36:03 hmm, yes thats an issue, but a very minor one Sep 17 09:36:08 So, in C++, or Rust, you can store 24 integers with the same memory that you can store 4 Integers (note the capital I) in Java. Disregarding collection overhead. Sep 17 09:36:26 of course, but you're never storing huge amounts of integers in arraylists anyway Sep 17 09:36:43 as soon as you have a single class all of that optimization would go away Sep 17 09:36:52 note that e.g. C# does the same Sep 17 09:37:07 That doesn't make it any better. Sep 17 09:37:25 Any yes, you do want to be generic over primitive types. Often. Sep 17 09:37:44 Maybe not for your usual enterprise website or thingy, but for many of what I wrote. Sep 17 09:38:16 I ended up using a template and specialising them via textual replacement (!?!) to compile them... Terrible. Sep 17 09:38:39 of course Sep 17 09:38:50 but if Java would be doing stuff like Rust... it would be niche like rust Sep 17 09:39:17 what I do like about the language is that the design is pragmatic enough that we can run software for years without having to redeploy whole server racks just on a whim of the language designer Sep 17 09:39:27 Hey , how can I boost load time of an activity with more than 60 buttons ? Sep 17 09:39:46 yes, lack of basic types coupled with generics is annoying for few use-cases when you actually need to do that Sep 17 09:40:06 but the tradeoff was giving up on GC and breaking backwards compatibility. Sep 17 09:40:07 Not saying Java did completely stupid choices, but as I understood, the only reason we have type erasure is backwards-compatibility. Sep 17 09:40:14 killer, I'd say that the issue is having 60 buttons :) Sep 17 09:40:46 flan3002, mhm, sadly backwards compatibility is commonly a huge crutch Sep 17 09:41:07 Mavrik: So, for basically anything. Want a Graph datastructure of ints? Well, specialise it. Now want it for floats? Yeah... Sep 17 09:41:34 yeah, that's the thing that pisses me off with Go for example, where is has the same issue Sep 17 09:41:36 Mavrik: Well, it's a list of buttons and each has to be clickable Sep 17 09:41:44 is it possible to run multiple instances of asynctask at the same time Sep 17 09:41:47 i always wondered about that Sep 17 09:41:55 And to be fair, Rust has an amazingly large community, given the fact that there's still no release candidate. Sep 17 09:42:05 I have seen some apps with good load time and about 50 buttons Sep 17 09:42:34 Mavrik: I looked at Go and immediately realized I won't be happy with it, after having used Rust... Sep 17 09:42:38 I didn't really have time to look at Rust yet, kinda stuck in C++/C/Java/Python world right now :) Sep 17 09:42:57 but I'm trying to avoid platforms until at least they get basic maturity Sep 17 09:43:27 Mavrik: You should. You'll be fighting the compiler, but once it compiles, idiomatic Rust doesn't really have runtime failure. Sep 17 09:43:44 I'm more worried about linking it against stuff :) Sep 17 09:43:56 What linking? Sep 17 09:44:05 other C/C++ libraries Sep 17 09:44:29 Well, you don't have to. And it's not /that/ hard. :) Sep 17 09:44:39 No, I actually have to :) Sep 17 09:44:47 Gamedev? Sep 17 09:45:01 None of my projects I did in C/C++ are without dependencies on other libraries and drivers Sep 17 09:45:12 No, large scale realtime video processing deployments :) Sep 17 09:45:20 And for other stuff I just prefer to use python anyway :) Sep 17 09:45:44 Too dynamic for me, but sweet for medium-size scripts. ^^ Sep 17 09:50:29 killer, post your adapter Sep 17 09:51:06 Often I find myself wanting to have a view with match_parent width and a proportional height that follows a certain aspect ratio. I haven't found an elegant way to do this yet on Android. Any recommendations? Sep 17 09:53:11 snowpong, you'd have to change it through code. like yourView.setHeight(yourView.getWidth()/2) or something like that. Sep 17 09:54:33 привет, ребята Sep 17 09:55:47 "C/C++" Sep 17 09:55:48 ... Sep 17 09:55:48 jvrodrigues: yeah I've found lots of different snippets that fixes this - I was hoping there might be a common solution or trick I had missed Sep 17 09:59:06 snowpong, not that i am aware of. If pretty neat code is what you want you can always create different layouts for diferent screen sizes and change the height according to each screen. Sep 17 09:59:18 finally Sep 17 09:59:25 here's the answer to my day-old question http://stackoverflow.com/questions/4028270/can-i-draw-outside-the-bounds-of-an-android-canvas Sep 17 09:59:39 you need to increase the rect of the canvas Sep 17 10:00:21 otherwise you get this https://dl.dropboxusercontent.com/u/19390574/Screenshot_2014-09-17-11-15-52.png Sep 17 10:06:46 jvrodrigues: the cleanest solution (for existing code) seems to be to introduce a custom aspectratiolayout (of course with the side-effect of increasing the viewtree depth) Sep 17 10:07:41 hmm that is pretty clean, but i wouldnt call it optimal Sep 17 10:07:56 you must be one of those OCD coders. God i love them, everything commented and in the right place Sep 17 10:31:44 I have a List. How can I use builder.setMultiChoiceItems(List, new boolean [] { false, false,false }, new DialogInterface.OnMultiChoiceClickListener() ? Sep 17 10:32:15 the setMultiChoiceItems is accepting only CharSequence[] Sep 17 10:33:07 i’m trying to understand the stats at the developer console… for “current installs by device”, it says 26.73% for “all apps in music & audio”… what does that mean? Sep 17 11:12:49 can I get a reference to Canvas from an Activity? i.e. I want to do a clipRect on a canvas that OnDraw of my class that extends View will use evrytime I want to draw something Sep 17 11:13:05 I want to get the reference to it so that I'd be able to set this clipRect() only once on it Sep 17 11:15:29 Hi, is there a Google API to verify if and Android ID, Device ID, IMEI, GSF ID, Subscriber ID or SIM Card Serial is valid and in use on Google's platform? I need to ensure that the access to my anonymous API from the device is legit and that the IDs are not just random numbers generated value by a spammer. Sep 17 11:16:13 mak``: no Sep 17 11:16:59 mak``: what was wrong with Google Auth? Sep 17 11:18:18 I made the mistake of requesting assets in certain dp size Sep 17 11:18:41 now my xhdpi asset has the numerical dimensions that I specified Sep 17 11:19:25 appel1, I guess it would not work on a device without a google account signed in? Sep 17 11:19:30 i am trying to upload image from phone to Parse Sep 17 11:19:40 I have the URI from phone like this, content://media/external/images/thumbnails/1023 - how do I get the bytes data from it ? Sep 17 11:20:01 mak``: true, are any of your potential customers likely to not use any google account? Sep 17 11:20:02 basically byte[] Sep 17 11:21:58 appel1, I need to estimate that case. I don't know if there are any stats around about that. I am evaluating options. Sep 17 11:59:16 hi Sep 17 11:59:21 can some one help me with this Sep 17 11:59:22 http://stackoverflow.com/questions/25889961/adb-device-not-found-glx-t1-kitkat-android-tablet-on-fedora Sep 17 12:01:21 jvrodrigues: haha! didn't see your OCD comment before now - too busy testing my solution :D Sep 17 12:02:25 molavy, -> #android-root Sep 17 12:08:51 ravilov: thanks Sep 17 12:33:31 hey guys Sep 17 12:33:35 so quick question Sep 17 12:33:43 how do i get gradle to build aar? Sep 17 12:34:12 i am allready applying plugin 'android-library' Sep 17 12:34:28 dont know if that is even relavent or not Sep 17 12:34:48 Any way to develop for android L without a device? Sep 17 12:34:57 Emulator? Sep 17 12:35:03 The emulator works however its incredibly unusable Sep 17 12:35:21 i.e. it's slow Sep 17 12:35:47 Well. Either you emulate or you don't. If you don't, you need a real device. There can't be any other way. What are you asking for? Sep 17 12:36:04 Are you using HAXM/KVM with an x86 image? Sep 17 12:36:34 I think so? Sep 17 12:36:53 I downloaded haxm through SDK manager, does that install haxm? Sep 17 12:36:58 Well, then you need a better pc. Sep 17 12:37:11 Well, you still need to explicitly use it. Sep 17 12:37:17 * flan3002 uses too much well. Sep 17 12:38:18 rymate1234: Is your CPU/ABI field in the AVD editor set to Intel Atom (x86)? Sep 17 12:38:40 Yes Sep 17 12:39:21 What processor do you use? Sep 17 12:40:04 Intel cote i5 4210u Sep 17 12:40:13 rymate1234: Does $ egrep '^flags.*(vmx|svm)' /proc/cpuinfo output anything? (Shame on you if you use Windows) Sep 17 12:40:30 Well I'm on windows Sep 17 12:40:49 Should I make a live USB and run the command Sep 17 12:40:51 xD Sep 17 12:40:52 rymate1234: https://software.intel.com/en-us/android/articles/installation-instructions-for-intel-hardware-accelerated-execution-manager-windows Sep 17 12:41:12 Install it and come back. Sep 17 12:41:28 ok Sep 17 12:43:22 flan3002: that worked! Sep 17 12:43:24 Thanks Sep 17 12:43:29 Good. :) Sep 17 12:45:59 Now to mess with android L Ui Sep 17 12:54:34 flan3002, well well well Sep 17 12:55:02 ravilov: Yeah? Sep 17 12:56:38 nothing, just being crazy, you should probably ignore me right now Sep 17 12:56:58 Being crazy, ravilov? What do you mean? Sep 17 12:57:36 Oh and telling me to ignore you is an example of reverse psychology that sadly still works for me. :P Sep 17 12:58:06 * flan3002 uses too much well. Sep 17 12:58:07 :p Sep 17 12:58:21 just messing Sep 17 12:58:22 Yeah, I do. :) Sep 17 12:58:29 Sometimes. Sep 17 12:58:32 ok, then Sep 17 12:58:48 you should never ignore me, ever! Sep 17 13:00:17 ...cool, I think that actually worked Sep 17 13:35:57 Hello guys, some advise needed. I want to use fragments in my app and I want to change ActionBar's title from a fragment. Do I need to retrieve the ActionBar object from the Activity so I can change the title or is there a better approach? Sep 17 13:38:05 I can also pass the ActionBar object in the Fragment's constructor. Anyway, I want the best programming pattern for this situation. I hope you can help. Sep 17 13:38:14 Number5: I would use an event bus such as auto to emit an actionbar title event out to whoever is listening. That way the two are completley decoupled. Sep 17 13:38:59 Number5: also don't pass things into fragment's constructors, they need to have a empty default constructor for the system to access. Sep 17 13:39:09 Ok, so like using an Interface with a method onChangeTitle(String title) or something. Sep 17 13:39:12 and by auto i mean otto* Sep 17 13:39:40 Yeah that's the simpiler way one sec let me see if I can find an example Sep 17 13:39:52 ok, i'll wait Sep 17 13:41:15 so http://developer.android.com/guide/components/fragments.html down in the section title creating event callbacks to the activity Sep 17 13:41:43 pieces029, this can also be done to change ActionBar's action menu right? But I found another example which I can override Fragment's method: public void onCreateOptionsMenu(Menu menu, MenuInflater inflater) and change there the menu. Sep 17 13:41:54 You have your activity implement an interface, and then ensure that when your fragment is attached to that fragment that it implments the interface you expect. Sep 17 13:42:00 This method is totally different to interface mechanism. Sep 17 13:42:14 yeah, that's how you would override what's in it. Sep 17 13:44:59 pieces029, thnx for your time my friend. Sep 17 13:45:29 no problem hope that helps Sep 17 13:58:24 Good morning all Sep 17 14:03:13 Goodmorning? It's like 16:00 h in Europe :P Sep 17 14:10:17 hello friends Sep 17 14:14:52 In AS, any idea why I can't access to a module's classes? Sep 17 14:20:19 gdrc: because you didnt add the module correctly? Sep 17 14:29:02 http://www.stereogum.com/1473772/watch-napster-documentary-downloaded/video/ Sep 17 14:31:28 does anyone know a way of getting the sha-1 key from code? Sep 17 14:31:31 not the hashkey Sep 17 14:31:45 for some reason the key in my debug.keystore does not match the key android-studio uses Sep 17 14:34:38 https://www.youtube.com/watch?v=ulShj4keKNw#t=135 Sep 17 14:39:52 Seems like everything is going to gaming in the market Sep 17 14:40:07 Prob because they have an app for everything now Sep 17 14:49:36 I'm getting an error while trying to get a UI Hierarchy snapshot with ddms Sep 17 14:49:44 Error obtaining UI hierarchy Sep 17 14:49:44 Error while obtaining UI hierarchy XML file: com.android.ddmlib.ShellCommandUnresponsiveException Sep 17 14:49:52 can someone help me figure out what is wrong? Sep 17 14:51:25 Hi guys. I found out that when I pass MediaStore.EXTRA_OUTPUT extra argument to the Camera app it ends up with creating actually 2 files. The first one is saved in the folder I specified when started Camera app (it has something ~1mb filseze and good quality) and the second one is saved in DCIM/Camera folder (and it has bad quality and fewer file size). And what I also noticed is that for the photo saved in my app-specific Sep 17 14:51:25 folder doesn't have a thumbnail even after I call MediaStore.Images.Media.insertImage() method to create a thumbnail for it. Sep 17 14:51:50 I tested it in both emulator and real device and it behaves identically Sep 17 14:52:32 My question is why does it work like that and how can I avoid creating of image in DCIM/Camera folder? Sep 17 14:54:34 In Android L, does Activities actually have transitions like this : http://material-design.storage.googleapis.com/videos/meaningfultransitions-visualcontinuity2_large_xhdpi.mp4 ? Sep 17 14:54:53 with the system bar moving Sep 17 14:55:28 run an L preview emulator and see Sep 17 14:57:17 I don't have one and just wanted a quick answer, if nobody answers that's what I'll do eventually Sep 17 15:00:03 Please help, file refuse to run after i change the package name in Android Studio... Sep 17 15:00:07 please help... Sep 17 15:00:43 Gradle Project failed, please fix your project and continue Sep 17 15:06:45 GrecKo: Vidéo can't be played due to corrupted file Sep 17 15:07:29 I currently have an ArrayAdapter, but I want to more dynamically add and remove items depending on what the searches for. So I don't know ahead of time how many "unfiltered" items there are Sep 17 15:07:41 oh, that's dumb, I can't cast my screen to chromecast on my 4.4.4 device? :( Sep 17 15:08:02 Should have bought a Nexus. Sep 17 15:08:33 Schoumi: it's not corrupted for me Sep 17 15:08:35 Should have gone to SpecSavers Sep 17 15:12:11 Mavrik, nah Sep 17 15:12:14 Gradle Project failed, please fix your project and continue Sep 17 15:12:15 stupid thing is too big Sep 17 15:15:51 Mavrik, I have nothing to use my chromecast on anyway Sep 17 15:15:54 'cept for my pc monitor Sep 17 15:16:09 mhm Sep 17 15:16:21 I use my chromecast alot, but I've never actually used the mirroring Sep 17 15:16:26 there's just... no usecase for me Sep 17 15:16:57 I need a new receiver Sep 17 15:17:19 http://www.amazon.com/Yamaha-RX-V775WA-Channel-Network-Receiver/dp/B00B981F2Y/ref=sr_1_1?ie=UTF8&qid=1410967030&sr=8-1&keywords=yamaha+775+receiver Sep 17 15:17:23 I want that receiver bad Sep 17 15:17:32 costco has it for $490 with the bluetooth module instead of wifi Sep 17 15:17:40 heh :) Sep 17 15:17:41 just need to convince the boss to let me buy it, haha Sep 17 15:22:52 johnflux androidhive.com Sep 17 15:24:20 johnflux androidhive.info Sep 17 15:29:55 what is the best way to send some application wide event. For example some thread throws out some event, and all fragments and/or activty's that has subscribed will recieve it ? Sep 17 15:30:32 Hi, I have an Android Studio question.. The autocomplete for my xml layouts is sort of broken. It only shows me autocomplete options from a library project, but no options from my own projects. Anyone know how to fix that? Sep 17 15:34:37 i am on version 0.8.9 Sep 17 15:37:30 vegetablesalad19: Observable/Observer ? Sep 17 15:39:49 vegetablesalad19, Otto Sep 17 15:48:14 GrecKo,MikeWallaceDev, Thanks will check them out! Sep 17 15:48:24 somebody help... how can i blur image in android Sep 17 15:48:51 i want to blue a image and place view on top it Sep 17 15:49:15 So with Retrofit callbacks, I guess there could exist the situation that the fragment it is used in is being destroyed right as a network call finishes which could cause a NPE right? Should I just isVisible check before I do anything in it? Sep 17 15:53:34 I think what I really need to do is implement Otto. But for the time being... Sep 17 16:00:38 mattblang: isAdded() Sep 17 16:01:10 i wonder why filter doesn’t support generics… Sep 17 16:05:03 should I be putting "final" in front of pretty much every parameter in every function? Sep 17 16:05:17 i wouldn’t go out of your way unless you have a reason to Sep 17 16:05:34 otherwise you’re gonna have lots of code wrapping Sep 17 16:05:41 but it can’t hurt Sep 17 16:06:21 JacobTabak thanks, so it does make sense that the case I mentioned could happen? Sep 17 16:06:30 yes mattblang all the time Sep 17 16:06:38 your fragment will be destroyed and your views will be null Sep 17 16:06:42 I have a runtime exception casued by a HttpMessageNotReadableException caused by a SocketTimeoutException/SSLException. If I do catch(RuntimeException e) { if(e.getCause() instanceof HttpMessageNotReadableException){handle} else throw e;} it should be handled right? Sep 17 16:07:08 Syzygy_: that should not be a runtime exception Sep 17 16:07:09 JacobTabak honestly, I am going to switch to Otto I think for one of our future versions. BUt I need to patch the existing one. Sep 17 16:07:17 Syzygy_, it shouldn't but it is. Sep 17 16:07:27 very weird Sep 17 16:07:43 spring framework? psssh Sep 17 16:08:49 yeah, something in REST/Spring does that Sep 17 16:09:38 Syzygy_: if HttpMessageNotReadableexception is a runtiem exception, you should be able to catch it directly Sep 17 16:09:44 without catching runtime and rethrowing Sep 17 16:09:56 JacobTabak, Syzygy_ my company uses Spring almost exlusivly. I must say, it is pretty nice to be able to almost copy, verbatum, the server side POJOs to my client to use with Retrofit and ORM Sep 17 16:10:28 im really not commenting on spring, it should just be an ioexception ... Sep 17 16:10:39 it seems to be a runtime exception according to the hockey stacktrace. Sep 17 16:10:48 aye thats what the documentation says Sep 17 16:10:50 but you can catch it directly Sep 17 16:11:11 with catch(HttpMessageNotReadableException e)? Sep 17 16:11:12 Syzygy_ why did you reference yourself in a message, haha Sep 17 16:11:21 Syzygy_: it should work Sep 17 16:11:43 mattblang, still a bit hungover from yesterdays office party i guess Sep 17 16:11:55 Syzygy_ hahaha, I thought maybe it was something like that Sep 17 16:12:05 800 bucks of booze and snacks Sep 17 16:12:10 Nice! Sep 17 16:12:16 What was the party for Sep 17 16:12:52 largest european convention related to our work interest in our city Sep 17 16:25:52 JacobTabak, can you explain to me why catching the HttpMessageNotReadableException directly would work? Sep 17 16:26:03 because it extends RuntimeException Sep 17 16:26:47 I don't see how that would change anything Sep 17 16:27:03 that is the exception being thrown, so you can catch it Sep 17 16:27:44 it's thrown at some point in the REST/Spring API, at the point where I can catch it it's only a runtime exception though Sep 17 16:28:19 i can’t explain Spring, but the exception you mentioned was a subclass of runtimeexception Sep 17 16:28:31 so i assumed it was being thrown directly Sep 17 16:29:28 if it were the HttpException I could catch it as both a RuntimeExcepton or the HttpException, but since it's a runtime exception caused by the HttpException I can't catch it Sep 17 16:29:44 sounds like an inconvenience in spring framework Sep 17 16:30:20 yeah Sep 17 16:30:49 but I tought I were clear on the fact that it's a runtime exception, so you telling me i can catc hthe httpexception confused me Sep 17 16:31:04 my point was that the httpexception IS A runtimeexception Sep 17 16:31:40 http://docs.spring.io/spring-framework/docs/current/javadoc-api/org/springframework/http/converter/HttpMessageNotReadableException.html = base class is runtimeexception Sep 17 16:31:48 sure, but you said i can catch it directly Sep 17 16:32:45 right, i dont know the inner workings of spring Sep 17 16:33:01 I'm not "in" spring Sep 17 16:33:07 but i assumed based on hte fact that the httpmessagenotreadablexception was a sublclass of runtimexception Sep 17 16:33:15 that it was in fact the exception being thrown Sep 17 16:51:16 windowSoftInputMode="adjustPan" seems to nearly work, but still a bit of the bottom part is hidden and it looks like it's exactly the height of the dictionary on top of the virtual keyboard Sep 17 17:08:46 hey, i'm having some trouble with intellij and gradle re: android projects...it's kind of intellij-specific, but i figure more than a few of you use android studio, which afaik is a customized version of intellij for android development Sep 17 17:09:45 basically, the IDE keeps running some gradle task in the background, and it keeps wiping the IDEA project's dependencies Sep 17 17:10:09 im assuming it's trying to refresh it from the build.gradle file, which i believe was generated by android studio...i apologize for not giving enough information, but i'm flailing :\ Sep 17 17:14:15 then delete build.gradle Sep 17 17:16:21 If it's a gradle projects, dependencies are defined in build.gradle. The idea project is set up from that automatically. Sep 17 17:18:58 diphtherial did you use the Project Structure dialog to add dependencies ? Sep 17 17:20:32 g00s: i attempted to, and they were removed when intellij executed some gradle refresh background task Sep 17 17:20:49 deleting the build.gradle isn't an option, since my collaborator uses both gradle and android studio Sep 17 17:20:56 yeah, to hell with that stupid dialog - just edit the build.gradle files directly and then do a gradle sync Sep 17 17:21:09 i see; i'll try that, thanks Sep 17 17:21:56 * capella sleep is for the weak ... :/ Sep 17 17:22:08 ok, so i'm getting the following error: Error:Could not find com.google.android.gms:play-services:5.0.98. Sep 17 17:22:11 Required by: Sep 17 17:22:47 diphtherial in the sdk manager, make sure you have ths support library repository Sep 17 17:22:56 and play services i guess Sep 17 17:23:07 urgh, that could've gone very badly, apologies. i've verified that that version of the play services library is present in the /extras/google/m2repository/ directory...maybe i'll post the build.gradle, too Sep 17 17:24:42 <_kw> I have an app with a webview where a user is logged in, how can I trigger a download for an image inside it? Sep 17 17:25:41 <_kw> going externally means logging in in the browser I guess Sep 17 17:25:45 diphtherial: make sure you are using the right version Sep 17 17:25:54 here's the build.gradle from the app directory: https://dpaste.de/3yK8 (there's a top-level build.gradle as well that's mostly empty, too: https://dpaste.de/Kv1g ) Sep 17 17:26:55 lasserix: fair point, i'll check again Sep 17 17:27:19 _kw use picasso Sep 17 17:27:22 generally, i wish i had a better idea of what intellij was doing behind my back Sep 17 17:27:39 diphtherial: i had that problem, so when i started it would say you need to update google play services on my phone Sep 17 17:27:51 just use gradle... Sep 17 17:27:53 then someone pointed out i was using the wrong version of play services in my build dependences Sep 17 17:27:55 i have expandableListView. populated by custom adapter with List inside it. when i remove items from the list and notifyDataSetChanged, i encounter weird behavior when some of the groups are expanded Sep 17 17:27:57 and don't configure your project using intellij Sep 17 17:28:35 pfn: well, i still like to use intellij for its code navigation and refactoring features Sep 17 17:28:43 groups are removed but children stay. to hide them i have to click the remaining groups few times Sep 17 17:28:58 if it's possible to manage the build process from the command line and just disable whatever automatic junk intellij's doing, i'd love to do that instead Sep 17 17:29:01 is this bug Sep 17 17:29:37 <_kw> picasso you say? Sep 17 17:29:38 anyhow, i'll figure this out; thanks for the advice so far, it's been helpful Sep 17 17:30:00 isn't this to be expected ? http://phandroid.com/2014/09/16/protip-swiping-away-recent-apps-killing-performance/ Sep 17 17:30:01 on a much easier to address note, is there something less verbose out there for doing simple http gets/posts than apache's HttpClient? Sep 17 17:30:24 if it kills the process, the cached process is gone. but is there more to it ? Sep 17 17:30:26 hello! i'm trying to set up my bluetooth to receive serial commands and use rfcomm and sdptools to set it up listening on channel 14 but connecting from android device via blueterm is not working Sep 17 17:30:30 has anyone an idea? Sep 17 17:30:52 sdptools browse local lists my SP at channel 14 Sep 17 17:31:10 diphtherial, so, do that, but don't change your project settings from intellij Sep 17 17:31:24 diphtherial, intellij only reads your gradle script Sep 17 17:31:30 it's not doing anything automatically... Sep 17 17:32:09 diphtherial, quit adding dependencies using intellij, do it in gradle Sep 17 17:32:15 is the iphone7,1 the iphone 6 or 6+ Sep 17 17:32:27 im pretty sure its the 6 Sep 17 17:32:35 pfn: understood, and i'd rather do that; i only attempted to add dependencies b/c it didn't seem to be picking them up from the gradle build script correctly Sep 17 17:32:53 EvilPenguin lol, wrong channel :) Sep 17 17:32:58 it does seem to be running some kind of gradle sync task; what is that, actually? is that a gradle build target? Sep 17 17:32:59 lol damn it. Sep 17 17:33:05 hehe Sep 17 17:33:23 g00s: im in too many damn childs. Sep 17 17:33:24 erm Sep 17 17:33:28 s/child/channels Sep 17 17:33:28 oh my Sep 17 17:33:35 yes you are Sep 17 17:33:40 at least that is now obvious :p Sep 17 17:33:48 awkward :) Sep 17 17:33:50 lol. Sep 17 17:33:52 EvilPenguin is Evil :) Sep 17 17:33:54 diphtherial, it's just reading the gradle script and loading dependencies Sep 17 17:34:01 g00s: and evil hacker :) Sep 17 17:34:47 evil? hacking? I sure hope children have nothing to do with this Sep 17 17:35:33 Is there a recommended way of privately hosting library projects on github/bitbucked and then pulling them from a application project? Sep 17 17:35:44 (this is what you get when you so much as mention the unholy device) Sep 17 17:36:26 actually now is a good time to get into iPhone again; swift kind of resets / levels the playing field Sep 17 17:39:53 pfn: it still seems to have trouble finding the play services library despite my verifying in the sdk manager that it's there (albeit it's for API level 19, not sure how to force it to install the one for 20, if it exists) Sep 17 17:40:09 huh? play services is not tied to an api level Sep 17 17:40:57 ah, sorry, i confused the rev. column and API column Sep 17 17:42:10 android-dev855: just generate an aar file and host it wherever you can add it as a local or push it into a maven repo Sep 17 17:48:26 http://pastebin.com/pUayNUF5 can someone try this code and/or tell me how I would catch this exception? Sep 17 17:48:32 g00s: dunno yo umight find this interesting http://hughewilliams.com/2012/04/19/ranking-at-ebay-part-1/ Sep 17 17:51:25 Syzygy_: what are you trying to do? Sep 17 17:53:05 A similar exception occurs somewhere in REST/Spring and I can not reproduce it. I'm unable to catch it either. So I'm trying to do it by causing a similar exception Sep 17 17:53:23 I guess you could replace the exceptions I use with some others Sep 17 17:53:26 You'll have to catch it in doInBackground Sep 17 17:54:28 doInBackground is not called from execute, it's done later in another thread Sep 17 17:55:14 ok, is there no way around that? Sep 17 17:55:31 No Sep 17 17:55:35 Productivity of meetings... https://www.youtube.com/watch?v=BKorP55Aqvg Sep 17 17:58:14 because i do believe the asynctask is nested within a library, so I can not change it to catch the exception. Sep 17 18:03:01 Then you're out of luck Sep 17 18:05:03 that's kind of bad considering i have nearly 400 crashes caust by this on a test build Sep 17 18:05:05 has anyone successfully created a serial connection via rfcomm? Sep 17 18:06:15 am i correct in thinking that gradle's looking for the google play library in /extras/google/m2repository ? Sep 17 18:07:31 hm, iOS emulator has support for BT. Google said they were going to add BT support to the emu years ago ... Sep 17 18:07:36 *iOS simulator Sep 17 18:09:00 yeah Sep 17 18:09:03 bad that they didnt yet Sep 17 18:11:26 diphtherial: yes Sep 17 18:13:49 diphtherial: can you run your project on your phone? Sep 17 18:13:53 so, here's the contents of that folder: https://dpaste.de/1xRH Sep 17 18:14:17 lasserix: i haven't been able to build it yet due to the project complaining that it can't find the play services library, although i swear it was able to find it the other day (not helpful, i know) Sep 17 18:14:27 ahh Sep 17 18:14:54 did you verify your android sdk manager has all its things? Sep 17 18:14:54 also, here's the main build.gradle and the app build.gradle, respectively: https://dpaste.de/Kv1g , app: https://dpaste.de/3yK8#L Sep 17 18:15:07 lasserix: yes, the play services library is installed Sep 17 18:15:28 it's listed as a subdir in the m2repository folder structure that i pasted earlier Sep 17 18:15:41 5.0.98 != 5.0.89 Sep 17 18:15:41 diphtherial: oh why dont you try using the android studio wizard to create a google play services project then copy the build file out of that? Sep 17 18:16:07 JakeWharton: ah, agh. @_@ thank you, thank you, and i feel awfully embarrassed Sep 17 18:16:51 well, needless to say, it works now Sep 17 18:17:56 thanks again Sep 17 18:19:36 sometimes you just need fresh eyes Sep 17 18:22:45 I need to find an Eye Replacement Center near me Sep 17 18:22:52 I could definitely use some fresh eyes Sep 17 18:23:01 how about z vision eyes? Sep 17 18:23:06 much better than x Sep 17 18:24:10 hrm. ok, hate to go on about this, but it's actually not automatically adding the depencies to intellij's "project settings->library" pane Sep 17 18:24:50 strangely the app is building (e.g. running make, which i assume runs the appropriate gradle targets, reports success), but the IDE insists that there are errors all over the place; i have no idea how that works Sep 17 18:26:08 if i insert 3 records into a sqlite table, then delete them, then insert 1 record, if the table was created with INTEGERPRIMERYKEYAUTOINCREMENT will that 4th record inserted have a primary key of 1 or 4? Sep 17 18:26:21 lasserix: 4; autoincrement doesn't fill in holes Sep 17 18:26:23 4 Sep 17 18:26:41 thanks Sep 17 18:26:54 which is a good thing, since it won't reconnect potentially orphaned foreign keys Sep 17 18:27:01 right Sep 17 18:27:05 just been a while ;p Sep 17 18:27:20 * diphtherial nods Sep 17 18:29:51 Do I only need both of these to develop android apps? http://developer.android.com/tools/index.html https://developer.android.com/sdk/installing/studio.html or do I just pick one Sep 17 18:30:46 you just pick one afaik; the ADT is a plugin for eclipse, whereas android studio is its own integrated thing Sep 17 18:30:58 you will need to install the SDK prior to installing either of those, though, again afaik Sep 17 18:31:09 ah, wait, der, never mind Sep 17 18:31:49 should I just go with android studio because its newer? Sep 17 18:31:53 but yeah, the choice, after downloading the SDK, is either the eclipse ADT or android studio Sep 17 18:32:08 i prefer android studio myself (well, i use intellij, of which android studio is a customized version) Sep 17 18:32:15 diphtherial: does autoincre start at 0 or 1? Sep 17 18:32:25 lasserix: pretty sure it's 1 Sep 17 18:32:44 in_deep_thought: either do intellij or android studio Sep 17 18:33:03 if you are brand spanking new to it to pick android studio Sep 17 18:33:45 ok I will. also is debian a good choice? or should I just go with ubuntu? I am looking for the least possible suprise errors to run into while starting to learn to develop for android Sep 17 18:33:57 like missing files/packages/etc Sep 17 18:34:05 like ia32 :) Sep 17 18:34:21 i believe google supports up to ubuntu 12 officially Sep 17 18:34:25 and it certainly works on others Sep 17 18:34:38 if you're a linux ninja, use whatever and make it work. If you're not, use ubuntu and follow the google directions to get what you need Sep 17 18:35:01 you'll definitely need multiarch ia32 Sep 17 18:35:38 is xubuntu as well supported as ubuntu? or will I run into more problems? Sep 17 18:37:21 you are being too cautios Sep 17 18:37:27 just intall it and see if you run a hello world app Sep 17 18:37:41 ;p Sep 17 18:38:05 lasserix: haha I tried that in debian and ran into an amazing amount of issues. Im wondering if its worth starting over with ubuntu Sep 17 18:38:15 yeah probbably then Sep 17 18:49:31 hello, I'm trying to do a imagebuttom blink by drawble..but in my logical, the button can recive 3 states, [off,on,blinking], off=0,on=1,blinking=2 ok...I did a xml witth drawble android duration... Sep 17 18:52:36 in_deep_thought, what issues are you running into with debian? virtually the only external dependency is Java... apart from that the Android Studio kit is self contained Sep 17 18:53:33 weirddan455: my avd’s keep freezing when I try and run apps on them. Sep 17 18:53:55 weirddan455: I actually overcame a lot of the issues but just makes me wonder if there will be more in the future than there need to be Sep 17 18:54:06 in_deep_thought, they can be very slow... i remember thinking mine froze and then just left it for 10 minutes and it finally came up Sep 17 18:54:44 weirddan455: did you have to keep clicking the “keep waiting” when you get the “taking too long to respond error”? Sep 17 18:55:44 in_deep_thought, that almost sounds like your window manager throwing that error... i don't think i ever got that it was just very slow Sep 17 18:56:16 in_deep_thought, i think if you run x86 images it's faster because with ARM it has to emulate the CPU Sep 17 18:57:42 Seems you can't have a table named table in sqlite database Sep 17 18:57:44 oh thats another thing. how do I know which cpu to choose. I thought behavioral differences pretty much just have to do with the difference in the phone and the android version. Should I be testing every possible cpu as well? Sep 17 18:57:59 in_deep_thought: Only if you use native code. Sep 17 18:58:07 And not really even then. Sep 17 18:58:18 flan3002: I use java, i think? Sep 17 18:58:24 i assumed everyone used java Sep 17 18:58:37 well if you use native code you'll have to recompile seperataly for every architecutre you intend to support Sep 17 18:58:52 lasserix_ don't use sql keywords :P Sep 17 18:58:56 is there any possibility that im using native code and don’t know it? Sep 17 18:59:06 in_deep_thought, there's the NDK that allows you to run native C/C++ code... you'd know if you were using it though Sep 17 18:59:59 in_deep_thought, only reason i said run x86 is because i think the emulator is faster... you just need to go into the SDK manager and download the x86 image Sep 17 19:03:17 in_deep_thought, Also FYI if you do decide to switch Xubuntu is identical to Ubuntu except for the default install UI. The packages are the same, repos are the same... you can even do a "sudo apt-get install xubuntu-desktop" on an Ubuntu install and get everything Xubuntu comes with. Sep 17 19:03:57 is it possible to install apk programatically? I don't mean in background, or something, I just want to launch the action like when user taps on .apk file... Sep 17 19:04:48 mrpyo: Yes, it is. Sep 17 19:05:10 mrpyo: What do you think happens when the user taps an .apk file? :P Sep 17 19:06:00 ACTION_VIEW Intent? Sep 17 19:06:32 < Ashiren> i have expandableListView. populated by custom adapter with List inside it. when i remove items from the list and notifyDataSetChanged, i encounter weird behavior when some of the groups are expanded Sep 17 19:06:44 < Ashiren> groups are removed but children stay. to hide them i have to click the remaining groups few times Sep 17 19:07:56 ashiren can you collapse all Sep 17 19:08:01 then notify dataset changed Sep 17 19:08:05 saving which you closed Sep 17 19:08:07 then open them again? Sep 17 19:08:26 that seems ugly Sep 17 19:08:50 i thought it should get rid of "orphaned" children by itself Sep 17 19:12:21 but it seems theres no other option Sep 17 19:12:26 thanks lasserix_ Sep 17 19:12:50 looks like otherwise you have to iterate through the list and remove the children yourself Sep 17 19:13:01 http://stackoverflow.com/questions/23136781/android-how-to-refresh-items-after-deleting-from-sqlite-database-in-expandableli Sep 17 19:14:23 well the item is deleted from other thread Sep 17 19:14:31 and my adapter know nothing about it Sep 17 19:14:46 i would need to track what item was deleted exactly Sep 17 19:14:47 Ashiren: let me see how i did this before Sep 17 19:15:06 Don't touch the adapter or its dataset from anything but the main thread Sep 17 19:15:24 then how do i update it Sep 17 19:15:43 Post new data to the main thread, update it Sep 17 19:15:49 your app would crash if you were calling datasetchanged from another thread Sep 17 19:16:00 i mean i "update" like this - i set the list to adapter everytime is updated Sep 17 19:16:06 and the list is synchronized Sep 17 19:16:39 from OnUiThread Sep 17 19:16:58 i know this, ive been there Sep 17 19:17:04 now the children stuff Sep 17 19:20:02 I wanted to ask this in the gradle room but there was only one person there .. can you attach a header to a repository { url 'https://whatever.com' } ? I want to pull something from a private github repo, but using the credentials block doesn't seem to work. I asm guessing that gradle is expecting to get a 401 back from the server and then it will try to basic auth. However, in this case the repository is github and t Sep 17 19:21:16 There's more than 1 in #gradle Sep 17 19:22:40 o .. my bad google told me the standard room was on codehaus .. just checked freenode (duh) .. i'll ask there Sep 17 19:23:03 how can i get logcat in the adnroid ddms view of intellij/androud studio to actually clear the backlog? Sep 17 19:26:45 what do you mean Sep 17 19:26:48 clear the log on device? Sep 17 19:26:53 logcat -c Sep 17 19:27:00 you don't get ddms to do it for you Sep 17 19:27:54 lasserix_ there is a little button like a garbage can Sep 17 19:28:10 yeah that doesn't work Sep 17 19:28:15 it just clears the screen Sep 17 19:28:22 lots of stuff doesn't work ;) Sep 17 19:28:22 as soon as another message appears, the whole backlog shows again Sep 17 19:28:31 oh thats weird Sep 17 19:28:36 this didn't happen when i used eclipse's logcat viewer Sep 17 19:28:54 pfn: no just within the viewer for logcat of intellij Sep 17 19:28:57 Doesn't happen here Sep 17 19:29:04 I never use the viewer in intellij Sep 17 19:29:05 * pfn shrugs Sep 17 19:37:51 Okay, here we go. "Please boss let me drop Android 2.3.x" part II Sep 17 19:38:08 * g00s gets popcon Sep 17 19:38:12 haha Sep 17 19:38:32 GA is showing us under 7% API Level 10 users for this month Sep 17 19:38:42 Down from about 9% I think a few months ago. Sep 17 19:38:54 We are still supporting iOS6, and that is under 4% actually. Sep 17 19:41:06 mattblang maybe he's just waiting for the moment the square root of GB percentage is > GB percentage itself :) Sep 17 19:42:13 g00s haha. really, its just that the app is for a government agency, and is available to the public. So it is much easier quantified, the number of users that won't get more features, while the development time lost is not as easily quantified Sep 17 19:44:21 hello guys Sep 17 19:44:53 lol @ support Android 2.* Sep 17 19:46:45 s/2/3/ Sep 17 19:52:48 do they still produce devices with 2.3.x Sep 17 19:53:41 Ahti333 i wouldn't be surprised - even if they do - if play services license is not granted Sep 17 19:53:55 oops Ashiren ; autocomplete ftl Sep 17 19:54:19 in other words, i bet you can't pass CTS now with 2.3 - but i'm not sure Sep 17 19:54:22 i hate hwo you can’t use R.id.xxxx for startAcitivityResult request codes Sep 17 19:54:47 java.lang.IllegalArgumentException: Can only use lower 16 bits for requestCode Sep 17 19:55:22 JacobTabak not sure how good idea it is to use res/ stuff for things that are not localized anyhow Sep 17 19:55:28 probably better to just make it a final int Sep 17 19:55:42 what do you mean? Sep 17 19:55:47 we use R.id everywhere in our apps Sep 17 19:55:53 gauranteed unique number Sep 17 19:55:59 without explicitly defining it Sep 17 19:56:07 would be really nice Sep 17 19:56:08 JacobTabak: just use your R.id but strip it down Sep 17 19:56:15 how to guarantee its unique? Sep 17 19:56:41 JacobTabak, g00s, Ashiren, flan3002, gdrc what would yall say is a reasonable threshold percentage for dropping support of an API version Sep 17 19:56:58 JacobTabak yeah ... i guess. would you rather see loader ID = 1000 or 78346873456783465 ? Sep 17 19:57:00 mattblang: Depends on your audience. Sep 17 19:57:07 mattblang: If you dev Sep 17 19:57:12 g00s: the nice thing about R.id is i dont need to see the number Sep 17 19:57:19 ever Sep 17 19:57:23 flan3002 just throw out a number, just curious, haha Sep 17 19:57:39 JacobTabak: what about using string and then taking hashcode? Sep 17 19:57:43 mattblang: If you dev for a company that only uses old devices, you don't have choice, if you want your users to pay, I'd go with 4.X any time. Sep 17 19:57:56 lasserix_: then i have to use context.getResources().getString() Sep 17 19:58:19 and im sure the hashcode could be anywhere from Integer.MIN_VALUE to Integer.MAX_VALUE which wouldn’t use the lower 16 bits anyway Sep 17 19:58:20 flan3002 it is a government app that is available for free to the public Sep 17 19:58:24 JacobTabak it might be time for a dingleberry :) Sep 17 19:58:35 lol Sep 17 19:58:58 ahh yeah Sep 17 19:59:03 time for more coffee Sep 17 19:59:09 Create a gradle task that generates your own id class Sep 17 19:59:13 Guaranteed unique ids! Sep 17 19:59:29 i just dont get why it can only be lower 16 bits Sep 17 19:59:34 mattblang: I don't know then. I guess that leaves you with a tiny bit more responsibility, but... Sep 17 19:59:37 maybe because request code + response code combined into 1 integer Sep 17 19:59:56 Because of fragments Sep 17 20:01:09 It uses the upper 16 bits to identify the fragment it's started from Sep 17 20:01:32 (if started from a fragment) Sep 17 20:01:46 ah Sep 17 20:01:52 well i think i’ll just use (short)R.id.xxx Sep 17 20:01:55 and see what happens Sep 17 20:03:29 how hard is it to follow an app tutorial that uses eclipse while using android studio? Once you get past setup, is anything that different? Sep 17 20:04:14 R.id.xxx & 0xFFFF would've been better, IMHO Sep 17 20:05:06 ravilov does casting it to a short do something different? Sep 17 20:06:08 hex looks more l33t Sep 17 20:06:17 you should make as few assumptions as posible about the size of primitive data types Sep 17 20:06:23 oh my god what is going on http://cl.ly/image/0r453S252U0C Sep 17 20:06:35 also i solved my expandableList problem by tracking when expanded items are Sep 17 20:07:23 JacobTabak: you have another class called Intent in the same file or imported Sep 17 20:07:54 i dont -.- Sep 17 20:07:59 JacobTabak i've had that problem before .... Sep 17 20:08:00 time to invalidate caches Sep 17 20:08:16 its compiling fine, ide is just freaking out Sep 17 20:08:25 it may be because you imported an inner class in intent, like "import Intent.Foo" Sep 17 20:08:51 restarting + invalidating caches fixed it Sep 17 20:08:55 never had these problems with eclipse :) Sep 17 20:09:18 because you'd never get this far Sep 17 20:09:48 I've occasionally had my eclipse mark all sorts of lines as having errors, while the file itself wasn't marked as having errors, and needless to say it compiled fine Sep 17 20:09:57 it just goes crazy sometimes, nothing a restart cannot resolve Sep 17 20:10:06 (of the IDE, not of the computer) Sep 17 20:10:12 eclipse goes crazy if you have a mismatched { } in a inner class Sep 17 20:10:24 but so does AS Sep 17 20:10:26 ... well now I have to try this Sep 17 20:10:46 actually i lied, it didn’t fix it Sep 17 20:10:52 liar Sep 17 20:11:02 the library is just like… confused Sep 17 20:11:03 http://cl.ly/image/072j09400J2y Sep 17 20:11:20 can someone tell me why java seems to make such ugly gui’s? even android studio seems to be using scroll bars and menu boxes from the 90’s. and android studio seems like it was very recently developed Sep 17 20:11:50 what is so '90s about AS's looks? Sep 17 20:12:01 JacobTabak try deleting all your imports Sep 17 20:12:09 ravilov its swing Sep 17 20:12:20 swing can be themed, can't it? Sep 17 20:12:36 JacobTabak assuming auto import is on, they should get recreated Sep 17 20:12:46 dunno, tried running AS on win7, didn't look anything like the '90s to me Sep 17 20:12:47 somethings whacked, it cant even find the correct classes Sep 17 20:12:51 and the import isn't ambiguous Sep 17 20:13:03 ok, maybe your gradle is fucked up now :) Sep 17 20:13:12 i changed the compile/target sdk version in the lib and it works now Sep 17 20:13:14 does AS have some sort of system console? Sep 17 20:13:14 like, stuff not in classpath Sep 17 20:13:15 -.- Sep 17 20:13:25 even tho it worked fine before.. Sep 17 20:13:39 (and was working fine, just showing up as unresolved symbol in AS...) Sep 17 20:14:04 ravilov: the scroll bars and menus seem to be from some windows 98 or xp program. none of the gloss or shine. sometimes it even looks like the text isn’t alighned in the boxes very well like off centered. Its like web development technologies made the design of default layouts a higher priority vs java/android Sep 17 20:14:09 g00s, I can't make my eclipse go crazy with mismatched braces, it just reports error(s) just like it should, and it goes back to normal as soon as I correct it Sep 17 20:14:38 in_deep_thought, maybe you're just nitpicking ;) Sep 17 20:14:53 the job of a good IDE is not to look pretty Sep 17 20:14:56 the main job( Sep 17 20:14:57 * Sep 17 20:15:07 ravilov: of course i am :) but i wonder why this is? it seems easy enough to make it better looking than it is Sep 17 20:15:12 the IDe is pretty in darcula :) Sep 17 20:15:19 stock theme is gross Sep 17 20:15:24 * ravilov 's eclipse looks exactly like the GTK+2 theme used, me likes Sep 17 20:15:42 eclipse uses SWT which wraps the native toolkit Sep 17 20:15:51 cocoa, gtk, etc Sep 17 20:15:57 JacobTabak, if pretty == dark as heck :p Sep 17 20:16:11 easier on teh eyes Sep 17 20:16:14 for me Sep 17 20:16:28 just imagine what turning the screen physically off can do for you then :p Sep 17 20:16:37 how would work g et done then Sep 17 20:16:43 i need hirelings Sep 17 20:17:21 use accessibility technologies such as text to speech :p Sep 17 20:17:28 much more tedious, but so very relaxing on your eyes ;) Sep 17 20:27:43 there are a lot better skins then dracula Sep 17 20:27:46 dracula was kinda drab Sep 17 20:28:02 you can find more : http://ideacolorthemes.org/home/ Sep 17 20:28:22 i like vibrant unknown (did a personal spin off from) that works well Sep 17 20:30:05 i just hope oracle fixes the Swing font AA on mac; almost everybody uses the apple jvm which uses quartz to render text - if you try running AS with the oracle JVM the text is eyebleeding Sep 17 20:30:24 but apple jvm days are numbered Sep 17 20:39:44 this is such a dumb question… is there an easy way to download an arbitry file to private storage? Sep 17 20:39:53 arbitrary* Sep 17 20:39:57 without usign download manager Sep 17 20:40:04 or httpurlconnection Sep 17 20:40:16 i hate writing custom code for stuff like this -.- Sep 17 20:40:31 maybe download manager is the way to go Sep 17 20:40:58 okhttp Sep 17 20:41:04 So do devs now need a physical address registered? Sep 17 20:41:18 okhttp 2.0? Sep 17 20:41:24 does it have a nice api for that? Sep 17 20:41:38 i still have 1.6 in my proj bc it didn’t work with L Sep 17 20:42:04 response.body().readAll(Okio.sink(file)) Sep 17 20:42:08 lovely Sep 17 20:42:22 thank you Sep 17 20:44:47 on that note, what do you all prefer to relatively simple http requests? apache's HttpClient is kind of verbose Sep 17 20:44:57 retrofit! Sep 17 20:44:58 *prefer for Sep 17 20:44:59 kohttp Sep 17 20:45:49 retrofit looks neat, will read more about it. what's kohttp? i can't find anything on it Sep 17 20:45:57 okhttp Sep 17 20:46:02 http://square.github.io/okhttp/ Sep 17 20:46:18 it's okhttp, but knocked out :p Sep 17 20:46:18 aha, thanks Sep 17 20:46:23 knock-out.. yeah. lol Sep 17 20:46:28 derp Sep 17 20:46:53 lasserix_, you should totally make a fork Sep 17 20:46:55 retrofit seems useful, but it's a little more structured than what i'd need for a simple http request Sep 17 20:47:14 diphtherial: yea retrofit is for mapping to objects Sep 17 20:47:25 and yeah, okhttp looks like exactly what i wanted, which was basically a builder pattern for constructing http requests Sep 17 20:48:38 yay for oneliners, right? Sep 17 20:49:01 indeed. and yeah, i'll likely use retrofit in some later projects :) thanks Sep 17 20:49:42 it’s response.body().source().readAll().sink() I guess? Sep 17 20:50:49 no, that will buffer the entire response in memory Sep 17 20:50:49 response.body().source().readAll(Okio.sink(file)) Sep 17 20:51:22 A Source is like an InputStream (but not stupid). A Sink is like an OutputStream (but not stupid). You readAll of a Source into a Sink. Sep 17 20:51:41 thanks, yes sorry typo in that question Sep 17 20:54:01 wow, never had a need to use eventbus so far so didn't look it up too much, for some reason I believed it's part of the standard android API Sep 17 20:54:41 is there an easy way to handle videoview when device rotates? Sep 17 20:55:21 Sixmsj: the easiest way is to handle onConfigurationChanged Sep 17 20:56:03 manifest: android:configChanges="orientation|keyboardHidden|screenSize" Sep 17 20:58:02 but that’s not viable in all activities obv, so ymmv Sep 17 20:59:02 you can probably save the mediaplayer to a setRetainInstance() fragment so it remains prepared on rotation, then reattached it to the newly created videoview as well Sep 17 20:59:04 haven’t tried it though Sep 17 21:17:19 very nice api for okhttp 2, worked really well. so it probably handles memory properly so no problem downloading 90+ small png files in a loop? Sep 17 21:17:42 or is there a better way to do that Sep 17 21:27:28 really not a fan of this coloured action bar thing hangouts are doing :( Sep 17 21:28:07 bankai_au let me update and try it out Sep 17 21:28:13 no no don't Sep 17 21:28:21 bankai_au too late. its your fault Sep 17 21:28:37 bankai_au haha, but really though, the new update has the ability to make calls on wifi so Sep 17 21:28:41 bankai_au been meaning to do it Sep 17 21:29:14 was looking at the material color palette yesterday, colors just got too bold Sep 17 21:29:30 weird shit like whit text on bright orange, yuck Sep 17 21:29:52 bankai_au did they material-ize hangouts ? Sep 17 21:29:58 i still haven't looked at L :/ when are we going to get a name for it so it doesn't sound like some weird anime character? Sep 17 21:30:08 g00s maybe they did, maybe they didn't. guess you'll just have to update Sep 17 21:30:18 :P Sep 17 21:30:26 anyone know if its kosher to do a bluetooth-classic inquiry scan at the same time as bluetooth-le scan ? Sep 17 21:30:29 no not materialise, they just coloured teh action bar and replaced the slideviewlayout to listview + new activity Sep 17 21:30:47 bankai_au: really? even on tablets? Sep 17 21:30:48 mattblang: his windows phone doesn't run hangouts :P Sep 17 21:30:55 haha Sep 17 21:30:59 haha Sep 17 21:31:00 JacobTabak: nfi don't use my tablet for anything but dev Sep 17 21:31:13 i just can’t imagine they have a fullscreen contact list Sep 17 21:31:15 yeah, my n7 is just for dev Sep 17 21:31:39 bankai_au im not liking that there is no separation at all between the tabs and the action bar Sep 17 21:32:46 i'm not sure what the thinking behind the tab was. To me tabs are there to flick between frequently used content. I almost never message anyone new, so i'm almost never going to use that tab and it takes up a fair bit of real estate Sep 17 21:32:57 well maybe not really, but it makes the actionbar look overly large Sep 17 21:33:11 and of course, you know, i could do 20000000 times better :/ -- i shouldn't bitch Sep 17 21:33:18 maybe thats why they got rid of the slide view Sep 17 21:33:45 the slideview made it feel fast(er) for me Sep 17 21:33:45 bc nobody messages new people Sep 17 21:34:17 using rxjava makes my code look so… weird Sep 17 21:34:53 JacobTabak not having lambdas makes the code look weird; once (if) we get that, it will be groovy Sep 17 21:35:17 mmmm lambdas will be nice :) Sep 17 21:35:42 it's still going to look a bit verbose, but i guess that's java for ya ;) Sep 17 21:36:11 i love that i can download a file in 2 lines Sep 17 21:37:33 ugh, startDiscovery() and startLeScan() are behaving differently on n7 and samsung devices Sep 17 21:38:09 i have to make sure there are no sharp objects or lethal weapons in my room when handling these things Sep 17 21:38:29 those apis are a bit rough around the edges, eh? Sep 17 21:39:04 JacobTabak there was a blog yesterday about android dev http://www.reddit.com/r/androiddev/comments/2gmmlm/why_its_hard_to_find_android_developers/ Sep 17 21:39:22 i glanced over it, didnt take much from it Sep 17 21:39:28 it's hard to find android developers? Sep 17 21:39:33 it also got me thinking - in cases were BLE is used, and the iOS version influences the GATT structure Sep 17 21:40:02 its very applicable to my job, i work for an NY-based startup that had an ios app first, and hired me to do the android app Sep 17 21:40:04 but yeah, I hate that android is typically designed after IOS Sep 17 21:40:05 which sucks Sep 17 21:40:17 90% of the time when something goes wrong: "it should work like ios" Sep 17 21:40:24 the android BLE APIS are a bit behind iOS so you wind up not being able to get the full functionality of the device unless you really design things carefully - that is - knowing the weaknesses of the android BLE apis Sep 17 21:40:25 "where's the documentation?" "look at the ios code" Sep 17 21:40:29 ... Sep 17 21:40:30 fuckers Sep 17 21:40:45 i personally like being able to take an ios app that’s hacked together Sep 17 21:40:51 and know how its supposed to work, see the code Sep 17 21:40:57 then architect an android app the right way, usign their mistakes Sep 17 21:41:04 as guidance Sep 17 21:42:37 JacobTabak are you in NY atm ? Sep 17 21:42:48 no but i’ll be there this weekend for droidcon Sep 17 21:42:53 oh yeah Sep 17 21:43:04 there is also a protest going on Sep 17 21:43:11 android developer protest? Sep 17 21:43:15 ha LOL Sep 17 21:43:32 speak up against the ios driven startup community Sep 17 21:43:40 http://peoplesclimate.org/march/ Sep 17 21:44:02 oh boy Sep 17 21:44:06 hope that doesn’t mess with my flight or anything Sep 17 21:46:51 hm, trying to think of a way around this. my user can have a BT classic or LE device - i should be able to interface to either. but i don't want a 'scan for LE device' and 'scan for classic' device action - thats lame, the user shouldn't care Sep 17 21:47:18 can you just do it serially and pretend it’s one operation Sep 17 21:48:02 yeah ... classic takes quite a bit of time. page scan followed by inquiry scan Sep 17 21:48:07 le goes fast Sep 17 21:49:11 JacobTabak did you figure out that kerning / view bounds issue ? Sep 17 21:49:38 no…. Sep 17 21:49:42 hmm Sep 17 21:49:47 just put in some padding/margin to make it look ok Sep 17 21:52:48 g00s, do you need to actually scan for BT or just get the list of what is paired? Sep 17 21:53:01 MikeWallaceDev scan Sep 17 21:53:06 ok Sep 17 21:53:24 foogetaboutit Sep 17 21:53:48 i was confused because after reading about the physical layer and radio, it seemed like it should be possible for a dual mode device to do both at the same time Sep 17 21:54:22 now i want to see the schematics :D Sep 17 21:55:27 MikeWallaceDev have you done any ble ? Sep 17 21:56:14 yeah, not that much, but yeah Sep 17 21:56:57 It's just that I had the same question as you, until I realized that I didn't need to scan, just list. Sep 17 22:00:54 MikeWallaceDev if i have a device in the broadcasting role, and the android device is the observer - basically looking for ADV_SCAN_IND packets; i'm noticing when i start a le scan i only get the first packet, not subsequent. so i have to keep start/stopping the scans. did you notice that ? Sep 17 22:03:41 I have a custom AccountAuthenticator that I want to use in multiple apps. Can I just include the same code in all my apps, all defined the same way in the manifest and rely on everything working properly Sep 17 22:04:02 or is there some other jiggery pokery I need to do? Sep 17 22:04:18 hi, i'm experincing an issue very similar to this one https://code.google.com/p/android/issues/detail?id=37484 , and I'm having trouble fixing it. i tried using teh code they provided, and i put in a Log.d() in it to see if it was used, and my log statement never gets called. Sep 17 22:05:10 does anybody have some pointers about how to either work around it or fix it completely? Sep 17 22:08:18 giantpune… sorry I’ve never work with that code Sep 17 22:12:42 giantpune: if it’s the same issue described in the bug, a fix is available Sep 17 22:13:14 what do you mean your log statement is never called Sep 17 22:14:11 i took the code from the second comment in that issue and put it in my project and added a Log.d() to see if it was getting called and that log statement never shows up Sep 17 22:15:11 did you make sure to reference the new class (Fixed…) rather than the old one from support lib? Sep 17 22:15:18 i also tried changing by build.gradle to use the android support 21.0.0-rc1 which they said fixes the issue, but then my project fails to build because they want me to compile for android L or something along those lines. Sep 17 22:16:17 a couldnt find a reference to "FragmentStatePagerAdapter" anywhere in my project to change over to the new subclass. Sep 17 22:17:37 so… you copied that code but you’re not using it? Sep 17 22:18:20 are you using a viewpager? Sep 17 22:19:18 yes Sep 17 22:19:26 what kind of adapter are you using Sep 17 22:21:01 i'll have to dig through the code to see which type it really is. its a subclass of something else. (not originally my code) Sep 17 22:21:43 well the fix in that issue was specifically for fragmentstatepageradapter Sep 17 22:22:09 public class TabsAdapter extends FragmentPagerAdapter Sep 17 22:22:19 so yea, thats not your bug Sep 17 22:22:27 what exception are you seeing exactly? Sep 17 22:23:00 its a classnotfound exception while unmarshmallowing this viw Sep 17 22:23:10 what class is not found? Sep 17 22:23:19 and when does it happen? Sep 17 22:23:21 on resume? Sep 17 22:23:25 on rotation? Sep 17 22:24:08 i'm not sure. i havent been able to trigger it. it happened once while i was out of eth house and all i got was a logcat a grabbed with anothre app. Sep 17 22:24:22 it happened when i resumes back to the app from another one. Sep 17 22:25:49 so fragment state pager adapter saves state of fragments whenever they’re removed from the viewpager Sep 17 22:25:56 and then restores their state Sep 17 22:26:02 when they’re restored Sep 17 22:26:10 fragment pager adapter just detaches them from your activity, they’re kept in memory Sep 17 22:26:32 so do you know which class was not found? Sep 17 22:28:17 nope. :( i tried for an hour to reproduce it and couldnt figure out how to trigger it Sep 17 22:28:52 ok well i can promise you that bug is a dead end Sep 17 22:29:01 the one in code.google.com Sep 17 22:29:11 doesnt apply to you Sep 17 22:30:24 damn. Sep 17 22:31:04 giantpune: should be easy to reproduce Sep 17 22:31:10 just disable ‘keep activities’ in developer options Sep 17 22:31:18 then switch apps, and come back Sep 17 22:34:20 heh. good call. that made it crash first try Sep 17 22:34:46 i keep that on, on my dev devices Sep 17 22:35:40 so pastebin the exception Sep 17 22:35:46 you probably are just trying to save something thats not parcelable or something Sep 17 22:35:58 actually probably not, but anyway im curious now Sep 17 22:36:03 i'm reinstalling now without classnames mangled Sep 17 22:36:04 When is best time to bind / unbind a service? onresume / onpause? Sep 17 22:36:18 onstart/stop Sep 17 22:36:29 thanks Sep 17 22:37:00 lasserix_: i recall reading in the docs recently, it only matters if you’re reusing the service in different activities/fragments Sep 17 22:37:13 http://pastie.org/private/3ep8kpsrejhs3holcsbw Sep 17 22:37:18 if you unbind in onpause, the service can be killed before onresume of the next activity Sep 17 22:37:41 yea giantpune looks like a bug on this 3rd party lib Sep 17 22:37:57 SlidingUpPanelLayout Sep 17 22:38:05 or not sure if thats 3rd party, maybe just custom Sep 17 22:38:52 either that or you’re just not parcelling correctly Sep 17 22:40:55 ahh i will be calling on start Sep 17 22:40:58 so that doesn't matter Sep 17 22:41:04 i just need the binding for messengers Sep 17 22:41:32 http://pastebin.com/FRuSzGJc Sep 17 22:42:15 dunno giantpune can’t get too deep in this right now Sep 17 22:42:19 at least you can reproduce Sep 17 22:43:06 thansk for teh point in teh right direction. Sep 17 22:43:45 giantpune: are you saving stuff to your fragments instance state? Sep 17 22:43:58 the fragments being dispalyed in teh viewpager Sep 17 22:44:12 like, try leaving the adapter empty and seeing what happens Sep 17 22:45:25 g00s: since onStop is not guarenteed to be called, is this bad in case service never gets stopped? Sep 17 22:45:50 lasserix_ it won't be called if your process is killed ;) in that case ... no need to worry anyhow :D Sep 17 22:45:57 k Sep 17 22:46:05 unless its in a different process! Sep 17 22:46:26 nah not using remote processes here Sep 17 22:50:22 If i am managing the service lifecycle outside of the binding, which binding flag should i use? Sep 17 22:50:57 BIND_AUTO_CREATE Sep 17 22:58:21 ahh i like 1.7 implict type for generics Sep 17 23:23:44 multicatch is pretty nice too Sep 17 23:38:58 ha hell has froze over Sep 17 23:39:00 i’m using fitxy Sep 17 23:43:07 Hey Everyone. I have a singleton class that lives in my Application.java and used for event tracking. The class needs a context. I want to pass the context into the class when it is created. A friend wants to create a global singleton context variable that points to getApplicationContext() and a static method that retrieves that variable. Can anyone give me a reason for or against this? Sep 17 23:44:31 shady, make your application a singleton Sep 17 23:44:43 and then create instances within the application Sep 17 23:45:00 you want as few static references as possible Sep 17 23:45:18 Application is a sivgleton Sep 17 23:45:56 right... Sep 17 23:46:04 so instantiate your tracker in your applications oncreate Sep 17 23:46:09 and store a reference to it in your application Sep 17 23:46:13 and access it through your application Sep 17 23:46:34 yes we are doing this. but that class needs a context. Sep 17 23:46:50 if you instantiate it in application onCreate, you can use ‘this’ as a context Sep 17 23:47:00 the question is to pass getApplicationConte Sep 17 23:47:06 xtext or Sep 17 23:47:26 pass it in to the constructor Sep 17 23:47:35 of your tracker Sep 17 23:47:35 store a global singleton that points to the global singleton referenced by getApplicationContext Sep 17 23:47:47 i’m trying to emphasize you don’t want global singletons Sep 17 23:47:53 yes i know that i what i want to do Sep 17 23:48:11 I am asking more for a why Sep 17 23:48:27 because it’s more testable Sep 17 23:48:30 I know the proper way to do it Sep 17 23:48:39 you can’t mock static objects Sep 17 23:48:46 ahh good point Sep 17 23:50:09 and they don’t work with DI Sep 17 23:50:29 guys some little help, its something easy but first time i use retrofit... im having this error: "must not have replace block.". im tryingt o use @GET("/url?param1={firstparam}¶m2={secondparam} ArrayList getObjects(@Path("firstparam") String firstparam,@Path("secondparam") String secondparam): Sep 17 23:50:42 can’t read , pastebin Sep 17 23:50:48 use @Query Sep 17 23:51:09 http://square.github.io/retrofit/javadoc/retrofit/http/Query.html Sep 17 23:55:38 JakeWharton is google using dagger in any of their android apps yet ? Sep 17 23:55:46 yes, for over a year now Sep 17 23:55:53 nice Sep 17 23:56:12 interesting Sep 17 23:58:38 JakeWharton are you able to say which ones? Sep 17 23:58:43 no Sep 17 23:58:54 secret sauce Sep 17 23:59:07 easy to find out if you have apks Sep 18 00:00:12 i don't have to know exactly. everything i've heard has been positive, but google using it in their own apps is another vote of confidence Sep 18 00:01:39 why? they also use RoboGuice in apps and it's fucking awful Sep 18 00:01:44 how can I register a broadcast reciever to reciever sharing intents, i.e. Action Send with mimeType="text/plain"? Sep 18 00:02:11 spoiler alert: google engineers are not different than engineers at every other company Sep 18 00:02:13 ankhwatcher: look up intent filter Sep 18 00:02:19 You can't, ACTION_SEND is for an activity Sep 18 00:02:24 for an activity intent* Sep 18 00:02:41 JesusFreke: ah, that would explain it Sep 18 00:02:58 http://developer.android.com/reference/android/content/Intent.html#ACTION_SEND Sep 18 00:03:06 JesusFreke: because this same intentfilter worked fine for my activity Sep 18 00:03:06 First two words "Activity Action" :) Sep 18 00:03:12 Yeah Sep 18 00:03:19 they're not interchangeable :) Sep 18 00:03:27 JakeWharton false, google engineers may at some point have to face the wrath of hackbod in person Sep 18 00:04:39 I wonder how brevos is doing this, maybe it's just hiding it's activity altogether Sep 18 00:04:55 ok without trying to dig into the source code, how come picasso.load() sizes an image differently than imageview.setimageuri() without any extra options? Sep 18 00:05:06 in an imageview with wrap_content x 2 Sep 18 00:09:15 I'm trying to improve this guy: https://github.com/Ankhwatcher/SharkShare by having the shared data enterpreted in the background and then the results overlayed for the user to choose from Sep 18 00:09:39 it's all inspired by the very cool bit.ly app brevos Sep 18 00:09:53 any ideas how I can do that? Sep 18 00:11:51 https://github.com/Ankhwatcher/SharkShare/blob/master/app/src/main/AndroidManifest.xml :) Sep 18 00:28:47 anyone here use recyclerview much? Sep 18 00:44:28 anyone know an api service for images? Sep 18 00:44:48 wat Sep 18 00:46:12 JacobTabak I want to eg fetch a list of images Sep 18 00:46:20 looking at Flickr etc Sep 18 00:46:33 oh. Sep 18 00:46:36 what kind of image slol Sep 18 00:46:39 imgur has a lot of images! Sep 18 00:47:34 JacobTabak: I looked at imgur so you've got a good question there Sep 18 00:47:40 I'm looking for HD images Sep 18 00:47:49 specifically automobiles and planes :) Sep 18 00:47:53 and motorcycles Sep 18 00:48:17 hi i'm new here. Sep 18 00:48:24 hi new here Sep 18 00:48:29 fresh meat Sep 18 00:48:35 yup taste me Sep 18 00:48:51 gawd I love the Internet Sep 18 00:50:25 is there a way to NOT uninstall the wear app when reinstalling the debug application on the phone? Sep 18 00:50:36 This did not use to happen a month ago >_< Sep 18 00:50:49 wuuu recyclerview is alive! http://g.recordit.co/eBnm1urrMG.gif Sep 18 00:51:19 looks awesome JacobTabak! Sep 18 00:51:47 is this a from scratch view you made or a customization of ViewPager? Sep 18 00:51:48 one problem is you can’t really determine the item with when implementing the decorator Sep 18 00:51:48 or something Sep 18 00:52:11 its an implementation of RecyclerView… the new listview that can go horizontal that is in v21 support lib Sep 18 00:55:55 thx, nice to know Sep 18 01:16:58 i want to implement a ‘flingback’ animation in my scrolling view, so it can only rest at certain places Sep 18 01:17:06 anyone done that before? Sep 18 01:22:21 JakeWharton, how would you compare Otto to GreenRobot's EventBus? Sep 18 01:22:30 they are != Sep 18 01:22:48 really? I thought they did the same things o.O Sep 18 01:23:03 Otto was designed for a specific use case and interaction Sep 18 01:23:10 EventBus was designed to be everything for everyone Sep 18 01:23:25 and having to name your methods a particular way is stupid! Sep 18 01:23:36 i wrote an annotation processor version of Otto two years ago Sep 18 01:23:50 it was ridiculously fast and also kind of stupid Sep 18 01:23:54 Oh okay. I was confused because they even compare the performance stats to Otto in their readme Sep 18 01:24:00 well, time to try Otto :D Sep 18 01:24:06 yeah that stuff doesn't matter much Sep 18 01:24:14 you can do something like this to save an image to the gallery MediaStore.Images.Media.insertImage(getContentResolver(), yourBitmap, yourTitle , yourDescription); is there an equivalent for videos? Sep 18 01:24:27 unless you have 20+ methods subscribing the performance differences are negligible Sep 18 01:25:07 Nope, I only have two lol Sep 18 01:25:19 thanks! Sep 18 01:25:37 we talked about adding more to Otto but it always felt like a poor fit Sep 18 01:25:43 moving across threads and such Sep 18 01:26:57 Ohh......right. Hmm, I just realized that's exactly what I'm doing. I'm posting a message from another thread to an activity to do something on the UI thread Sep 18 01:28:26 https://github.com/square/otto/issues/38#issuecomment-11014961 Sep 18 01:29:06 except change post(event) in the Runnable to AndroidBus.super.post(event) and save yourself a stack frame Sep 18 01:31:31 g00s, thanks (your message last night), but you're 7 days early ;) Sep 18 01:31:52 canadiancow :) Sep 18 01:33:17 JakeWharton, thanks a lot! Sep 18 02:22:44 Hey guys, can't seem to google what I should change to customize the actionbar when it is in "back" state. I am using a navigation drawer, have modified actionbar fine for normal views, but whenever I pop up a new activity the actionbar uses a different layout. Don't know if I'm clear engouh. Sep 18 02:47:16 SASDOE, the action bar menus are set in OnCreateOptionsMenu so you'd either set them in a base class or set them in each activity (or both) Sep 18 02:53:36 Afzal: cool thanks **** ENDING LOGGING AT Thu Sep 18 03:00:00 2014