**** BEGIN LOGGING AT Tue Oct 14 02:59:59 2014 Oct 14 03:08:27 how in the hell you are able to customize anything with this documentation http://developer.android.com/reference/android/R.attr.html#prompt Oct 14 03:10:25 any one know if there's a ButterKnifey way to do setOnGroupExpandListener on an expandableListView? Oct 14 03:12:17 otherwise I guess I'll do it the ol' fashioned way :) Oct 14 03:13:41 there isn't Oct 14 03:14:18 no worries, thanks Oct 14 03:18:11 does propmt extends from a dialog or something? Oct 14 03:18:19 what are its attributes Oct 14 03:18:23 :/ Oct 14 03:20:20 cliffreich you have to look at the xml files for the most part Oct 14 03:20:37 attr.xml, styles.xml, themes.xml, etc Oct 14 03:21:05 which themselves will reference a bunch of other stuff, dimens, layouts, drawables, etc Oct 14 03:23:17 wait seems like its only the to use an indirect string Oct 14 03:23:37 thre* Oct 14 03:24:04 not like i can customize the spinner dialog or anything Oct 14 03:26:45 can i setup a layer for spinner ui items and other for dialog items? Oct 14 03:29:05 any LoaderCallbacks experts? Oct 14 03:29:33 Are there any good ways to intercept ALL requests from a webview? Oct 14 03:30:11 I have an activity that is loading a fragment that implements LoaderCallbacks, via fragmentManager.beginTransaction() Oct 14 03:30:31 I never get onLoadFinished Oct 14 03:30:50 then I read "Note that normally an application is not allowed to commit fragment transactions while in this call, since it can happen after an activity's state is saved" Oct 14 03:31:01 what alternatives do I have? Oct 14 03:34:57 if anyone knows the answer http://stackoverflow.com/questions/26352206/how-do-i-set-up-a-spinner-using-a-different-layer-for-spinner-items-and-another Oct 14 03:35:11 :) Oct 14 03:36:01 use booleans as state indicators and make the transaction in onPostResume and not in on Oct 14 03:36:14 onActivityResult* Oct 14 03:36:31 viperZ28 Oct 14 03:36:39 yes Oct 14 03:37:06 what i said, transactions in onPostResume Oct 14 03:37:38 hum, then put fragment element in layout resource file? Oct 14 03:38:27 you mean replace a view for a fragment? then yes Oct 14 03:38:31 all your fragment transactions Oct 14 03:38:45 so you wont have state loss issue Oct 14 03:39:11 will try that Oct 14 03:39:14 thanks Oct 14 03:44:27 Can anyone tell me what is wrong with this? I am trying to authenticate against a form based auth site, but it systematically returns a 403 saying I need to accept cookies. I have set the cookieStore, but it is always empty. I tried adding a get request first, doesn't work either. https://gist.github.com/SASDOE/5f10ed552ce90bc75e63 Oct 14 04:07:55 SASDOE if you set your cookie store properly you shouldn't need to do anything Oct 14 04:08:04 you shouldn't need to iterate over them, attach them to the request Oct 14 04:08:44 The problem is that the website seems to thing I am not accepting cookies Oct 14 04:08:55 well in the code you pasted you're not specifying a cookie store Oct 14 04:08:58 i'm not an expert on this - its been a while Oct 14 04:09:02 but you need to set a cookie store Oct 14 04:09:49 localContext.setAttribute(ClientContext.COOKIE_STORE, cookieStore); this is setting it. No? Oct 14 04:10:31 you should be using HttpUrlConnection not Apache client for one Oct 14 04:12:03 Ok well I'll try with that then, be back if I still have issues. Cheers Oct 14 04:12:38 any reason why you are using Default http client? Oct 14 04:12:58 Umm it worked for basic auth! Oct 14 04:13:41 http://developer.android.com/reference/java/net/HttpURLConnection.html Oct 14 04:13:46 read the section about cookies Oct 14 04:14:02 also, read this old article (from 3 years ago) about the difference between the two clients http://android-developers.blogspot.com/2011/09/androids-http-clients.html Oct 14 04:16:03 Cheers mate Oct 14 04:17:14 Jesse hates that post Oct 14 04:19:11 http://loopj.com/android-async-http/ Oct 14 04:19:18 ^ that is your new best friend Oct 14 04:20:12 persists cookies automatically in SharedPreferences Oct 14 04:20:24 Cool as well! Oct 14 04:20:27 is that even a good idea ? Oct 14 04:20:40 g00s: why not? Oct 14 04:21:09 it's fine Oct 14 04:22:28 man, i've rewritten so much stuff today Oct 14 04:22:37 it's crazy how shitty my code was 3 months ago :P Oct 14 04:24:41 it's kinda therapeutic, especially delete slabs of useless code :) Oct 14 04:25:02 best code is the code that isn't there :D Oct 14 04:25:06 lol Oct 14 04:25:10 actually, i added code :P Oct 14 04:25:18 ViewHolders Oct 14 04:25:38 no more 1000 findViewById() calls in getView() Oct 14 04:36:55 * Fall grins Oct 14 04:37:05 How can I go about detecting when an ACTV is open and showing results? Oct 14 04:40:45 hm, maybe one day i'll figure out the caveats of rx on android https://github.com/ReactiveX/RxAndroid/issues/12 Oct 14 04:40:58 so far most of this has been in my services, no need to worry afaik Oct 14 04:41:09 component lifecycle strikes again :) Oct 14 04:44:14 How do I get layout params if I want to return a textivew in the getView of an adapter? Oct 14 04:46:52 listview adapter? Oct 14 04:46:54 gridview adapter? Oct 14 04:47:00 whatever the adapter view is, thats the type of layout params you need Oct 14 04:50:01 lasserix_: hi Oct 14 04:52:42 g00s: easy fix: keep them out of the lifecycle Oct 14 04:53:35 JakeWharton the only thing from rxandroid that i've used so far, are the 2 schedulers ... do you use any of the other stuff ? Oct 14 04:53:46 i've only used the main thread one Oct 14 04:57:16 g00s viewobservables are nice Oct 14 04:58:51 hey niru Oct 14 04:58:54 listview Oct 14 04:59:08 JacobTabak: can you be more specific? Oct 14 04:59:25 ListView has its own layoutparams? Oct 14 04:59:58 what i need to check - i have a few hot observables in a bound service; activity.onStart subscribes those to a CompositeSubscription. activity.onStop removes the subscriptions. i have to make sure tere is no leak there Oct 14 05:00:05 lasserix_ AbsListView.LayoutParams Oct 14 05:03:23 lol man i spent all day trying to get robolectric to work with burst because i wanted to use it in a test, but there's no way i'm going to bother with integrating it into my project right now :\ Oct 14 05:03:33 funny how things like that work Oct 14 05:08:22 JacobTabak i get sidetracked too , only to find out mayb weeks later i'm off course :D Oct 14 05:08:36 such is curiosity Oct 14 05:08:51 maybe was worth it Oct 14 05:08:52 we'll see Oct 14 05:09:01 i learned how to debug apps with 2 different class loaders \o/ Oct 14 05:12:47 lasserix_: yesterday the compound view worked fine when I changed to findViewById Oct 14 05:13:34 I am trying with adding set methods for the views in the controller class and accessing them in the MainActivity Oct 14 05:14:26 But here problem is I am not able to see those methods through MainActivity Oct 14 05:17:53 cool Oct 14 05:18:57 lasserix_: sorry now I am able to see those methods :) Oct 14 05:19:14 I dont understand what was missing. Oct 14 05:19:43 I am writing a simplified clone of the old windowing system of the win32 api. I use bitmap controls ven for text. Can I use the standard input keyboard here? I would rather not write the whole android text input system again Oct 14 05:20:53 niru you probably used the wrong cast Oct 14 05:21:00 if you cast it linear layout you won't see them Oct 14 05:21:01 hmm Oct 14 05:22:37 Anohter thing; my debugger is very moody, does anyone know how to open a debug console in the IDE and write a "live debug log" too that? Oct 14 05:27:15 lasserix_: But I am only able to see the textviews Oct 14 05:27:24 http://pastebin.com/sJMR3Ufk Oct 14 05:27:25 make sure the access modifiers Oct 14 05:27:35 I think I have given visible=gone Oct 14 05:27:51 private TextView mtext; Oct 14 05:27:51 private EditText medittext; Oct 14 05:27:57 oh yeah Oct 14 05:28:01 that you were talking about code Oct 14 05:29:09 damn setTextSize causes lots of gc Oct 14 05:29:29 as many views I define in the activity_main that many rows get generated? Oct 14 05:29:45 This is so weird, i am getting gc like crazy, when setTextSize in a subclass textview Oct 14 05:29:50 *onDraw method Oct 14 05:30:18 isnt there a warning whenever you do anything in ondraw Oct 14 05:30:27 but i am just setting text size Oct 14 05:30:27 lasserix_ Can I override the draw method in a textview and still use the "built-in" text input "sytem"? Oct 14 05:30:28 no Oct 14 05:30:35 what? Oct 14 05:30:44 text input system? Oct 14 05:30:47 wtf are you talking about? Oct 14 05:30:53 but yeah Oct 14 05:30:54 lasserix_ Sorry, new to the API Oct 14 05:31:12 sorry your question makes no sense to me Oct 14 05:31:31 what does draw and input have to do? Oct 14 05:31:43 or i could generally say, as long as you call super? Oct 14 05:33:06 I will try harder: The "standrd" android keyboard (i know you can install several but for sinplicity), that I want to use, but I want to override all drawing methods as the "GUI" is to be used to a game. Oct 14 05:33:21 But yes, they dont really have anything to do with each other Oct 14 05:35:25 thanks lasserix_ its working in all sense now Oct 14 05:36:16 lasserix_ I assume that the "normal" control; textboxes, statics and so on are drawn with vector graphics as are all modern GUIs I know off, and I want to draw them using bitmaps that can be drawn very quickly by hardware blits. Oct 14 05:36:20 so whatever methods I need for a particular view I have to define them in the controller class only and just access them in MainActivity Oct 14 05:42:03 lasserix_: Oct 14 05:50:37 cph_dk: you want to replace the soft keyboard? Oct 14 05:50:48 niru you can define the methods whereever you want Oct 14 05:55:36 lasserix_: checkout the other tab Oct 14 06:25:34 there isnt xml designer other than the android studio one Oct 14 06:25:45 damn thing goes crazy and change values from other views Oct 14 06:56:36 i think i'm going to struggle with material design. even the basic stuff. like when i go into Play and scroll, the different papers move in a concerted but independent fashion - it doesn't even make sense to me. its like fucking with my brain, nothing in the real world would work like this. Oct 14 06:59:05 Which is funny, because material design is supposed to imitate an organic experience and feel more natural Oct 14 07:01:11 i'm skeptical, but maybe it will grow on me Oct 14 07:03:38 thats why it's called *organic* lol Oct 14 07:13:49 JesusFreke there ? Oct 14 07:16:20 hi all, I'm getting the following error while trying to get current gps location: http://pastebin.com/k8AD4Hf6 (the exception happens on line 28)...Any ideas how to solve it? Oct 14 07:28:00 java.lang.SecurityException: invalid package name: com.example.mypackage.test Oct 14 07:28:01 meital, did you look at the exception? SecurityException Oct 14 07:28:15 I didn't understand what it means Oct 14 07:28:44 http://stackoverflow.com/questions/16176862/securityexception-when-calling-getlastknownlocation Oct 14 07:28:46 you are probably missing the permissions needed in your manifest Oct 14 07:28:47 google it Oct 14 07:29:11 I did and I read this link Oct 14 07:29:15 it's not the same issue Oct 14 07:29:17 yes Oct 14 07:29:22 did you add permissions? Oct 14 07:29:22 they are talking about "com.google.android.gms" Oct 14 07:29:25 yes Oct 14 07:29:33 are you using a custom rom? Oct 14 07:29:40 yes Oct 14 07:30:21 that could be it Oct 14 07:30:42 how do you think that it's related Oct 14 07:31:28 read the 3rd and 2nd to last comment under the question: http://stackoverflow.com/questions/16176862/securityexception-when-calling-getlastknownlocation Oct 14 07:32:57 I'll try running it on nexus in a sec, but I really don't think that this is the problem... Oct 14 07:34:29 they didn't say what the actual problem was Oct 14 07:34:36 it is a starting point Oct 14 07:36:30 ok, thanks Oct 14 07:38:43 same error on nexus Oct 14 07:43:37 iunno man Oct 14 07:43:57 you try changing the name? Oct 14 07:44:08 or starting a new blank project Oct 14 07:48:51 meital: pastebin your manifest Oct 14 07:55:07 or not, up to you Oct 14 07:59:00 Morning All Oct 14 08:00:16 Leeds: http://pastebin.com/wjbygHup Oct 14 08:03:51 What's a really typical way to get a user to click something without interrupting the workflow? Oct 14 08:03:58 I dont' want dialogs and all that shit ^^ Oct 14 08:04:14 meital: that's got a different package name in it... Oct 14 08:05:47 its possible to avoid that one activity ever goes to pause? just need it to destroy and recreate Oct 14 08:06:43 and create* Oct 14 08:08:13 Leeds: it's the same one, I modified it before sending it to you ;) Oct 14 08:08:29 well, good luck with that then... Oct 14 08:09:07 oh wait I did copy the wrong one :O Oct 14 08:20:55 hmm, learning about actionbarsherlock is fun Oct 14 08:21:36 ~fun? Oct 14 08:21:59 https://code.google.com/p/styled-action-bar/source/browse/trunk/res/menu/main_menu.xml Oct 14 08:22:12 basically the menu is tag is confusing Oct 14 08:22:56 I have been trying to include style and inside items inside it Oct 14 08:23:10 dammit quality of code decrease severely when i avoid to rest Oct 14 08:23:12 but I just learned that I have been trying it wrong Oct 14 08:23:16 or need more caffeine Oct 14 08:23:53 i thought sherlock isnt very useful now that android provides native backwards compativility Oct 14 08:23:58 bility* Oct 14 08:25:20 http://pastebin.com/CjZfzNUx Oct 14 08:25:39 so, all I wanted to add was a default background colour Oct 14 08:25:46 is it too much to ask Oct 14 08:28:51 is it possible to pass in two types of params to an Oct 14 08:28:54 AsyncTask Oct 14 08:29:46 hmm Oct 14 08:29:54 onats, no, it only takes one generic Params type - you'll have to make your own class to hold your params Oct 14 08:30:01 any good reliable ways to wait for layout in custom views? Oct 14 08:31:05 anyone know how i flatten a curve? Oct 14 08:31:27 AbigailBuccaneer: can I pass in arraylists to it? Oct 14 08:31:41 ie i am generating a scaling factor min-1 (min is greater than 0), the values decline too sharply, how can i flatten it? Oct 14 08:32:17 yeah onats, Params could be ArrayList Oct 14 08:32:29 (or, preferably, List) Oct 14 08:36:24 AbigailBuccaneer: it’s complaining I ahve to implement its doInBackground Oct 14 08:37:33 please ignore Oct 14 08:40:34 lasserix_: hammer and/or exercise is your friend depending where curve is Oct 14 08:40:35 lasserix_, that's a fairly generic question - what properties of the curve do you want to keep? Oct 14 08:40:54 hmm Oct 14 08:41:22 i mean, the easiest way to "flatten" a curve is to replace it with f(x) = 0, but intuitively that's probably not what you want :) Oct 14 08:42:17 if you apply a monotonically increasing function to the output of the curve - say sqrt(x), then it'll keep the sign of its derivative Oct 14 08:42:37 (ie. everywhere the original has peaks and troughs and inflections, so will the the new curve) Oct 14 08:42:47 basically i am trying to find a scaling function for Oct 14 08:43:10 the size of a dot on the face of a tube Oct 14 08:43:42 where orenitation the tube so that it revolves about an axis perp to your faces' forward vector Oct 14 08:44:02 and the greatest point (1) is at the center, so as the dot travels to the top or bottom (pole) it decreases Oct 14 08:44:36 maybe it's just because i haven't had breakfast yet but i really don't follow, lasserix_ Oct 14 08:44:38 right now i am just using the projection on the y axis as the ratio of the diff between the center and the relative position of the dot Oct 14 08:45:38 like if you had a bike tire, and you spun it so its axis of revolution is perpindicular to you Oct 14 08:45:45 and text was written on the rim, Oct 14 08:46:11 holding the tire in front of you so your eyes are coplaner with the center of the tire, so that the text on the rim closest to is the largest Oct 14 08:46:29 as you spin it the text goes farther away from you Oct 14 08:46:40 and decreases in size (relative to what you see) Oct 14 08:46:49 okay, i understand so far Oct 14 08:47:24 so like i said, project the point onto the y axis (its the up vector, z is into you, x is horizontal) Oct 14 08:47:34 take the abs diff between it and the center point Oct 14 08:47:53 then the ratio of that and the center point gives you 0 - 1 where values at the top and bottom become 0 and closer to the center become 1 Oct 14 08:47:57 but it's too steep Oct 14 08:49:54 oh this is the android dev Oct 14 08:49:54 heh Oct 14 08:50:01 it's for a list view i am trying to make circular Oct 14 08:50:05 looking Oct 14 08:50:28 yeah i did think this was a slightly odd channel to ask the question on, but it sounded fun :P Oct 14 08:51:37 oh Oct 14 08:51:41 i was asking in the math got confused Oct 14 08:51:53 which channel, but its for a listview to make it appear "circular" Oct 14 08:52:05 by which you mean cylindrical, right? Oct 14 08:52:06 anyways is there a way to do a Z rotate in canvas? Oct 14 08:52:08 yeah Oct 14 08:52:12 clindrical Oct 14 08:52:20 heheh, I got the background colour changed Oct 14 08:52:30 the activity's colour has changed Oct 14 08:52:54 and the area of actionbar where there is text or icon/drawable Oct 14 08:55:10 lasserix_, android.graphics.Canvas does deal with android.graphics.Matrix Oct 14 08:57:47 lasserix_, it looks like somebody's already solved this fairly elegantly :) https://github.com/renard314/ListView3d/blob/master/src/de/inovex/mtc/ListView3d.java Oct 14 08:58:39 solved something Oct 14 08:58:42 not what i need but it helps Oct 14 08:58:43 thanks Oct 14 08:58:47 fuck Oct 14 08:58:53 this is way too much work for this damn effect Oct 14 09:00:57 lasserix_, even if it's not the exact effect you want, it shows you a pretty nice strategy Oct 14 09:04:18 wot Oct 14 09:04:49 is there a good guide to code effects/graphic manipulation on android Oct 14 09:06:22 http://b2cloud.com.au/how-to-guides/3d-rotations-on-views/ Oct 14 09:07:43 something like that ty Oct 14 09:08:34 heh Oct 14 09:08:39 that was not direction at yo Oct 14 09:08:51 oh weird Oct 14 09:08:57 z is XY Oct 14 09:10:11 argh Oct 14 09:10:14 left handed system Oct 14 09:10:18 heh Oct 14 09:12:22 love it when you can dump a bunch of code because you switch the wya you do it Oct 14 09:12:30 no longer need to scale text if i just rotate it out o the visible plane Oct 14 09:13:31 Hello! Oct 14 09:15:01 pfn: Why do I still get this: "Failed to parse : must be an integer number or codename." ? I thought that this could be a string strating from 1.3.x? Oct 14 09:16:53 marrrk: where did you hear that it could be a string? Oct 14 09:17:14 Here: https://github.com/pfn/android-sdk-plugin Oct 14 09:17:17 L? Oct 14 09:17:21 hehh Oct 14 09:17:39 It says: minSdkVersion and targetSdkVersion are now SettingKey[String] and no longer SettingKey[Int] (support android-L) Oct 14 09:18:18 wha if its android-L Oct 14 09:18:22 are you writing scala? Oct 14 09:18:30 Yes Oct 14 09:21:39 yeah, you need to be clear about that up-front... someone else might be able to help Oct 14 09:22:30 say, pfn :) Oct 14 09:25:56 Okay, sorry I thought the "pfn:" at the beginning was enough to imply that :) Oct 14 09:28:14 I have another, more general question, though. How do I share data between fragments? I have this facebook session and I want it to be accessible in different fragments. Should I put it in the containing activity or can I pass it around in some other way? Oct 14 09:29:28 aaahhh, that feeling when your actionbar looks the way you want it to Oct 14 09:29:36 make a Facebook state manager and inject is as a dependency into all fragments that need it Oct 14 09:29:51 read about dependency injection if you don't know what I'm talking aout Oct 14 09:35:18 I am reading about this now. Oct 14 09:52:59 What would cause gradle not to include transitive dependencies? Oct 14 09:55:44 If you're depending in a specific packaging (e.g. if you've added @aar) Oct 14 09:56:16 why would that cause gradle to not include the dependencies? Oct 14 09:57:59 Ask those who made gradle Oct 14 09:58:56 Is the right way to explicitly set transitive=true for that dependency or is ther any other way? Oct 14 09:59:25 Is it possible to relase an update of your already published app but the update is only for say 4.+ devices ? Oct 14 09:59:37 While the old version remains for the older devices ? Oct 14 10:00:26 No idea, try it Oct 14 10:01:00 what if you use version detection for the new features Oct 14 10:01:34 I just had that issue back when gradle was new and aars was in addition to apklibs. Everyone moved to wars now anyway Oct 14 10:02:10 wars? Oct 14 10:02:12 for android? Oct 14 10:02:58 neredsenvy, yes Oct 14 10:03:03 It's just the UI I like the option of having FlowLayout but I guess I'll redo the layout with what was available in 2.3 I really don't have anything complex on the UI 2 buttons one is a gps even other opens a drop down list Oct 14 10:03:04 aars Oct 14 10:03:17 neredsenvy, even though, I think you need to do multi-apk release to keep the listing for older apps Oct 14 10:03:35 neredsenvy, you'll see the entry in apk list and those that are available for download will be shown as such Oct 14 10:03:51 Alright i'll give it a shot Oct 14 10:04:17 so, alternating the row layout (in a listview) can only be achieved using division by two operation ? ? Oct 14 10:05:16 Uhm, if you want every other layout to be different, that's the easiest way to do it, yes Oct 14 10:06:19 yup, alternate rows Oct 14 10:06:37 if (position % 2 == 0) { setDark} else {set light} Oct 14 10:06:38 done :) Oct 14 10:07:42 hello! anyone got experience with product flavours ? Oct 14 10:07:48 Are there any cool/hip/just not shit ORM libraries out there at the moment? D: Oct 14 10:08:06 want to install same apk twice on the same phone, but still got Failure [INSTALL_FAILED_CONFLICTING_PROVIDER] Oct 14 10:08:09 error Oct 14 10:09:55 Mavrik: I thought that pros might have some elegant solution :-P Oct 14 10:10:18 AlternateRowProducerFactory ;) Oct 14 10:10:38 :-P thats what I am talking about :-P Oct 14 10:10:39 which extends RowProducerFactory which extends ProducerFactory Oct 14 10:11:40 gordon_: multiple apps can't have the same content provider authorities Oct 14 10:17:12 SimonVT: http://tools.android.com/tech-docs/new-build-system/applicationid-vs-packagename Oct 14 10:17:22 "With the new Android Gradle build system, you can easily build multiple different versions of your app; for example, you can build both a "free" version and a "pro" version of your app (using flavors), and these should have different packages in the Google Play store such that they can be installed and purchased separately, both installed at the same time, and so on. " Oct 14 10:17:57 I'm having an issue with ViewPager... I'm using one to display a gallery of items that can be either images or videos. The problem is that, because the ViewPager loads items up-front, I get the "Video loading" progress indicator on the image before the video. Unfortunately, setOffScreenPageLimit has a minimal value of 1, so I'm not sure how to prevent the video from loading in advance here Oct 14 10:19:14 gordon_: content provider authorities still have to be unique Oct 14 10:23:32 SimonVT: so it's impossible to install app twice ? Oct 14 10:23:57 what I want to achieve is, master build from google store and sometimes tests integration build on the same device Oct 14 10:24:04 No.. Just make sure they have their own content provider authority.. Oct 14 10:24:41 isnt possible by changing app package? Oct 14 10:24:42 Use two devices if you want to install the same app twice, or change the package name Oct 14 10:28:28 i wonder whats wrong Oct 14 10:29:23 SimonVT: thank you. I think I know how I can change it Oct 14 10:29:36 it's in AnroidManifest in section ? Oct 14 10:29:55 so i am tyring to build an interface where first there will be an image slideshow (using ViewPager for this) and a fixed chat box at the bottom. you can tap on the text box and write comments/chat messages that will start showing up below the viewpager. so I guess this will require scrolling hence my hierarchy is like this ScrollView > RelativeLayout > ViewPager + TextView - but ViewPager doesn't appears Oct 14 10:30:06 whats the right way to actually build this interface i wonder Oct 14 10:31:17 Yep, in the manifest. The content provider also uses it for uri matching Oct 14 10:31:28 has anyone used this app called hot or not? Oct 14 10:31:32 Doesn't ViewPager work with sliding/scrolling too, this is just a guess but your issue is probably that you're using 2 scrolling elements inside eachother shmoon Oct 14 10:31:34 =it has a similar interface Oct 14 10:31:38 shmoon probably by learning a bit more about it all and practice …. I mean what the relative layout for ? Oct 14 10:32:00 VnM: viewpager is horizontal scrolling, i am talking about the entire activity vertical scrolling Oct 14 10:32:31 StingRay_: i need multiple Views in the page, so wrapped them in a relativelayout and put in the scrollview since scrollview can have only 1 child Oct 14 10:32:54 ok yeah you really need some practice and less assumption Oct 14 10:33:24 viewpager > (page) viewGroup/scrollView +> many views Oct 14 10:34:15 I think that is more what you want…cause logically the way you have it…even if you could get it to work… it kinda means you want to horizontally scroll while keeping the previsou page vertical position when transitioning ... Oct 14 10:34:24 and not sure thats really what you want Oct 14 10:36:16 so in the page, first i want a slideshow of images. at the bottom there should be a chat box.if i swipe up with a nice transition the slideshow should go up and show me all the comments below the slideshow. Oct 14 10:36:40 slideshow has horizontall scrolling, entire thign has vertical scrolling and i want some nice transition when the swip up is done Oct 14 10:38:25 this is sort of what i want to achieve http://youtu.be/zumrT8abD5Y?t=1m28s Oct 14 10:38:28 in terms of UX Oct 14 10:39:10 btw Oct 14 10:39:16 if I'm doing some image filtering Oct 14 10:39:31 I need to use NDK or I can do something on java side? Oct 14 10:39:36 no experience in that field Oct 14 10:40:18 gordon_, I dont see why you could not do it in Java. Oct 14 10:40:58 memory Oct 14 10:41:04 with filtering I think it really depends on ,what,how and time constraints if RT Oct 14 10:41:05 well, I will try :) Oct 14 10:41:17 as to what to do and where Oct 14 10:41:56 if i create something usefull i will open source it Oct 14 10:42:09 maybe someone will add more options Oct 14 10:58:53 complicated shit Oct 14 10:58:57 apps arn't easy Oct 14 10:59:12 Life isn't easy! Oct 14 10:59:21 First app isn''t easy Oct 14 10:59:30 after that it gets a lot easier Oct 14 10:59:38 in my experience Oct 14 10:59:55 first app took me 6 months to make Oct 14 10:59:58 second app, 2 days Oct 14 11:00:32 shmoon, Oct 14 11:01:02 One of the shortest full fledged apps I built took 22 days from scratch to launch Oct 14 11:01:28 pifon: lol Oct 14 11:01:43 made more money on the second app than the first one Oct 14 11:01:51 so it wasnt shit either Oct 14 11:01:55 well Oct 14 11:02:11 i've spent 2 months on this and am thinking i have taken toooooo long, been depressed for the past few days due to this fact Oct 14 11:02:24 your first app? Oct 14 11:02:30 Then I worked for two product companies an observed one very annoying thing in both of them (then I started up) Oct 14 11:02:30 ya first mobile app, first android app. Oct 14 11:02:47 shmoon, wait, I'll send you something that will ease you Oct 14 11:02:54 how can I see the memory my app is hogging Oct 14 11:03:08 more like a benchmarking question Oct 14 11:03:09 yeah first app is the hardest becuase you have to figure out everything from scratch Oct 14 11:03:49 smallest of errors need hours of googline Oct 14 11:03:52 googling Oct 14 11:04:05 shmoon, 18weeks! http://www.kinvey.com/blog/2086/how-long-does-it-take-to-build-a-mobile-app Oct 14 11:04:26 Looney, There was a tool called LittleEye which would do that Oct 14 11:04:43 otherwise you've got to be pretty much of a data scientist to make sense of the things profiling returns :P Oct 14 11:05:17 hehehe, data scientist and yours truly Oct 14 11:05:19 :-P Oct 14 11:05:31 :P Oct 14 11:05:57 Is it only me who thinks profiling should have been easier on Android? Oct 14 11:06:10 heck, there are so many things that needed to be super easy but they aren't Oct 14 11:06:19 for example, custom fonts on TextViews Oct 14 11:08:10 Yeah just let us use Comic Sans! Oct 14 11:08:15 well, Google should've gone with C/C++ in lieu of Java Oct 14 11:08:35 but that is my opinion Oct 14 11:08:59 and here I am abusing Java to earn my bread and butter Oct 14 11:11:12 Looney, em Oct 14 11:11:14 Looney, Android Studio Oct 14 11:11:17 bottom right corner Oct 14 11:11:21 "Memory monitor" Oct 14 11:11:24 click on it. Oct 14 11:12:24 sheikhaman, hmm, I think profiling is easy enough, the quality of information needs improvement tho Oct 14 11:12:36 clicking a button in DDMS isn't really rocket science Oct 14 11:12:44 I am better without profiling if I have to use AS or Eclipse Oct 14 11:13:15 huh Oct 14 11:13:31 what kind of stupidity is this? You have a tool for profiling and you refuse to use it and then whine here? Oct 14 11:13:57 you have also command line dumps available since kitkat Oct 14 11:15:42 hmm, I once tried to set AS for a project which was created in AS Oct 14 11:15:55 took me quite sometime Oct 14 11:16:08 first it demanded some particular gradle version Oct 14 11:16:17 which I tried to provide Oct 14 11:16:23 Hey guys I am making something like a game. So lets say that people are split into groups of 3 or 5 persons and they have to be viewing each member of their group moving on Google Maps. This means that I have to find a good way to update each clients map real time or pseudo-real time at least. Oct 14 11:16:26 but couldn't please it Oct 14 11:17:07 and ended up making project ant compatible Oct 14 11:17:12 Would a server + Database approach work or should it be developed based on the peer-to-peer logic? Because Image that there could be thousands of people playing it and all of them are going to belong in groups Oct 14 11:19:26 Mavrik, Right, profiling is easy, but gives too much "data" out, too less "information" Oct 14 11:19:45 sheikhaman, hmm, yeah, I think I know what do you mean Oct 14 11:20:09 the process is not that tough, what you have in the end is a dump, and you have to dig through it to make sense out of it Oct 14 11:20:21 sheikhaman, for what kind of data? Oct 14 11:20:48 Mavrik, I'd be shamelessly comparing it with XCode's profiling tools :D Oct 14 11:21:01 in what way are those different? Oct 14 11:21:09 How much memory, battery and network, "my app" is eating, I am just concerned with that Oct 14 11:21:16 I only used them a few times, but they had pretty much same facilities available. Oct 14 11:21:36 I haven't made the switch to AS yet, and if its easier there, then my whining is out of place certainly Oct 14 11:21:38 (note that that was at times of Xcode5) Oct 14 11:21:59 well AS isn't a magic bullet, but there are rather useful tools out there Oct 14 11:22:09 XCode6, when you launch an app and go to some mode (I guess debug), has a live graph of the resource usage of your app Oct 14 11:22:11 ADM (while fugly as hell) is quite useful Oct 14 11:22:12 that's it Oct 14 11:22:27 you don't have to enable or take a dump of anything Oct 14 11:22:31 its just right there Oct 14 11:22:44 well, that's what my point is. Oct 14 11:22:50 Google is an engineering driven company Oct 14 11:22:59 what they make, is engineering marvel for sure Oct 14 11:23:12 but a normal developer, is not a marvel engineer most of the times :D Oct 14 11:23:50 mhm, I agree Oct 14 11:23:58 if I put out resource usage of all the threads in a system before developers, from which very few even understand the threading completely (I also am not a pro there), they are less likely to make sense out of it Oct 14 11:24:05 even though, I see alot of people complaning when they just forgot to even check the tools they have available Oct 14 11:24:11 result : they won't be taking pains in improving the performance of the apps Oct 14 11:24:35 while true, I really don't see just how in this world can you write a mobile app without understanding threasd Oct 14 11:25:18 Oh, I live in a country where you find thousands of people like that, at least in the beginning stage :D Oct 14 11:28:12 would you people kill me if I implement a class as runnable for network related tasks ? ? Oct 14 11:30:54 Looney, hehehe.. Oct 14 11:31:40 Mavrik, but I'll surely check what AS has to offer in Memory Monitor, thanks for it! Oct 14 11:32:30 That's great, the S3mini does not mirror the activities automatically after setting the language to Arabic Oct 14 11:32:36 I mean.. Oct 14 11:32:44 I cannot find anything about this, why don't they mirror it Oct 14 11:32:56 I will also try to re-install AS Mavrik Oct 14 11:33:08 to see how it is better than Oct 14 11:33:14 *Eclipse Oct 14 11:35:25 platzhirsch, is it possible that the Android on that is too old to support RTL? Oct 14 11:35:39 Mavrik: no, it's a 4.2 device Oct 14 11:37:05 What would be more efficient speed-wise; inheriting Android built-in GUI class and overiding their draw methods or writing a whole new GUI system? Oct 14 11:37:18 es Oct 14 11:37:33 s3mini is 4.1.2 isn't it? Oct 14 11:37:52 wait what Oct 14 11:38:03 oh damn it, you are right! Oct 14 11:38:07 at least there isn't an update available for mine Oct 14 11:38:18 Zharf: thanks, I was really believing that I am on 4.2 with this device Oct 14 11:38:25 Mavrik: so yeah good point Oct 14 11:38:55 cph_dk, depends on what exactly are your requirements Oct 14 11:39:02 and what do you mean by "GUI system" Oct 14 11:44:34 Interesting in knowing more about writing a GUI system :D Oct 14 11:46:53 Eclipse Related : Has it happened with anybody where the editor (Java/XML) shifts to extreme right for some reasons leaving a big blank between project explorer and the editor? Oct 14 11:48:09 sp should be preferred font size for screen-size and screen-dpi agnostics Oct 14 11:49:01 missed "?" in last message Oct 14 11:54:08 Looney, yes - SP is DP scaled with users font size settings Oct 14 11:54:18 so you should use it for prettymuch most-all text Oct 14 11:54:45 looks like my memory is serving me well :-D Oct 14 11:54:51 sheikhaman, maybe your files encoding is set to Arabic? not sure if that would cause that but sounds like it switched to RTL Oct 14 11:55:03 motivation to get started on the UI/data rendering part Oct 14 11:56:16 Maybe you had opened (and altered) your files in a different program and it changed the encoding Oct 14 11:57:58 VnM, not at all buddy, its like i am working on it since hours, and switching between maximized and restored positions, and suddenly once during that it decides to act funny Oct 14 11:58:40 when I try to resize the project explore to just check if there's anything in the middle, there's something square-ish, in the middle of the worskpace, but is blank (grey).. Oct 14 11:58:49 that's quite ghost like though Oct 14 11:58:56 Scary Oct 14 12:00:20 And it's only happening in some files? not all? Oct 14 12:01:32 I'm having issues with Eclipse + Android http://i.imgur.com/HJaGBCg.png Oct 14 12:02:06 it started out of the blue and i tried to nuke the ~/.eclise and /usr/share/eclipse Oct 14 12:14:38 is it possible to fix the width of textview such that it only has space for 4 digits/characters? Oct 14 12:14:59 pick a good size? Oct 14 12:15:07 most fonts change widths Oct 14 12:15:19 Hi all. I have a custom view and want to create a 'customTextColor' parameter, so that I can set this in xml. How can I do this? Oct 14 12:15:54 mikedg: I think that default font is sans serif and has fixed width Oct 14 12:20:30 VnM, I'll notice what files it's happening with. that's a good point! Oct 14 12:21:28 FrancescoV, You should be looking at defining and picking attributes in attribs.xml Oct 14 12:21:32 (or attrib.xml) Oct 14 12:23:35 sheikhaman: ok thanks, didn't know the corrent name to get this done Oct 14 12:23:42 :) Oct 14 12:27:25 on an unrelated note, is absence to women in Tech (generally) and Android (specifically), to a large extent, an omnipresent phenomenon? Oct 14 12:27:49 Whats women? Oct 14 12:28:48 hahaha.. got you, VnM ;-) Oct 14 12:38:29 free tip : if your layout viewer is throwing an exception while displaying layout, it might be because it's API level is set to 20 (wear), and your layout has an edit text Oct 14 12:38:45 AndroidWear doesn't have an EditText, and it throws error there Oct 14 12:38:52 (on eclipse) Oct 14 12:56:35 should I pass view to my class implementing runnable Oct 14 12:57:01 or retrieve data from the class instance Oct 14 13:11:42 anybody here who has used the Calligraphy library? : https://github.com/chrisjenx/Calligraphy Oct 14 13:23:13 hello. is it possible that any other method is called before onCreate when start new activity? Oct 14 13:23:48 slani: http://developer.android.com/reference/android/app/Activity.html Oct 14 13:23:58 check the "Activity Lifecycle" paragraph Oct 14 13:24:29 If you're getting errors before your onCreate slani, then there is probably something wrong in your Manifest file Oct 14 13:26:02 the problem is that I make and app with navigation drawer and onNavigationDrawerItemSelected is called before onCreate when I open app. Oct 14 13:30:22 That's very unlikely Oct 14 13:32:19 SimonVT: yes. What can be wrong? Oct 14 13:36:25 No idea. I'm not convinced that it's called before onCreate Oct 14 13:43:40 hey guys Oct 14 13:46:44 I am starting to repent my implementation of thread as a class which extends runnable Oct 14 13:46:49 :-P Oct 14 13:48:27 Im being argued here over something with Android Oct 14 13:48:29 to hell with it, I am going to pass ListView to my class Oct 14 13:50:14 and then use post() Oct 14 13:50:17 wish me luck Oct 14 13:50:44 Best of Luck :D Oct 14 13:53:23 from your grinning, I am assuming that what I am trying to achieve is not one of the possibilities Oct 14 13:53:45 I have a listview with items Oct 14 13:53:57 they're talking about wanting to edit/delete/etc on these items Oct 14 13:54:06 is swipe the usual way to go? or longpress? Oct 14 13:54:23 Odaym thats a design issue Oct 14 13:54:28 I know Oct 14 13:54:28 depends on you and your implementation Oct 14 13:54:36 but there must be a guideline, no? Oct 14 13:54:37 Odaym well ask a designer Oct 14 13:54:41 Odaym no Oct 14 13:55:45 Odaym if you follow all of google guidelines and …well.. have google design your app…just look at the google apps and "copy" Oct 14 13:56:07 but most of the time, better to do some actual "design" and proper consideration Oct 14 13:56:45 why do I feel that passing ListView to my class is a bad idea :'( Oct 14 13:57:48 Looney dunno, but you certainly are keep is up to date with your real time feelings and thoughts Oct 14 13:57:50 Because threads often outlive activities? Oct 14 14:00:14 alright, well let me tell you where this implementation is in question Oct 14 14:00:37 man i havent been back to this channel in awhile Oct 14 14:00:39 there's a navigation drawer that pops out from the right, inside it there is an expandablelistview and in one of the groups there is a listview that opens down Oct 14 14:01:13 it's what to do with those items inside that inner listview Oct 14 14:01:26 mini overflow? :P Oct 14 14:01:26 whether to make them editable by swiping them to the right, or longpressing them Oct 14 14:01:37 google loves their mini-overflows Oct 14 14:01:45 long-press is pretty dead Oct 14 14:01:50 as it has no call to action Oct 14 14:02:03 swipe has a similar issue Oct 14 14:02:12 not to mention it complicates viewpager/drawer code Oct 14 14:02:26 which is why google has started pushing overflow buttons Oct 14 14:02:39 I don't really understand what you're saying Oct 14 14:02:50 go look at the google music app Oct 14 14:02:58 those 3 dot mini overflow buttons Oct 14 14:02:58 I dont want to look at google-specific things Oct 14 14:03:09 I'm asking about this considering the setup I have (described above) Oct 14 14:03:10 well you are asking design questions Oct 14 14:03:27 Android people will refer design questions to Google's design patterns Oct 14 14:03:47 not all TachyonDev Oct 14 14:03:48 sounds like you are just trying to get support for your own train of thought Oct 14 14:03:53 StingRay_, fair Oct 14 14:03:55 google is not really good at design Oct 14 14:04:09 StingRay_, fair enough Oct 14 14:04:43 annoyingly apple is lots better for the most part :( Oct 14 14:04:47 well, the setup I have is described above, feel free to imagine it and answer with what you think would be suitable for it. Again, nav drawer from the right with listview inside one of the expandable listview's groups Oct 14 14:05:07 0daym if you really want design feedback, you should get some screenshots Oct 14 14:05:33 you want to know how to surface actions in an expandable listview in a drawer, right? Oct 14 14:05:39 for a given row Oct 14 14:05:51 long press vs swipable rows Oct 14 14:07:49 yep Oct 14 14:08:01 they want "just as with iOS" Oct 14 14:08:15 and iOS has right-to-left swipe, but that is already taken by the drawer I suspected Oct 14 14:08:56 here let me get the screenshot Oct 14 14:09:08 #1 Android Design Rule: Don't Copy iOS Oct 14 14:09:25 I know, it isn't me who's arguing this Oct 14 14:09:31 you cannot tell people that :P Oct 14 14:09:34 they won't be convinced simply Oct 14 14:09:42 heh you havent been doing this very long then Oct 14 14:10:00 i refuse to work for any company that has designers that can't see that Android requires its own design Oct 14 14:10:01 no they don't trust what I say Oct 14 14:10:01 hence here Oct 14 14:10:25 copying iOS design requires massive amounts of custom view code that because a monster to maintain Oct 14 14:10:39 if you have a small android team you cannot let designers push massively custom views Oct 14 14:10:39 https://www.dropbox.com/s/sz58tlethwyhkpv/Screenshot%202014-10-14%2017.09.47.png?dl=0 Oct 14 14:10:58 Speaking of design feedback, https://www.dropbox.com/s/ywt7ikw2ag64w01/Screenshot_2014-10-13-02-39-37.png?dl=0 halp Oct 14 14:11:06 Heh Oct 14 14:11:07 guh Oct 14 14:11:25 Darklust wtf is that ? Oct 14 14:11:38 its scary Oct 14 14:11:49 :/ I'm aware it's not pretty Oct 14 14:11:59 my old DJ app is pretty scary though Oct 14 14:12:00 lol Oct 14 14:12:19 https://play.google.com/store/apps/details?id=com.djtachyon.android.VirtualTurntableFree Oct 14 14:12:21 Darklust not that it's not pretty, it's more like erm…well…. I'm gonna shut up Oct 14 14:12:26 hah Oct 14 14:12:33 No tell me! Oct 14 14:12:41 anybody for some nice brownish orange? Oct 14 14:12:49 link isn't opening here! Oct 14 14:12:49 I wanna see! Oct 14 14:12:58 isnt opening? Oct 14 14:13:07 I know the 9-patch crap in the actionbars gotta go Oct 14 14:13:19 I just like purple Oct 14 14:13:25 https://lh4.ggpht.com/worzVQI0m0rEskvUX7ZKWKuvFhypqUYMF7TPY3DrB3PL7s2DgMqdKJwgLtyMSnBUeQ=h900-rw Oct 14 14:13:32 does that link work to the screenshot in the store? Oct 14 14:13:32 wow its 900am and my day is already complete Oct 14 14:13:37 Droid Turbo release date is finally known Oct 14 14:14:04 I'm not touching another Android phone until an Snapdragon 810 device comes out Oct 14 14:14:07 StingRay_: I want brutal honesty! Oct 14 14:14:25 TachyonDev: rumor is, its 805 :) Oct 14 14:14:33 xximjasonxx, exactly Oct 14 14:14:40 TachyonDev: is Nexus 6 going to have 810 Oct 14 14:14:41 looks like an Atari game Darklust Oct 14 14:14:45 Darklust well, dont forget not only am I a designer/artist…I'm quite critical too Oct 14 14:14:54 xximjasonxx, not from what ive heard, i heard its 805 too Oct 14 14:15:02 Lol oh geez Oct 14 14:15:02 where's the boss Oct 14 14:15:02 lol Oct 14 14:15:02 did you see my screenshot StingRay_ , TachyonDev ? Oct 14 14:15:04 is there a device that is going to have an 810? Oct 14 14:15:11 right above Darklust's Oct 14 14:15:19 Darklust and I would fire any designer that even thought that was good or passable … :( Oct 14 14:15:40 Well what needs to be different Oct 14 14:15:41 0daym: what are the calls to action on those rows supposed to do? Oct 14 14:15:45 why so gloomy Darklust Oct 14 14:15:54 I had the iPhone 6 for a week, ended up going back to my Maxx. Which I had known the Turbo was coming, I wouldve used my Verizon upgrade for htat Oct 14 14:16:00 live and learn i guess Oct 14 14:16:01 ;0; I suck at UI Oct 14 14:16:02 Darklust all of it Oct 14 14:16:20 only on the Alternate Tuning group are you supposed to be able to edit the listview's items Oct 14 14:16:21 colors, spacing alignment …. well… all of it Oct 14 14:16:29 the items in lighter gray Oct 14 14:16:38 edit, delet Oct 14 14:16:38 +e Oct 14 14:16:45 Hm Oct 14 14:17:07 I thought borders to the listviews would help Oct 14 14:17:29 i take pride in my designer working with my on the android app, i got the app featured on the android wear store :) Oct 14 14:18:24 Another issue is how much space I have I don't know what to do with Oct 14 14:18:39 That screenshot is from a 10 inch device Oct 14 14:19:16 so how about not using a damn drawer Oct 14 14:19:30 why do people hate activities so much these gays Oct 14 14:19:31 days Oct 14 14:19:33 lo Oct 14 14:19:57 O.o Oct 14 14:20:19 square doesnt even like fragments now Oct 14 14:20:29 http://corner.squareup.com/2014/10/advocating-against-android-fragments.html Oct 14 14:20:58 but the overuse of drawers is getting a little crazy Oct 14 14:21:17 Yeah, I still have to finish reading their article Oct 14 14:21:21 they do make some sense Oct 14 14:21:35 although fragments made a few of the things easier, they added a lot of complexities too Oct 14 14:21:39 TachyonDev thats what you get when there are few leaders and free thinkers who can design, and you just have armies of followers and copiers :) Oct 14 14:21:49 heh yeah Oct 14 14:22:20 square may have gone a bit far with the whole encasulated unit testable custom view stuff Oct 14 14:22:47 where the whole app is one activity views are just moved around Oct 14 14:23:20 it sounds nice, except nobody is doing it Oct 14 14:24:46 Odaym: basically, the amount of effort you are putting into this custom drawer implementation, you could have been done if you did it in an activity Oct 14 14:25:06 man... Oct 14 14:25:13 what are you talking about man Oct 14 14:25:40 trying to stuff all this implementation into a drawer is restricting the hell out of you Oct 14 14:25:51 well it's already been done Oct 14 14:25:57 my question does not lie there Oct 14 14:26:02 then add a button to each row Oct 14 14:26:07 with a popupwindow Oct 14 14:26:15 done Oct 14 14:26:17 so no swipe, no longpress Oct 14 14:26:33 you can make longpress bring up the popup window too Oct 14 14:26:35 if you want Oct 14 14:26:45 but there is no visible call you action with either thing Oct 14 14:26:54 what does clicking on the row do? Oct 14 14:27:01 that is correct Oct 14 14:27:03 the last sentence Oct 14 14:27:13 it chooses that tuning and populates the activity with it, closing the drawer Oct 14 14:27:19 places a checkmark as the one you see too Oct 14 14:27:26 how about this Oct 14 14:27:43 clicking on it bring up a spinner/chooser that lets you either "Select, Edit, Etc" Oct 14 14:27:54 the call to action? Oct 14 14:27:57 not the thing itself Oct 14 14:28:08 so take over the whole rows click Oct 14 14:28:26 nah, I dont want to have to go through 2 steps to select the tuning Oct 14 14:28:27 and let them chose to either set that as the current choice or edit or delete Oct 14 14:28:27 that's a no Oct 14 14:28:58 so little room over there Oct 14 14:29:09 okay so how about a whole new screen for editing/deleting with the same info in there Oct 14 14:29:18 why try and cram all the editing stuff in there Oct 14 14:30:12 perhaps the header "alternate tuning" has the edit pencil icon on it Oct 14 14:30:34 and clicking that brings you to a new activity where you can edit and delete and do whatever you want without being restricted by the drawer Oct 14 14:30:52 hmm Oct 14 14:31:07 no way they'd go for that Oct 14 14:31:11 or any of what you mentioned so far Oct 14 14:31:13 nor would I Oct 14 14:31:26 but thanks Oct 14 14:31:30 you are coding this for a phone right? Oct 14 14:31:35 yes Oct 14 14:32:09 good luck man Oct 14 14:32:31 thanks Oct 14 14:32:52 wait, can you edit your current selection? Oct 14 14:33:10 or can you only edit ones you dont have selected? Oct 14 14:34:15 i was trying to figure out how to have a popup button along with the checkmark, but if they are mutually exclusive, you can put the mini-overflow icon on any row that isnt currently selected Oct 14 14:34:16 you can edit anything Oct 14 14:34:29 hmm Oct 14 14:34:52 do you think you could highlight that checked row without a checkmark? Oct 14 14:35:26 https://lh5.ggpht.com/gc3q2j461-5W5QMBsULoyjkrFBW6x5T368OCuaw8KMtQi21Wj7MTpNQNde_AiApZYA=h900-rw Oct 14 14:35:41 though google music crams multiple icons in Oct 14 14:36:03 so yeah that my suggestion, mini-overflow on each row with the options in it Oct 14 14:36:18 figure out how to fit everything else Oct 14 14:36:32 long press can also launch the popup Oct 14 14:37:25 Odaym, good luck Oct 14 14:38:52 I will have to confess, I love coming to this chat room :) Oct 14 14:39:04 This was one thing missing in my life since quite some time! Oct 14 14:39:27 hello, anyone can help check out why i always got cursor.getCount = 0? Oct 14 14:39:47 semanty: that question is massively vague Oct 14 14:39:53 semanty cause there are no records presented in that cursor Oct 14 14:40:08 * StingRay_ loves solving things :) Oct 14 14:40:12 lol Oct 14 14:40:39 TachyonDev: StingRay_ then how to solve? Oct 14 14:40:40 semanty: either you database doesnt have the data you think it does or your projection/selection is incorrect Oct 14 14:41:00 semanty: what content provider are you querying? Oct 14 14:41:12 a system one or your own? Oct 14 14:41:25 TachyonDev: i use project as null, there is no exception either Oct 14 14:41:31 projection Oct 14 14:41:40 semanty: what content uri Oct 14 14:41:41 TachyonDev: my own Oct 14 14:41:45 okay Oct 14 14:42:00 and are you sure you actually inserted the data? :P Oct 14 14:42:14 or have your database setup right? Oct 14 14:42:38 TachyonDev: this is my content provider https://dpaste.de/R3dx Oct 14 14:42:42 for query Oct 14 14:43:14 TachyonDev: i have a refreshservice to inser the data Oct 14 14:43:23 looks generally right, when you did you insert did you log how many were inserted? Oct 14 14:43:28 TachyonDev: so there is no wrong with database Oct 14 14:43:36 semanty, im not going to debug it Oct 14 14:43:47 semanty, generally looks like okay architecture .. Oct 14 14:43:52 TachyonDev, sup home slice, long time Oct 14 14:44:04 pfn: hey bro .. yeah Oct 14 14:44:14 pfn: not sure why im here lol Oct 14 14:44:16 TachyonDev: i can see in the log, that the data are correctly inserted Oct 14 14:44:28 TachyonDev: it is just in refresh service Oct 14 14:45:01 so you query(uri, null, null, null, null) and get 0? Oct 14 14:45:06 TachyonDev, you've gotten bored or something Oct 14 14:45:31 pfn: yeah maybe .. we just rewrote all our APIs so now I have to take a perfectly working product and redo it all :P Oct 14 14:45:39 pfn: though we are on retrofit now which is amazing Oct 14 14:46:26 TachyonDev: yes, Oct 14 14:46:49 TachyonDev: oh, i see some Oct 14 14:47:17 I have this problem: i launch an activity B (with startActivity()) from activity A without any particoular flags when i click on a button after 10 second. If i click button and put A in background and open another app (Telehone for example) i have two different behaviour: on Android <= 4.3 B doesn't appears; on Android >= 4.4 B appears. Why ? Oct 14 14:47:39 TachyonDev, everyone seems to say that Oct 14 14:47:46 pfn: lol Oct 14 14:47:55 I haven't yet tried retrofit Oct 14 14:48:03 pfn: well i mean the POJO thing does not like proguard whatsoever Oct 14 14:48:21 pfn: so you end up opening yourself up a bit to api analysis Oct 14 14:49:19 pfn I second what TachonDev said (amazing) Oct 14 14:49:37 I'll give it a whirl some day, maybe Oct 14 14:49:41 on my next java project Oct 14 14:50:22 semanty: figured it out? Oct 14 14:51:14 ech0s7: so you are saying that your activity stack is not being preserved which switching applications? Oct 14 14:51:41 ech0s7: try setting the 4.4 device to not save activities in the debugging settings, you may see the same behaivior Oct 14 14:51:58 no, TachyonDev i'm saying that from Android 4.4 if my activity is inactive (not in foreground) could launch activity in front! Oct 14 14:51:59 ech0s7: the 4.3 device may be memory managing Oct 14 14:52:19 ech0s7, you're communicating pretty poorly, demonstrate this effect in a video Oct 14 14:52:25 adb screenrecord => post it somewhere Oct 14 14:52:29 imgur or youtube it Oct 14 14:52:32 srsly Oct 14 14:52:39 ok pfn one moment Oct 14 14:52:56 oh wait he says he is launching an activity from the background with a timer? Oct 14 14:53:05 okay i dont want any part of this lol Oct 14 14:53:14 TachyonDev: i am not sure, i do not set up the view, because i found it may be due to the asyn task, there are two times, at 1st, no values but at 2nd, it has Oct 14 14:53:18 yes TachyonDev i launch activity in background with a timer Oct 14 14:54:04 ech0s7: why the hell would you do that Oct 14 14:54:31 in any case, launching an activity from the background has always worked the same way Oct 14 14:54:50 but yeah, your app might be getting killed, as TachyonDev says Oct 14 14:55:07 pfn, no! before android 4.4 if you start activity from background the second activity doesn't appears Oct 14 14:55:43 with android >= 4.4 if you launch an activity from another one in background, the second one appears Oct 14 14:55:48 i'm recording a video Oct 14 14:56:05 ech0s7: are you sure its the android version doing that and not that its a faster phone with more memory to hold the activity stack? Oct 14 14:56:08 no, it doesn't work for *you* Oct 14 14:56:29 hehe Oct 14 14:57:02 TachyonDev: yes, i have two phone Samsung Note3, one with android 4.3 and one with 4.4.2 Oct 14 14:57:12 i have found this post Oct 14 14:57:12 http://stackoverflow.com/questions/20676691/flag-activity-multiple-task-doesnt-work-on-android-4-4-kitkat Oct 14 14:57:29 see 1 Answer, pfn, TachyonDev Oct 14 14:57:49 interesting Oct 14 14:57:53 it says In Android 4.3, Activity started from inactive Activity is also inactive. But in Android 4.4, Activity started form inactive Activity will be brought to front. Oct 14 14:58:08 I just verified that it is so Oct 14 14:58:39 huh, yuck Oct 14 14:58:41 what ugly design Oct 14 14:58:57 pfn what design ? Oct 14 14:59:02 if you are running into this probably then it sounds like your architecture needs tweaking Oct 14 14:59:09 are you talking about ugly design in Android 4.4 ? Oct 14 14:59:12 launching an activity from the background like such Oct 14 14:59:36 that wasn't ever documented as working like that, iirc Oct 14 15:00:09 i hate splash screens Oct 14 15:00:17 what a waste of everybody's time Oct 14 15:00:43 but it may happen that for some reason you click a button that starts an activity and you are faster to click the home button of the terminal then system to run startActivity() line of code :) Oct 14 15:00:56 or your application thread is scheduled to run after Oct 14 15:01:02 should customadapter always extend arrayadapter Oct 14 15:01:09 Looney, no Oct 14 15:01:28 is a very remote chance, but there is no need to have an application with a timer (and ugly design) Oct 14 15:01:33 It should never extend arrayadapter Oct 14 15:01:52 pfn: i give you another example Oct 14 15:01:52 SimonVT: +1 Oct 14 15:02:19 pfn: imagine that you are waiting for a response from server, and when response arrive you have to start an activity Oct 14 15:02:29 ech0s7, doing it wrong Oct 14 15:02:33 lol Oct 14 15:02:39 indeed Oct 14 15:02:50 firing context changes from a server call? Oct 14 15:02:55 yikes Oct 14 15:03:07 TachyonDev: not on the same thread Oct 14 15:03:14 ech0s7: thats not the problem Oct 14 15:03:27 ech0s7: server calls have an arbitrary length of time Oct 14 15:03:39 ech0s7: UI responses should be immediate Oct 14 15:03:50 yes, TachyonDev Oct 14 15:04:19 having a random activity interrupt what im doing, up to seconds later, does not sound fun Oct 14 15:06:04 SimonVT: should it extend BaseAdapter then? Oct 14 15:06:15 Looney: where is the data coming from? Oct 14 15:06:27 Yes Oct 14 15:06:31 every SEO-ed tutorial seems to be favouring arrayadapter Oct 14 15:06:34 Or CursorAdapter Oct 14 15:07:37 Looney: arrayadapter is for very basic datasets.. basically for tutorial purposes to show you how adapters work heh Oct 14 15:07:39 found http://developer.android.com/guide/topics/ui/layout/gridview.html Oct 14 15:09:03 Looney: right, baseadapter doesnt limit the type of structure you are iterating through Oct 14 15:09:42 cool Oct 14 15:09:51 cursoradapter is built for iterating through the content provider, which is what most apps end up doing Oct 14 15:10:40 well, iterating though a database with a cursor ..doesn't have to be a CP Oct 14 15:11:28 eh fair Oct 14 15:12:23 TachyonDev: https://dpaste.de/QTzF could you help on this? Oct 14 15:12:51 but i think he is talking adapters for listview presentation Oct 14 15:13:46 CursorAdapter + Loaders = Bees Knees Oct 14 15:13:58 i’ve imported my android project into intellij.. how do i setup the android sdk though? Oct 14 15:14:10 so if I extend BaseAdapter, all I have is to .add() my data to it Oct 14 15:14:13 mylord: dont use intellij Oct 14 15:14:32 https://developer.android.com/sdk/installing/studio.html Oct 14 15:14:34 TachyonDev: why? and what’s better? eclipse? it’s kind of slow on osx, btw Oct 14 15:14:48 Android Studio is a custom version of intellij Oct 14 15:15:01 eclipse is garbage Oct 14 15:15:25 or I should pass it arraylist Oct 14 15:15:26 TachyonDev: i just don’t like gradle Oct 14 15:15:30 it is? Oct 14 15:15:31 Gradle is amazing Oct 14 15:15:36 what are you talking about Oct 14 15:15:38 semanty: You're changing your adapters dataset without calling notifyDataSetChanged Oct 14 15:15:55 IntelliJ is leaps and bounds ahead of Eclipse Oct 14 15:16:13 there is no excuse not to be using Android Studio and Gradle at this point in time Oct 14 15:16:23 TackyonDev: i’ve had more trouble with it compared to eclipse when importing projects Oct 14 15:16:41 of course, because most projects are already in eclipse Oct 14 15:16:42 TackyonDev: ok, i’ll give it another shot Oct 14 15:16:58 it takes some front loading to get setup, but it is way more efficient Oct 14 15:17:10 i dont have a single line of imported source code.. its all pulled in with gradle Oct 14 15:17:35 pfn, TachyonDev have you seen video ? Oct 14 15:17:38 not to mention being able to support multiple build flavors and manifest merging Oct 14 15:17:39 SimonVT: do you know how to add into my code? https://dpaste.de/oLqk Oct 14 15:17:40 (test case) Oct 14 15:17:44 not interested since you're doing it wrong Oct 14 15:19:36 ech0s7: you should NOT be doing that Oct 14 15:20:25 once your application is switched from, that is it Oct 14 15:20:34 what you CAN do is fire a notification Oct 14 15:20:37 semanty: Not gonna look through that.. EVery time you change your adapters dataset you HAVE to call notifyDatasetChanged Oct 14 15:20:38 So, do that Oct 14 15:20:44 the behavior is different on 4.3 and 4.4 .... Oct 14 15:20:45 then the user can CHOOSE to go back to your application Oct 14 15:20:53 TackyonDev: I’m doing PhoneGap (cordova actually). With eclipse I import the prepared platform directory. How would I do that with AS? Oct 14 15:20:56 ech0s7: it doesnt matter if it is different Oct 14 15:21:01 mylord: NOOOOOO Oct 14 15:21:04 TachyonDev: why not ? Oct 14 15:21:07 ech0s7: because it is WRONG Oct 14 15:21:13 I really think so Oct 14 15:21:26 ech0s7: You should NOT fire activities from your application after switching context Oct 14 15:21:40 what if you are trying to text someone because a text came in, you are going to interrupt the user? Oct 14 15:21:49 that is a sure why to get unisntalled Oct 14 15:21:59 and I think it is correct to start an activity after a period of time within your app .... Oct 14 15:22:11 TackyonDev: why NO? I needed x-platform and I now also like having my entire app configurable server-side (especially to circumvent apple app updates 4+ days) Oct 14 15:22:27 TachyonDev: In fact, it is wrong behavior from Android 4.4 Oct 14 15:22:48 ech0s7: interrupting the user after they have switched away from your application is WRONG Oct 14 15:22:59 TachyonDev: YES! Oct 14 15:23:13 and before Android 4.4 was not possible.... Oct 14 15:23:21 ech0s7: so why do you care about the behavior of 4.4/4.3 if you shouldnt be doing it Oct 14 15:23:28 except new_task, etc... Oct 14 15:23:43 yes it was possible, you are just saying the activity stack is different Oct 14 15:24:05 TachyonDev: ...read except new_task, etc... Oct 14 15:24:15 okay im done Oct 14 15:24:19 next Oct 14 15:33:34 SimonVT: sorry, i tried, but don't konw the way Oct 14 15:34:38 What part of it is confusing? Oct 14 15:37:59 it was "possible" because you were using flag_activity_multiple_task Oct 14 15:38:05 abusing Oct 14 15:38:57 ech0s7: okay fine, one more word on this, I see what crazy shit you are doing now. You should NOT be trying to launch activities into a stack in the background so that when you go back to the application the stack is preserved. You have no idea what the device is going to do onPause of your application. It could recycle it or whatever. You may have been doing this before, but it is an UNSUPPORTED paradigm. The correct way to Oct 14 15:38:57 do this is to fire the activity immediately, put in a loader or spinner, and start the API call. If the user switches away, the activity is still there and preserved, and onResume() you load the data you got from the API call. Oct 14 15:39:32 pfn: took me awhile to figure out what he was doing Oct 14 15:39:51 pfn: he is building background activity stacks that before 4.4, wouldn't surface. Oct 14 15:39:58 yeah, that's completely wrong Oct 14 15:40:32 TackyonDev: http://stackoverflow.com/questions/24936689/import-cordova-project-in-android-studio Oct 14 15:41:19 mylord: dude, no idea how Android Studio works with those cross platform monsters Oct 14 15:41:32 mylord: thats allll you Oct 14 15:42:59 mylord also this is android dev, phongap and cordova have own channels Oct 14 15:43:09 k Oct 14 15:43:50 TachyonDev: could you help that,? i search onsite for much Oct 14 15:44:05 TachyonDev: Oct 14 15:44:08 TachyonDev: https://dpaste.de/oLqk Oct 14 15:45:06 semanty: why are you using a loader just to query a content provider? Oct 14 15:45:14 Hey guys I am making something like a game. So lets say that people are able to create groups/parties between them and I would like to give them the ability to view each member of their group moving on the Google Maps. This means that I have to find a good way to update each clients map's markers real time or pseudo-real time at least. Oct 14 15:45:21 Is the approach with a DBMS fine? Like every client should be responsible to update the DB every so with their current coordinates and thus everyone can make queries to view every other party member's location. Making this with a small time interval will look like real time. Or would a peer-to-peer approach be better? Like every group/party runs on its own by letting players exchange info between them containing everyone's new Oct 14 15:45:21 coordinates every so. I think the second approach is better because it does not give load to any server, but there may be drawbacks. Oct 14 15:45:44 cnap: is that an android question? Oct 14 15:45:58 TachyonDev: do yo mean there is no need? i learn using cursorload is better async task Oct 14 15:46:18 TachyonDev: could be.. Because if there is a lib or something that android community provides for good p2p stuff it should help Oct 14 15:46:47 semanty: the loader system is bound by the activity/fragment lifecycles Oct 14 15:47:10 TachyonDev: so? Oct 14 15:47:17 Also I am asking here because everyone in here has way more experienced in mobile development and any opinion is very welcome Oct 14 15:47:30 semanty: are you planning on putting that data into a listview? Oct 14 15:47:42 TachyonDev: yes Oct 14 15:48:22 TachyonDev: i have ever used simplecursor, but it seems not support separateAdaptorlist, that is y i change to simple cursor Oct 14 15:48:31 simple adapter Oct 14 15:48:35 semanty then you should be using a cursor adapter Oct 14 15:48:45 it takes care of all of this for you Oct 14 15:48:55 you just set the cursor in the adapter Oct 14 15:49:35 no need to pull all the data into structures like that Oct 14 15:49:47 TachyonDev: i am not familiar with that, does it support getViewType? Oct 14 15:49:51 yes Oct 14 15:49:57 it supports everything Oct 14 15:50:01 If you use resourcecursoradapter it even handles the holders Oct 14 15:50:04 it's the jesus of adapters Oct 14 15:50:08 mikedg: <3 Oct 14 15:50:36 running some ideas in my head... can I draw paths as dotted lines with multiple colours or solid lines with multiple colours side by side? Oct 14 15:50:44 semanty: do some research on cursor adapters Oct 14 15:50:46 on google maps that is Oct 14 15:51:00 bing cursor adapters Oct 14 15:51:06 TachyonDev: i saw it support, Oct 14 15:51:29 TachyonDev: but there is no way to change the current code? Oct 14 15:52:08 semanty: what is the current code doing Oct 14 15:52:34 TachyonDev: get data from two tables and list them in list view separately Oct 14 15:52:59 and what wrong with it? Oct 14 15:54:19 also cursors are initialize to -1 Oct 14 15:54:34 so you can do while(cursor.moveToNext()) {} without moving to first Oct 14 15:54:47 TachyonDev: The content of the adapter has changed but ListView did not receive a notification. Make sure the content of your adapter is not modified from a background thread, but only from the UI thread. Make sure your adapter calls notifyDataSetChanged() when its content changes. [in ListView(-1, class android.widget.ListView) with Adapter(class com.lw.android.mysite.SeparatedListAdapter)] Oct 14 15:56:03 right, so at the end of onLoadFinished() call stepAdapter.notifyDataSetChanged() and eventAdapter().notifyDataSetChanged() Oct 14 15:56:47 your steps data changed, so you have to notify all the adapters Oct 14 15:57:15 again, this is all automatic with cursor adapters Oct 14 15:59:21 TachyonDev: i do put o the load finished end, but i still the same Oct 14 16:01:02 omg http://jsharkey.org/blog/2008/08/18/separating-lists-with-headers-in-android-09/ Oct 14 16:01:03 lol Oct 14 16:01:16 i remember trying to do separated lists back in 2008 .. Oct 14 16:01:52 TachyonDev: wow, you really have much experience on that Oct 14 16:02:05 semanty: yup, been doing this for that long Oct 14 16:02:35 semanty: don't worry, Android has had sectioned listview problems for 6 years too Oct 14 16:02:38 you arent the only one Oct 14 16:03:26 they are finally fixing it in L Oct 14 16:03:31 TachyonDev: that is y i think that separatelist adaptor will be out o fdate soon Oct 14 16:03:42 TachyonDev: i don not use L one yet Oct 14 16:04:33 i havent even tried RecyclerView yet Oct 14 16:04:50 pfn, mikedg SimonVT: anybody tried RecyclerView and sectioned lists? Oct 14 16:05:08 recyclerview is for noobs Oct 14 16:05:49 lul Oct 14 16:06:15 TachyonDev: using just while(cursor.moveToNext()) {} without moving to first, it is the same Oct 14 16:06:35 i mean my current sectioned list implementation utilizes a custom loader with a matrix cursor and a custom cursor adapter Oct 14 16:06:48 semanty: yeah it should be the same, i was just pointing it out Oct 14 16:07:09 I made my own LayoutManager Oct 14 16:07:14 A grid with headers Oct 14 16:07:16 SimonVT: <3 Oct 14 16:07:40 TachyonDev: so the final way is just talk with my contentprovider or cursoradapter? Oct 14 16:07:55 semanty: eh .. not that simple when it comes to sectioning Oct 14 16:07:58 semanty: try RecyclerView Oct 14 16:08:13 SimonVT: I managed to do that multiple apks :) Oct 14 16:09:32 okay i need to consume at least one new API today Oct 14 16:09:39 tapping out for now Oct 14 16:10:24 TachyonDev: in the past,without recyclerview, that is the method? Oct 14 16:10:36 Hello Members. I have an app just published to the playstore. The app can be run on test devices from eclipse but when uploaded to the playstore it crashes on an error “Unable to instantiate application com.ClickApp.click.ClickApp” ClickApp Spcifically is a class that extends Application and is part of the primary package. Not an include. Has anyone else run into this or knows why this is happeneing? THANK YOU Oct 14 16:10:58 semanty: there is no standard method Oct 14 16:11:00 thats the problem Oct 14 16:11:52 ChuckingRocks: you should read the next few lines of the error Oct 14 16:12:13 TachyonDev: i feel sad, it takes me one week to do what i want, i have to work and not it is 00.12 Oct 14 16:12:24 now Oct 14 16:13:19 I can post the whole issue but i do not want to spam this chat. It is looking in “java.lang.ClassNotFoundException: Didn't find class "com.ClickApp.click.ClickApp" on path: DexPathList[[zip file "/system/framework/com.google.android.maps.jar", zip file "/data/app/com.ClickApp.click-1.apk"],nativeLibraryDirectories=[/data/app-lib/com.ClickApp.click-1, /vendor/lib, /system/lib]]” Oct 14 16:13:50 ChuckingRocks use pastebin Oct 14 16:13:54 How can it not find a class in the package com.ClickApp.click Oct 14 16:13:58 ok no prob Oct 14 16:14:40 http://pastebin.com/mMVE6Pr3 Oct 14 16:16:00 I will provide the manifest as well if that helps? :-) Oct 14 16:16:30 semanty: what you can do is have a single adapter and have a steps entry that is a "header" Oct 14 16:16:44 semanty: just an extended base adapter Oct 14 16:17:10 semanty: so when you need a section break, you add that into the steps data and have a row type for section header Oct 14 16:19:04 Since this seems to be a pathing issue the correction might be a small one. But i am blind to the solution. I am not using proguard either. Oct 14 16:19:51 TachyonDev: sorry i can not completely understand Oct 14 16:22:15 semanty: if you add a row type to your steps data, then you can can have a type for Section Header. Then you only need a single adapter. Oct 14 16:23:03 Which had nothing whatsoever to do with his issue, which is simply that he's not calling notifyDatasetChanged when he's changing the dataset Oct 14 16:23:14 SimonVT: he said he added that Oct 14 16:23:20 SimonVT: and it didnt do anything Oct 14 16:23:31 I wouldn't trust that Oct 14 16:23:36 SimonVT: lol Oct 14 16:23:50 At the very least he'd be getting another error Oct 14 16:24:03 i told him to notifyDataSetChanged on all his adapters at the end of load finished Oct 14 16:24:33 but im pretty sure he is using jsharkey's 2008 blog post as a reference Oct 14 16:25:20 TachyonDev: you mean i have to use baseAdapter Oct 14 16:27:01 TachyonDev: in fact, i don't care about header much, what i want is to separate step and event Oct 14 16:28:35 TachyonDev: a project has several steps, each step has many events, i want to show a project, several steps and many events on the same page, the height is automatically reset by its data count Oct 14 16:28:46 I'm trying to convert my bean/pojo to org.json.JSONObject, like so: JSONObject jsonObj = new JSONObject(bean); but it's not compiling 'Cannot resolve constructor for Bean. Now I've read the documentation and it says that my bean should have public getters, and it does. It's a bunch of private fields that are encapsulated Oct 14 16:29:00 If anybody can shed some light on this matter it would be much appreciated, thanks. Oct 14 16:29:26 TachyonDev: i firstly use fragment, but it is good for project only, and disaster for step and event, so i have to use separate section Oct 14 16:31:40 hmm, pop-quiz time Oct 14 16:31:45 TachyonDev: i ever thought join into one table, it seems hard Oct 14 16:31:54 putting together a shopping list for the company, what are your must-have test devices? :) Oct 14 16:32:34 Samsung devices, and a htc one Oct 14 16:32:38 Isn't that all people have anyway? Oct 14 16:32:41 Mavrik: Galaxy S2, S3, S4, S5, Note 2, 3, 4 Oct 14 16:32:45 heh yeah Oct 14 16:33:02 SimonVT, hmm, not really Oct 14 16:33:09 but yeah, some samsungs are on the list Oct 14 16:33:37 http://www.appbrain.com/stats/top-android-phones Oct 14 16:33:46 looking for outliers, like SE WT18i Oct 14 16:33:46 can someone help me with this Oct 14 16:33:47 https://gist.github.com/anonymous/0076787f7c359c55dbf6 Oct 14 16:34:05 every one is samsung in the top 10 according to appbrain Oct 14 16:34:08 having 5 Samsungs proved to be rather pointless the last run around Oct 14 16:34:32 Mavrik: whats your minimum android version requirement? Oct 14 16:34:42 15 right now Oct 14 16:35:12 ok, Galaxy Tab 3 for the Intel CPU Oct 14 16:35:56 15? lol Oct 14 16:35:58 must be nice Oct 14 16:36:35 hmm, do we have any 512MB / really lowend devices with 4.0.3 or 4.1? Maybe the Galaxy mini crap or something? Oct 14 16:37:15 that make me crazy Oct 14 16:37:47 MediaMetadataRetriever return first frame only Oct 14 16:38:26 molavy, even with time parameter? Oct 14 16:38:52 Mavrik: what? Oct 14 16:39:06 getFrameAtTime Oct 14 16:39:10 Switching over to Android Studio. Running into this error: “Error:SDK location not found. Define location with sdk.dir in the local.properties file or with an ANDROID_HOME environment variable.”. Studio put local.properties into the project automatically, with the path set to the sdk bundled with the Studio. I even downloaded the latest SDK and changed the path to that. It still doesn’t work. What am I screwing Oct 14 16:39:10 up? Oct 14 16:39:29 that code must work, but i don't know why it return only first frame Oct 14 16:39:37 TachyonDev, I haven't touched recycler view yet Oct 14 16:39:38 Mavrik: https://gist.github.com/anonymous/0076787f7c359c55dbf6 Oct 14 16:40:20 please look above link and tell me where i should changes Oct 14 16:40:50 molavy, metadata retriever takes time in microseconds Oct 14 16:40:58 you look like you're passing time in milliseconds Oct 14 16:41:13 also, it'll return the closest keyframe (position of which is video dependnat) Oct 14 16:41:40 shanen, did you run the SDK manager and actually download tools + platform? Oct 14 16:42:03 Mavrik: i=i+1000 in for loop, then getFrameAtTime(i*1000,...) Oct 14 16:42:30 Mavrik: Gah, I’m an idiot. Need. moar. coffee. Thanks sir! Oct 14 16:42:48 * genii slides everyone fresh mugs of coffee Oct 14 16:43:00 1000×1000=1000000, that is micro second Oct 14 16:43:34 molavy, so you're trying to retrieve 1000 frames in first second of video Oct 14 16:43:42 even though there might not even be two keyframes there? :) Oct 14 16:44:28 looks like go 1.4 gets some preliminary Android support Oct 14 16:44:29 no get length in milisecond then i just get each second a frame Oct 14 16:44:51 Mavrik: i just want get 1 frame from each second Oct 14 16:45:24 so fix your time calculation. Oct 14 16:45:40 g00s, in what way? shared libraries? easy bindings? Oct 14 16:46:35 whats the memory used for each pixel on the camera? isn’t it 1byte per pixel? Oct 14 16:46:47 for a camera preview Oct 14 16:47:51 Mavrik: what should i do? Oct 14 16:48:16 learn what SI prefixes mean. Fix your time parameter. Oct 14 16:48:48 in_deep_thought, of course not, that would mean only 255 total colors :P Oct 14 16:49:10 Mavrik: why is that? Oct 14 16:49:30 in_deep_thought, um, because 1 byte can only describe 255 states Oct 14 16:49:33 256, sorry. Oct 14 16:49:44 right? ;) Oct 14 16:49:59 Mavrik: of course how silly of me to forget Oct 14 16:50:07 camera preview is usually a planar YUV 4:2:2 format Oct 14 16:50:13 with average 12 bits per pixel Oct 14 16:50:33 what are you trying to do really? :) Oct 14 16:50:59 Mavrik: http://faculty.ksu.edu.sa/Kayed/PublishingImages/SI%20Prefixes.gif Oct 14 16:51:21 i can't understand you, please give me one line code Oct 14 16:51:25 Mavrik: figure out how big an uncompressed picture is taken with the camera, based on how many megapixels the camera has Oct 14 16:51:55 hi Oct 14 16:52:15 in_deep_thought, ah, yeah, 12-bits per pixel is a good rule-of-thumb for raw capture Oct 14 16:52:38 in_deep_thought, but remember, when you display it (e.g. on SurfaceView or Bitmap or whatnot) it has to be converted to ARGB which is 4 bytes per pixel Oct 14 16:52:51 (simplified, sometimes you can render YUV directly) Oct 14 16:53:20 Mavrik: ah ha. yes, good to jog my memory like so Oct 14 16:54:42 man I suck at android Oct 14 16:55:23 yeah, well anything camera related is rather complex :) Oct 14 16:55:27 not just on Android :) Oct 14 16:55:27 Mavrik: ok so if I want to show a camera preview without running into a out of memory error, should I be concered about the 12 bits per pixel or 4 bytes per pixel? Oct 14 16:55:41 in_deep_thought, honestly no Oct 14 16:55:50 in_deep_thought, especially if you just set preview directly to a Surface or TextureView Oct 14 16:55:54 why Oct 14 16:55:56 not Oct 14 16:56:22 because the camera driver will render the preview directly to rescaled memory area of the size you're showing Oct 14 16:56:24 Mavrik not really sure http://dotgo.sourcegraph.com/post/99652962343/brad-fitzpatrick-on-the-future-of-the-go-programming Oct 14 16:56:36 so i guess thats what fitzpatrick has been up to Oct 14 16:56:37 so even if you have full-screen preview it'll just be 1920x1080 on largest phones Oct 14 16:56:43 which is "only" 8MB of raw memory :) Oct 14 16:56:45 I am gettin a java.lang out of memory exception even when doing the simple preview class on googles camera tutorial: http://developer.android.com/guide/topics/media/camera.html Oct 14 16:57:04 in_deep_thought, hmm, that probably means there's an issue somewhere else Oct 14 16:57:12 you should probably pastebin the relevant preview capture code :) Oct 14 16:57:27 in_deep_thought, you can also do a heap dump and use Eclipse MAT to see what's taking up so much memory Oct 14 16:57:46 (are you keeping several raw frames aroud? are you trying to load full 12 Mpix image into memory?) Oct 14 16:58:02 Mavrik: any idea? Oct 14 16:58:25 g00s, honestly I'd still be rather they'd give us Java 8 than to throw at us another problematic bridging and glue layer Oct 14 16:58:38 Mavrik yeah, me too Oct 14 16:58:51 Go will use JNI no matter what Oct 14 16:58:56 which will make debugging anything a hell on earth Oct 14 16:59:09 Mavrik: this is the activity that uses my preview class: https://bpaste.net/show/9ef6a61e581e Oct 14 16:59:24 and honestly, I haven't really been affected by Dalvik performance for awhile, it's just the Java 6 cruft that annoys me now Oct 14 17:00:28 in_deep_thought, that code is funny, what are you trying to do? O.o Oct 14 17:00:50 you're also keeping like 3 copies of the same image in memory which is causing you problems :) Oct 14 17:02:05 I'm trying to convert my bean/pojo to org.json.JSONObject, like so: JSONObject jsonObj = new JSONObject(bean); but it's not compiling 'Cannot resolve constructor for Bean. Now I've read the documentation and it says that my bean should have public getters, and it does. It's a bunch of private fields that are encapsulated Oct 14 17:03:40 sounds like you're missing a public constructor (since it's complaining about constructors), or a constructor fingerprint it accepts Oct 14 17:06:19 there is really no idea? Oct 14 17:07:39 oh FF 33 Oct 14 17:13:22 Mavrik: what should i do, i check max in debug mode and it is about 5000, Oct 14 17:14:09 Mavrik: are you there? Oct 14 17:14:33 hello Oct 14 17:23:31 Hello, i have a drag and drop process between 2 listviews, but it works only when i use the OnItemLongClickListener(then it start dragging the selected item of the listview). I cant make it works with, for example, OnItemClickListener Oct 14 17:25:50 i just want to make sure, is it possible to convert a free to paid app (and vice versa) ? Oct 14 17:26:14 can some one help about this problem: https://gist.github.com/anonymous/0076787f7c359c55dbf6 Oct 14 17:26:18 its not Oct 14 17:26:20 g00s no Oct 14 17:26:22 Last I checked (which was a while ago) once you make the app free, you can't charge for it. Oct 14 17:26:39 g00s free to iap/paid yes Oct 14 17:26:41 once you go free, you never go be Oct 14 17:26:48 :) Oct 14 17:26:59 but not paid to fee Oct 14 17:27:07 ? Oct 14 17:27:12 You can go paid to free Oct 14 17:27:19 You can't go free to paid Oct 14 17:27:26 ah k Oct 14 17:27:48 just have problems with the avd and the emulator for intel (mac), “hax not installed”. If you want to solve it, go to http://developer.android.com/tools/devices/emulator.html and follow it. Oct 14 17:29:15 For an infinite viewPager ( w/ a fixed number of pages - using FragmentPagerAdapter ) all implementations/libs/stackOverflow answers make use of A_BIG_NUMBER for getcount(). What if I check if fragment for position already exists and return it from getItem(position) ? Oct 14 17:30:05 do you mean one that would go back to the first fragments after youv'e gone through them all? Oct 14 17:30:11 yes Oct 14 17:31:11 each fragment is only created once, and its created based on index Oct 14 17:31:42 i think viewpager will freak out if you try to put two of the same item at different positions Oct 14 17:32:07 i don't understand why you'd use a viewpager for this thought, someone should create a custom viewpager ot handle this case properly Oct 14 17:32:43 Mavrik: why is it funny? it opens and the countdown begins, 3,2,1. then the captured image is shown in a new activity and the captured image is shown full screen. It seems like a standard thing to do to me but I couldn’t find any tutorials or examples of it online. Oct 14 17:33:04 in_deep_thought, yeah but I don't get what you're trying to do in that code Oct 14 17:33:10 why do you save everything into file Oct 14 17:33:13 then reload it to drawable Oct 14 17:33:17 then copy it to another bitmap Oct 14 17:33:21 and then save it again Oct 14 17:33:29 without even checking if you have right color space? :) Oct 14 17:33:39 and you keep all three copies in memory Oct 14 17:34:17 well I want to be able to add stuff to the captured image. like a cool border design. this is why its drawable Oct 14 17:34:42 so in other words its probably my camera activity that needs work, not my camera preview? Oct 14 17:34:44 but why the whole intermediate save to disk step? Oct 14 17:34:53 the capture callback yes :) Oct 14 17:35:01 JacobTabak, I can set the 3rd ( of 5 pages ) as current item and use setOffscreenPageLimit(2) to keep 2 fragments from each side Oct 14 17:36:02 in_deep_thought, take the data[] array, use RenderScript or whatever to convert it to ARGB888 and then initialize Bitmap directly from that data buffer Oct 14 17:36:14 then you can just create Canvas on that Bitmap and draw on it as much as you want Oct 14 17:38:39 Hey guys Oct 14 17:39:39 So i'm learning fragments currently and I'm a bit confused about the communication between them. If I have an activity and 2 fragments, but only one fragment needs to communicate to the other one do I need an interface for the receiving fragment? Oct 14 17:39:45 Mavrik: can you help me about this problem? Oct 14 17:40:17 Mavrik: if I just want to do the first part, without drawing on the image, the simplest possible camera app in other words, I just need to save it to a bitmap right? and then save the bitmap to a file Oct 14 17:40:17 ? Oct 14 17:40:31 WhoNeedszZz you should probably use some sort of event bus, fragments shouldn't be aware of eachothers life cycles Oct 14 17:41:10 WhoNeedszZz or you could just grab fragmentA from fragmentB and call whatever method using fragmentManager Oct 14 17:42:04 how specifically would you grab fragment a from fragment b? Oct 14 17:42:48 get the fragment manager and find by tag Oct 14 17:43:18 ah. i suppose. i'd prefer to not have any knowledge of the other fragment Oct 14 17:43:29 in_deep_thought, note that there's a difference between capturing preview and doing full-resolution image capture :) Oct 14 17:43:31 it's not … just grab instance call method Oct 14 17:43:34 :) Oct 14 17:43:39 in_deep_thought, full-resolution image capture can be done directly into JPEG file Oct 14 17:43:40 just starting with android studio, you know on eclipse if you havent imported something you can right click and it will import for you etc, how do u do that on Android Studio? Oct 14 17:43:45 StingRay_ you'd have to cast Oct 14 17:43:49 and check if it's attached Oct 14 17:43:52 all sorts of ugly stuff Oct 14 17:44:03 Mavrik: ok I just want to capture the prevew, not full resolution. Does this mean I need to use a bitmap? Oct 14 17:44:10 cast yeah.. no check really if your flow design is proper Oct 14 17:44:38 in_deep_thought, that's a strange question, you don't have to, but I have no idea how you'll compress it to JPEG or something usable otherwise :) Oct 14 17:45:28 Mavrik: so then the camera preview can be saved into a file onPictureTaken without every having to worry about bitmaps? what file type would it be saved as? Oct 14 17:45:36 in_deep_thought, your data[] array is in format set by camera parameters (see getPreviewFormat() in Parameters) Oct 14 17:45:44 and you need to get that somehow into a JPEG :) Oct 14 17:45:48 or whatever you want Oct 14 17:46:28 in_deep_thought, it's just raw YUV picture data Oct 14 17:46:33 it would be useless for anything :D Oct 14 17:49:43 Mavrik: ok so lets take a step back. I want to make the simplest possible camera app. One that shows the preview, starts a countdown and takes the image, then shows the image just taken. Where would I start? I haven’t seen the google camera tutorial reference anything you have mentioned: YUV picture data, get PreviewFormat(), ARG888 RenderScript, etc. Oct 14 17:50:59 in_deep_thought, that's basically that: http://developer.android.com/guide/topics/media/camera.html#capture-picture Oct 14 17:51:11 you setup preview and call "takePicture" on camera object to grab the picture into a JPEG file Oct 14 17:51:25 Mavrik: thanks anyway Oct 14 17:51:42 ok sorry let me explain what I'm trying to do. It is a very simple excercise to learn fragment communication. So I'm modeling a "microwave" where it has 2 fragments: 1 for the "door" (which just changes bg color to indicate "on") and 1 for the controls (a start and stop button). When I press the start button I want it to toggle the bg color of the door to yellow and put a toast message saying it is on. When you click the stop button it should Oct 14 17:51:43 toggle the door bg color to gray and a toast message saying it is off. Oct 14 17:51:52 in_deep_thought, see that you get a "jpeg" callback on the takePicture method which gives you a JPEG compressed data you can just dump directly into a file Oct 14 17:52:09 in_deep_thought, note that that grabs picture in full camera resolution, not the preview resolution Oct 14 17:52:12 but that's the simplest way :) Oct 14 17:52:25 anyone else just wanna time travel to tomorrow? Oct 14 17:52:29 (you control the resolutions used by camera via the camera parameters class before calling startPreview) Oct 14 17:52:32 The controls fragment needs to communicate to the door fragment, but the door doesn't need to communicate anything back. Oct 14 17:52:56 What's the proper way of accomplishing this? Oct 14 17:53:00 WhoNeedszZz, I suggest you grab one of the messagebus libraries (EventBug or Otto) and use that Oct 14 17:53:05 it'll make your app way cleaner and keep you sane Oct 14 17:53:41 that way you just do "EventBus.getDefault().post(TurnMicrowaveOnEvent())" in one fragment Oct 14 17:53:47 and subscribe to that event in the other framgnet Oct 14 17:54:01 and that will also absolve you of all the issues with orientation changes, state changes and other crap Oct 14 17:54:12 if you subscribe on a bus in onResume and unsubscribe in onPause Oct 14 17:54:55 Mavrik, I'm doing it without external libs Oct 14 17:55:05 Don't, it's stupid. Oct 14 17:55:17 Don't have a choice... Oct 14 17:55:24 It's for a class Oct 14 17:55:32 Well then roll your own. Oct 14 17:55:52 or just make a spaghetti where fragment calls the activity and activity then passes data on to the other fragment and back Oct 14 17:56:02 you'll just have to be careful to keep state sane Oct 14 17:56:23 I don't need it to communicate back Oct 14 17:56:27 Just need one way Oct 14 17:56:40 Controls frag -> activity -> door fragment Oct 14 17:57:09 I'm just struggling to understand how frag comm works Oct 14 17:57:32 there really isn't any communication API (that's why we use messages buses : )) Oct 14 17:57:46 you just do ((YourActivity)getActivity()).openDoor() Oct 14 17:57:54 and implement openDoor in YourActivity which passes state on Oct 14 17:58:09 I was following this: http://developer.android.com/training/basics/fragments/communicating.html Oct 14 17:58:18 Which doesn't use any message bus Oct 14 17:59:14 yes, thatćs what I said now. Oct 14 17:59:39 it causes terrible spaghetti in larger apps tho :) Oct 14 18:00:31 I'm using Android Studio and added fragments with the auto generated stubs. It has an interface, which is supposed to be implemented by the activity. My question is since I don't need the door to communicate back does that obviate the need for the inteface in the door fragment? Oct 14 18:00:45 Mavrik: thanks. is the PictureCallback called after the picture is taken. Or whenever the camera preview is running? Oct 14 18:01:20 in_deep_thought, the callbacks in takePicture call are only called once after you call that method Oct 14 18:01:42 and the exact order is in javadoc :) Oct 14 18:01:52 ah so once I call onPictureTaken Oct 14 18:02:38 the PreviewCallback is called on each preview frame though, but I strongly suggest against using it for your usecase :) Oct 14 18:02:44 gets complicated really quickly :P Oct 14 18:04:07 so takePicture, grab the JPEG from jpeg callback and dump it on the disk :) Oct 14 18:05:50 yeah i don’t need that. Im gonna try replacing my onPictureTaken with google’s PictureCallback and see what happens with the memory crash thing Oct 14 18:24:07 can some one help me with this Oct 14 18:24:08 http://stackoverflow.com/questions/26367706/mediametadataretriever-getframeattime-return-first-frame-only Oct 14 18:29:57 hi ! I got this webview, and I use "shouldInterceptRequest" and "onPageFinished" to handle let some specific webpage treated differently. But when a specific webpaged is called with a 302 redirection from a not specific webpage... shouldInterceptRequest don't detect it Oct 14 18:30:08 any idea why ? Oct 14 18:31:21 ex: http://A is a 302 redirection to http://B. shouldInterceptRequest will be triggered with "url = A", and onPageFinished will be triggered with "url = B" Oct 14 18:31:41 but I want "shouldInterceptRequest" to be triggered when redirected Oct 14 18:35:53 again, I will reiterate how much I hate wear development Oct 14 18:36:01 everything is terribly slow >_< Oct 14 18:39:48 Afzal the compile / build / test process? Oct 14 18:40:49 g00s, just the way everything is handled. Out-of-date google play services on the watch? resync all apps. Moto 360 debugging is only over bluetooth so the apps take a minute to copy over. Oct 14 18:47:24 i wonder how well the wearable stuff is selling Oct 14 18:51:03 on par with vr headset of the 90's I would expect Oct 14 18:51:25 ok, in fact "WebViewClient.shouldInterceptRequest()" isn't called on a 302 redirect... Oct 14 18:51:34 any idea how to do it anyway ? Oct 14 18:52:31 blusky not sure many care about web stuff in here…well, speaking from the point of observing people and answers etc Oct 14 18:53:01 well, how to perform an instagram oauth without it ? :D Oct 14 18:53:11 (that's what i'm trying to do :) ) Oct 14 18:53:22 they dont have api ? Oct 14 18:53:29 it's web based Oct 14 18:53:40 and it need to be performed using their web interface Oct 14 18:53:53 so figured that i'll need to do it with a webview Oct 14 18:54:10 i jsut want to put the instagram auth on the webview, and retrieve the token Oct 14 18:54:22 to send it to my app's API Oct 14 18:54:31 but the token is sent after a 302 redirect Oct 14 18:55:56 how to actionbar size in xml? Oct 14 19:04:23 JesusFreke ragel 6.9 is out, but looks like only some small changes (ocaml) Oct 14 19:11:00 it's selling pretty well considering the Android Wear app downloads Oct 14 19:18:44 is there a generic way to tell a view to update it's drawable to reflect new state (i.e. pressed, checked, etc..) Oct 14 19:19:11 selector Oct 14 19:19:54 http://developer.android.com/guide/topics/resources/drawable-resource.html#StateList Oct 14 19:20:07 gonna say selector is a tag Oct 14 19:20:25 stateListDrawabe is what you are after Oct 14 19:21:04 wow this libgdx thing aint too bad is it ? Oct 14 19:21:13 well…for a beginner to play with :) Oct 14 19:22:08 StingRay_: Not all views use state list drawables in the same way. Was hoping there was an abstract construct in place already for telling a view to update it's drawables Oct 14 19:22:34 well thats due to not all view having anything like the same states Oct 14 19:23:04 enabled != pressed != active != selected Oct 14 19:23:41 but based on state changes you should not have to call anything to update a view Oct 14 19:23:48 so thats confusing :) Oct 14 19:24:13 Question : anyone done anything in libgdx ? Oct 14 19:30:07 StingRay_: oh, thats right Oct 14 19:34:49 :/ waste 4 hours because google play services on the watch doesn't install from the device Oct 14 19:34:51 FUUUUUUUUUU Oct 14 19:35:05 3 weeks out of Android and it feels more like 3 months Oct 14 19:37:43 Hello, I have a strange issue that I have not been able to solve. My app works fine when installed on a device from eclipse or run on a VM but once uploaded to the playstore it crashes immediately as it cannot find a class in the Package. This class is in the main App package. The class with the error extrends Application. Any help would be greatly apprecieated. ERROR = http://pastebin.com/mMVE6Pr3 Oct 14 19:41:15 hi Oct 14 19:41:23 can someone help me with this: Oct 14 19:41:49 http://stackoverflow.com/questions/26367706/mediametadataretriever-getframeattime-return-first-frame-only Oct 14 19:44:31 molavy no idea but I would be worried about "(not necessarily a key frame)" Oct 14 19:44:43 implies that it would/could/should be a key frame Oct 14 19:45:28 JakeWharton: i just saq a Square commercial on TV but you weren't in it. Did you leave? Oct 14 19:45:36 i'm in the back Oct 14 19:45:43 a bit blurry Oct 14 19:45:53 why you blurry? you got ebola? Oct 14 19:47:22 is there other channel except #android-dev that can help me about this probelm Oct 14 19:47:25 problem Oct 14 19:47:35 i was just reading about that bug that shouted on the bus "i have ebola" and took a mask off Oct 14 19:47:47 err, bug -> guy Oct 14 19:48:18 http://www.washingtonpost.com/news/to-your-health/wp/2014/10/14/yelling-i-have-ebola-on-a-bus-can-get-you-arrested/ Oct 14 19:48:21 dont do that :) Oct 14 19:56:40 So there's random links on the web for the latest play services apk, but they are always on the sketchiest looking sites. Where do they get their apks from? Is there any legit Google source I can get them from, or build the latest version myself from the open source? Oct 14 19:58:48 they are closed source Oct 14 19:59:46 That's what I figured. I don't understand where all the websites are getting their apks from Oct 14 19:59:56 yiati, find one from android police Oct 14 20:00:01 They pull it from devices Oct 14 20:00:24 Ahh Oct 14 20:16:28 Hey guys.. anyone ever get 14 digits back instead of 17 for a mac address? Oct 14 20:16:37 specifically in the paired bluetooth devices.. Oct 14 20:17:40 I’m seeing a crash report that had an indexOutOfBoundsException where I take a substring of the device info.. Oct 14 20:17:46 it should be a 17 character mac address Oct 14 20:18:01 but apparently something that’s 14 characters got in there Oct 14 20:18:36 is it an extra bit? Oct 14 20:19:17 print everything first Oct 14 20:19:21 then get substring Oct 14 20:19:24 (by character, I mean that I am counting the “:”) Oct 14 20:19:39 how would that help? Oct 14 20:20:07 This app is in production, it’s the only crash like it in the reports. from a HTC One M8 and M7 Oct 14 20:22:20 parsing strings from something according to what you expect should be there is really not good Oct 14 20:23:11 where you gonna start to look now Oct 14 20:23:18 you lawst! Oct 14 20:23:54 well it’s a MAC address.. Oct 14 20:24:05 I mean where you gonna start debugging Oct 14 20:24:10 I'll tell you what Oct 14 20:24:36 push a new version that writes logs to a logfile, implement that and a button somewhere that says "Submit report" Oct 14 20:24:50 when the user presses that button, send yourself that file and look for what went wrong Oct 14 20:25:16 they’re all xx:xx:xx:xx:xx:xx Oct 14 20:25:26 http://www.hastebin.com/qohijerito.java Oct 14 20:25:29 use that Oct 14 20:26:07 oh I have one Oct 14 20:26:09 a button Oct 14 20:26:21 how are you supposed to name an interface in android land, nobody uses the "IName" Oct 14 20:26:21 every time there is a stacktrace it prints to a logfile Oct 14 20:26:35 do you just name the interface a basic name then customize the implementations? Oct 14 20:26:37 http://en.wikipedia.org/wiki/MAC_address "Another convention used by networking equipment uses three groups of four hexadecimal digits separated by dots (.) (e.g. 0123.4567.89ab )" Oct 14 20:26:41 There you go, 14 characters Oct 14 20:27:15 hey...that's not 17! Oct 14 20:27:18 and what does "GMS" stand for in google play services Oct 14 20:27:23 nice.. thanks Oct 14 20:27:28 you should've known that! Oct 14 20:27:30 google map services Oct 14 20:27:33 JacobTabak using I is like hungarian notation Oct 14 20:27:34 lol thats funny Oct 14 20:27:42 yeah, i don't do it Oct 14 20:27:46 is that common on HTC devices? Oct 14 20:27:55 or does that depend on the device that is paried Oct 14 20:27:58 paired Oct 14 20:28:05 why are you using bluetooth to begin with Oct 14 20:28:13 use bluetooth low energy, now Oct 14 20:28:16 tricknology wants pain ! Oct 14 20:28:19 peripherals Oct 14 20:28:23 old code Oct 14 20:28:32 BLE is more problematic than classic on android, actually Oct 14 20:28:41 G00s would know Oct 14 20:28:42 nah man, BLE is my bitch Oct 14 20:28:48 been working with it for 9 months now Oct 14 20:28:56 g00s meet Odaym you guys should talk, lol Oct 14 20:29:04 I'm Sicp Oct 14 20:29:09 lol, this is my registered nick Oct 14 20:29:19 the other is linked Oct 14 20:29:21 to use BLE, one does not need BLE peripherals? Oct 14 20:29:26 of course, yes Oct 14 20:29:30 well no point lol Oct 14 20:29:32 for me anyway Oct 14 20:29:35 Odaym great, quick q. i've had problems sending multiple reads / write to the gatt server so i queue them and wait for the response before sending the next. have you had that problem too ? Oct 14 20:29:37 didnt know you were working on something old, thought you're inventing something Oct 14 20:29:42 yes Oct 14 20:29:59 it will do already pending Oct 14 20:30:20 no escaping, you have to do like you do. I didn't do like you did though, I use a flag that gets set when onCharacteristicWrite is done Oct 14 20:30:29 SimonVT wtf is this? lol Oct 14 20:30:33 it would be nice if I could see what you did, or how you did the queue Oct 14 20:30:37 I didnt know how Oct 14 20:30:49 or more properly I wasnt given the time to queue, we needed to publish Oct 14 20:31:11 Odaym and that supid startLeScan shit not working with 128 bit UUIDs, crazy stuff everywhere Oct 14 20:31:22 well we work with our own devices Oct 14 20:31:24 didnt see that part Oct 14 20:31:37 so I know exactly what to look for and exactly the device name and everything Oct 14 20:31:43 but does your own device have a custom UUID for at least one service ? Oct 14 20:31:44 i had it a little easier Oct 14 20:31:50 yea, the only one Oct 14 20:32:01 I set notification on it, and write to it everytime Oct 14 20:32:13 so some writes respond, some just get written and I write more of them Oct 14 20:32:41 I used google's sample app heavily at first to verify all of this Oct 14 20:32:56 funny, i had problems with that app Oct 14 20:33:06 which exactly? Oct 14 20:33:12 there seems to be a lot of discrepency between samsung and nexus devices Oct 14 20:33:13 you're supposed to see 1 characteristic being filled Oct 14 20:33:32 tricknology: Seems like you didn't rtfm on mac addresses Oct 14 20:33:42 well no it depends on what you're connecting to, I never tried with anything else BLE, except my Flex Oct 14 20:33:43 well for one, on some devices you will receive multiple broadcast packets, on others only the first one Oct 14 20:33:44 and the app saw that Oct 14 20:34:23 so i got different results on sgs4 and nexus 7 Oct 14 20:34:26 but by far the biggest problem (and this is a known bug) is the "already has pending command) Oct 14 20:34:32 " Oct 14 20:34:44 on iOS we don't even take a pause between each command Oct 14 20:35:23 i don't pause, but i do wait Oct 14 20:35:32 same Oct 14 20:35:33 (for the pending thing to complete) Oct 14 20:35:33 wait Oct 14 20:35:42 yeah Oct 14 20:36:01 but you said you queued Oct 14 20:36:06 yes Oct 14 20:36:10 full proper queue or logically queuing them? Oct 14 20:36:16 meaning with datastructure? Oct 14 20:36:28 well atm i am using some internal Android classes Oct 14 20:36:36 like HierarchicalStateMachine Oct 14 20:36:39 is this proprietary? can I see that? Oct 14 20:36:43 advanced shit brah.. Oct 14 20:36:47 SimonVT.. yeah I guess. Nor the guy that wrote this bit Oct 14 20:36:58 Odaym if you look at almost any networking code in android, its based on that :) Oct 14 20:37:04 Wifi, Bt. etc Oct 14 20:37:22 what's your use case? Oct 14 20:37:26 why do you need to send so much Oct 14 20:38:20 I need to send every time I hear audio Oct 14 20:38:37 there isn't much sending. but i wanted to separate the component that parsed characteristics and formatted requests for writing -- with the overall BT lifecycle like service discover, connection states, etc Oct 14 20:38:57 well I do the discover and connect once the activity is up, and that's it Oct 14 20:39:10 well, scan and connect also when the device disconnects Oct 14 20:40:05 we're part of the Internet of things, though, we should be getting cooler Oct 14 20:40:53 Odaym quick q, again docs don't say ... for all the BluetoothGattCallback methods ... the first param is always BluetoothGatt Oct 14 20:41:14 can one listener get different objects for that ? Oct 14 20:41:25 (or is it always the same object) Oct 14 20:42:13 Odaym also, what is the Flex ? Oct 14 20:42:19 fitbit Oct 14 20:42:26 oh nice Oct 14 20:42:40 did you work on the official fitbitapp ? Oct 14 20:42:47 no Oct 14 20:43:19 what devices have you tested on, and which ones were the most "interesting" or troublesome ? Oct 14 20:44:17 I didn't really understand what you asked above, but well what I have in the service is the callback with all the onSomethingDones and 1 connectionStateChanged. inside the activity I have mGattUpdateReceiver which has connect, disconnect, services discovered Oct 14 20:44:37 and DATA_AVAILABLE, of course Oct 14 20:44:56 this is the core of what makes it run so well, I know when anything happens Oct 14 20:45:53 well I haven't worked with anything but the device we make, Roadie tuner Oct 14 20:46:25 its BLE chip is from NRF in Japan, they give you also sample apps that let you communicate with the device easily, but source is closed. contacted them about it and they wouldnt give Oct 14 20:46:33 so I wrote something BETTER! Oct 14 20:47:25 ah the first param is BluetoothGattCallback, yes Oct 14 20:47:33 yes of course, same, why would you want to change it Oct 14 20:48:18 from those methods, you do broadcastUpdate, you listen in the activity Oct 14 20:48:34 Odaym yeah nRF has a nice app Oct 14 20:48:43 did yu use 'master control panel' ? Oct 14 20:48:45 yea Oct 14 20:48:48 sweet Oct 14 20:48:51 that's the one I envied, so I copied it Oct 14 20:48:57 lol Oct 14 20:49:00 as a first prototype, got it to do exactly as they did Oct 14 20:49:08 and then posted on the forum that I dont need you anymore I did it Oct 14 20:49:11 you didn’t decompile the jar and see for yourself? Oct 14 20:49:17 actually, even if you use that and change some settings, a fialog pops up to alert behavior can be wonky Oct 14 20:49:23 we BOUGHT so much of that chip from them, they wouldnt release the source, imagine Oct 14 20:49:34 yea connection doesnt work everytime Oct 14 20:49:41 sometimes I go out and back in for it to connect Oct 14 20:50:01 I am experiencing this too Odaym Oct 14 20:50:03 used to know how to do that, tricknology, but now dont have patience for such process anymore Oct 14 20:50:08 Odaym oh yean, next question :D when you use connectGatt, do you use true or false for autoconnect ? Oct 14 20:50:36 false Oct 14 20:50:38 this is why I love this channel Oct 14 20:51:06 I connect(address) (from activity) everytime I discover device with device.getName() like I want Oct 14 20:51:14 so Im scanning all the time Oct 14 20:51:27 Odaym oh YEAAAH! next question: how many devices can you reliably connect to ? Oct 14 20:51:38 i think its either 4 or 7 depending on android version Oct 14 20:51:44 one at a time with my implementation Oct 14 20:51:50 you can only work with 1 Roadie Oct 14 20:51:59 ah k, i have to connect to all the things :D Oct 14 20:52:13 so i have a service with a list of clients; these clients are all HSMs Oct 14 20:52:23 and send stuff back to the UI via RxJava Oct 14 20:52:36 one thing I faced not long ago and didnt attend to, was RSSI Oct 14 20:52:37 Warning, Odaym - from my experience device.getName() does not always return a name Oct 14 20:53:05 Odaym i was wondering, if asking for the RSSI was just like any other command that needed to be queued (to avoid the already pending). Oct 14 20:53:25 depending on how many devices are paired.. Oct 14 20:53:28 well right now I rely on onLeScan :\ to give it to me Oct 14 20:53:41 so naturally there is no constant update of it Oct 14 20:53:44 you get it once Oct 14 20:54:11 in the other device tester app, I need to connect to the closest and disconnect as it goes farther away from the phone Oct 14 20:54:22 so that's something I havent figured yet Oct 14 20:54:34 would be nice if you could get signale strength.. Oct 14 20:54:51 maybe ping the device? Oct 14 20:55:18 well, if you look at the BLE protocol, the parameters for link connection timeout are customizable but we don't have access to that Oct 14 20:55:18 ..doubt that ms will be enough resolution to determine meters though Oct 14 20:55:33 it could just be that the link layer hasn't timed out to disconnect or report back up to tthe HCI Oct 14 20:55:34 what I've seen so far that this works perfectly (in a final outcome kind of way) if you have the device you're working with be loaded with all the information you'll ever need Oct 14 20:55:49 the communication protocol must know that when some characteristic is written to me, I should reply with this Oct 14 20:56:01 make the BLE be responsible for as little as possible Oct 14 20:56:34 no idea what I would've thought if the device's firmware was written also by us and fully customisable Oct 14 20:56:48 wasn't* Oct 14 20:57:04 ahhhh Oct 14 20:57:08 Odaym can you explain that again? seemed important, but i didn't follow Oct 14 20:57:17 there was something I faced with Samsung, it wouldn't connect Oct 14 20:57:35 then I made the connect happen through a Handler, and it worked just fine Oct 14 20:57:42 silly as that, but it was noticeable Oct 14 20:58:04 as in, UIHandler.sendEmptyMessage(connectBLE) Oct 14 20:58:18 weird Oct 14 20:58:23 that's the most reliable to do it, Im sure Oct 14 20:58:25 which model was that Odaym? Oct 14 20:58:28 S4 Oct 14 20:58:32 it was major Oct 14 20:58:33 which though Oct 14 20:58:36 T-Mobile? Oct 14 20:58:42 oh, well here we dont have like that Oct 14 20:58:56 the store buys phones and they all just say..Samsung Oct 14 20:58:59 aah, I forget #android-dev is international sometimes Oct 14 20:59:04 there's no data plan and stuff Oct 14 20:59:05 well you can get the model number Oct 14 20:59:15 like Oct 14 20:59:20 d2spr is Sprint Galaxy S3 Oct 14 20:59:24 yea there's an excel sheet somewhere, 1 sec Oct 14 20:59:37 thanks :) Oct 14 20:59:50 Odaym i have to check again, but did you notice that callbacks weren't happening on the main thread ? Oct 14 21:00:03 (BluetoothGattCallbacks) Oct 14 21:01:02 Odaym.. do you have super.handeMessage(msg) in your UIHandler by any chance? Oct 14 21:01:11 that's not possible, do you have the intent filter for them in the activity? Oct 14 21:01:26 there's 5 of them Oct 14 21:01:45 the bluetooth SIG has some silly wrapper stuff but i didn't use it (application accelerator) Oct 14 21:01:56 i wonder if they thought, Oct 14 21:02:06 no tricknology, just switch (message.what) Oct 14 21:02:28 that is weird, then.. Oct 14 21:02:41 why? Oct 14 21:04:16 that S4, when you do android.os.Build.MODEL it returns exactly Samsung Galaxy S4 Oct 14 21:04:47 google have a list Oct 14 21:05:31 https://storage.googleapis.com/support-kms-prod/E11293177BF13755A0AFDD6B02245D516890 Oct 14 21:06:01 yea no that's not it Oct 14 21:06:37 it's completely changed, it wasn't like this before Oct 14 21:06:53 maybe it's just this shitty Pocket client Oct 14 21:07:02 Devices are listed in the following format - Marketing Name (Build.os.DEVICE/Build.os.MODEL). Oct 14 21:08:50 what I meant about making BLE responsible for as little as possible was that for example, the device I work with has a BLE and the communication protocol (dont know how its written exactly) but I do know that when I write characteristic with XYZ bytes, it knows to respond Oct 14 21:09:12 before, I didnt know this, and used to write to characteristic, wait for it to get populated, then write that I wanted to read that characterstic Oct 14 21:09:29 turns out that it already replies, I just need to write (with writeDescriptor) to setNotification on it Oct 14 21:09:41 and I would get the answer back automatically Oct 14 21:09:51 with ACTION_DATA_AVAILABLE, an array of 20 bytes Oct 14 21:10:42 I have googled high and low and can't seem to find a way how to style the checkbox and radiobutton of a submenu in an ActionBar. Does anyone here have any input? http://i.imgur.com/PXG6Rxc.png Oct 14 21:10:53 i'm going to be excited when BLE is supported Oct 14 21:10:56 BLE 4.1 Oct 14 21:11:06 that pending shit needs to be resolved Oct 14 21:11:13 have you seen the actual code that causes it? :P Oct 14 21:11:20 4.1 brings connection oriented channels, which is very nice Oct 14 21:11:23 Are we allowed to utilize little bits of code from the examples provided on the developer website in properitary apps? Oct 14 21:12:09 Odaym also L lets you put the phone in peripheral mode Oct 14 21:12:17 yea you cant now, right Oct 14 21:12:48 but i wish there was some bulletproof way to ask the HW, how many connections can I make? just letting the user attempt but then fail once limit is reached is lame Oct 14 21:12:51 I only heard about the 2 modes at the very beginning when I was trying to understand what it was Oct 14 21:13:07 the issue now is that the internet is bare about this topic Oct 14 21:13:12 there's nothing out there, only problems Oct 14 21:13:45 i got the TI SensorTag app on my device, it crashes right away Oct 14 21:13:51 (on attempting to connect) Oct 14 21:14:01 why? Oct 14 21:14:06 no idea Oct 14 21:14:14 the crash what's it say I mean Oct 14 21:14:26 i have to see if the source code is available Oct 14 21:14:52 i can't believe TI is putting lamesauce 8051 controllers on their CC2541s Oct 14 21:15:03 https://code.google.com/p/android/issues/detail?id=62833 Oct 14 21:15:36 so the command you send is just dumped Oct 14 21:15:55 hey guys. i use the Maps API. how do i get animateCamera to call back even if the camera's where i want it? since i use left padding on the map, marker.getPosition() isn't the same as map.getCameraPosition(). Oct 14 21:15:58 "If previous writeCharacteristic has not finished, it will ignore (simply discard without callback) the new one." Oct 14 21:16:04 yep Oct 14 21:16:09 as if nobody's business Oct 14 21:16:19 there should be a way to ask if anything is pending Oct 14 21:16:25 I know! Oct 14 21:16:26 otherwise, you have to maintain and sync this state Oct 14 21:16:35 if you fuck up your counter or flag, its all shit Oct 14 21:16:43 and you explicitly get a log when you disconnect, saying "device disconnected, clear_pending_flag" Oct 14 21:16:50 so the flag is there Oct 14 21:17:10 Odaym the guy that led the BLE team left a while back Oct 14 21:17:14 (at google) Oct 14 21:17:24 how did you know that? Oct 14 21:17:26 lol Oct 14 21:17:49 his linkedIn :D Oct 14 21:17:54 ohh Oct 14 21:18:16 anyone here who still works on android team? Oct 14 21:18:23 doesnt Simon? Oct 14 21:19:16 there should be someone, this is a live version of Stackoverflow, god damn it Oct 14 21:19:22 should be getting a lot of attention Oct 14 21:19:37 i'll flag it too Oct 14 21:20:15 the thing about bluedroid, is when they introduced it in 4.2, it fucked up bluetooth classic as well. we needed firmware updates Oct 14 21:20:24 (for the bluegiga module) Oct 14 21:21:01 well good night for me Oct 14 21:21:05 talk tomorrow Oct 14 21:21:10 gnight Oct 14 21:22:41 if i was doing this for iOS i'd be down by now Oct 14 21:22:45 *done Oct 14 21:29:36 Hei guys, I want a tutorial on how to build an android app, but one more dealing with actually importing and handling 3d objectss. Know of any? Oct 14 21:29:49 besides compatibility issues should an app that was working with v4 fragments work with non-v4 fragments? I am running into an issue, I has to switch from SupportMapFragment to MapFragment, I was am using a Fragment to inflate a view, then using FragmentManager to replace with a MapFragment. However, I am getting an error "No view found for id 0x7f070059 Oct 14 21:29:49 (org.openpilot.androidgcs:id/map_fragment_container) for fragment GoogleMapFragment{43269228 #0 id=0x7f070059}" Oct 14 21:30:19 ssarah: http://www.learnopengles.com/android-lesson-one-getting-started/ Oct 14 21:33:02 hm, tomorrow a big day ? http://www.androidpolice.com/2014/10/14/google-posts-sweeeeet-video-teasing-androids-l-release-again/ Oct 14 21:34:20 lol that video is cute Oct 14 21:35:09 is there a way to modify the layout of a soft-keyboard dynamically? Oct 14 21:36:50 Hi! On my 2012 Sony Xperia L with Android 4.2 and the latest Chrome, why do YouTube videos embedded with jQuery Embedly play back so slow? I'm getting like 1 frame per second: http://jsbin.com/fecum/1/ Oct 14 21:36:54 If I, on the same device, go straight to https://youtube.com/watch?v=R7T_2RVQBv0 however, playback is perfectly normal. Why is this? Oct 14 21:47:57 * bankai_au wonders if anyone is actively using kotlin :/ Oct 14 21:56:32 bankai_au i also wonder about Ceylon Oct 14 21:56:43 *googles* Oct 14 21:57:38 afaik they were planning android support Oct 14 21:57:38 "You can also run Ceylon programs on a JavaScript virtual machine" <-- that's kinda cool Oct 14 22:05:26 bankai_au new rxjava talk a little more advanced https://speakerdeck.com/daschl/reactive-programming-with-rxjava-for-efficient-data-access Oct 14 22:14:22 argh java 8 :( Oct 14 22:14:49 yeah lamdas everywhere Oct 14 22:15:13 disgusting, isn't it ? Oct 14 22:15:42 the way that code is formatted in that presentation - i think its just a matter of getting used to Oct 14 22:16:28 i meant disgusting they keep rubbing it in our faces ( that we can't use it ) Oct 14 22:17:01 most of the java world doesn't give a shit about android Oct 14 22:17:17 because its not java :D Oct 14 22:17:32 hello everyone. My company wanted me to develop an android app and without even asking my opinion they decided i'd do it in Xamarin. I'd like to know some of your opinions on Xamarin Oct 14 22:17:47 ekhidna: best not to ask that q here :) Oct 14 22:18:09 ekhidna if youguys are a c# i guess that could make sense Oct 14 22:18:25 why? is not an app dev question? Oct 14 22:18:47 i don't think many people here use it. i've never heard anyone discuss it Oct 14 22:19:06 now if you want to discuss robovm, i'm your man :) Oct 14 22:19:17 ok. thanks Oct 14 22:19:43 g00s: this is the second time i've seen netflix talk about their code. I find it quite interesting. Nice to see they are so open about it too Oct 14 22:20:13 is there a video bankai_au g00s Oct 14 22:20:20 bankai_au i think the video is coming soon Oct 14 22:20:26 bankai_au why can't you use lambdas? Oct 14 22:20:31 are you anti dingleberry too? Oct 14 22:20:47 i didn't specifically mean lambdas, i meant all the java8 fun Oct 14 22:20:49 i'm anti shit-that-can-break-my-build-system Oct 14 22:20:55 lol Oct 14 22:21:12 my shit dont build anyway Oct 14 22:21:15 cant get worse Oct 14 22:21:27 ClassNotFoundException: Didn't find class "org.mockito.internal.matchers.Any" raaaaage Oct 14 22:21:30 i hope when L is announced we get some new tooling Oct 14 22:23:30 Anyway just another question. hope it fits here. If all my app to use a specific font, let's say forex: Arial, Can I do it in XML, or must i do it programmatically. If it's an extended awnser just reply in xml or programmatically and indicate a link please Oct 14 22:24:41 my answer - theme appropriately Oct 14 22:25:43 so i should work in themess. thaks bankai Oct 14 22:26:16 bankai_au have you seen any examples of Sheduler.trampoline ? Oct 14 22:26:52 mighthave to write a custom scheduler, hmm Oct 14 22:26:58 I would like to start a service with minimal state, so I can do specialized logging from anywhere in my activity. Can the service be started before the activity? I tried starting it an instance initializer, but that failed. Oct 14 22:26:58 nup :S Oct 14 22:27:19 start it from your application context Oct 14 22:29:34 nivag, anywhere in activity, or activities? if singular, just log in your activity, why the service Oct 14 22:31:34 or why the service at all. what is specialized logging Oct 14 22:33:46 whatitis, pfn: Logging user interaction, and other stuff. Not only during development when device contacted to my dev box, but also in the field. Using internet connection not an option due to privacy & confidentiality requirements - apart from internet often not available. This is used to record eye test results. Oct 14 22:33:46 Nivag i do some specialized logging over mqtt/0mq. i put it in a service. model it like AsyncQueryService in AOSP Oct 14 22:34:57 Hi, I have a question about gridlayout, is there a way to fix the column width of the each column in the grid? Oct 14 22:34:59 which is basically putting things onto a queue, start service, services empties queue, stop until queue is filled up Oct 14 22:36:11 g00s: not familiar with mqtt/0mq & AOSP, do you have URL's I can look up? Oct 14 22:36:43 g00s: not heard of them, am quite new to Android dev Oct 14 22:41:40 anywho nivag, starting&binding to a service is pretty fast. I remember yesterday you were trying to use the instance of your service before it actually binded. did you ever change your code&it was still too slow? Oct 14 22:42:10 don't bother binding Oct 14 22:42:13 binding is unnecessary Oct 14 22:43:48 is there more to that, thatd be useful to him? Oct 14 22:43:53 whatitis: I have solved yesterdays problem. Oct 14 22:44:41 whatitis: I am probably trying to overoptimise unecessarily! I am partly persisting in an attempt to understand things better. Oct 14 22:45:06 guys as some of you already noticed i'm quite a new to Android. I must to a management app in android. It happens the company has about 30 000 clients. I will call a web service that reponds me in JSON with the info of Client Name and Telephone Number and i will map it to a class. That will be displayed in a ListView. Do you think Android will have memory to store a Custom Adapter with about 30 000 Items? Oct 14 22:45:40 pfn. I am not binding the log servioce now. Oct 14 22:46:36 pfn: I am looking at the ApplicationContec=xt option, that seems the way I should gvo for loging. Oct 14 22:47:41 * Nivag wonders if anyone else here types slower than I do, to avoid typos! :-) Oct 14 22:49:29 Nivag, 4 finger typist for 40 years ;P Oct 14 22:49:37 ekhidna: can't see that ending well for many reasons :/ perhaps you should re-think your approach Oct 14 22:50:53 ObrienDave: been using one one finger for about 45 years Oct 14 22:51:17 2 on a good day! Oct 14 22:52:14 and I thought I was slow ;P Oct 14 22:52:56 me too I already made some experiences and it surprised me bcause it repondond well utinl about 16 000. But I know this is not a good aproach. Problem is right now I can't remeber any way around. Must find a way to work better in this question Oct 14 22:53:21 even using a viewholder, merely scrolling such a list would be painful Oct 14 22:53:23 bankai_au also saw this https://speakerdeck.com/mttkay/reactive-soundcloud-tackling-complexity-in-large-applications Oct 14 22:53:41 not really sure what he's talking about in his Propeller thing though Oct 14 22:53:55 Matthias - i see him pop up in a lot of searches Oct 14 22:54:11 I get the impression storing stuff in static variables is highly NOT reccommended? Oct 14 22:54:17 slide 26 Oct 14 22:56:35 well that you guys. keep up the good work :) Oct 14 22:58:08 g00s: i did a very very very heavy WS a few years ago and knowing then what i know now about RX would have made everything SOOOO much faster Oct 14 23:12:22 <_avatar> i'm working with a large application that has over 65k symbols, even after proguard. i'm trying to use --multi-dex and https://android.googlesource.com/platform/frameworks/multidex/ to split into, then load multiple dex files at runtime. new problem: if a class that processes annotations is in a different dex file than the class that declares the annotations, i get an IllegalArgumentException at runtime. Oct 14 23:12:25 <_avatar> i've tracked the exception down to this block of native code: http://code.metager.de/source/xref/android/4.1.1/dalvik/vm/oo/Resolve.cpp#100 ... does anyone know how to work around this problem? Oct 14 23:48:28 hey Oct 14 23:48:42 is there any library or something ready with all the sharing stuff? Oct 14 23:49:04 like share to *insert hundreds of used social neworks and apps* Oct 14 23:49:24 for my actionbar Oct 14 23:52:16 http://developer.android.com/training/sharing/shareaction.html maybe something like this ? Oct 14 23:52:58 bankai_au thats it ty Oct 14 23:53:19 anyone have experience with loopj async http client? Oct 14 23:53:31 i don't think it handles chunked encoding right Oct 14 23:53:37 i'm just getting an empty response Oct 14 23:57:54 is there a way to modify the order of the keys in a soft-keyboard dynamically? Oct 15 00:21:51 a soft keyboard of your own? sure Oct 15 00:22:45 or rather, is there a way to have a completely custom layout? pfn: how? Oct 15 00:23:04 by editing the keys object programmatically? Oct 15 00:23:31 keyboard Oct 15 00:23:56 uhm, that's what I mean, but how? Oct 15 00:24:06 ah, duh, sorry, I misread Oct 15 00:25:50 well, yeah, how.. I don't see a way to do this Oct 15 00:29:43 use your own Keyboard class implementation Oct 15 00:39:31 I see now, you are right, thanks pfn Oct 15 00:49:31 anyone taking bets on what L is gonna be ? Oct 15 00:49:42 I say it's 'Lardy Cake' Oct 15 00:55:38 lol such an english dish, love it! Oct 15 01:03:25 hi Oct 15 01:03:53 L is Lollipop Oct 15 01:05:24 if it's LOLlipop they better play that song Oct 15 01:17:44 _avatar, probably by loading the annotation as part of the main app? Oct 15 01:18:23 So Google wants you to disable ssl 3.0 completely. If you do, the default email client on 4.4.x fails Oct 15 01:18:40 https://www.openssl.org/~bodo/ssl-poodle.pdf Oct 15 01:19:26 how to reproduce? Oct 15 01:19:31 by default, dovecot uses 'ssl_cipher_list = ALL:!LOW:!SSLv2:!EXP:!aNULL' if you change it to 'ssl_cipher_list = ALL:!LOW:!SSLv2:!SSLv3:!EXP:!aNULL' the default email client in android fails Oct 15 01:20:41 enable TLS Oct 15 01:20:53 where? Oct 15 01:21:09 in dovecot? Oct 15 01:21:48 perhaps you are missing the point Oct 15 01:22:12 TLS is enabled on the server side Oct 15 01:22:12 it is the client that is failing Oct 15 01:22:33 maybe Oct 15 01:22:38 don't particularly care, not a dev question Oct 15 01:22:43 ALL will allow tls, and disable !* Oct 15 01:23:29 I'd imagine no better place to discuss this than the dev channel Oct 15 01:24:03 nope Oct 15 01:24:13 dev doesn't give a shit about aosp Oct 15 01:24:17 try #android-root or xda Oct 15 01:25:18 i don't understand POODLE **** BEGIN LOGGING AT Wed Oct 15 01:30:49 2014 Oct 15 01:41:43 nexus 6 & 9 tomorrow Oct 15 01:42:00 well, pre-order Oct 15 01:42:34 i'm still betting lollipop, because of all the flavors :D (watch, car, phone, etc) Oct 15 01:43:25 LemonCurdPie Oct 15 01:43:30 just rolls off the tongue, huh ? Oct 15 01:45:05 not as well as Key Lime Pie ;P Oct 15 01:45:32 what i'm really lusting over is 12" ipad though Oct 15 01:45:44 pdf ebooks will be great Oct 15 01:47:35 What is meant by "be careful when taking user input for password" Oct 15 01:47:38 in this example Oct 15 01:47:49 http://stackoverflow.com/questions/10782187/how-to-encrypt-file-from-sd-card-using-aes-in-android Oct 15 01:56:26 any way to tell AS Observable is rx, not android.database ... and stop asking me which one ? Oct 15 02:06:50 so, I'm stuck on a bug, it seems like it should be easy enough, but being as I'm ssh'd into my box at home and working from somewhere else, I can't reliably pastebin, so I need help. I'll post the one-line error in the next message. Oct 15 02:06:52 system/core/adb/file_sync_service.c:71:13: error: too many arguments to function 'selinux_android_restorecon' Oct 15 02:07:10 anyone have any pointers I can use to try and figure this out? Oct 15 02:08:05 croey! Oct 15 02:10:05 got the relevant part in a paste (should have been easier) Oct 15 02:10:07 http://paste.ubuntu.com/8562294/ Oct 15 02:10:20 anyone have any insight on this? Oct 15 02:23:19 whats the best way to update to android studio 8.12? There isn't an installer is there? Oct 15 02:27:29 0.8.9 is the latest afaik Oct 15 02:28:33 just now checked for updates Oct 15 02:30:39 canary is 8.12 has the new avd manager was going to check it out be there isn't a downloadable installer just a zip file Oct 15 02:31:07 I guess 8.12 isn't going to be available as a patch update because of something they had to change Oct 15 02:32:52 I just didnt know how to update to 8.12 from 8.9 if I just needed to over write all the files in the install directory or what Oct 15 02:36:17 what the hell... the android apk has a bug but when i use build variant as release there is no bug Oct 15 02:44:06 wow 8.12 is a lot faster Oct 15 02:45:07 Hi Oct 15 02:45:38 Can anybody help me with this? I dont know whats wrong. It says unfortunally application closed or something Oct 15 02:45:40 http://paste.ubuntu.com/8562396/ Oct 15 02:46:30 "or something" ? Oct 15 02:46:41 drdreed yeah, their patching mechanism was busted :D Oct 15 02:46:48 Unfortunattly Matte spel has stopped Oct 15 02:46:50 so just get the full zip Oct 15 02:47:05 yeah i got the zip and just replaced the current files Oct 15 02:47:05 peyam: what does logcat say Oct 15 02:47:36 one of the plugin files was a duplicate so I just had to delete the old one and everything is working fine now Oct 15 02:48:07 bankai_au, http://paste.ubuntu.com/8562419/ Oct 15 02:48:47 peyam: sweet :) all fixed then, right ? Oct 15 02:48:59 sorry? Oct 15 02:49:17 did you read the stacktrace? Oct 15 02:50:11 bankai_au, http://pastebin.com/efAR5W9d Oct 15 02:50:16 but they are declareed Oct 15 02:56:25 Found the problem Oct 15 02:57:11 android.view.menu? Oct 15 02:58:14 yes Oct 15 02:58:17 I deleted it Oct 15 02:58:26 Now other problem occured Oct 15 02:59:36 what's the other problem? **** ENDING LOGGING AT Wed Oct 15 03:00:00 2014