**** BEGIN LOGGING AT Fri Apr 03 02:59:59 2015 Apr 03 03:04:33 TacticalJoke and pfn, thanks both for the help. The gradle thing solved it and got my project up on github without issue! Thanks again Apr 03 03:05:57 g00s, maps has location history Apr 03 03:06:06 g00s, part of gms Apr 03 03:07:06 Hillatio: The joys of Gradle. :) Apr 03 03:07:48 God, I hate my 2.4ghz wifi Apr 03 03:07:54 it's so garbage Apr 03 03:08:14 spectrum too crowded? Apr 03 03:08:31 no, my router is a pos Apr 03 03:08:56 I have a diff 2.4 on another network and it works great Apr 03 03:09:12 the problem is it's on my external net Apr 03 03:11:04 interestingly, and this makes total sense, the reception might be better if the power level is turned down because of the amplifier adding distortion Apr 03 03:11:25 less power less distortion, as long as you aren't getting run over by everyone else :) Apr 03 03:13:36 nah, signal strength on it is relatively low Apr 03 03:13:57 i'm using 2.4 because my n7 2013 is terrible with 5ghz Apr 03 03:15:10 I only use 2.4 because of the chromecast and my thermostat Apr 03 03:19:42 my macbook pro doesn't support 5ghz so i'm stuck with 2.4ghz even though there are like 25 routers in my apartment complex Apr 03 03:20:05 and every channel is crowded Apr 03 03:20:11 liuwenhao, whu? macs were some of the best for supporting 5ghz, I'm very surprised you don't have it Apr 03 03:20:35 liuwenhao, apple says all new ones as of 2012 support 5ghz Apr 03 03:20:54 mines a 2010 Apr 03 03:20:57 2.4ghz only Apr 03 03:21:01 i've tried everything to get it to work on 5ghz Apr 03 03:21:06 not that this is "apple hardware help channel", but we always loved macs when doing wifi deployments because we can shove them up onto 5ghz and forget it Apr 03 03:21:06 to no avail Apr 03 03:21:07 ah Apr 03 03:21:28 yeah. well if you don't have the radio, isn't going to happen. 2010 is fairly old at this point, guess that doesn't surprise me Apr 03 03:21:29 Will an AsyncTask automatically free itself when it reaches onPostExecution? Apr 03 03:21:44 I need to get a new laptop but no money Apr 03 03:22:44 my laptops poop too Apr 03 03:23:16 it's really slow, but I think if I ever bought an ssd for it it would help as much as buying a new laptop Apr 03 03:23:56 i put a ssd in my desktop and it made such a huge difference for coding Apr 03 03:24:01 used to take me 45 seconds to boot android studio Apr 03 03:24:06 now it's down to 3-4 seconds Apr 03 03:24:19 and build times used to take me over a minute Apr 03 03:24:38 glad i'll never have to use a 5400rpm drive again Apr 03 03:26:15 * g00s has 5400 rpm drive ;) Apr 03 03:26:33 Really? Apr 03 03:26:39 2009 MBP Apr 03 03:26:48 get a cheap ssd Apr 03 03:27:05 and put your 5400rpm in an external enclosure Apr 03 03:27:12 meh, i need a new machine. topuchpad, battery, and superdrive have / are failing Apr 03 03:27:18 only 4G emory too Apr 03 03:28:05 ummm does Android's java implementation support switch statements on Strings? Apr 03 03:28:12 in java 7 Apr 03 03:28:20 yes Apr 03 03:28:29 so I can use Java 7 with Android? It is converted to dex just fine? Apr 03 03:29:02 its in the docs ;) Apr 03 03:29:12 :D Apr 03 03:29:15 *reads* Apr 03 03:29:48 compileOptions { sourceCompatibility JavaVersion.VERSION_1_7 targetCompatibility JavaVersion.VERSION_1_7 } Apr 03 03:32:07 phix: You probably won't need that. Apr 03 03:32:21 Even java8-lookalike is converted to Dex just fine (retrolambda) Apr 03 03:33:16 TacticalJoke: I do for String's in switch statements Apr 03 03:34:27 Do you have a Java-only module and are you using JDK 8? Apr 03 03:34:34 AFAIR, that's the only time I've had to use that. Apr 03 03:34:56 Otherwise, I just use Java 7 language features. Apr 03 03:35:37 Personally, I'm tempted to use retrolambda. Professionally, I don't think my teams are ready for this, even though I've been slowly introducing Rx Apr 03 03:36:01 TacticalJoke: I am using java version 7 Apr 03 03:36:01 CedricBeust i thought there was a problem with RL and Jack/Jill ? Apr 03 03:36:04 well, 1.7 :) Apr 03 03:37:01 phix: Have you tried *not* using the compileOptions thing? Apr 03 03:37:03 It should just work. Apr 03 03:38:04 g00s: Can't comment on that, I've never tried it Apr 03 03:39:46 yo Apr 03 03:40:15 got a new machine, setting up a dev environment Apr 03 03:40:25 cool story bro ! Apr 03 03:40:45 so, Im looking at building games for the new shield console Apr 03 03:41:22 Im just not sure if I should use the nvidia TADP (tegra android dev pack) or unity Apr 03 03:42:01 do you only cre about the shield console? Apr 03 03:42:07 or do you plan to run on other android devices Apr 03 03:42:42 where are the temporary apk's saved to? Apr 03 03:43:39 no other mobile really Apr 03 03:43:52 androidtv, pc desktop, web browser Apr 03 03:44:14 Im specifically focusing on console style gameplay Apr 03 03:44:32 Storyteller, probably the only major deciding factor then is how well supported it is / is there a community. There's probably more community support for unity. tadp might be better. Apr 03 03:44:38 'if I had a million dollars' Id build games for ps4 Apr 03 03:44:39 I don't know the answer, but that's how i'd approach it Apr 03 03:44:43 fprophet: if you build a signed apk in android studio it should be under your project folder Apr 03 03:45:02 yeah, Im torn on this Apr 03 03:45:09 app->build->outputs->apk Apr 03 03:46:22 TacticalJoke: The reason I brought it up is because I was receiving a compiler error about it Apr 03 03:46:49 nvidia community is pretty silent on this... Apr 03 03:46:58 is anyone here using unity? Apr 03 03:47:12 or the TADP ? Apr 03 03:47:43 Storyteller: ummm wrong channel? Sounds like you should be in #ubuntu :) Apr 03 03:47:52 unless unity exists in android Apr 03 03:48:03 what does ubuntu have to do with anything? Apr 03 03:48:29 unity, its a game engine Apr 03 03:48:39 phix, unity is a very popular game engine on android Apr 03 03:48:42 wait, everyone thinks its unity3d Apr 03 03:48:59 because unity.com was taken, so they bought unity3d.com instead Apr 03 03:49:08 its just 'Unity' Apr 03 03:49:11 Storyteller, unity is also the window manager ubuntu uses, but, i knew what you meant. Apr 03 03:49:18 aah, ok Apr 03 03:49:22 I don't use it. but i know what you meant. Apr 03 03:49:42 anyone use the TADP or nsight from nvidia? Apr 03 03:54:02 dragorn: ah ok :) Apr 03 03:54:19 Storyteller: it is also the windows manager in ubuntu :) Apr 03 03:54:38 [22:49] Storyteller, unity is also the window manager ubuntu uses, but, i knew what you meant. Apr 03 03:54:56 yup just read that :P Apr 03 03:55:18 still, any feedback on unity and the nvidia dev tools would be welcome Apr 03 03:56:02 Storyteller, you may find more people awake during US daytime hours Apr 03 03:56:03 I have never done game programming before Apr 03 03:56:26 yeah; i've done a lot of work w/ nvidia chipsets, but never in terms of doing game dev on them Apr 03 03:56:39 It is also a public holiday :) Apr 03 03:57:54 Im just interested in the ide/dev environments myself Apr 03 03:58:08 Storyteller: as am I, let me know what you find out Apr 03 03:58:17 I may well do that Apr 03 04:02:44 Storyteller: unity is highly regarded Apr 03 04:02:59 your game will also run on desktops, etc Apr 03 04:03:09 yeah, I use it a lot, Im very familiar with it Apr 03 04:03:17 like, since 1.4-1.6 Apr 03 04:03:26 although ARC/NaCl is making me seriously think about doing a game in straight android sdk Apr 03 04:03:34 well and libgdx Apr 03 04:08:25 Storyteller: ok, sorry but what is your question then? Apr 03 04:09:12 Im just looking for feedback or the two environments for developing for android Apr 03 04:09:17 pros and cons Apr 03 04:09:43 Id gather that the TADP would be more direct for android Apr 03 04:10:04 Im interested in if there are a lot of problems exporting from unity Apr 03 04:10:09 things like that Apr 03 04:10:38 ah ok, unfortunatley i don't know anything about any of that :/ Apr 03 04:11:26 <_genuser_> myke: how's multi-identiy web browing going? Apr 03 04:11:49 _genuser_: thanks for that! tho i haven't had a chance to try it out. i noticed yahoo just deployed the same thing! Apr 03 04:12:18 <_genuser_> myke: heh, yahoo just can't let go of big ads, flash ads, and other disruptive items. Apr 03 04:12:41 and they're back to their own search engine, and it's now the firefox default as well Apr 03 04:13:05 <_genuser_> I remember when Yahoo was the "it" web portal. Apr 03 04:13:19 _genuser_: did you see my android async example? Apr 03 04:13:30 <_genuser_> I thought google was dumb looking front page with no webmaster skillz to put tons of stuff on front page. Apr 03 04:13:42 <_genuser_> myke: no, I didn't. was it something worth seeing? Apr 03 04:14:08 really? i always appreciated google's ultra-minimal page...that came out right when "portals" were a big thing, and altavista was all crapped up Apr 03 04:14:31 <_genuser_> yeah, it took me a while and then I just really liked google's minimalist approach. still do. Apr 03 04:14:39 _genuser_: no! lol. i do appreciate any feedback on it Apr 03 04:14:50 g00s said it made a lentil come out of his nose or something Apr 03 04:15:08 <_genuser_> lentils, eh? Apr 03 04:15:14 _genuser_: https://github.com/fkspro/android-tinyasync Apr 03 04:15:24 have a look and let me know what you think, thanks in advance Apr 03 04:15:41 <_genuser_> I've been wasting time trying to get people to friend me on facebook. I'm sending friend requests and waiting to see if anyone accepted it, lol. Apr 03 04:16:05 i'm *pretty* sure this doesn't leak any references, but maybe someone who actually knows java could confirm Apr 03 04:16:59 i genuinely think this may be of use to noobs Apr 03 04:17:08 just to understand the entire async web request life cycle Apr 03 04:18:16 what i really wanted: try { activity_reference.some_method(foo); } catch { log_error("Oops, this activity doesn't have some_method() ??"); } Apr 03 04:20:08 what i really *really* wanted: in onCreate() or whatever: $result = yield(getUrl("https://api.example.com/some/resource")); Apr 03 04:20:10 hehe http://imgur.com/IMPLLmS Apr 03 04:20:25 i haven't tried any coroutine libs with android yet Apr 03 04:20:33 <_genuser_> lol lot of my asyncs are done like this. class mega_turd{ ....; class worker extends AsyncTask<>{ ...; mega_turd_var=blah; } } Apr 03 04:20:53 _genuser_: and i've formalized that! Apr 03 04:20:59 <_genuser_> lol, nice! Apr 03 04:21:08 and no asynctask, because you don't need a worker queue Apr 03 04:21:20 <_genuser_> I've had a "avoid" relationship with java. even tho, that's teh programming language I've used the most in the last year doing android coding. Apr 03 04:21:47 yep i come from mostly a dynamic type background: perl, php, python ... the P culture Apr 03 04:22:03 i know just enough C++ for it to mess up my reasoning about java :( Apr 03 04:22:12 <_genuser_> I've shifted to camel-casing method names, var names (except public vars acting as properties). who has the time or energy to write getters/setters. Apr 03 04:22:27 why use a getter/setter when you can just public? Apr 03 04:23:13 <_genuser_> right. only time I use those is when a property is read-only or when it needs to be assigned a value in a certain way, so the code can format the input correctly. Apr 03 04:23:56 <_genuser_> oh, so I see, your TinyAsync is just using regular java threads... Apr 03 04:24:11 yep yep Apr 03 04:24:17 i cut out the middleman, basically Apr 03 04:24:27 <_genuser_> you just insulted java... Apr 03 04:24:37 i did, i really really did, very much so! Apr 03 04:24:38 <_genuser_> java is all about middleman, if anything. Apr 03 04:24:51 i'm pretty sure this code will keep me from ever getting hired as any kind of java developer Apr 03 04:24:54 which is ok by me Apr 03 04:25:45 <_genuser_> it's an approach. but your java code is neatly organized. you understand that it's against java mindset, etc. etc. I think that makes you a java dev. Apr 03 04:26:06 <_genuser_> I switch from c/c++ originally where method names were mostly 8 chars. so were vars. Apr 03 04:26:09 i still stand by my claim that 1) most coders (who are average by definition) are still struggling with doing everything in a pure functional or reactive style and 2) although this is gross as heck to look at it, it does not leak activity references and isn't unduly inefficient overall Apr 03 04:26:23 <_genuser_> then to VB where you would uppsercase them and have ReallyLongNamesThatCompletelyDescribedTheMethod() Apr 03 04:26:27 it lets you keep your activity's logic code with your activity's ui code Apr 03 04:26:56 ok, thanks for looking at it, and understanding what i'm doing Apr 03 04:26:57 <_genuser_> yeah, I dislike the non-activity related stuff literred in activity class. Apr 03 04:27:00 i do appreciate that Apr 03 04:27:05 <_genuser_> notification interfaces etc etc. Apr 03 04:27:12 g00s is just all lolwat and lentils Apr 03 04:27:31 whats going on? Apr 03 04:27:38 i haven't started eating my lentils yet Apr 03 04:27:40 <_genuser_> heh, I guess a community developers a certain mindset and if you digress, it will make it harder for you to connect with others Apr 03 04:27:55 _genuser_: the idea here is you would build your app bottom-up...you wouldn't call urls directly like i do, but have a lib that does...but the lib is entirely synchronous, the TinyAsync class deals with that part of it Apr 03 04:28:00 <_genuser_> g00s: we should hope not. he said you just pushed them out your nose earlier.... Apr 03 04:28:26 oh i think myke made me laugh whil ai was swallowing some lentils and then went down the wrong pipe Apr 03 04:28:29 hurt like hell Apr 03 04:28:44 i'm sure rx android is a great way but...does it really make sense for everyone to drop their actual work and learn an entirely different core architecture? Apr 03 04:28:50 <_genuser_> myke: I had a lot of custom even notification setups when I first started android. overtime, I've learned to just do it the java way even if it sounds wrong to me. Apr 03 04:28:50 eventually became airborne :) Apr 03 04:28:55 <_genuser_> g00s: lol Apr 03 04:28:56 i know them's fiting words right there Apr 03 04:29:22 <_genuser_> myke: for example, why do you have to implement an interface and pass a reference to yourself to be notified of an event. that's just dumb. Apr 03 04:29:25 g00s: sorry Apr 03 04:29:38 <_genuser_> myke: you shoudl be able to give a function reference and say, call this if raise event X. Apr 03 04:29:45 <_genuser_> just like in _any other normal language_ Apr 03 04:29:45 _genuser_: i know right? but otherwise you leak activity references >:( Apr 03 04:30:11 _genuser_: or in python or whatrever, you just call your methods direclty on a reference you get back, and catch exceptions at runtime Apr 03 04:30:16 <_genuser_> .NET, you just assign an eventhandler, but it's really just a function pointer. Apr 03 04:30:41 but can't do that in java, at least not witout creating an even worse object that uses reflection to sidestep the limitations of static typing Apr 03 04:30:49 <_genuser_> and that function runs in the instance of the object that assigned that eventhandler. and done. Apr 03 04:31:10 _genuser_: don't use .NET, is it static or dynamic typed? Apr 03 04:31:16 <_genuser_> yeah, that's why I think java should have been doing more to fix itself instead of complaining about Microsoft early 2000s. Apr 03 04:31:29 this is a classic design pattern problem, this is how i solved it, i know it's inelegant and gross - but it works! Apr 03 04:31:35 ok I am having issues with shared preferences. I have created a PreferenceFragment that reads app_preference.xml file. I then set preferences. I try to access the preferences from another activity (in same project / package) and I am not getting the stored valueds Apr 03 04:32:12 one problem with pubsub, all the ones i've seen use your object's class to figure out if you should get a msg or not Apr 03 04:32:24 but what if you only want a specific instance to get the msg? Apr 03 04:32:35 now you have to implement that inside your msg object...also gross Apr 03 04:33:34 well it's not a matter of "fixing itself", it's a direct result of the core design decisions about java's type system Apr 03 04:33:48 myke: The limitations you are talking about have nothing to do with static typing and everything to do with the language Apr 03 04:33:51 <_genuser_> myke: I think all .NET is static typed now. Apr 03 04:33:55 One thing I should add is I use a library that defines a preferences file too. However it is in a different package Apr 03 04:33:57 CedricBeust: ok Apr 03 04:33:58 any ideas? Apr 03 04:34:19 <_genuser_> back in VB, you could do var blah. or just blah=2. but with .NET that has changed in VB.NET Apr 03 04:34:39 well that's true, C++ templates let you do some things i wanted to do in this code but couldn't, not in java anyway Apr 03 04:35:10 <_genuser_> myke: for instance getting messages, I like this. ObjectReader has event, onDataRead. Object Listner1 says reader.onDataRead=method1(); and Listener2 says nothing. Apr 03 04:35:27 <_genuser_> myke: so Listener1 will get the message and Listener2 won't. Apr 03 04:36:12 <_genuser_> phix: are you able to read them in the first activity? Apr 03 04:36:29 <_genuser_> phix: or is only your preferenceFragment able to see and set the values correctly. Apr 03 04:36:34 _genuser_: ok I found the issue, constant was named differently than the key :/ Is there a way to link the XML preference key with a constant? or vice versa? Apr 03 04:36:53 I hate having to repeat myself and keep these things insync manually Apr 03 04:37:00 _genuser_: hmm gotta contemplate that Apr 03 04:37:12 most likely i just haven't sussed out the finer points of using an event bus Apr 03 04:37:14 I wish I could use C or C++ instead of java to write my app in but the NDK never seems to work :( Apr 03 04:38:24 Why would you prefer C++ over Java, except maybe familiarity? Apr 03 04:39:07 type inference Apr 03 04:39:18 <_genuser_> myke: all windows programming languages have been like that forever. VB6, Delphi, .NET (VB/c#), these are the ones I've used. Apr 03 04:39:29 for me, it's just because I am more familiar with it Apr 03 04:39:41 <_genuser_> phix: I guess you could store it's name in a key or string value in the resources. Apr 03 04:40:00 fprophet: Then spend some time learning Java, familiarity is transient, it comes and goes, but in the end, Java is just a more productive language Apr 03 04:40:55 hey, i have a cust list fragment within an activity, and I pass the activity as the context to the listadapter. now when i press on any views in the activity, the listadapter getView gets called. How can i get around this? Apr 03 04:41:14 <_genuser_> phix: I just put it in a string in strings. then I do: context.getSharedPreferences(getString(R.string.preference_file_key), Context.MODE_PRIVATE); Apr 03 04:42:18 man google services documentation is total shit Apr 03 04:42:36 wth is this https://developer.android.com/reference/com/google/android/gms/analytics/HitBuilders.HitBuilder.html#setNonInteraction(boolean) Apr 03 04:43:03 when are grown ups going to start documenting things ? Apr 03 04:44:09 _genuser_: ok that might work. Can I also use @string references for the values that a PreferenceList uses? Apr 03 04:45:26 jubjub: not sure i understand, are you registering the fragment's getView on your activity (accidentally) ? Apr 03 04:46:50 <_genuser_> phix: meaning? oh you're going to populate a PreferenceList from an "array" or strings from your strings resource? Apr 03 04:47:09 myke: i think so. If i click a button in the activity it calls the listadapters getView. Apr 03 04:47:21 ok, next question, I have a custom widget that uses the following xml schema: xmlns:app="http://schemas.android.com/apk/res-auto". I would like to add a style in styles.xml that refers to the app schema. Is this possible? if so, how do I add the schema into the file? Apr 03 04:47:30 jubjub: can you pastebin or whatever your code? Apr 03 04:47:35 _genuser_: yes, unless that is a bad idea? Apr 03 04:48:51 myke: here is the code for the listfragment and adapter http://pastebin.com/QGsmpCCB Apr 03 04:49:28 <_genuser_> phix: no, it should actually be a good idea. you can declare a string array in your resources. and then in PreferenceList use that. Apr 03 04:50:07 <_genuser_> phix: do populate a PreferenceList you'll need to extend a BaseAdapter. this BaseAdapter could initialize it's data from the string array in the constructor. Apr 03 04:50:42 _genuser_: That sounds like a good idea Apr 03 04:51:46 <_genuser_> I have a generic SelectionAdapter extends BaseAdapter. in constructor, public SelectionAdapter(Context Ctx, String[] value); I just pass it the list of strings to use and the context to use. :) Apr 03 04:51:56 _genuser_: How are you at styles? I am using a library that provides custom widgets (for changing the font and specifying it via layout xmls). I would like to create a style for these custom widgets and define the font Apr 03 04:52:16 the font is in a different XML schema that I cant seem to access in the styles.xml file Apr 03 04:52:25 <_genuser_> phix: this allows me to re-use the same adapter for all my list based selections in preferences dialog. Apr 03 04:53:14 does everyone use android studio? Apr 03 04:53:27 sohrab: yes, except for the people that still use eclipse :) Apr 03 04:53:28 i'm just getting started... what are some good forums for android dev? Apr 03 04:53:52 phix, i see, thanks Apr 03 04:54:06 Google Apr 03 04:54:18 :) Apr 03 04:54:21 <_genuser_> phix: UI, is I think the weak point of many a developer. I don't know enough about the UI xml layours and schemas to offer any "valuable" comments. :) Apr 03 04:54:40 i'm doing the sample app at developer.android.com. is that a good place to start? Apr 03 04:54:45 <_genuser_> sohrab: I still use eclipse. but remember ADT plugin doesn't support anything higher than API level 18. so basically you have to move to AS. Apr 03 04:55:12 jubjub: sorry i'll check in a minute Apr 03 04:55:20 <_genuser_> sohrab: pick any place and follow thru some lessons. then write an app that does something. complete it with all necessary elements. it will force you to learn stuff. Apr 03 04:55:47 _genuser_, k, cool Apr 03 04:55:47 ^^ Apr 03 04:56:02 there's feee online classes from courser, etc Apr 03 04:56:07 is wearable development significantly different than regular android dev? Apr 03 04:56:32 ok so any one here awesome at UI xml layouts and styles?? :) Apr 03 04:56:42 myke, do you recommend any one in particular? Apr 03 04:56:46 myke: np :) Apr 03 04:58:01 sohrab: cousera has a 6-part soup to nuts one Apr 03 04:58:07 try that, it's free Apr 03 04:58:11 i'm doing it Apr 03 04:59:54 myke: https://www.coursera.org/course/androidpart1 ? Apr 03 05:02:17 -- sohrab: is wearable development significantly different than regular android dev? Apr 03 05:02:24 you mean poor docs? no Apr 03 05:02:43 sohrab: yes, all 6 courses are linked at https://www.coursera.org/specialization/mobilecloudcomputing2/36?utm_medium=courseDescripTop Apr 03 05:02:59 i read last week the tesla guys complained the docs were bad Apr 03 05:05:03 myke, cool, thanks Apr 03 05:05:33 github a bit laggy for everyone or just me? Apr 03 05:05:42 sohrab: the sample code is useful as well: https://github.com/aporter/coursera-android Apr 03 05:05:45 best of luck! Apr 03 05:06:05 github was under attack earlier this week, dunno if that might have something to do with it Apr 03 05:11:40 hmmmm, no one awesome? or are awesome people modest? Apr 03 05:11:59 myke: did you get a chance to take a look? Apr 03 05:13:08 jubjub: doing so right now, a bit confused actually Apr 03 05:13:29 jubjub: could you put your app on appetize.io maybe? Apr 03 05:13:46 jubjub: i'm pretty n00bish so bear with me Apr 03 05:14:57 myke: i have no idea how to use that haha. I mean i made a work around for the issue by check if the ViewHolder is null. But not sure it that's the best method. Apr 03 05:20:42 jubjub: i'm a noob but check out http://developer.android.com/reference/android/app/ListActivity.html Apr 03 05:20:51 it does not have a getView(), just a getListView() Apr 03 05:21:27 jubjub: how are you creating the fragment's instance? Apr 03 05:21:48 myke: private FragmentList fragmentList = new FragmentList(); Apr 03 05:22:04 or wait sorry that's an arrayadapter down there Apr 03 05:22:51 hmm! Apr 03 05:22:58 private static class ViewHolder{ Apr 03 05:23:02 did you really want that? Apr 03 05:23:10 i may be too tired/clueless to help Apr 03 05:23:28 but why are you using a class member instead of an instance member there? Apr 03 05:23:56 i'm pretty sure something is wronb because you set it in the constructor Apr 03 05:24:14 so the last instance will override the textview registered for all of them - is that really what you wanted/intended to happen? Apr 03 05:27:53 myke: that's not quite what the issue is, but i fixed it, so thanks :D Apr 03 05:28:15 myke: i forgot to set the Tag on the view, silly mistake. Apr 03 05:31:25 jubjub: lol ok :) Apr 03 05:59:46 groxx there ? Apr 03 05:59:52 * g00s struggling Apr 03 06:00:24 * phubarr struggles too Apr 03 06:00:38 googling "android screenviewbuilder fragment" == not encouraging Apr 03 06:04:35 liuwenhao, which city do you live in? Apr 03 06:06:29 g00s, if ga is that bad why not use other analytics? Apr 03 06:06:56 this is the first i've looked at. well, Fabric had almost no documentation at all Apr 03 06:07:04 Fabric Answers Apr 03 06:07:22 iprime what do you recommend ? Apr 03 06:11:11 can anyone help me with my nexus 5? Apr 03 06:11:46 fuck off then Apr 03 06:12:02 g00s, oh i'm just wondering if you've tried and discarded other options.. i've used different ones in different project... like Flurry, but I cannot say how good it was for those handling the analytics data Apr 03 06:14:46 iprime thanks, looking at that now Apr 03 06:15:43 woah Apr 03 06:15:50 i guess it's been a while since i've last used flurry Apr 03 06:16:01 now i'm reading yahoo bought them for $200m last july Apr 03 06:16:46 <_genuser_> why is it so hard to do quality work???? Apr 03 06:17:19 also mixpanel Apr 03 06:17:45 _genuser_ because android is too complex Apr 03 06:17:47 * g00s ducks Apr 03 06:18:10 too much work for too little reward in almost all areas Apr 03 06:19:15 Hello Apr 03 06:19:27 So should I avoid using a PreferenceActivity whenever I can? Apr 03 06:19:51 * regan shrugs Apr 03 06:19:55 just use PreferenceFragment, unless you need that dual pane headers crap Apr 03 06:20:14 which doesn't even kick in unless you're on 9" i think, and who has that :| Apr 03 06:20:24 hard to do quality work? not really Apr 03 06:20:38 I am trying to build an ndk app Apr 03 06:20:59 I am getting a warning Apr 03 06:21:01 WARNING [Project: :app] Current NDK support is deprecated. Alternative will be provided in the future. Apr 03 06:21:06 g00s, I can't do this without the activity? http://1.bp.blogspot.com/-36qvdU6NgLo/UQeSVcEmwOI/AAAAAAAAPbc/ZuJ4nIDdkRk/s1600/Multipane.png Apr 03 06:21:31 Fragments..? Apr 03 06:21:44 by hand Apr 03 06:21:46 cbot` no, then yuo need PregetenceActivity. sorry Apr 03 06:22:02 heh, PreferenceActivity + Headers Apr 03 06:22:02 alright well who needs that dual pane header crap anyway Apr 03 06:22:36 yeah i think that is 9" landscape, not even sure if its portrait Apr 03 06:22:53 i dont need that many inches for what im doing Apr 03 06:23:06 I am attempting to follow ndk instructions and it give the warning above.. The tute is http://www.shaneenishry.com/blog/2014/08/17/ndk-with-android-studio/ Apr 03 06:23:06 4" is big enough Apr 03 06:23:08 <_genuser_> g00s: eh, I meant in general. Apr 03 06:23:11 right men? Apr 03 06:23:23 regan, just ignore the warning for now Apr 03 06:23:32 <_genuser_> g00s: I'm taking a break from my proj to quickly put donw a music idea I had and it's just sounding like crap. Apr 03 06:23:36 thanks pfn Apr 03 06:23:45 regan, there is no alternative at the moment Apr 03 06:23:47 regan fo into ctyosleep for another 2 years, maybe android will be ready for you then Apr 03 06:24:11 dont' forget to wake up though Apr 03 06:24:16 I need something to demo in a job interview on Tuesday Apr 03 06:24:27 what happens if you wake up and windows phone has 90% market haha Apr 03 06:25:03 <_genuser_> g00s: lol, that would be strange and cool all at the same time. windows phones. Apr 03 06:25:29 g00s is so very annoyingly anti android Apr 03 06:25:45 5 Windows phones apps combined has as many downloads in 1 year as just one of my iOS apps on the on the very first day Apr 03 06:25:47 he really should just leave and get in on the ground floor of Windows phone Apr 03 06:25:51 i'm just jojing, no way WP will have 90% Apr 03 06:26:20 pfn i'm waiting for it to take off, i'm sure like many other mobile devs ... including android ones Apr 03 06:26:21 regan, why demo ndk Apr 03 06:26:46 becuase employers are stupid and demanding Apr 03 06:27:09 regan i'd do iOS again too but i'm so out of the loop there. seems like competition is crazy, app store saturated, etc Apr 03 06:27:33 well android app store is saturated too these days Apr 03 06:27:37 every app store is saturated Apr 03 06:28:03 iOS 8 changed everything… constraints and the screen size of regular and compact Apr 03 06:28:17 if it's worth writing apps for, it will have apps already written Apr 03 06:29:03 Windows Phone is not saturated… I have users email saying thank god someone wrote something for it.. no apps, no users, no apps, no users, rinse and repeat Apr 03 06:29:20 regan is xcode behaving well these days ? Apr 03 06:29:36 xcode crashes like a mother Apr 03 06:29:41 haha Apr 03 06:29:51 and they’ve put in secret menus Apr 03 06:30:26 was watching wwdc and I was like… “how is anybody supposed to find that..?” Apr 03 06:30:30 regan: in what situation does xcode crash for you? Apr 03 06:30:34 i read some blogs recently bemoaning the dip in apple software quality, specifically OS X Apr 03 06:30:55 regan, have you tried appcode? Apr 03 06:31:12 I have a simple GridView with a custom BaseAdapter. I am using the basic ViewHolder pattern. convertView is often null when scrolling down. Why would convertView keep being null when I inflated the layout already? Apr 03 06:31:29 xcode crashes often under many conditions I’ve got 20 of my own apps in the app store and I’ve worked on half a dozen of other peoples apps and it seems to crash more often with every release Apr 03 06:31:29 where should i create and set periodic sync adapter ? eg application class or my dashboard or starter activity ? Apr 03 06:31:47 xcode 2 or 3 when interface builder was separate was pretty solid Apr 03 06:32:44 I have a build error when attempting to compile code under the NDK in Android Studio Apr 03 06:33:33 It seems like it is not getting the filename passed on the command line to the compiler Apr 03 06:33:36 http://pastebin.com/XLBHFGyb Apr 03 06:33:48 <_genuser_> g00s: quite racist of you .... Apr 03 06:33:55 <_genuser_> g00s: to say WP will never have 90% market Apr 03 06:34:49 it doesn't look good Apr 03 06:35:17 no users, no apps, no users, no apps … forever and ever amen Apr 03 06:36:28 <_genuser_> g00s: look as in UI? or as in it doesn't look good for its establishement Apr 03 06:36:33 iprime: most employers demand xcode and thus I stick to what is in use… I don’t use Xamerine or whatever its called either.. Apr 03 06:36:39 _genuser_ establishment Apr 03 06:37:06 regan, it's not xamarin related, appcode would just be an aid, it still uses xcode Apr 03 06:37:14 <_genuser_> g00s: having done more android, I don't want that effort to go to waste. But having coded on windows forever, I really do like the windows way of doing things. Apr 03 06:37:51 <_genuser_> g00s: and I'm making an assumption here that WP coding would be similar. if it's much too different, then I suppose I don't want to learn another skill just for the sake of it. Apr 03 06:37:58 _genuser_ i think they have more experience building a platform that isn't sitting on a server Apr 03 06:38:13 guys Apr 03 06:38:36 <_genuser_> g00s: I'm split right now between wanting google docs type stuff or microsoft word, which is a powerhouse and can put out amazing quality stuff. Apr 03 06:38:46 <_genuser_> SaaS or Saa(desktop application) Apr 03 06:38:49 g00s: That doesn't exactly equate to "They have experience in mobile operating systems"... Apr 03 06:39:20 <_genuser_> CedricBeust: they may have consistently missed the mark. but their goal is always noble. Apr 03 06:39:43 They want to write awesome products, like... everybody else? Apr 03 06:39:54 <_genuser_> they wanted to make mobile computing about productivity. it wasn't until iphone/instagram and facebook became popular with pre-teen girls who were addicted to it, taht mobile computing took off. Apr 03 06:40:22 <_genuser_> the HP tablets were 10years ahead of the ipad. however, they allowed you actually do work. so they never took off. Apr 03 06:40:34 <_genuser_> now ipads allow you to look hipster and check your facebook and everyone has one. Apr 03 06:40:47 <_genuser_> *hp tablets with winxp. Apr 03 06:41:00 ndk does not seem to pick up file names… http://pastebin.com/gMgeL1JL Apr 03 06:42:09 i have a weird problem getContentService(). methods red and says cannot be resolved, Apr 03 06:42:10 http://postimg.org/image/3sqm4ao9f/ Apr 03 06:45:44 Actually, ignore that, it is bulding the C++ code. Apr 03 06:45:49 I’ll shut up now. Apr 03 06:47:49 _genuser_ regan i think things could change with windows 10; if there is demand for a work /productivity experience across form factors MS should be in a good position Apr 03 06:48:11 Nobody is going to make a dent in the mobile OS for a solid decade Apr 03 06:48:53 <_genuser_> g00s: hard to be super productive on phones tbh. can't really do much other than "like" stuff. Apr 03 06:49:06 <_genuser_> but surface pro I find to be the best of tablet/laptop/desktop worlds. Apr 03 06:49:32 I have a friend who is a die hard MSFT fan boy who worked for MSFT sponsored companies and he is trying to put on a brave face but even he admits things are not hopeful for them Apr 03 06:49:39 <_genuser_> only if they address the heating issue, it could be pretty nice little thing. Apr 03 06:50:06 Hey the Raspberry PI can run Windows 10 according to him… its going to be a special cut down version. Apr 03 06:50:10 _genuser_ it would be a bit easier for them to address that in the pro when they upgrade to broadwell Apr 03 06:50:21 sci-fic: The object "getContentService()" is returning doesn't have the method getIsSyncable(Acount, String). Can't put it any plainer than what the IDE already said. Apr 03 06:50:43 <_genuser_> ultimately entire civilizations die off. microsoft is just a corporation. they could just go away after a whiel, or they could switch gears and go into other areas... Apr 03 06:51:03 they have enough money and r&d going to last a while i suppose Apr 03 06:51:08 surf2b1:almost 30 methods and interfaces like this Apr 03 06:51:17 but when i set periodic sync it works Apr 03 06:51:19 thats weird Apr 03 06:51:21 <_genuser_> g00s: I have the pro and I really like it. mostly use it for drawing. but I can see how it can be used for all office needs. including coding. Apr 03 06:51:31 sci-fic: Make sure your API level matches when that method was introduced Apr 03 06:51:42 _genuser_ john skeet uses it for coding Apr 03 06:51:50 <_genuser_> g00s: provide employees with a surface pro3, and a dock. they bring it in dock it and bam just like the laptops it becomes desktop. Apr 03 06:51:55 <_genuser_> g00s: who's john skeet? Apr 03 06:52:09 i always see him answering stuff on SO :D Apr 03 06:52:17 <_genuser_> g00s: oh. Apr 03 06:52:35 <_genuser_> I do'nt have a dock setup and my desktop does more powerful stuff with a dual monitor, so I just use tablet for digital drawing. Apr 03 06:52:45 <_genuser_> ultimately for music, I'd like. Apr 03 06:52:54 <_genuser_> but SP3 can drive dual monitor setup easily. Apr 03 06:52:59 talk about rep on SO: http://stackoverflow.com/users/22656/jon-skeet Apr 03 06:54:09 hm, mixpanel is free for 200K events/month Apr 03 06:54:17 <_genuser_> mixpanel? Apr 03 06:54:31 oh i'm looking at analytics libraries now Apr 03 06:54:44 choose between GA, mixpanel, maybe apselar Apr 03 06:54:49 <_genuser_> oh, I was thinking of incorporating crashlytics into my app. Apr 03 06:55:02 <_genuser_> so you've some experience with researching these libs? Apr 03 06:55:15 some of these are more geared to installs & retention, other more about whats going on inside the app Apr 03 06:55:25 <_genuser_> I had originally planned on flurry. but then my brothers company uses crashlytics now. Apr 03 06:55:28 _genuser_ not really, just looking at them now Apr 03 06:55:51 yeah i have to look at flurry too Apr 03 06:56:02 <_genuser_> he said crashlytics was pretty cool. all logs from all users can be uploaded to the "dashboard". you just log in and stare at the amazing ways people break your app Apr 03 06:56:17 :D Apr 03 07:23:50 derp! Apr 03 07:24:59 * g00s launches lentil barage at myke Apr 03 07:30:31 * ivali activates shield Apr 03 07:31:05 hehe Apr 03 07:34:49 is google analytics still batching stuff in sqlite on the main thread? Apr 03 07:35:03 guess i'll have to measure with strictmode Apr 03 08:10:33 Hello. Apr 03 08:11:19 Is there something that can provide variables or key-value pairs on an application level? Apr 03 08:11:54 Or do I need to pass the data from one activity to the other each time I want to do activity-activity communication? Apr 03 08:19:31 Rashad - you can extend android.app.Application or use singletons or just use a separate class: http://stackoverflow.com/questions/1944656/android-global-variable?lq=1 Apr 03 08:20:49 Thanks. Apr 03 08:27:56 ploops Apr 03 08:46:11 I had my fragments, added to backstack, and everything was working fine. Now the back button no longer brings the backstack fragments but exits the app. I have no idea what caused the change. Any tips? Apr 03 08:52:49 my sync adapter works really weird anyone here experienced ?, i need to get a summary what happened in sync, for example if it runs really close time to each sync request, even if it updated 10 rows in db counter returns me 20 Apr 03 08:53:34 i already clear preferences of that count before every sync request and write to prefs after sync is done Apr 03 08:58:02 hello i recently switched to android studio (to try it) however my app no longer looks the same (in one of my views at least) this is using the view getting included is a custom view and has some parameters on it. however it appear that the view is NOT using the parameters when in "include" it acts as the default view without any parameters on it .when in eclipse it works as expected Apr 03 08:59:12 so many questions Apr 03 08:59:26 it's 3am here in montana...probably not a lot of US people awake, sorry Apr 03 09:01:27 xml here: http://pastebin.com/DKEzd2cJ Apr 03 09:01:56 myke: such is life nothing that can be done about that Apr 03 09:02:41 fanno: not until we get new physics! Apr 03 09:03:03 myke: hehe you might need to go to anothe runiverse for that =P Apr 03 09:03:31 maybe, we don't know that Apr 03 09:04:11 funny thing is if i do not use include and add the view directly it works as expected but i use include because the main view is always the same but only the one included changes depending on phone layout Apr 03 09:17:14 i'm still using eclipse Apr 03 09:42:04 danijoo: Apr 03 09:44:01 I have found the solution to my previous problem. I was going for API 11, so I had to use getSupportFragmentManager and had to change all my fragments to android.support.v4.app.Fragment s. Apr 03 09:50:18 sci-fic, Apr 03 09:58:58 Hi people i have a question regarding error messages Apr 03 10:01:06 Firstly i have a lot of input fields that the user is required to fill out during registration. These messages are validated during input and apear as red and a corresponding error message when the user has filled out something wrong. Apr 03 10:03:44 Then when the user presses continue (the user can choose to do so whenever he wants to) i have a method parsing this input to JSON that later will be sent to an API. The JSON parsin method will throw and exception when the user is missing input or have inputted a correct value. I then want to display an error message to the user in a dialog. conatining information regarding what went wrong. At the current stage i attach a user rea Apr 03 10:03:44 dable message to the exception message and this works fine but i can help but think that maybe an exception message shouldnt be readable for the user but instead should be something for my loggs Apr 03 10:04:09 Is user readable exception messages bad practice? Apr 03 10:08:22 i would say it is bad practice indeed Apr 03 10:09:15 Do you have a suggestion to how i should do it instead? Apr 03 10:09:24 Error codes perhaps? Apr 03 10:09:39 they appear red right? Apr 03 10:09:53 also you may disable buttons as long as something is wrong Apr 03 10:10:33 Yeah but this checking is done before the continue button is pressed. The exceptions are thrown is the user preemtively presses continue and should inform the user of where the fields are still red Apr 03 10:11:09 A problem with disabling it is that the user might not understand what is wrong and just get frustrated Apr 03 10:11:23 i really want to just display a dialog telling the user what is wrong Apr 03 10:11:45 so do it Apr 03 10:11:58 if its readable Apr 03 10:12:54 Yeah but how will i know what is wrong unless i throw an exception in the continue method with a user readable message. Since people have claimed user readable exceptions are bad practice how should i do it otherwise? Apr 03 10:13:29 so dont let the continue happen Apr 03 10:13:57 But then when should i display the dialog? Apr 03 10:14:56 My question is more about how exception handling should work when you want to display a desciptive message to the user Apr 03 10:15:14 oh Apr 03 10:15:41 duno :v Apr 03 10:17:44 Sorry if i was a bit unclear. My current aproach is i have a class called UserInputException that i throw if the input didnt pass validation and i attach a user readable message to this exception. Then when i catch this is display a dialog with the exception message. But if user readable exceptions are bad practice how should this methos change in order to work with good practice Apr 03 10:20:16 So my code works but i dont want to use things that are considered bad practice Apr 03 10:25:37 arkaros, so you mean you don't show stack traces or java related messages? Apr 03 10:25:45 if so, then you're actually good Apr 03 10:27:12 I also use this method for network exceptions too but in that case i catch the IO exception and use: throw new Exception(, ) Apr 03 10:27:49 So the old exception is still retained and wrapped in a user readable message Apr 03 10:28:29 Is this still fine? Apr 03 10:31:44 I mean, if the user doesn't get anything too technical it's probably fine (e.g. "you didn't enter a name" vs "com.something.FieldException"..) Apr 03 10:33:16 Yeah thats what im doing and im trying to wrap the old exception so i can use the getCause method if i need it later Apr 03 10:34:37 i think you're good then, maybe overthinking good practices ) Apr 03 10:37:00 Yeah perhaps Apr 03 10:37:09 Thanks for all of your help :D Apr 03 10:37:15 I am trying to work on a Android application , which needs Floating Action Bar from the Material Design , Apr 03 10:37:43 After lots of research i was able to find out Github Library for FAB Apr 03 10:38:27 now , I want to know that if i use the github library in my application , then will it be wrong if i make my application for commercial purpose? Apr 03 10:39:49 jackhum: depends on licencing of what you are using. Apr 03 10:40:09 *licence Apr 03 10:40:53 claint, : https://github.com/futuresimple/android-floating-action-button Apr 03 10:41:23 jackhum: https://github.com/futuresimple/android-floating-action-button/blob/master/LICENSE.txt Apr 03 10:41:26 claint : so you are saying that i can use code from github? and make my application commercial? Apr 03 10:41:42 jackhum: come on, you can search for word licence on the page and read it Apr 03 10:42:09 does anyone have an android 5 tablet around? Apr 03 10:42:26 the android 5 genymotion emulator I use pretends android cannot understand 'audio/flac' Apr 03 10:42:27 which licences repo's are okay to copy? Apr 03 10:53:49 that's like asking, which stars in the universe are type M stars? But, to answer, most free (as in freedom) licences are okay to use. You can look up wikipedia for a comparison of free software licences. Apr 03 11:41:24 I am getting an unsatisfied link error thrown when attempting to call ndk code. Apr 03 11:47:39 I am confused about how to build ndk code in Android Studio Apr 03 11:50:43 *Bump* hello i recently switched to android studio (to try it) however my app no longer looks the same (in one of my views at least) this is using the view getting included is a custom view and has some parameters on it. however it appear that the view is NOT using the parameters when in "include" it acts as the default view without any parameters on it .when in eclipse it works as expected, xml here: http://pastebin.com/DKEzd2cJ Apr 03 12:05:12 I am getting native method not found ….. path.to.class.funcName() I see the .so files on the file system so either it is not packaging them in the apk or the export of “JNIEXPORT void JNICALL myInit( JNIEnv*, jobject) … is bad.. Apr 03 12:08:15 Is JVM unit testing working in Beta 1? (Not sure if want to spend loads of time trying it to see.) Apr 03 12:08:39 zipinfo on the apk file shows that the .so files are packaged for each abi… so I must have fork()ed up the export somehow or the call from the Java code. Apr 03 12:14:22 okay if I run strings on one of the .so files I see the function, but if I run nm on it I see nothing. Apr 03 12:26:27 hi guys Apr 03 12:27:27 I have some programming experiences with java, python and php, would like to learn android development and rapidly prototype, any good tools that one can use? Apr 03 12:27:47 is it complicated to get an app to the google play store? Apr 03 12:28:01 compared to Apple no Apr 03 12:28:10 about the same as windows phone Apr 03 12:28:44 Playsatation VITA… forget it… even if they were still accepting submissions Apr 03 12:29:02 i see Apr 03 12:29:09 Hello guys. My boss asks me to make a possibility to change application font size (enterprise) in Settings of Application. How I can implement it? :/ Apr 03 12:29:27 I don't want to make setTextSize for all text fields :( Apr 03 12:29:28 any good starting literature on building android apps with web component, user login, user profile etc. ? Apr 03 12:29:40 google it Apr 03 12:30:25 then go to the local (city or university) library Apr 03 12:31:08 The export is bad… everything is correct but the export … I am convinced of it. Apr 03 12:36:33 resolved it… don’t stay up all night programming or you make stupid errors and typos. Apr 03 12:39:36 is there any low cost and low energy way to locate a users location to show him localized information other than wlan and gps? Apr 03 12:40:16 is there a way to harness the lte cell information to localize him? Apr 03 12:41:24 since you are on both channels, CoreLocation in iOS uses a combination from gps, wlan and triangulation of cell towers, from what a coworker told me once. Apr 03 12:42:15 Not so sure about Android Apr 03 12:43:22 is there a way to indirectly add a layout parameter to a textview programmatically without explicitly using its public API method? Apr 03 12:43:41 O.o Apr 03 12:43:46 if "m" as class attribute prefix means that variable is a "member" of the class... what means the "s" prefix ? Apr 03 12:43:58 string ..? Apr 03 12:44:12 static? Apr 03 12:44:13 static? Apr 03 12:44:14 i.e. textView.setAllCaps() => textView.addLayoutParam(Styleable_..._textAllCaps, shouldAllCaps) Apr 03 12:44:26 oh yes.... seems static Apr 03 12:44:29 thanks :) Apr 03 12:44:47 where is the emoicon for shrug.? Apr 03 12:55:59 http://stackoverflow.com/questions/26838730/the-application-icon-does-not-show-on-action-bar and "getSupportActionBar().setLogo(R.drawable.ic_launcher);" does not work hmpf :/ - ( compileSdkVersion 22, buildToolsVersion "22.0.1", minSdkVersion 7, targetSdkVersion 22, compile 'com.android.support:appcompat-v7:22.0.0') Apr 03 13:00:54 any one upgraded to gradle 2.3, android plugin 1.0.0. I see that I have multiple projects and one of project is unable to find symbol in another project file though the methods are defined. Apr 03 13:01:14 same works fine on old gradle and plugin Apr 03 13:01:26 I have made migration related changes in things Apr 03 13:01:33 nop, still on officially supported 2.2.1 with 1.1.0 plugin Apr 03 13:04:58 any dev here Apr 03 13:10:16 no. Apr 03 13:16:16 lmao Apr 03 13:16:27 * sammyboy slaps Mavrik with a huge dick Apr 03 13:17:03 :' Apr 03 13:18:32 feathersanddown: Those aren't needed, IMO. Apr 03 13:19:11 prefix? m, s, ..... Apr 03 13:19:29 Yeah. Apr 03 13:19:54 Back in windows 3.11 days “sz” prefix was a string… Apr 03 13:19:55 Hello Apr 03 13:19:55 let me try 2.2.1 Apr 03 13:20:07 I've got some issues with asynctask Apr 03 13:20:08 ah, for me neither Apr 03 13:20:53 * regan doesn’t care as long as the code base is consistent Apr 03 13:21:00 but outside on internet there are tutorials and guides that use it, i don't know what means "s", just to know :) Apr 03 13:21:03 for example, I've got some error messages when I compile, as: error: illegal start of type Apr 03 13:21:15 feathersanddown: Ah, okay. Yeah, it usually means 'static'. Apr 03 13:21:20 or error: unexpected type Apr 03 13:21:35 cristian_c: It sounds like you have Java syntax errors. Apr 03 13:21:51 when I declare the class that extends the Asynctask class Apr 03 13:22:05 I'll post the code on pastebin Apr 03 13:22:48 I've done some searches in the web, I've also tried to change the syntax, but I've not found the trick yet :( Apr 03 13:24:37 TacticalJoke, http://pastebin.com/KVDMJSsC Apr 03 13:26:46 Which line has the error? Apr 03 13:26:49 TacticalJoke, http://pastebin.com/TpVURStk Apr 03 13:26:54 the build log Apr 03 13:27:06 some errors, due to asynctask Apr 03 13:27:45 you can't use primitive types in generics Apr 03 13:28:20 extends AsyncTask Apr 03 13:28:21 ? Apr 03 13:28:57 Have you tried Build > Make Project again? Those error messages are not related to the code you posted. Apr 03 13:29:17 :O Apr 03 13:29:30 fails with 2.2.1 too Apr 03 13:29:34 not sure whats the issue Apr 03 13:29:43 same works fine with older versions of plugin Apr 03 13:30:46 TacticalJoke, you are right Apr 03 13:30:48 trying 1.1.0 now Apr 03 13:30:52 line don't match Apr 03 13:30:56 I build again Apr 03 13:31:07 *lines Apr 03 13:33:03 TacticalJoke, http://pastebin.com/fFQKPebS Apr 03 13:33:43 TacticalJoke, the previous log matches code before changes Apr 03 13:34:10 same results Apr 03 13:34:14 bye Apr 03 13:35:50 cristian_c, you need to brush up your java :) Apr 03 13:36:21 Some of your "Void"s need to be "void"s. Apr 03 13:36:32 Check out the example on this page: http://developer.android.com/reference/android/os/AsyncTask.html Apr 03 13:36:36 TacticalJoke, previously they were 'void' Apr 03 13:38:49 TacticalJoke, http://pastebin.com/Zfq62syz Apr 03 13:39:10 TacticalJoke, when I wrote the code, I looked also at that Apr 03 13:39:39 'return type void is not compatible with Void' Apr 03 13:39:44 cristiano, Void for doInBackground, void for all your other async task methods... Apr 03 13:39:50 Not *all* of them need to be 'void'. Apr 03 13:40:10 'AsyncTask's generic types Apr 03 13:40:10 ' Apr 03 13:41:01 ah, ok, 'void' for onpostexecute Apr 03 13:41:36 and onProgressUpdate :) Apr 03 13:42:32 ok, 'boolean' in place of 'Boolean' Apr 03 13:42:43 cristian_c: you can't use primitive types in generics Apr 03 13:42:59 cristian_c, no, has to be Boolean Apr 03 13:43:03 what icedp said Apr 03 13:43:09 cristian_c: so inside <..> should be Void or Boolean Apr 03 13:43:36 cristian_c: but you can return void or boolean or int if you with - the will be casted Apr 03 13:43:54 s//wish Apr 03 13:43:59 actually, it is: AsyncTask Apr 03 13:44:24 then: protected void doInBackground(Void... params) Apr 03 13:44:54 and: protected void onProgressUpdate(Boolean... values) Apr 03 13:45:17 and: protected void onPostExecute(Void result) Apr 03 13:45:22 :D Apr 03 13:46:47 protected void doInBackground(Void... params) { Apr 03 13:46:49 return type void is not compatible with Void Apr 03 13:47:41 cristian_c: The compiler is literally telling you where to use "void" and where to use "Void". Apr 03 13:49:43 Convert void to Void Apr 03 13:50:52 arkaros, Some of your "Void"s need to be "void"s. Apr 03 13:50:58 :| Apr 03 13:51:06 cristian_c: Just do as the compiler says. :) Apr 03 13:51:27 Yeah pretty much Apr 03 13:51:40 compiler says: error: Unmount.AddUmountTask is not abstract and does not override abstract method doInBackground(Void...) in AsyncTask Apr 03 13:51:55 And why does it say that? Apr 03 13:52:07 It has a problem with 'void' there. Apr 03 13:52:38 I mean the return type, not the parameter type. Apr 03 13:52:41 your doInBackround should return a Void not a void Apr 03 13:53:05 Capital V is an object regular v is a primitive Apr 03 13:53:25 TacticalJoke, ok, 'void' is the returned value Apr 03 13:53:36 arkaros, ok Apr 03 13:53:36 So: AsyncTask Apr 03 13:53:42 and then: Apr 03 13:53:59 protected Void doInBackground(Void... params) Apr 03 13:54:00 arkaros, it's just so Apr 03 13:54:18 actually, it is: AsyncTask Apr 03 13:54:45 then: protected void doInBackground(Void... params) Apr 03 13:55:01 arkaros, ah, ok Apr 03 13:55:03 sorry Apr 03 13:55:15 Yeah but you should use: protected Void doInBackground(Void... params) and it will work Apr 03 13:55:17 :( Apr 03 13:55:26 I try Apr 03 13:55:44 remember to return null in the doInBackground too Apr 03 13:55:46 case closed! Apr 03 13:55:54 I remember I've already done this Apr 03 13:56:01 Otherwise it will complain about not returning anything Apr 03 13:56:09 Nice :D Apr 03 13:56:40 arkaros, there is already return null in DoBackground Apr 03 13:56:48 cristian_c: let me share with you how do I solve problems Apr 03 13:56:53 cristian_c: https://www.google.com.ua/search?hl=en&q=AsyncTask%3CVoid%2C%20Boolean%2C%20Void%3E#hl=en&q=%22AsyncTask%3CVoid%2C+Boolean%2C+Void%3E%22 Apr 03 13:57:00 Nice then it should work perfectly Apr 03 13:57:06 cristian_c: click StackOverflow link - and you're done Apr 03 13:57:11 or Apr 03 13:57:21 icedp, when I erote the code, I did many searches Apr 03 13:57:29 On a side note, I don't think 'void' is a primitive, since the JLS says that it's not a type at all. I think it's just a keyword. Apr 03 13:58:27 ah, ok Apr 03 13:58:29 cristian_c: but you haven't tried mine Apr 03 13:58:42 only DoInBackground needs Void for the returned value Apr 03 13:58:48 Capital V is an object regular v is a primitive Apr 03 13:59:13 cristian_c: I don't think 'void' is a primitive at all. Apr 03 13:59:17 'boolean' is. Apr 03 13:59:19 icedp, now, from 4/5/6 errors only to 1 error Apr 03 13:59:25 'void' is not a type. Apr 03 13:59:31 void is a void Apr 03 13:59:41 Primitives are listed here: https://docs.oracle.com/javase/tutorial/java/nutsandbolts/datatypes.html Apr 03 13:59:43 yeah, this is the mean, obviously Apr 03 14:00:37 So the problem with 'void' was that you were using a keyword where a type identifier was expected. The problem with 'boolean' was that you were using a primitive type where a reference type was expected. Apr 03 14:00:39 err nmind Apr 03 14:00:54 TacticalJoke, icedp arkaros issue seems solved: http://pastebin.com/Ln4M5b0c Apr 03 14:01:05 TacticalJoke: you seem to be right. void is a keyword, Void is a class Apr 03 14:02:08 cristian_c should java first Apr 03 14:02:18 cristian_c: Which method? Apr 03 14:02:26 Yeah, myke is right. This is basic Java stuff. ;o Apr 03 14:02:39 TacticalJoke, onpostexecute Apr 03 14:02:50 but the previous type errors are solved Apr 03 14:02:51 :) Apr 03 14:03:01 i was gonna offer my example to all noobs as a way to understand async Apr 03 14:03:10 but gotta have a better handle on java first Apr 03 14:03:27 cristian_c: Just do as the compiler says. Apr 03 14:11:55 hi Apr 03 14:12:06 yo Apr 03 14:12:18 ciao Apr 03 14:12:31 omelette du fromage Apr 03 14:12:52 Damn you made me hungry Apr 03 14:12:54 Ashiren: hein, encore un francais ici? Apr 03 14:13:12 i know french only from dexter Apr 03 14:13:22 Il y a plusieurs. Apr 03 14:13:41 TacticalJoke: i never knew , Apr 03 14:13:56 Wait, that should probably be "Ils sont plusieurs" or something. Apr 03 14:14:02 It's been years. :D Apr 03 14:14:27 "omelette AU fromage" please Apr 03 14:14:42 still hungry Apr 03 14:14:45 :p Apr 03 14:15:00 thats cool Apr 03 14:15:20 just make sure every letter is silent, et voila, parle vous francais Apr 03 14:16:23 oh, putain, Apr 03 14:16:26 i will fix this sentence too, but this is not the right place for that :) [...] et voila, tu parles comme un français Apr 03 14:16:28 * regan smacks bionic Apr 03 14:16:50 I hate when people say "wallah". Rly? Apr 03 14:16:56 TacticalJoke, BUILD SUCCESSFUL Apr 03 14:16:59 I mean in text. Apr 03 14:17:05 wololo Apr 03 14:17:32 TacticalJoke, thanks fot the help Apr 03 14:17:40 De rien. Apr 03 14:23:20 Hey guys, I just started with Android. I implemented data "layer" for my app, so now I have data from service waiting for me in array. Let's say I have users. Now I want to display all users retreived. How do you do that, I have no idea what to google. Is appropriate to create fragment for user and then somehow dynamiclly add them to XML? How should I do that Apr 03 14:24:24 ewooy, look at listview and listview activity Apr 03 14:24:38 dragorn: thanks, exactly what I needed Apr 03 14:25:04 ewooy: And adapters to bind data to a view Apr 03 14:25:18 like a listview, etc Apr 03 14:26:01 yiati, yeah, i figure that will be in any listview tutorial Apr 03 14:27:58 getView() returns null even after onCreateView(). This happens when fragments are created and dynamically added to a container from an Activity. Strange. :( Apr 03 14:28:43 isn't it possible to have a class A exist in both debug and main src dirs? Apr 03 14:29:07 alexfu, no, but you can have it exist in production and debug Apr 03 14:29:13 alexfu, or whatever your other build type is Apr 03 14:29:38 dragorn: anything but in main? Apr 03 14:29:55 alexfu, yeah, it'll always build main and get cranky that there's two copies Apr 03 14:30:52 alexfu, just make a release/java/whatever/foo.java and debug/java/whatever/foo.java and you'll be set Apr 03 14:31:45 dragorn: what if i need a test specific version of class A.. Doesn't tests use debug builds? Apr 03 14:34:08 alexfu, bad idea... Apr 03 14:34:31 think of another, testable way to solve the problem Apr 03 14:36:34 pfn: i'd like to override a dagger binding w/ a mock version of the object. been trying to figure out a way to do it but nothing so far. Apr 03 14:36:48 Problem: meh.... Instead of streaming, the web server is serving the music file as an attachment. MediaPlayer won't play it. Apr 03 14:38:38 * pfn doesn't use dagger Apr 03 14:38:39 * pfn shrugs Apr 03 14:38:45 welp Apr 03 14:38:50 alexfu, throw a different dagger setup in your testing dir. look at u2020 from jakewharton, he's got examples for how to do it Apr 03 14:39:10 but should be doable Apr 03 14:39:13 alexfu, you just throw your testing injector into your setup Apr 03 14:39:21 dragorn: was looking at that project... Apr 03 14:39:28 alexfu, pretty much the canonical reference Apr 03 14:40:14 dragorn: injecting twice doesn't seem to work Apr 03 14:40:38 alexfu, make the debug inject an override of the same provider, as I recall Apr 03 14:41:40 alexfu, put overrides = true in your @Module and provide the same @Provides as your production one Apr 03 14:41:44 dragorn: this has to happen BEFORE the graph is created, correct? Apr 03 14:42:05 the good thing about standards is there's so many of them :P Apr 03 14:42:10 alexfu, then, if you follow jakes model, in modules.java put your test stuff at the end Apr 03 14:42:23 alexfu, it'll override your production providers just fine Apr 03 14:43:10 dragorn: this is what you're referencing? https://github.com/JakeWharton/u2020/blob/master/src/internalDebug/java/com/jakewharton/u2020/DebugU2020Module.java Apr 03 14:43:21 can I add fixed footer views to a listview? Apr 03 14:43:26 sorry, i mean https://github.com/JakeWharton/u2020/blob/master/src/internalDebug/java/com/jakewharton/u2020/Modules.java Apr 03 14:43:28 as in always visible on the bottom of the screen Apr 03 14:43:35 for a material drwaer Apr 03 14:43:59 Couldn't you just place a View underneath the ListView? Apr 03 14:44:16 that'd be my fallback Apr 03 14:44:24 but I'd like to do it in the listview if possible Apr 03 14:44:31 ListView header/footer views scrolll with the other list items. They cannot be always visible. Apr 03 14:44:40 damn Apr 03 14:45:00 alexfu, i'm referring to https://github.com/JakeWharton/u2020/blob/master/src/internalDebug/java/com/jakewharton/u2020/ui/DebugUiModule.java which overrides the UiModule providers Apr 03 14:45:15 alexfu, which is then included from modules.java Apr 03 14:45:58 Darkwater, you can emulate something like it, but it's nontrivial. I've only seen emulating sticky headers, come to think of it Apr 03 14:46:12 Darkwater, the problem with a sticky footer is you might be half off the bottom of the screen Apr 03 14:47:00 yeah I'll just add some extra views Apr 03 14:47:19 Darkwater, you could slap a frameview over it, or shove it up and put a footer outside the listview and jsut deal w/ the partial last item thing Apr 03 14:47:23 really wish the developer training was updated for material already Apr 03 14:48:11 dragorn: so how does this work for testing? I can't have a Modules class in both debug and androidTest. Apr 03 14:48:34 alexfu, I don't use android test, if I'm building a test I use a flavor. I have no idea how it interacts under your setup. Apr 03 14:50:56 dragorn: how does that work? Apr 03 14:51:14 Hello guys, Is it possible to display cyrillic caracters on an android app ? Apr 03 14:51:24 alexfu, i have a testing flavor which sets the package name, turns on aggressive log output, and enables all the test cases / test options in the ui Apr 03 14:51:34 alexfu, which may, or may not, have debuggable set to true depending on what I'm doing Apr 03 14:51:38 mtR_: don't see why not Apr 03 14:52:05 mtR_, yes. Apr 03 14:55:11 https://www.youtube.com/watch?v=qGtU14vdhX0 -- man, this is a shitty bug that I can't figure out wtf is happening Apr 03 14:55:43 hrm Apr 03 14:55:47 * pfn shrugs Apr 03 14:59:25 i'm wondering how the build process different for a product flavor vs a build type Apr 03 15:00:10 both can contribut to src and res and have their own package names. Apr 03 15:17:09 I'm trying to save to /storage/emulated/0/Pictures, and it seems I do save (no errors) but I don't see it when I plug my phone into my computer. Apr 03 15:17:49 I go to "Internal Storage/Pictures" and I don't see my stuff there. Apr 03 15:20:05 you need to request a media rescan Apr 03 15:25:44 hi all, does anybody know why my Android device / app resolves Hostname using PTR and not DNS ? through PTR I'll get IPV6 address which is wrong (server doesn't respond on that address) Apr 03 15:26:07 happened just after upgrade to 4.4.2 Apr 03 15:27:15 how can i put a searchview in actionbar ? Apr 03 15:27:17 so you need the newest jdk and api version for the latest source and also older jdk and api versions if you want to build older ? Apr 03 15:27:23 igoogled but a lot of junk Apr 03 15:27:48 does anyone know if it is possible to not have the image being opened in a tab by Android studio after a seimple "screen capture" (screenshot)?? Apr 03 15:30:04 sci-fic, http://stackoverflow.com/questions/14137141/android-actionbar-customize-search-view Apr 03 15:30:30 ty Apr 03 15:40:47 dragorn: i see nothing on my activity Apr 03 16:03:24 Help! I ran the 'am' command on: http://muzso.hu/2014/05/11/manually-running-a-media-scan-on-android and now when I connect my phone to my computer, I no longer have a "Internal Storage" folder Apr 03 16:19:33 This is good: "A UI is like a joke. If you have to explain it, it sucks." Apr 03 16:19:44 i like it Apr 03 16:20:28 I got it from here but rephrased it: http://www.reddit.com/r/ProgrammerHumor/comments/318xzi/a_ui_is_like_a_joke/ Apr 03 16:26:01 hey, is there a way of adding a mapfragment programatically to a container instead of using the xml component as described on the developer site Apr 03 16:42:07 does anyone have recommendations for micro usb cables. I've gone through 2 in the past month Apr 03 16:43:21 I use adb wirelessly most of the time. Apr 03 16:43:25 Have you tried that? Apr 03 16:43:27 why do you need to be so picky on your micro usb cable ?? Apr 03 16:43:53 bpe: well when they disconnect from the adb all the time Apr 03 16:44:17 lol Apr 03 16:44:17 tacticaljoke: I had no idea that was a thing... Apr 03 16:44:23 not sure it's the cable fault Apr 03 16:44:41 but it's a good thing to verify connectic at first Apr 03 16:44:55 Details here: http://stackoverflow.com/a/14357876 Apr 03 16:45:17 Aside from that, I've have longevity with even cheap ($3) cables. Apr 03 16:45:49 Solarnum, amazon's house brand is reasonably decent. I just carry 2 or 3 incase one goes stupid. Apr 03 16:45:57 question about android studio, I have a function in my build gradle file that calls Python, I an getting an error that it can't find it. However, if I run gradlew, the function works correctly. Where does Android Studio pick up it path from? (Mac) Apr 03 16:45:58 monoprice is usually solid too Apr 03 16:46:20 bpe: Yah I was sure it was the nexus 7s connectivity at first but I tried a new cable and had no problems for like 2 weeks Apr 03 16:46:48 tacticaljoke: thanks I will look into it Apr 03 16:48:19 tacticaljoke: that was stupid easy... Apr 03 16:48:38 thankks a lot Apr 03 16:48:43 It's a good idea to use it, IMO. It's probably not good for our batteries to charge them constantly. :) Apr 03 16:49:11 Though batteries might be smart enough not to be affected by it too much. Apr 03 16:49:22 i would like a brand new gold `adb' please Apr 03 16:49:49 I'm wondering when was the last adb update... Apr 03 16:50:48 https://android.googlesource.com/platform/system/core/+log/master/adb Apr 03 16:50:55 Hi, im trying to get a elevation effect on prev21 devices. I was reading about use shadows. I have this: http://paste.ideaslabs.com/show/fEtR83h2A7 and i cant get it work :S i didnt see a shadow. Help would be appreciate :) Apr 03 16:51:04 19 hours. ;o Apr 03 16:52:29 yeah but it has not been released Apr 03 16:52:38 like https://android.googlesource.com/platform/system/core/+/50184062b87a121458fff7d8bda6656c59c71c77 i just tested it Apr 03 16:52:41 *not here* Apr 03 16:53:55 i still have to use alt+3 (num pad) on windows to have a ctrl-c like-effect (it makes a heart -> root@generic_x86:/ # ♥ ) Apr 03 16:54:02 + enter Apr 03 16:56:30 Its a good practice to use a View for simulate shadow in pre-v21 devices?? Apr 03 17:02:22 *Bump* hello i recently switched to android studio (to try it) however my app no longer looks the same (in one of my views at least) this is using the view getting included is a custom view and has some parameters on it. however it appear that the view is NOT using the parameters when in "include" it acts as the default view without any parameters on it .when in eclipse it works as expected, xml here: http://pastebin.com/DKEzd2cJ Apr 03 17:03:02 https://github.com/pfn/android-sdk-plugin/commit/166857911711605a48090954060d1aa7699c245b Apr 03 17:03:07 yay, fixed my retrolambda support Apr 03 17:08:38 Got a shitty problem going on here. Apr 03 17:08:51 I needed to punch a hole in a shape using porterduff clear Apr 03 17:09:08 it only drew black, so I turned off hardware acceleration Apr 03 17:09:30 but now, that I need to animate drawing this, it's dropping frames left and right. If I bring back the issue, it doesn't drop frames Apr 03 17:09:56 so now I'm back to square one. Is there a way to prevent PorterDuff.Mode.CLEAR from drawing black? Apr 03 17:16:22 why onStartCommand should not be overriden from an IntentService? Apr 03 17:17:06 ollien: yes Apr 03 17:17:26 Solarnum: how can I do this? Apr 03 17:17:56 ollien: try this, canvas.drawColor(0x00000000, PorterDuff.Mode.CLEAR); Apr 03 17:18:59 Solarnum: no dice. That just draws a giant black square Apr 03 17:19:27 http://i.imgur.com/EboLHMl.png Apr 03 17:19:45 ollien: it should draw transparent color Apr 03 17:19:54 icedp: correct, it should. Apr 03 17:20:09 ollien: let me look into it. I did this last year. Apr 03 17:20:25 Solarnum: thanks so much! Apr 03 17:20:40 ollien: what is your bg color? Apr 03 17:20:51 icedp: I don't believe i've set one. Apr 03 17:21:04 background color in the canvas? Apr 03 17:21:48 yes the first drawColor in onDraw Apr 03 17:22:07 icedp: ah. I didn't do one. However, when I set it to the aforementioned line, that happened. Apr 03 17:22:29 and what happens if the IntentService gets killed? as I can't handle onStartCommand and return START_NOT_STICKY or something Apr 03 17:24:37 gotta run. Got a prior engagment. Solarnum let me know if you figure it out :) thanks so much once again Apr 03 17:24:46 MyWay: https://android.googlesource.com/platform/frameworks/base.git/+/master/core/java/android/app/IntentService.java Apr 03 17:25:04 I've found the answers, by default if it gets killed it's behaviour is like START_NOT_STICKY unless you call a method on the constructor Apr 03 17:25:06 thank you JesusFreke :P Apr 03 17:25:11 MyWay: it will return START_REDELIVER_INTENT for you, if you've called setIntentRedelivery(true) Apr 03 17:25:27 couldn't find it Apr 03 17:25:59 The api has been there since api 5 Apr 03 17:26:00 Anyone got a tutorial for writing video decoder/encoder ? Apr 03 17:26:07 http://developer.android.com/reference/android/app/IntentService.html#setIntentRedelivery(boolean) Apr 03 17:26:44 and what if I don't want onHandleIntent to get queued while it's running? is there a way to do it? Apr 03 17:27:01 that... doesn't make much sense Apr 03 17:27:06 why? Apr 03 17:27:14 that's the whole point of an IntentService Apr 03 17:27:31 if you want to do something different, use Service and do whatever you want :) Apr 03 17:32:32 Is 1.2 Beta "stable"? Apr 03 17:33:09 is anything stable? :D Apr 03 17:33:30 build titles don't lie Apr 03 17:33:31 it's beta Apr 03 17:35:11 Do y'all know whether JVM unit testing is working in 1.2 Beta? I couldn't get it working in 1.2 Preview 1. Apr 03 17:35:42 it makes sense, thank you JesusFreke Apr 03 17:35:43 It just wouldn't recognise my test files as test files. Apr 03 17:39:23 just download it and find out Apr 03 17:44:26 Hello, fellas. I have a question about Eclipse and issues with an android project. It would be great if you could help me Apr 03 17:45:11 here's the screenshot of my project: http://imagizer.imageshack.com/img913/1622/b54g68.jpg Apr 03 17:45:17 app of the day! https://play.google.com/store/apps/details?id=com.cc.babyiq&hl=en Apr 03 17:45:23 as you can see, it has an error, but I can't find the source of the issue Apr 03 17:45:40 cthulchu: the source is its eclipse Apr 03 17:45:46 cthulchu: you should use android studio Apr 03 17:45:55 did you try cleaning the project? Apr 03 17:46:13 lasserix, I know I should, but I'd like to know how to fix this in Eclipse. Apr 03 17:46:30 Hi, Apr 03 17:46:31 clean the project, restart eclipse Apr 03 17:46:46 make sure your dependencies are in order Apr 03 17:47:05 what do you mean by cleaning the project? Apr 03 17:47:07 ollien: I just wondering. Why can't you draw with background color? Apr 03 17:47:07 i can't remember quite, but i remember having a problem that looked like that, and it had to do with dependencies Apr 03 17:47:19 cthulchu: you are using eclipse and you seriousily don't know clean function? Apr 03 17:47:20 hm Apr 03 17:47:22 goto project-> clean Apr 03 17:47:33 or wherver the build function is Apr 03 17:47:34 I'm trying to make an app using android studio, but sometimes, my app freezes for no appearent reason, with this being logged: Apr 03 17:47:35 WaitForGcToComplete blocked for 9.884ms for cause Background Apr 03 17:47:38 lasserix, I'm a newbie :( Apr 03 17:47:47 thanks, will try Apr 03 17:48:14 dragorn: what I ended up doing is implementing a custom test runner and overwrote newApplication to return a custom Application that adds my test modules into the modules list. not sure if this is the best way, but it works. Apr 03 17:48:19 Why are you using Eclipse? If you're a newbie, Android Studio is ideal. Apr 03 17:48:21 That line isn't logged when the app is running normally. Apr 03 17:49:26 Maybe I should also mention that it only freezes like that when I start a thread to do some calculations in the background. Apr 03 17:49:42 oh hell... Apr 03 17:50:11 I thought the only people using Eclipse for Android development now were old-timers who haven't moved to Android Studio for emotional or logistic reasons Apr 03 17:50:25 TacticalJoke: my thoughts exactly Apr 03 17:50:53 This line is also in the output: WaitForGcToComplete blocked for 10.686ms for cause HeapTrim Apr 03 17:50:54 stockholm syndrome, obviously Apr 03 17:51:04 Is that related to that other line? Apr 03 17:52:02 Ben_g: when you say some calculations... Apr 03 17:52:03 Ben_g: 10ms isn't enough to be visible Apr 03 17:52:20 e.g. at 60 fps, frames are every 16ms Apr 03 17:52:58 Well, it's a process that first does some calculations, then waits a while, then does more calculations. Apr 03 17:53:06 (glossing over details a tiny bit, of course) Apr 03 17:53:40 Ben_g: are you sure you're not blocking the main thread? :) Apr 03 17:53:54 It sounds like you aren't running the thread in the background Apr 03 17:54:36 lasserix, thank you very much. It turns out when I imported the project from a flash drive there was a mistake. Thank you! Apr 03 17:55:13 new Thread(new Runnable(){@Override public void run{/*Some code here*/}).start() Apr 03 17:55:45 that shouldn't block the main thread, does it? Apr 03 17:56:27 (Sorry if I sound like a noob. This is the first thing I do in Android) Apr 03 17:57:29 Is anyone familiar with the effects of applying an application ID in Gradle than is different from the package name in the Manifest and its effects on BroadcastReceiver names? Apr 03 17:57:56 I am trying to mimic INSTALL_REFERRER intents from Google Play using adb am. Apr 03 17:58:07 BTW: Buttons keep working, so if they're handled in the main thread, I know that only the background task freezes. Apr 03 17:58:48 What I am seeing, is that my BroadcastReceiver is not able to receive an explicit INSTALL_REFERRER intent unless the component specified is in the format “applicationID”/“packageName”.”broadcastReceiverName”, which I’m not expecting since I would assume “packageName” should actually be “applicationID”. Apr 03 17:58:54 ben_g: then it's definitely not blocking the ui thread Apr 03 17:59:17 is the ui thread seperate from the main thread? Apr 03 17:59:28 the ui thread is the main thread Apr 03 17:59:47 ben_g: no Apr 03 18:00:00 ok, then I know that only the background thread is freezing. Apr 03 18:00:09 no freezing, working Apr 03 18:00:13 s/no/not/ Apr 03 18:00:43 ben_g: didn't you say that in the thread you wait and then do more work Apr 03 18:01:17 yes, but it should show something on the screen right before each wait cycle. Apr 03 18:01:24 ahh Apr 03 18:01:37 ben_g: you need to use a handler to post the updated ui info Apr 03 18:01:53 I do use a handler for that. Apr 03 18:02:31 It's working perfectly most of the time, it just occasionally freezes and shows those lines in the console. Apr 03 18:03:40 http://novaember.com/s/686484244.png Apr 03 18:03:42 bullshit Apr 03 18:06:02 The app I'm making is a game in which players place symbols on a grid to form a certain combination. Making such a combination grants the player an extra turn. Apr 03 18:06:44 Ben_g: You might have to post code to get more feedback. Apr 03 18:06:55 The code I'm running in the background is AI code that checks what it should to to get as much turns as possible Apr 03 18:08:00 so it should always finish quickly if it can't score any points on a single turn (since it doesn't have to calculate extra turns then) Apr 03 18:08:13 And even in that situation it sometimes freezes. Apr 03 18:08:20 But I'll paste the code on pastebin. Apr 03 18:08:23 hm, is a listview even a good widget for something that will always have two static items? Apr 03 18:08:36 What are the items? Apr 03 18:08:46 settings and help & feedback Apr 03 18:08:52 it's for a drawer Apr 03 18:09:20 http://pastebin.com/jevgGttW Apr 03 18:09:40 The background task code starts at line 298. Apr 03 18:10:30 Darkwater: why can't you make those 2 views siblings to the ListView? Apr 03 18:10:53 hm, haven't thought about that actually Apr 03 18:11:16 still though, wouldn't solve my problem (probably) Apr 03 18:11:44 why not? Apr 03 18:11:52 nope Apr 03 18:12:00 my problem is that the listview isn't there Apr 03 18:12:05 or wait, I think I know the problem Apr 03 18:12:10 what? Apr 03 18:12:17 does a listview always need an adapter? Apr 03 18:12:27 Darkwater: yeah Apr 03 18:12:29 I just used addHeaderView to test it quickly Apr 03 18:12:30 ah Apr 03 18:12:40 that explains, derp Apr 03 18:12:47 It should still exist without the adapter, though. Apr 03 18:12:49 Just without items. Apr 03 18:13:23 I'm not sure how to explain that screenshot you posted. The ListView appeared to not exist at all. Apr 03 18:13:38 hm Apr 03 18:13:49 I'll try adding an adapter Apr 03 18:13:49 addFooter/addHeader must occur BEFORE setting the adapter though Apr 03 18:14:10 got it Apr 03 18:14:26 yup, view's there now Apr 03 18:14:36 widget, you strange Apr 03 18:14:55 so, is there a cleaner way than using a listview? Apr 03 18:15:37 Darkwater: if you just want a header view, then using addHeader is the way to go Apr 03 18:15:56 assuming you dont want any fancy stuff like sticky headers. Apr 03 18:16:12 there's the obvious way of just using two relativelayouts or whatever and make them clickable, but I'd like to have the ripple effect as well Apr 03 18:16:29 I just want two sticky footer items Apr 03 18:16:30 Ben_g: You're calling Bitmap.createScaledBitmaps in View.onDraw. Apr 03 18:16:32 like uhm Apr 03 18:16:37 That sounds extremely slow. Apr 03 18:16:41 i would avoid using relative layous Apr 03 18:16:46 I guess you're forgetting to set bitmapsInitialised to false. Apr 03 18:16:51 s/false/true/ Apr 03 18:17:18 http://material-design.storage.googleapis.com/publish/v_2/material_ext_publish/0Bx4BSt6jniD7WjVXWld2TS1oNlk/patterns_navdrawer_settings1.png Apr 03 18:17:21 like that Apr 03 18:17:28 settings and help/feedback are always visible on the bottom of the screen Apr 03 18:17:41 I indeed forgot to set bitmapsInitialized to true; Apr 03 18:18:39 But since it only needs to render a frame after user input or after an AI calculation, the speed isn't really important. Apr 03 18:18:54 Darkwater: use a LinearLayout w/ orientation vertical. add ListView and two buttons as children. set ListView height to 0dp and it's layout_weight to 1. button heights can just be wrap_content. Apr 03 18:19:11 But still thanks for pointing that out. Apr 03 18:20:41 Darkwater: you can add ripple effect by setting background to drawable Apr 03 18:21:52 ben_g: that's not true. In the while loop you call post.invalidate Apr 03 18:22:08 instance.invalidate(); Apr 03 18:22:10 ** Apr 03 18:22:26 ben_g: right? Apr 03 18:22:45 Yes Apr 03 18:22:50 relativelayouts are bad?? Apr 03 18:22:59 But that isn't really speed-critical. Apr 03 18:23:28 Ben_g: View.draw gets called a lot while a View is on screen. Apr 03 18:23:36 ben_g: Just pointing out that while the AI move is being calculated onDraw is being called repeatedly. Apr 03 18:23:50 anyone know if GoogleAnalytics trackers can be used from multiple threads ? Apr 03 18:23:53 not just after Apr 03 18:24:12 That's true. Apr 03 18:24:55 alexfu: don't think I can make buttons look like how I want them to, though Apr 03 18:24:58 So on android, scaling images is so slow that I should try to always preload them? Apr 03 18:25:06 ben_g: it could be a memory problem Apr 03 18:25:14 (I assume you can't put views inside buttons) Apr 03 18:25:18 since you're working with scaled bitmaps Apr 03 18:25:23 Darkwater: of course you can. set background to selectableItemBackground, add a drawableLeft Apr 03 18:25:37 add some drawablePadding Apr 03 18:26:12 hmm Apr 03 18:26:13 using a single button with a drawableLeft is much better than a LinearLayout with an ImageView and TextView. Apr 03 18:26:22 lemme try thart Apr 03 18:26:25 sounds interesting Apr 03 18:27:10 But I now only generate the scaled bitmaps on the first frame, and the freeze still occurs. Apr 03 18:27:44 ben_g: bitmaps initialized is always false also. It is set only in the field scope. Apr 03 18:27:58 such bugs Apr 03 18:28:11 ben_g: so those bitmaps are not only generated once Apr 03 18:28:26 Solarnum, I added bitmapsInitialized = true; at the end of the if. Apr 03 18:28:30 icedp I can, but porterduff clear produces all black for some reason Apr 03 18:28:47 Solarnum: you didn't happen to find anything did you? Apr 03 18:29:04 ollien: I messaged you Apr 03 18:29:12 Solarnum: didn't see that thanks Apr 03 18:29:21 ben_g: ah, in the paste bin that wasn't set Apr 03 18:29:40 ollien: no, but why do you need Mode.CLEAR at all? pass you theme's background color and draw with it everything you need to appear clear Apr 03 18:29:48 Yeah, I only noticed it after you told me about it. Then I immediately added it and recompiled. Apr 03 18:30:11 Hi all. Quick design question. How do you all feel about apps that ask for a review only once (on second or third open)? Is this too intrusive? Apr 03 18:33:54 liuwenhao, As long as it doesn't keep bugging you, I personally wouldn't mind that, though after using it only once or twice it might still be a bit soon for a review. Apr 03 18:34:01 icedp: seems like a bit of a hack to me, but it could work. Apr 03 18:35:17 liuwenhao: whatever the core functionality of the app is, make sure the user has used that. That's the only time i really get annoyed is when it asks me to review before i've even used the thing. Apr 03 18:35:32 Does it cost money to be able to upload a freeware app to the android market? Apr 03 18:35:44 ben_g: 25$ for a dev account Apr 03 18:35:47 You still need to buy a dev license Apr 03 18:36:04 Solarnum: makes sense, I don't want to be aggressive about reviews Apr 03 18:36:29 alexfu: what would you suggest to make the font of button look like a list item? Apr 03 18:36:42 ie. same font family and no allcaps Apr 03 18:37:15 Hello, I want to perform an HTTP request which receives an audio Apr 03 18:37:30 Darkwater: use whatever styles you're using for the list items Apr 03 18:38:03 the response shows as MIME type : audio/mpeg Apr 03 18:38:05 ah right, there's probably an attr for default font Apr 03 18:38:17 I want to play that sound as its received Apr 03 18:38:23 I wonder what class should I use Apr 03 18:38:32 I wont do any longer operations with the "audio" Apr 03 18:38:37 should I use Audio class? Apr 03 18:39:14 Is there a different popular platform for redistributing android software for free? (I want to make my programs available for free and ad-free, but I don't want to pay to be able to do that) Apr 03 18:39:44 there definitely is, but I can't think of any right now Apr 03 18:40:07 play store 1 time fee is like $20 isn't it? Apr 03 18:40:16 25 I think, yeah Apr 03 18:40:44 yeah $25 Apr 03 18:41:21 I hate how you can only use a credit card for that, though Apr 03 18:41:29 I don't own one, nor do I ever plan on getting one Apr 03 18:41:47 unless you can buy free prepaid credit cards or something like that Apr 03 18:42:49 Darkwater: Google Wallet for you then Apr 03 18:45:30 alexfu: I can use paypal through google wallet in the play store Apr 03 18:45:40 but to register as developer I need to use a creditcard Apr 03 18:45:56 Hi guys Apr 03 18:45:57 there is a more or less legitim reason why they use a credit card Darkwater, they inherit from the banking security checks and verification Apr 03 18:46:04 but in any doubt, they will ask your real ID anyway Apr 03 18:46:30 I'm learning Android and I could use some help here Apr 03 18:46:31 (or maybe not even need to be in doubt, even on some wave/roll schedule) Apr 03 18:46:51 I'd guess it's for that reason, but there are other ways to verify ID Apr 03 18:47:07 and it's a shame they're only supporting such an easy/cheap way Apr 03 18:47:32 I'm trying to add a view to a layout dynamically. The textView doesn't show up, and a found a lot of stackoverflow post where the syntax is the same as my code Apr 03 18:47:50 it costs almost 0 computing resource to store (badddd) and check if a card is valid Apr 03 18:47:55 The code if someone see something... Apr 03 18:47:56 http://pastebin.com/V0jFuezE Apr 03 18:48:06 and it's pretty efficient compared to other check method Apr 03 18:48:31 but i understand your frustration Apr 03 18:48:50 xDevol: are you sure R.string.no_work isn't empty? Apr 03 18:48:56 you can still distribute your apk outside of the playstore in the meantime, especially if you go to deliver a free app Apr 03 18:49:04 Oh my Apr 03 18:49:08 but you will need the playstore visibility too Apr 03 18:49:14 Such a shame Apr 03 18:49:34 yeah, thankfully I have until july/august to worry about distribution Apr 03 18:49:36 I'm sorry, you're right Apr 03 18:49:44 xDevol: happens to all of us :v Apr 03 18:49:55 Darkwater, what are you building? Apr 03 18:50:04 Well thanks Apr 03 18:50:10 happened to me like two days ago, was wondering why the item I just added wasn't here - it was just empty because I forgot to set the label :v Apr 03 18:50:21 iprime: a schedule app for schools Apr 03 18:50:28 neat Apr 03 18:50:46 something like that can also happen if your theme's def text color is the same as your parent layout's background :) Apr 03 18:51:06 http://novaember.com/s/304621714.png Apr 03 18:51:11 to free students of that horror Apr 03 18:52:44 Darkwater: looks pretty user friendly to me Apr 03 18:53:32 except it's ugly as fuck and also not mobile-friendly Apr 03 18:53:50 plus, push notifications are cool Apr 03 18:54:49 How can I keep my android gui code organized? I always end up with a big pile of unstable code with "setContentView" everywhere... Any tips? Apr 03 18:55:09 stackoverflow is fundementally flawed Apr 03 18:55:37 Flaghacker: setContentView everywhere? Apr 03 18:55:49 you should only be calling it like once per activity lifecycle Apr 03 18:55:49 lots and lots of content views ... Apr 03 18:55:55 lasserix, I mean spread out thoughout my code Apr 03 18:56:10 Flaghacker: custom views, compound views Apr 03 18:57:55 Mmmm... When I have a layout with a bunck of textfields that display inforamtion, and another layout that shows any problems that have occured Apr 03 18:58:11 Should I combine them into one view? Apr 03 18:59:12 Flaghacker: nicer to have them scroll/fill separately imho Apr 03 18:59:28 myke, what do you mean? Apr 03 18:59:47 you can always use a parent layout that shows both Apr 03 19:00:11 and how can I toggle between them? Apr 03 19:01:29 liuwenhao: It makes the app look bad to me. Apr 03 19:01:34 The kind of app that begs and pesters the user. Apr 03 19:03:30 it does not disturb me if it is one or two times, it's not even shaddy since it's so important to have many nice review/note Apr 03 19:04:07 i'm wondering on app which asks for ratings, what is the percentage of success rate on rating, and what is the percentage of bad vs good rating on that too Apr 03 19:04:25 also suprisingly, it's not forbidden, like it's forbidden to incite ppl to click on ad Apr 03 19:05:15 adq: typically w/ ads, there's revenue involved Apr 03 19:05:35 which is my guess as to why it's a no no to tell ppl to click your ads Apr 03 19:05:49 also, on "IRL products" , feedback is often almost hidden (in case the proba is high to get complain lol), or totally put in front like "our customer feedback matters" (bullshit sometimes, just they need this attitude) Apr 03 19:05:57 everything is so blur, everywhere Apr 03 19:06:04 Flaghacker: two tab buttons across the top? or a spinner? ideally whatever material design says Apr 03 19:06:20 myke, what is material design? Apr 03 19:06:26 myka Apr 03 19:06:32 alexfu, yeah i imagined the scenario where the app could be free with ads, free with IAP, or simply a paid one Apr 03 19:06:37 and I meant how to toggle from java? Apr 03 19:06:47 for sure, it looks far more legit on an open-source free app lol Apr 03 19:06:57 liuwenhao are you asking about users rating your app ? Apr 03 19:08:33 Imagine a desktop app that asked the user to rate it (or do something similar). Take Firefox as an example. How would users react? Apr 03 19:08:55 It seems to me that people are way more tolerant of annoying stuff on mobile (for some reason). Apr 03 19:09:13 i dont think its a terrible idea, but the timing has to be perfect. you can do it like this https://medium.com/circa/the-right-way-to-ask-users-to-review-your-app-9a32fd604fca Apr 03 19:09:14 Imagine if Firefox had a perma-ad. Apr 03 19:09:29 that said, i dont do it either Apr 03 19:10:36 Flaghacker: from java just inflate the appropriate layout, this may also be a good use of a fragment Apr 03 19:11:46 Yeah, that article makes a good point: showing the prompt in a non-modal way is far less annoying. Though I doubt I'd ever do it too. Apr 03 19:12:13 and Solarnum pointed something important too, not the first time you launch the app lol Apr 03 19:12:25 myke, I'll do some research I guess :D Apr 03 19:12:48 noob question, when i fill in a form field on a static layout with a do-nothing activity and rotate, the form data sticks around. how? does the new layout attach to the same view objects in the application, or do the view objects belong to the activity and use some other mechanism to persist? Apr 03 19:13:15 Flaghacker: there's a lot of ways to do it, you could also make a combined view like you said Apr 03 19:13:37 any suggestions for adding padding on top of a listview? Apr 03 19:13:38 The activity persists certain view data (e.g., the text of an EditText) for views with IDs. Apr 03 19:13:45 thank you Apr 03 19:13:47 myke, k ty Apr 03 19:13:50 if I use regular padding, there's like a blind spot when scrolling Apr 03 19:13:54 Also stuff such as the scroll position of a ListView. Apr 03 19:14:01 it should act more like a margin on the first item Apr 03 19:14:08 Doesn't preserve visibility, though (which I find weird). Apr 03 19:14:35 TextView text is persisted if and only if freezesText is set. Apr 03 19:14:40 (Unless it's an EditText.) Apr 03 19:18:27 Darkwater, add an 'empty' header Apr 03 19:18:40 the new version of Keep has a nice FAB button on pre-5.0 devices Apr 03 19:19:01 that messes with the positioning though, for click handlers Apr 03 19:19:08 is there a nicer way than hardcoding -1? Apr 03 19:19:22 make the header not clickable? Apr 03 19:19:41 the first real item would still be on the second position Apr 03 19:19:59 Doesn't padding work on a ListView? Apr 03 19:20:07 I don't think I've tried, but I'd expect it to. Apr 03 19:20:08 it does, but not nicely Apr 03 19:20:14 Oh, wait: I'm thinking of margins. Apr 03 19:20:15 I'll make a screenshot Apr 03 19:20:20 How about padding in the container? Apr 03 19:20:21 well actually never mind Apr 03 19:20:26 it's pretty much the same as a margin Apr 03 19:20:34 Give the parent padding. Apr 03 19:20:39 never mind that, I'll take a shot anyway Apr 03 19:20:44 Okay. lol Apr 03 19:21:18 http://novaember.com/s/874901431.png Apr 03 19:21:45 the list are clipped within the padding Apr 03 19:21:48 list items* Apr 03 19:23:04 Darkwater, are you sure that making it not clickable doesn't work? Apr 03 19:23:20 oh wait.. the term is "not selectable" Apr 03 19:23:27 Darkwater: I don't understand the problem. Isn't that what padding does: add a gap? Apr 03 19:23:42 I want the text to be visible there though Apr 03 19:23:53 what I want is more like a top margin on the first item Apr 03 19:24:00 but I figure that's a bit of a hack Apr 03 19:24:07 so I was wondering if there's a cleaner way Apr 03 19:24:44 Oh. You could do that in an adapter. When returning the first item, add top-padding (or something). Apr 03 19:25:12 yeah, guess that's he best option Apr 03 19:25:29 Darkwater, you mean you've tried the non-selectable header hting? Apr 03 19:25:39 yes Apr 03 19:25:55 IMO, the ListView header/footer thing is a huge hack. Apr 03 19:26:02 If you've seen the code... ugh. Apr 03 19:26:13 I'll pass :v Apr 03 19:26:37 oh well Apr 03 19:26:54 is it better to let uncaught excepts be handled by Play or GA ? Apr 03 19:26:59 *exceptions Apr 03 19:27:38 AFAIR, it wraps the adapter in another adapter, and synthetically adds views. It's weird. (And it's also got weird bugs on certain Android versions.) Apr 03 19:28:00 oh fuck that's even worse than what I thought of Apr 03 19:30:38 I guess you could do two things: (1) have your adapter return padding for the first child or (2) have your adapter return an extra, non-clickable first child. Apr 03 19:30:52 s/return/add/ Apr 03 19:30:57 Use Realm feels like I'm playing an RPG for some reason. Apr 03 19:31:10 I'm sending my objects to the battlefield in the 'Realm'. Apr 03 19:31:16 Fight for glory my objects! Apr 03 19:31:27 Using Realm* Apr 03 19:31:35 ooh, I like the idea of the adapter returning a fake item Apr 03 19:31:45 I was running my first app from cli from the point mentioned in https://developer.android.com/training/basics/firstapp/running-app.html Apr 03 19:31:47 since I just realized I need to mess with layout params to add margin which I hate Apr 03 19:31:52 but did not find gradlew in project directory Apr 03 19:31:59 ** Objects surprised you! ** Apr 03 19:32:27 Darkwater: With the thing I'm saying, you wouldn't need to use margins. Just padding for the first ListView child. Apr 03 19:32:51 I think, anyway. Apr 03 19:33:05 that'd make it ugly when pressed Apr 03 19:33:07 I guess it depends on the nature of the ListView children. Apr 03 19:33:09 Ah, okay Apr 03 19:33:13 ie. the background would he too high Apr 03 19:35:24 Just remember to add 1 to Adapter.getCount and so on. Apr 03 19:35:33 And offset getView positions by 1. Apr 03 19:41:10 and add another view type Apr 03 19:41:17 so much to do, so much to see Apr 03 19:41:53 more than just a little chanve :v http://novaember.com/s/529158067.png Apr 03 19:41:55 chance Apr 03 19:41:58 change god damnit Apr 03 19:42:05 hmm TacticalJoke Darkwater why do you consider wrapping an adapter in an adapter bad? Apr 03 19:42:23 why would you Apr 03 19:42:33 Well, it's not necessarily bad, but I do think it's bad in the case of ListView headers/footers. Apr 03 19:43:42 I think that code should be inside the custom adapter, and not a feature of ListView. Apr 03 19:44:11 TacticalJoke isn't that the way you would do it with recyclerview ? Apr 03 19:44:14 Yeah. Apr 03 19:44:47 TacticalJoke: well, but you can Apr 03 19:46:07 I just think addHeaderView/addFooterView are an attempt to do something that the adapter itself should be doing. Apr 03 19:46:14 What if you want a middle (not a header/footer) view? Apr 03 19:46:20 Then you have to implement it in your adapter. Apr 03 19:46:56 I know you were asking about adapters wrapping adapters, but this is a similar topic. :D Apr 03 19:47:04 not much point debating it, its been there for a long time. if its there, may as well use it heh Apr 03 19:47:12 It's buggy, though. Apr 03 19:47:24 might as well use deprecated APIs as well, right Apr 03 19:47:26 TacticalJoke, what if your listview have to show differnet adapters with the same header/footer Apr 03 19:47:27 they're there anyway Apr 03 19:47:42 On some versions (I forget which), you have to call AbsListView.addFooter/HeaderView *after* AbsListView.setAdapter. Apr 03 19:47:43 Hey, how i can change the button background without losing the shadow??? Apr 03 19:47:52 Wait, I think it's before, not after. Apr 03 19:48:03 codekK: are you trying to make it transparent? Apr 03 19:48:14 it's before Apr 03 19:48:22 otherwise, try specifying elevation manually Apr 03 19:48:22 Darkwater, nop Apr 03 19:48:41 yeah, before, that was before kitkat Apr 03 19:48:50 Darkwater, i just defined at selectionr with three green colors but now my button lost the elevation Apr 03 19:50:52 Darkwater, maybe android:clipToPadding is what you're looking for Apr 03 19:52:02 iprime: Hmm, that might be a case where adapter composition is a solution. I'd have to try it, though (never have). Apr 03 19:53:15 what if you don't have control over the adapter, there might be many what ifs Apr 03 19:53:30 iprime: *cries internally* Apr 03 19:53:31 thank you Apr 03 19:53:36 iprime: In any case, if you move to RecyclerView then you have no choice. You have to then do it all in the adapter. Apr 03 19:54:29 well, maybe the clipToPadding magic will work for RecyclerView too Apr 03 19:54:56 Why would you not have control over the adapter, BTW? I'm trying to imagine the scenario. Apr 03 19:55:51 i guess it's a possibility.. like you use some lib or something, idk, but also it might be tiresome to do composition for every new adapter Apr 03 19:56:40 you ever used the common guy's merge adapter? Apr 03 19:59:06 i wonder how the whole thing (header/footer/adapter item) will play with having to dynamically turn the header/footer on and off Apr 03 20:00:27 surprised there is no cwac-merge for recyclerview Apr 03 20:03:31 Ah, didn't know about clipToPadding. Handy! Apr 03 20:03:36 Hello, I have installed android studio and then I opened SDK manager and installed packages too but I am wondering how to open ADB and shell Apr 03 20:04:51 jennie: in your shell's config (.bashrc or .profile or whatever) add the paths in the sdk you need, for adb, android, etc Apr 03 20:05:01 then you can invoke adb directly Apr 03 20:05:25 i usually have a symlink "sdk" in my home dir linked to the actual sdk dir Apr 03 20:05:42 i can only see program files Android directory Apr 03 20:05:51 and Android studio 64.exe Apr 03 20:05:55 where is shell :-S Apr 03 20:05:58 oh are you on windows? Apr 03 20:06:02 yes Apr 03 20:06:20 i would strongly recommend a unix shell, like the msysbash (?) that comes with git Apr 03 20:06:33 you can use cmd.com i guess Apr 03 20:06:52 but if you're doing any cli work a replacement shell is worth it Apr 03 20:07:26 otherwise, in your windows system prefs is PATH, tack the sdk dirs onto that Apr 03 20:07:27 cli on cmd? ouch Apr 03 20:07:35 Ashiren: i know rite?? Apr 03 20:08:59 i think there are windows-style cmd.com replacements as well but i don't dev on windows Apr 03 20:09:16 there's "Terminal" in Android Studio Apr 03 20:09:29 which uses the system terminal, cmd in windwos Apr 03 20:09:35 :v Apr 03 20:09:48 For simply running "adb shell", it probably doesn't matter. :D Apr 03 20:09:49 didn't know...but you still just get cmd Apr 03 20:09:57 yah Apr 03 20:09:57 true Apr 03 20:10:17 so much you can do with adb tho Apr 03 20:10:52 ALright, got it :-) thank you very much Apr 03 20:16:25 jennie: cygwin Apr 03 20:19:19 there has to be a way to do this - in google analytics i have seperate release and debug properties in my console. so debug builds should log to debug property, and release builds log to release property. Now i have custom dims - but to work, the indexes need to be the same across profiles Apr 03 20:19:51 but i can't figure out how to change the index of a custom dim - if this isn't possible, i have to delete the whole profile from scratch and recreate it ... Apr 03 20:19:58 which is so absurd, i have to be doing something wrong Apr 03 20:22:12 lasserix: if you only need bash then gitbash is a better way to go Apr 03 20:22:24 no no Apr 03 20:22:27 cygwin is much better Apr 03 20:22:43 ultimatly you will need more than bash Apr 03 20:22:55 cygwin is different, cygwin is a full posix layer that bash and etc run on top of Apr 03 20:23:05 hi Apr 03 20:23:20 if you don't need all that because you're only using adb then it's unneccesary clutter Apr 03 20:23:50 iirc that gitbash comes with a bunch of the standard unix tools as well Apr 03 20:24:17 otoh no job control and many unix things are missing Apr 03 20:25:14 and there's always http://unxutils.sourceforge.net/ Apr 03 20:35:08 Whats best test framework to use for integrated / functional testing? Apr 03 20:35:17 integrated/functional <-> ui Apr 03 20:39:05 Does anyone know how to copy your Realm object to the 'Realm database' when the primary key already exists? It says that when you have a primary key you have access to createOrUpdate() but I don't see anything in the docs besides createOrUpdateAllFromJson which I don't have. I'm creating or updating from the actual object itself.. >.> Apr 03 20:40:09 createOrUpdateObjectFromJson() also exists* Apr 03 20:49:03 how do you sort your methods and fields in a class? Apr 03 20:49:49 alphabetically or in some magic order? Apr 03 20:50:10 the ide can give you alphabetical breakdown Apr 03 20:50:21 I'm asking for what *you* are doing Apr 03 20:50:24 like the people here Apr 03 20:50:29 i do what makes sense ;) Apr 03 20:50:33 (yay ambiguous language) Apr 03 20:50:36 Not alphabetically. Apr 03 20:50:37 but what makes sense? Apr 03 20:50:57 like, onCreate first, then the rest of the lifecycle callbacks for activities, then custom methods? Apr 03 20:51:16 that's what I do pretty much Apr 03 20:51:24 // this method location chosen by a fair roll of the dice: 8 Apr 03 20:51:45 lol Apr 03 20:51:46 i do mostly override firsts Apr 03 20:52:00 One pattern I like is to define, say, a public method and then define private methods that it calls directly beneath (and methods that they call directly beneath, and so on). Apr 03 20:52:22 I'm not keen on the idea that methods should be grouped by visibility. Apr 03 20:52:45 and how do you group variables Apr 03 20:52:56 i try to do variables at beginning Apr 03 20:53:03 Yeah, fields at the start. Apr 03 20:53:07 but many times i put them above methods when needed Apr 03 20:53:09 If there are many fields, I probably need to refactor. Apr 03 20:53:11 Though that depends. Apr 03 20:53:13 and forget to move them Apr 03 20:53:41 I group View objects together, then Context and other stuff separate (EventBus, whatever) Apr 03 20:53:44 one of the hard things about using GA is that since it was web product initially, almost all the help is web-centric. like 99% Apr 03 20:54:35 wait, some people actually put variables amongst methods? Apr 03 20:54:51 I *always* put them on top, like, religiously Apr 03 20:55:12 Yeah, I've been doing that forever, regardless of language. Apr 03 20:55:20 Well, assuming that it's an OO language. Apr 03 20:55:30 (Putting them at the top, I mean.) Apr 03 20:56:52 Hi. Anyone know of a way to change default locale for an application? Something like LC_ALL=... in linux Apr 03 20:56:54 I'm glad we don't have the header/implementation distinction in Java. I hated that in C++. Apr 03 20:57:01 And Delphi. Apr 03 20:57:31 TacticalJoke: Why would you hate it? Apr 03 20:57:39 well it makes sense Apr 03 20:57:48 the R.ids that are generated are static right? Apr 03 20:57:48 I mean the two files: foo.h and foo.cpp. Apr 03 20:57:49 plus, it's not always needed Apr 03 20:57:51 It was so tedious. Apr 03 20:58:12 Java did a good thing for the world by eliminating that, IMO. Apr 03 20:58:17 ROFL Apr 03 20:58:41 Most IDEs provide a shortcut to switch between the two. Apr 03 20:58:51 TacticalJoke: it has some advantages Apr 03 20:59:01 for example, you only need the headers of libraries you use Apr 03 20:59:18 How is a library's API specified in Java? Apr 03 20:59:41 in the library itself, which includes the implementation Apr 03 20:59:53 oh god Apr 03 21:00:18 the header / source split of C is probably the worst feature ever Apr 03 21:00:20 I take it you've never noticed the .jar.h file? Apr 03 21:00:30 nope Apr 03 21:00:39 (including the whole JavaScript) Apr 03 21:00:54 There never was a split. It's how C/C++ compilation is done. Apr 03 21:01:22 Doesn't make it any more tedious or annoying to support and develop. Apr 03 21:01:23 it's also entirely unnecessary Apr 03 21:01:40 ROFL Apr 03 21:01:52 kralyk: Are you really rolling? Apr 03 21:01:58 I am not. Apr 03 21:02:08 why are you lying then Apr 03 21:02:12 There a whole bunch of reasons for header files Apr 03 21:02:13 you say "rofl" like there's no alternative to header/source files, when many languages do just fine without them Apr 03 21:02:17 For instance, C++ templates Apr 03 21:02:32 kralyk, the bunch of reasons can be shortened into "we had to keep C compatibility" Apr 03 21:02:47 It is true you can do without header files. Apr 03 21:03:06 how is C++ copmatible with C Apr 03 21:03:11 For example, Rust does this. But it's not without drawbacks. Apr 03 21:03:30 all non-C languages do that Apr 03 21:04:03 Which of those non-C languages could possibly replace all of C/C++ usecases? Apr 03 21:04:08 Darkwater, it's a (kinda) strict superset Apr 03 21:04:17 kralyk: aside from a small increase in binary size (if you choose to have any externally-visible methods), what's the drawback? Apr 03 21:05:29 1) Harder compilation (see how Rust strugles with that), 2) It's nice to have a library interface defined in a file rather than to have to dig it out of a binary by some magic Apr 03 21:05:40 what Apr 03 21:05:47 the header system wrecks havoc with C++ Apr 03 21:05:58 What havok? Apr 03 21:06:07 it's one of the main reasons why C++ compiles so slowly - the way template compilations work Apr 03 21:06:14 the only con of headers I can see is having to maintain two files Apr 03 21:06:29 the resolution and deduplication of template instantiantions in object files is the reason why it can take minutes to hours for large projects to compile Apr 03 21:06:35 Mavrik: That's because it's hard to compile templates, that's got nothing to do with header files Apr 03 21:06:35 groxx in analytics, do you have separate debug and release properties in admin console like iosched - so that debug builds track to one property, release builds to another ? Apr 03 21:07:36 kralyk: because headers aren't magic? besides, you could probably pull them out with `strings` if forced to. and I _completely_ fail to see how it makes compilation harder. Apr 03 21:07:39 kralyk, actually it does - or more precisely - it has to do with the way how C and C++ define compilation units as units that compile Apr 03 21:07:54 g00s: ¯\_(ツ)_/¯ Apr 03 21:07:55 kralyk, and header files are a side effect of that isolated compilation unit approach Apr 03 21:08:12 g00s, what's the point of tracking debug builds? :) Apr 03 21:08:30 Mavrik: Yes. I still don't see any havoc. Compilation units enable incremental compilation which _speeds up_ compilation. Something that Rust lacks. Apr 03 21:08:31 Mavrik to test analytics and not conflate data between the 2 Apr 03 21:08:42 kralyk, not in case of templates Apr 03 21:09:01 Mavrik: Why not, Apr 03 21:09:02 ? Apr 03 21:09:05 templating engine actually significantly prolongs compilation because the object files are isolated Apr 03 21:09:10 if anything, having separate .h files that have to be #include / #define + #ifdef / #pragma once'd is _massively_ more complicated than compiling a single file. Apr 03 21:09:24 and it instantiates them for each inclusion via header file and then has to go through and resolve all the duplicated templates when linking Apr 03 21:09:33 which is one of the main causes for long compilation times Apr 03 21:09:41 LTO being the other if you use it :P Apr 03 21:09:58 didn't most compilers create precompiled headers like 10 years ago ? Apr 03 21:10:10 i dont thin its a performance problem any more Apr 03 21:10:16 also having separate compilation units significantly intereferes with cross-compilation unit optimization Apr 03 21:10:20 Mavrik: So your definition of "havoc" is "somewhat longer compile times"? Apr 03 21:10:26 "can create" and "people are using" are different things, but I think so g00s Apr 03 21:10:28 it took us until what... 2010 to get a working LTO that's portable? Apr 03 21:10:47 Besides, compilation unit still doesn't need recompiling if it and its dependencies are not altered since last compilation. Apr 03 21:10:52 kralyk, dude, larger projects compile in terms of several tens of minutes to hours Apr 03 21:11:00 even with precompiled caches Apr 03 21:11:00 Mavrik: Yes. So? Apr 03 21:11:03 Mavrik: Yes. So? Apr 03 21:11:16 kralyk, you don't know the meaning of the word "somewhat" Apr 03 21:11:32 we're talking about 10^3 - 10^5 orders of magnitude slower compilation Apr 03 21:11:35 Mavrik: What would you do speed up the compilation then? Apr 03 21:11:44 Avoid templates? :D Apr 03 21:11:48 what pretty much all other languages do Apr 03 21:11:56 and that's...? Apr 03 21:11:57 (and did since) Apr 03 21:12:10 compile to javascript! Apr 03 21:12:15 use a package oriented approach where compiler is aware of other source units Apr 03 21:12:20 not just the linker Apr 03 21:12:31 that's why you get Java to compile a 10x bigger project in about 1/100th of time Apr 03 21:12:38 which means faster CI cycles and development Apr 03 21:12:40 groxx: :D Apr 03 21:12:47 I mean c'mon Apr 03 21:12:59 EVERYONE improved on that fail feature of C Apr 03 21:13:06 every single language since 1975. Apr 03 21:13:09 including pascal. Apr 03 21:14:09 I hate how everything has to target the JS-VM these days. We need a revolution! Apr 03 21:14:15 You can't really compare compiling to Java bytecode. Try compiling to native code and you'll find out your approach is actually slower. Rust has been exactly there...7 Apr 03 21:14:36 kralyk, just... stop. Apr 03 21:15:05 machine code generation from intermediade code is a very short step on both LLVM and GCC toolchains Apr 03 21:15:25 never knew android devs were so passionate about c/c++ headers & related issues :D Apr 03 21:15:28 TacticalJoke: and the revolution shall be to compile JVM to JS-VM, and compile JS to Java, running on a JS-VM JVM! Apr 03 21:15:40 also Go generates "native" code and has one of the fastest compilers out there Apr 03 21:15:52 g00s, well, I've been developing C and C++ software for years now Apr 03 21:15:59 LLJVM Apr 03 21:16:02 g00s, at some point you get bitter from all the dumb bullshit :P Apr 03 21:16:03 just thought i'd throw that in there Apr 03 21:16:10 g00s, you know, just like doing Android dev for double jackpot :P Apr 03 21:16:16 TacticalJoke: that'll blur the line between Java and JavaScript, further confusing literally everyone in the world Apr 03 21:16:21 I'm not an Adnroid dev Apr 03 21:16:29 It sounds like the logical next step. Apr 03 21:16:42 we need to port node.js to jvm Apr 03 21:16:48 why compile java to bytecode targeting a c-based vm when you can compile c to bytecode targeting a java-based vm? Apr 03 21:16:53 I'm just not a fan of the conecpt of "let's do what Java does, because Java does stuff better..." It doesn't :D Apr 03 21:16:59 everything is everything now Apr 03 21:17:07 We could create Noderoid. Apr 03 21:17:07 kralyk, you do what works best Apr 03 21:17:12 Mavrik vert.x is pretty close to node.js on jvm Apr 03 21:17:15 everything is so much all the time, I can't even Apr 03 21:17:16 and this type of translation is only going to become more common and used Apr 03 21:17:30 kralyk, if you don't want to use a better solution because Java used it... well... I'd rather not insult you here :) Apr 03 21:17:31 I can't. I just can't. Apr 03 21:17:40 I don't even go far as evening. Apr 03 21:17:42 g00s, good, and then we port .NET to JavaScript Apr 03 21:17:44 as^ Apr 03 21:17:47 Java's generics aren't really comparable to C++'s anyway... Apr 03 21:17:52 and we can run .NET apps in JavaScript running on node.js on jvm Apr 03 21:17:56 on a MIPS android device. Apr 03 21:18:40 Does Go even have generics? Apr 03 21:18:44 everyone here has seen http://bellard.org/jslinux/ yea? Apr 03 21:19:14 Mavrik: Go has generics? Apr 03 21:19:24 Detected 3.333 MHz processor. Apr 03 21:19:33 The JavaScript situation is like a huge The Onion newsletter. Apr 03 21:19:44 javascript is a terrible language Apr 03 21:19:53 The PHP situation, too. Apr 03 21:19:54 naturally it's what everyone wants to target, now Apr 03 21:20:01 kralyk, nope Apr 03 21:20:10 lol Apr 03 21:20:13 Have you seen what the creator of PHP says about himself? He basically admits that he doesn't have a clue about programming. Apr 03 21:20:23 myke: yeah :D also EM-DOSBOX: https://archive.org/details/softwarelibrary_msdos_games Apr 03 21:20:43 TacticalJoke, that's unsurprising, that's why he created an effective language for people who don't have a clue about programming :P Apr 03 21:20:44 Mavrik: I don't get why you compared it to C++'s compilation of generics then but whatever... Apr 03 21:20:44 I'm not a real programmer. I throw together things until it works then I move on. The real programmers will say "Yeah it works but you're leaking memory everywhere. Perhaps we should fix that." I’ll just restart Apache every 10 requests. Apr 03 21:21:03 http://en.wikiquote.org/wiki/Rasmus_Lerdorf Apr 03 21:21:09 kralyk, that's because I didn't. We were talking about header/source file split Apr 03 21:22:18 I thought we agreed templates is the source of slow C++ compilation. You probably know well C/C++ compilers don't actually have any concept of header file. Apr 03 21:22:18 TacticalJoke, https://www.xkcd.com/1495/ Apr 03 21:22:34 :D Apr 03 21:22:48 They just #include whatever you tell it to #include Apr 03 21:23:01 kralyk, templates in connection to compilation units being isolated Apr 03 21:23:06 templates as themselves are ok. Apr 03 21:23:06 groxx: that looks cool unfortunatley it seems the remote library of games it tries to pull in is down/gone :( Apr 03 21:23:47 TacticalJoke: ironically php's runtime is excellent compared to trying to run perl, python, etc from apache Apr 03 21:23:53 myke: hm, Sim City worked fine for me Apr 03 21:24:02 i don't recall ever having php leak memory Apr 03 21:24:08 as does Prince of Persia Apr 03 21:24:15 groxx: maybe i'ts me, are you in the US? Apr 03 21:24:24 myke: yep Apr 03 21:24:41 weird, ok, i'll try again later, thanks for the link Apr 03 21:25:47 Mavrik: The trouble is compiling packages as a whole doesn't really help you in that regard. That's what Rust does: It compiles the whole library/executable as one, but that means it has to scan it's entire code and figure out the dependency tree of the whole thing including generics. Which, in combination with the fact that this approach makes incremental compilation difficult, actually prolongs compile Apr 03 21:25:48 time. Apr 03 21:26:33 I like that the world has gotten to the stage where using a nice and neat language to create a modern app is viable. Kotlin is an example. :) Apr 03 21:26:39 C/C++ projects typically compile for a long time the first time, but subsequential compilations are typically much faster. Apr 03 21:27:08 TacticalJoke but you stillhave to use kotlin on android, which is hardly nice and neat Apr 03 21:27:10 In the past, it was way more difficult to choose, and many choices were bad. Apr 03 21:27:21 I am using Sugar ORM library in my app and I want to find the maximum value of column(int) in a table. How can I do that? Apr 03 21:27:32 True, but compare that to something like C++ on Win32. I'd pick Kotlin on Android. :D Apr 03 21:27:37 Or C on Win32. Apr 03 21:28:18 TacticalJoke: we were already there with python + gtk bindings Apr 03 21:28:33 g00s, well I heard we now have Qt support :P Apr 03 21:29:31 I haven't used Python much, but I do recall badly missing static typing when I did. ;| Apr 03 21:29:41 man talk about crazy, i remember Qt and their moc compiler Apr 03 21:29:50 moc is pretty nuts Apr 03 21:30:11 I used QtCreator when going through a crazy Linux-only phase. Was kinda neat. Apr 03 21:30:23 anyone here use vim? Apr 03 21:30:28 It certainly made C++ feel less horrible. Apr 03 21:30:40 Not for years, Darkwater. I recovered. :D Apr 03 21:31:04 QtCreator is IMO the best IDE out there... Too bad it's only viable for C/C++ projects... Apr 03 21:31:22 TacticalJoke: what made you stop? Apr 03 21:31:26 I used to like Visual Studio a lot (though it needed Resharper for godo refactoring). Apr 03 21:31:28 good* Apr 03 21:31:39 Darkwater: I found that I was struggling to use basic text boxes (as in "use them efficiently"). Apr 03 21:31:41 I use it occasionally, just not for most coding Apr 03 21:32:03 what do you mean? Apr 03 21:32:18 My brain was in vim mode all the time. It was a struggle to use anything non-vim. Apr 03 21:32:26 oh like that Apr 03 21:32:30 I see Apr 03 21:32:45 And I started thinking "It'd be cool if all text fields were vim-like", and I started to realise I had a problem. :D Apr 03 21:33:03 hehe Apr 03 21:33:26 In Linux most of the text fields are a little shell-like, ie. you can use Ctrl+U to clear them etc. But it's not a huge difference Apr 03 21:33:35 Do you not use shortcuts in as? Apr 03 21:33:36 I think I'm on a pretty decent level with vim, but I don't have that problem tbh Apr 03 21:33:44 even my zsh is still in emacs mode Apr 03 21:34:09 http://ibin.co/1x6FJJEz5Kj6 Apr 03 21:34:12 everything old is new again Apr 03 21:34:31 (kralyk: it's emacs-like - you can also set your shell to be more like vi) Apr 03 21:34:33 modV: I don't see anything for getting calculated columns :\ you probably need to `getSugarDb().getDb().rawQuery("select max(whatev) from your_table");` or whatever the syntax looks like. Apr 03 21:34:40 lol Apr 03 21:34:48 Darkwater: That's cool. One problem I have is that I like something and get obsessive about it, trying to use it everywhere. Apr 03 21:35:07 Darkwater: Ah ok, I pretty fine with the default shell behaviour... Apr 03 21:35:13 I'm* Apr 03 21:35:20 I kind of have that too, but strangely enough I seem to have no problem with places where I can't use vi stuff Apr 03 21:35:28 well most text fields in linux use gnu readline, which is more like emacs Apr 03 21:35:30 Darkwater: I like vim fwiw Apr 03 21:35:34 though I do have an extension for chrome that opens a textbox in vim :v Apr 03 21:35:39 (though I barely use thta) Apr 03 21:35:41 Hello, Im using async task in order to donwload a file and do something with it Apr 03 21:35:43 groxx: Do you use the vim plugin for AS? Just curious. Apr 03 21:35:52 i use a browser called conkeror which brings emacs keybindings to the web...in any textbox you can edit like minimal emacs, which is great Apr 03 21:35:54 the point is that I dont really want to store it to sd or any storage Apr 03 21:36:08 I just want to keep it in memory a couple secs, do something with it and forget it Apr 03 21:36:22 TacticalJoke: sometimes. it gets confused at times, and doesn't play well with manually selecting stuff with the mouse (which is often easier when you need a specific range), which tends to make me disable it and forget to re-enable Apr 03 21:36:27 groxx from where should i call the method getSugarDb? Apr 03 21:36:30 groxx: thanks for demonstrating why i don't like ORMs Apr 03 21:36:31 so I dont want to use FileOutputSream Apr 03 21:36:37 99% of the time though it's pretty nice Apr 03 21:36:44 diegoaguilar: You can't just use a byte array or something? Apr 03 21:36:44 that way, I should have to write to sd and then care to delete it ... Apr 03 21:37:00 myke: heh. yeah, they leak :) but you can get a long way with it. Apr 03 21:37:15 diegoaguilar, that's why ByteArrayOutputStream exists :) Apr 03 21:37:56 I hate how AS is resource-hungry. Consumes gigabytes of RAM and is slow as hell... Apr 03 21:37:59 modV: there's a method with that name on SugarContext, not sure how that's used in the app though. do whatever you normally do to get your SugarDb instance, and call `getDb()` on it to get the underlying SqliteDatabase, which will let you do whatever you want Apr 03 21:38:07 groxx: you've read http://blog.codinghorror.com/object-relational-mapping-is-the-vietnam-of-computer-science/ yea? Apr 03 21:38:16 myke: I think so Apr 03 21:38:36 Mavrik, whats different from using that class to using a native byte array datatye Apr 03 21:38:38 datatype Apr 03 21:38:39 ? Apr 03 21:38:43 groxx let me try that Apr 03 21:38:53 TacticalJoke: sometimes I accidentially close a tab in chrome because I'm used to ^W deleting the last word Apr 03 21:38:54 myke: what about the beans approach? I liked that one. Apr 03 21:39:01 diegoaguilar, besides them being two different things :P Apr 03 21:39:02 none Apr 03 21:39:06 kralyk: welcome to every IDE ever. when emacs was the go-to ide people denigrated it as "Eight Megs And Constantly Swapping" Apr 03 21:39:08 myke: my personal preference is for relatively simple ones that just reduce boilerplate (because there is a LOT of boilerplate with most SQL-like interactions), and provide an easy "escape hatch" to just do what you need by hand. when they get big and fancy, they get difficult. Apr 03 21:39:26 Ok, thanks Mavrik Apr 03 21:39:36 diegoaguilar, you can do either :) Apr 03 21:39:39 oh... btw what I'm receiving its actually an mp3 Apr 03 21:39:43 ByteArrayOutputStream uses a byte[] under the hood. It just facilitates appending and stuff. Apr 03 21:39:58 myke: There are much leaner IDE's like the QtC mentions or sublime... But QtC's for C/C++ only and Sublime is basically just a better editor... Apr 03 21:40:01 groxx, also not having to write damn joins by hand Apr 03 21:40:01 And of course is an OutputStream. Apr 03 21:40:02 which, I want to play once and forget about it Apr 03 21:40:12 groxx: that's the problem, the easy boilerplate you can whip up a few quick methods for yourself, the hard stuff the ORM can't handle or generates horrible inefficient sql for Apr 03 21:40:23 what class should I construct once I got the byte array ready? Apr 03 21:40:30 myke, except they're never quick and even less simple Apr 03 21:40:37 especially when you have to manage the DB as a resource Apr 03 21:40:43 Mavrik: then those are the ones the orm won't do well with either Apr 03 21:40:55 of course it will Apr 03 21:41:01 myke: yeah, but you have to whip those methods up every time, and customize them to your table structure. hence a library. Apr 03 21:41:02 every project led by people with your mindset Apr 03 21:41:06 ended up writing their own shitty ORM Apr 03 21:41:10 Mavrik: i can tell from the solution groxx pointed that that orm is probably using a singleton holding a db handle Apr 03 21:41:13 not hard to do that yourself Apr 03 21:41:37 right. Apr 03 21:41:56 no, but say you could generate all your cursor-parsing code. is it worth it? Apr 03 21:42:00 until you find out your home baked ORM deadlocks / is slow / causes long queries because it's so simple Apr 03 21:42:12 and then it gets a bit more complex... until it's a horrible monstrosity :) Apr 03 21:42:23 Mavrik: for once I agree completely :ú Apr 03 21:42:25 :) Apr 03 21:42:50 I made my own ORM once, using xml as the table defs, and a *huge* xslt template to do the c# codegen Apr 03 21:42:53 it was glorious Apr 03 21:43:03 heh Apr 03 21:43:09 I think we all made an ORM once Apr 03 21:43:13 this is why I've been enjoying GreenDAO - it caches compiled queries, has a pretty straightforward mapping to tables, and writes my cursor-to-object code for me. less painful for 99% of my DB code. Apr 03 21:43:25 Yeah, GreenDAO is pretty neat Apr 03 21:43:40 Changing the tables is a bit annoying though Apr 03 21:43:52 yeah. and the "generator" is weird. Apr 03 21:43:56 I also liked Realm, it's fast as hell and has a reasonable API Apr 03 21:44:00 and the cache isn't threadsafe (but can easily be turned off) Apr 03 21:44:05 that's always the issue. django handles it by making you define your schema as python code instead of doing it directly in sql. Apr 03 21:44:12 and offers a migrator tool Apr 03 21:44:36 south is about the most useful piece of software I've ever seen Apr 03 21:44:42 I'd buy that guy a barrel of beer Apr 03 21:45:05 You guys know RedBeanPHP? It's pretty much the only good thing to have happened in the PHP world... Apr 03 21:45:20 Mavrik: yeah, realm seems neat, minus the "migrations are experimental" thing, the insane "prototype" object thing they do, and the fact that all the objects have to stay on the same thread. it pushes you to do main-thread for everything. Apr 03 21:45:50 groxx, hmm, yeah, I've only used it for minor stuff Apr 03 21:45:56 notification storage, etc. Apr 03 21:46:24 Mavrik: also, I'm very curious how it does its "always up to date" thing, and if it's actually truly threadsafe. seems inherently unpredictable. Apr 03 21:46:45 yeah, need to test it more really Apr 03 21:46:59 but I do mean it - it's a neat project, it's nice to use, it just has some shady / weird edges that mean I'll probably never use it for an important project :| Apr 03 21:48:00 they've been awfully quiet about how the internals work too, preferring to include breathless "omg realm is so threadsafe it'll knock your socks off" quotes Apr 03 21:48:02 the other side of this is, many devs use sql as a simple data store. you can go hard the other way and push the interface into your db server with stored procedures Apr 03 21:48:36 groxx, the thing is, mostly for smaller stuff I just need a dumb easy storage on disk Apr 03 21:48:49 Mavrik: which it's quite nice for, yeah :) Apr 03 21:48:52 on Android you practically never do any "real" DB work Apr 03 21:48:58 and all ORMs are frigging annoying to setup still Apr 03 21:49:47 I don't understand why some Android APIs are so DB-ish, like the contacts or calendar. Why can't it just be accessible through regular OO approach. Apr 03 21:50:22 depends on how you define "real". there's a fair amount of sqlite backing my for-work project, and a db that will probably store ~100k rows + eventually full text search on my side project. it's not "real" in terms of millions or billions of rows, but it's not trivial either. Apr 03 21:51:05 kralyk: because they run through a ContentProvider, which is a pretty thin (and weirdly crippled) cross-process API for SQLite Apr 03 21:51:33 most of them have a nearly 1:1 mapping for uri + params -> table + where clause Apr 03 21:51:56 groxx: Yeah, ContentProvider, that's what I meant. It's a PITA to work with... Apr 03 21:53:09 just think of it as a database. one where you can't know what the structure actually looks like, or enforce transactions, or even query its capabilities. what could go wrong!? Apr 03 21:53:39 :D Apr 03 21:55:26 What's the best way to determine which activity to show on startup? Let's say the user hasn't logged in yet (auth token not saved somewhere), you need to send him to the Guide Activity. Else you guide him to the Main Activity. Apr 03 21:55:43 I can come up with some ways to do it, but they all are clunky and bad code imo. Apr 03 21:56:09 Muchoz: Maybe send him from inside MainActivity.onCreate? Apr 03 21:56:26 Then MainActivity is always first. Makes things pretty simple. Apr 03 21:56:45 Then there would be some kind of animation or change, wouldn't there? I would feel weird, I want it to not be noticable. Apr 03 21:57:31 Muchoz: fwiw if you startActivity from within onCreate of another activity, you won't even see the first one. you can invisibly start (or .finish() will work too) as many as you want, animations etc will still work. Apr 03 21:57:49 groxx, oh alright. Thank you for the information. Apr 03 21:57:58 Glad it's so easy. Apr 03 21:58:01 groxx: The documentation says that, but IME you do see the transition. Apr 03 21:58:06 .finish() in onActivityResult() will also let you pop down through multiple layers Apr 03 21:58:16 The documentation for startActivityForResult says this: "As a special case, if you call startActivityForResult() with a requestCode >= 0 during the initial onCreate(Bundle savedInstanceState)/onResume() of your activity, then your window will not be displayed until a result is returned back from the started activity" Apr 03 21:58:18 TacticalJoke: hm, I've done it a few times, never had problems Apr 03 21:58:34 I can't remember which versions I tested on, but I could never get rid of the visible transition. Apr 03 21:58:58 have a main activity that checks for a valid token, and issues an Intent for either the main page of the logged in activity, or the login start activity Apr 03 21:58:59 if you don't do setContentView and to overridePendingTransition(0,0) nothing will show Apr 03 21:59:03 that's what i'm planning to do anyway Apr 03 21:59:06 TacticalJoke: maybe conflicting animations? Apr 03 21:59:12 I tend to leave them default Apr 03 21:59:15 overridePendingTransition(0, 0) will hide the animation every time Apr 03 21:59:18 Yeah, they were default. Apr 03 21:59:23 that's how Google apps fake half of material animations Apr 03 22:01:25 omg its friday, i blew all week fighting google analytics for mobile Apr 03 22:01:33 Making custom seekbar with nine patch, however the background/progress are always too thick, anyone know how to size thumb-background/progress relative to eachother? Apr 03 22:02:08 Btw, is there a setting that when you create a layout file it automatically goes to the Text tab instead of the design tab? Apr 03 22:02:36 Yeah, that setting is in the design tab (funnily enough). Apr 03 22:02:55 The gear icon. Apr 03 22:03:58 ah, thanks for reminding me. I changed that in eclipse, hadn't remembered to do so for AS Apr 03 22:04:45 I remembered from g00s having the same issue before. :D Apr 03 22:05:57 Have been doing ContentProvider later. Actually it's not that hard, but I've got a huge feeling something is not right. SyncAdapter pulls from URLs a JSON that is almost like DB columns where it's being put. And loader asks ContentProvider by the URI that much resembles original address where I've got JSON from in the first place. And that data mostly unchanged gets to Adapters... So.. (URI) -> Retrofit Apr 03 22:05:59 objects -> ContentValues -> table columns -> (thru URI) -> binding to Views Apr 03 22:08:36 Hello... I just downloaded Android Studio and am getting message "No Android SDK found." But I had downloaded the bundle, including SDK, and entered the SDK location during Android Studio installation. Apr 03 22:13:25 PeterSJC: Are you able to open Android Studio and see the "Quick Start" screen? Apr 03 22:13:37 from ui thred, how can I know when an async task has ended Apr 03 22:13:41 ? Apr 03 22:13:48 If so, click Configure > Project Defaults > Project Structure and see what "Android SDK" says. Apr 03 22:13:56 diegoaguilar: onPostExecute Apr 03 22:17:50 TacticalJoke: I have it open but do not know what the Quick Start screen is. Apr 03 22:18:07 It shows "Welcome to Android Studio". Apr 03 22:19:13 If you're not there, just go File > Project Structure > Android SDK location (or whatever it is). Apr 03 22:21:51 so - i haven't really been using eventbus lately, but i can see a good fit there with analytics; events percolate up to the app to get tracked - don't really want the trackers in my lower layers Apr 03 22:22:21 TacticalJoke: Not showing that. It has some code, a hello-world program that was created automatically. Also getting message "failed to find target android-22...." Oh, wait, I just saw where it wants me to install missing platform and sync. I will try that. Apr 03 22:29:00 so Solarnum I should make onPostExecute receive ... a message from doInBackground? Apr 03 22:29:03 is there any way I could manage the SQLite databases of my app, inside the emulator? Apr 03 22:30:19 saapas, connect to an external sqlite server? :P Apr 03 22:30:20 TacticalJoke: I apparently installed SDK but still get "No Android SDK" message. I clicked the Configure link close to that message, and then did Build|Clean Project. Now getting message:Error:A problem occurred configuring project ':app'. Apr 03 22:30:20 > failed to find Build Tools revision 22.0.1 Apr 03 22:30:35 diegoaguilar, do sqlite servers exist? Apr 03 22:30:37 diegoaguilar: that's how it works. Whatever variable you return in doInBackground gets sent to onPostExecute Apr 03 22:31:32 diegoaguilar: but if you want to notify the ui thread that the asynctask was completed you could use a listener and just fire it in onPostExecute Apr 03 22:31:46 saapas: I just pulling the database file, editting it and pushing back. It can be automated by writing a small shell script Apr 03 22:31:49 "Microsoft Engineer: Open Source Windows Is 'Definitely Possible'" Apr 03 22:32:00 Solarnum, can you guide me a bit with the listener thing? Apr 03 22:32:13 PeterSJC: Why not look at File > Project Structure > SDK Location? Apr 03 22:32:30 icedp, right, ok, that's one.. could there be an app for managing them? Apr 03 22:32:57 SQLite Manager says google. Apr 03 22:33:02 saapas: there are, but I don't use them, just a sqlite3 shell Apr 03 22:33:23 icedp, why? :) Apr 03 22:33:52 then it's another thing to have it installed on the emulator >_< Apr 03 22:34:41 saapas: don't need to do it often. I also that way I don't let myself forger plain SQL commands =) Apr 03 22:34:50 ok Apr 03 22:37:24 TacticalJoke: It shows the SDK in the location where it exists. Apr 03 22:39:09 mobile dev & cars http://www.infoworld.com/article/2904897/application-development/coding-for-cars-next-generation-mobile-apps.html Apr 03 22:39:21 Hmm, I'm not sure, PeterSJC. Never encountered this issue before. Apr 03 22:39:24 Is it 1.1.0? Apr 03 22:39:49 Yes Apr 03 22:40:53 Can you open the SDK Manager (via Tools > Android)? Apr 03 22:41:25 BTW, totally off-topic but . Apr 03 22:41:31 diegoaguilar: http://pastebin.com/SA67pcvF one way you could do it Apr 03 22:43:24 TacticalJoke: Yes, open now. Shows some stuff installed: SDK Build Tools, SDK Platform, etc. Apr 03 22:46:07 thanks so much Solarnum Apr 03 22:48:16 hey guys, question: Assuming you would make a launcher app like androids launcher with 5 pages and max 20 imageButtons on each page. Apr 03 22:48:19 Would you use some caching strategy for the bitmaps on these buttons? Would you recycle the bitmaps? i`m not sure if i should/need to do this, since i dont develop for devices below honeycomb and since the bitmaps itself are not pretty large Apr 03 22:48:31 diegoaguilar: no problem! Apr 03 22:50:32 Solarnum, I checked it Apr 03 22:51:00 so , launchTask and the MyListerner interface Apr 03 22:51:08 are supposed to be inside the activity code? Apr 03 22:51:37 diegoaguilar: launch task is just an example method for wherever you start your asynctask Apr 03 22:51:56 soo I am having problems with a java.lang.OutOfMemoryError when loading bitmaps, I know the problem is the size the thing is I can't reduce their size, and my app has some activities, I suspect my other activities are not releasing bitmap memory how can I be sure Apr 03 22:52:12 diegoaguilar: the interface can be its own class but you need an instance of wherever you intend on using it. Apr 03 22:52:34 Ok, thanks Solarnum Apr 03 22:52:35 :) Apr 03 22:55:17 floWenoL, how are you loading and keeping references of bitmaps? are you using a cache of anytime or lazy loading etc? using a list, or just a view images? Apr 03 22:55:51 and why can't you reduce size? what size are the images, and what size are they being displayed Apr 03 22:59:19 whatitis: i dont keep the references, they go into bitmapdrawables, they are 1080 width and displayed at that same width Apr 03 22:59:49 Have you considered down-sampling the images? Apr 03 22:59:59 With BitmapFactory.Options.inSampleSize. Apr 03 23:00:25 if I downsample them they will show up too small Apr 03 23:00:36 the screen has 1080 and I need them with 1080 Apr 03 23:01:35 it is an image view i am extending, i just load the bitmap make an imagedrawable and set it Apr 03 23:01:49 more specifically a bitmapdrawable Apr 03 23:04:13 adkins: you can look at code of stock launcher I guess Apr 03 23:07:58 floWenoL: you can check you memory usage with Memory Monitor in AS. Press cause GC button Apr 03 23:08:29 floWenoL: also you could use library like Picasso which handles everything for you Apr 03 23:09:05 ^^ Apr 03 23:09:13 picasso is good Apr 03 23:09:30 floWenoL: definitely do. android universal image loader is a good one Apr 03 23:10:38 icedp: good idea, thank you Apr 03 23:11:20 icedp: where can i find the source of stock launcher? Apr 03 23:11:39 floWenoL: I typically just use https://github.com/chrisbanes/Android-BitmapCache to cache my images. Apr 03 23:13:22 will check those Apr 03 23:13:39 adkins: I'm not sure. Here is something oldish https://android.googlesource.com/platform/packages/apps/Launcher/ Apr 03 23:15:54 icedp: thx Apr 03 23:36:37 I'm using IntelliJ + Gradle to develop my App. It /seems/ to me that when I edit a layout file, and click Run, it doesn't seem to recompile the xml file. Am I crazy? If "Clean" the project, it certainly does. Apr 03 23:37:39 :( I saw that occasionally in Eclipse, never in Android Studio (yet). Not sure about IntelliJ-proper. I assume you're using the android plugin? Apr 03 23:37:54 groxx: I am. Apr 03 23:38:38 definitely saving before running? (dunno if that's necessary / configurable, but just checking) Apr 03 23:39:17 I'll have ot pay more attention to be sure, but i think so. Apr 03 23:40:03 :S that was super annoying in Eclipse. dunno how to check / fix / etc, but I feel your pain at least :( Apr 03 23:40:47 oh god I hate the maven plugin so much Apr 03 23:41:23 I wonder if I can update the plugin without all hell breaking loose, I forgot that the older version totally doesn't work with aars that have nested impl.jars Apr 03 23:41:43 Whats the best testing framework to use for automated, integration/funtional testing (/andor ui)? Apr 03 23:43:56 floWenoL: look at inBitmap in bitmapfactory options Apr 03 23:45:08 lasserix: robotium? https://code.google.com/p/robotium/ Apr 03 23:46:37 pfn you're using that because of the client Apr 03 23:46:39 ? Apr 03 23:47:08 i guess you tried convincing them to do things the normal way Apr 03 23:47:59 so sad :( Apr 03 23:49:32 why does WebView extend AbsoluteLayout when it was deprecated in API 3? Apr 03 23:49:53 I wouldn't be using maven otherwise Apr 03 23:49:57 that plugin is so poorly written Apr 03 23:52:06 I have a LinearLayout (Veritcal). If want the last control to be flush-bottom, always. What's the right way? Apr 03 23:57:27 I'd just use RelativeLayout. Very easy then. Apr 03 23:58:06 It might be possible in LinearLayout by use of View.bringToFront, but I'm not sure. Apr 04 00:03:37 or put a Space with height 0dp and weight=100, but probably RelativeLayout, yes Apr 04 00:10:27 /c Apr 04 00:14:32 hi guys Apr 04 00:14:38 hy Odaym Apr 04 00:14:41 and gerls Apr 04 00:14:42 =) Apr 04 00:14:46 Waiting for an apk in beta to propogate on the play store is the worst Apr 04 00:14:55 can I specify a branch in the include line of a library that Im using? Apr 04 00:14:59 in the compile line* Apr 04 00:15:14 compile 'me.panavtec:drawableview:0.5.0:specific_branch' ? Apr 04 00:15:18 Solarnum i heard it can take 1-2 days now for apps because of the human filtering by google Apr 04 00:15:39 g00s: I havn't experienced that with publishing apps recently Apr 04 00:16:45 g00s: but yah I heard about that too... fingers crossed Apr 04 00:17:06 I think it's been slow long before too Apr 04 00:18:35 It's usually around 2-4 hours for me Apr 04 00:20:29 I just wish you didn't have to publish apps to the play store to test in-app purchases Apr 04 00:24:25 our last release was a lot slower than normal :/ Apr 04 00:24:50 4 or 5 hours, I think, instead of ~1-2. Apr 04 00:25:25 still, a few hours is so much better than a week. Apr 04 00:26:07 yah no kidding. Apr 04 00:33:43 so which apps are taking 1-2 days now, new ones, or ones in a specific genre? because from what i've read this is happening Apr 04 00:34:17 maybe anything related to showing content, vide players, music perhaps Apr 04 00:34:32 the ones that flagged themselves for nazi content in the rating questionaire? Apr 04 00:34:38 lol Apr 04 00:46:48 :'( I wonder if we'll ever be able to use android studio's built in decompiler to decompile whatever bytecode exists on the device, instead of pulling from the local android jar. Apr 04 00:47:09 at the moment, if you uninstall "sources for SDK" and try to debug, you get stuff like this http://cl.ly/image/0Y1x3d0g3b2C Apr 04 00:50:48 groxx: well, there's https://github.com/JesusFreke/smali/wiki/smalidea :D Apr 04 00:51:39 will that give me the smali running on the device, or will it switch to local sources when I debug into a platform method? Apr 04 00:52:24 well, you would have to pull the framework dex files from the device Apr 04 00:52:36 and if they're odexed, deodex them Apr 04 00:52:44 :'( Apr 04 00:52:48 although, that would be a neat feature to add :) Apr 04 00:52:50 groxx yeah there is a bunch of android sources not in the android sources :) Apr 04 00:53:11 what I really want is something to read the in-memory framework, because sometimes OEMs / carriers add crazy shit, and I'd really like to know what (if anything) they were thinking. Apr 04 00:54:16 I'm guessing I can't get at the framework dex files without root, and some devices aren't really rootable :( Apr 04 00:54:24 you can Apr 04 00:54:31 the system partition is fully readable Apr 04 00:54:47 but, they're usually odexed Apr 04 00:54:56 so you also have to deodex Apr 04 00:55:36 plus, art deodexing isn't currently supported in baksmali Apr 04 00:56:06 not sure why I never tried that, but you're right. cool. Apr 04 00:57:03 on a totally different topic: anyone using ARC Welder, the Chrome OS android runtime thing? how the hell do you get application run / crash logs out of it? Apr 04 00:57:27 it's all "test your app!" -> kablewy -> and then you can't do anything to fix it. Apr 04 00:57:53 is it just me or is ui testing kind of besides the point / redundant Apr 04 00:58:08 integration i can see, but testing if like a button goes to the next activity? Apr 04 01:00:44 unpacked-libs\as_compatibility-v7-mediarouter_19.1.0\res\values\styles.xml:28: error: Error retrieving parent for item: No resource found that matches the given name 'Widget.AppCompat.Light.ActionButton'. Apr 04 01:00:46 ugh, come on maven... Apr 04 01:01:24 lasserix: if there's enough behavior around it that that may not be certain: sure, why not? but if you're checking if "button.onClick { startActivity(ur activity) }" works, you're pretty much testing the OS, not your app, which is indeed pretty redundant. Apr 04 01:02:07 groxx can you give me a practical example? Apr 04 01:03:00 say activity A has a form to fill out, and depending on how, sends you to activities B, C, D, E, or F. fill out the form, click the button, assert that the right activity has started. Apr 04 01:03:43 you'd probably be better off testing the logic that got you to that point rather than the end state, but odds are something hacky like that will still be _useful_, and more beneficial than a non-existent-but-perfect test. Apr 04 01:03:45 so basically if the programmer has done their job you are testing a switch statement functions properly? Apr 04 01:04:10 and couldn't that be done with a unit test? Apr 04 01:04:30 is there any way to override maven's dependency tree? Apr 04 01:04:37 unit: possibly. there's also stuff like making sure that your UI doesn't change, since you can pull screenshots and compare them to previous runs. Apr 04 01:04:44 e.g. someone uploaded a pom, but it doesn't have the correct transitive libraries Apr 04 01:04:50 (so what I got above happens, I think) Apr 04 01:04:51 hmm Apr 04 01:05:06 so I could say dependency A depends on dependency B Apr 04 01:06:44 i'm sure i did that with maven 5 years ago Apr 04 01:07:16 actually 2008 ;) Apr 04 01:07:49 oh man, the nostalgia - anyone use maven 1? omg ... Apr 04 01:25:18 wooo statisics are up for my app. 125 installs, 17 reviews all 5 star :DDDDD Apr 04 01:25:29 when will facebook buy me out? Apr 04 01:26:14 g00s, maven really shouldn't support that kind of mechanism Apr 04 01:26:24 g00s, just thought it'd be nice if it could Apr 04 01:27:22 DadFoundMy: well done! Apr 04 01:27:46 I find it hard to get good reviews, all my 5 star reviews say "good app" or something nice, and my 1 star reviewers never even write a comment Apr 04 01:29:08 duncannz: only 6 of them are written reviews, but they are pretty positive Apr 04 01:29:32 one kid totally understood the point of the app which is nice to see "Awesome I like how simple it is and how it doesn't require my identity, call info, credit card number, my life story etc. Gg op" Apr 04 01:29:34 DadFoundMy: your app must be good to not get a single 1 star that's for sure Apr 04 01:30:01 duncannz: to be fair the only place ive shown it to yet it very friendly so that probably helped Apr 04 01:30:04 what's the app? if you don't mind I'm curious now Apr 04 01:30:13 https://play.google.com/store/apps/details?id=com.dylanredfield.agendaapp2 Apr 04 01:30:36 on the play store it shows as 1-5 downloads but in my console its ~125 Apr 04 01:31:13 yeah there's lots of lag involved Apr 04 01:31:16 and major rounding Apr 04 01:31:19 can't really get this project converted over to gradle because there are so many apklibs being used :( Apr 04 01:33:26 duncannz: yeah even the console took a a whhile to update Apr 04 01:33:38 for me it updates every 24 hours I think Apr 04 01:33:45 I don't see why it's so slow but whatever Apr 04 01:37:12 has anybody got a good guide for a $1 "remove ads" IAP? Apr 04 01:37:33 The TrivialDrive sample app is good Apr 04 01:38:49 Solarnum: this? https://github.com/congenialmobile/TrivialDrive Apr 04 01:38:53 does anyone know what file android studio uses to show on the tab in in the window manager Apr 04 01:39:24 duncannz: It's one of the sample projects you can download alongside the sdk. It's a google sample app Apr 04 01:40:22 topic needs updating as sdk 5.1.1 has been out a few days! Apr 04 01:40:22 looks like that is the same thing Apr 04 01:41:55 where is this file located??? http://i.imgur.com/tyP0nk8.png Apr 04 01:47:05 DadFoundMy: what do you mean by "window manager"? Apr 04 01:47:21 see my screenshot above Apr 04 01:47:36 yeah. dunno though, is that a linux build? Apr 04 01:47:41 yeah Apr 04 01:48:01 my project compiles correctly, the only marker i see is in that tab Apr 04 01:48:05 dunno. my title always shows the currently-editing file. Apr 04 01:49:43 hello Apr 04 01:59:17 hello Apr 04 02:05:46 in android studio there is no way to open a second project, right ? Apr 04 02:06:39 of course go to file and reopen project Apr 04 02:06:41 u will see Apr 04 02:07:00 feathersanddown: I tend to use file -> open recent, and it'll ask if you want to open in "this window" or a "new window" Apr 04 02:10:34 ooohh Apr 04 02:10:44 new window, plop, easy! :) Apr 04 02:20:06 groxx there? last Q about GA for a while i promise :) Apr 04 02:20:20 heh, we'll see :) Apr 04 02:21:12 ok 2 questions (1) are trackers thread safe and (2) What thread do you .send() from ? Apr 04 02:22:37 How do I check for a enter key press? Apr 04 02:22:40 an* Apr 04 02:23:14 g00s: trackers, dunno. I .send from main, never noticed any issues. Apr 04 02:23:32 groxx ok thanks, so no strictmode violations ? Apr 04 02:23:47 IHideKeys: as in, detect a newline, or detect the "done" button on a keyboard? Apr 04 02:23:50 was reading it was writeing to sqlite to queue things up in batches before transmission Apr 04 02:24:04 g00s: lemme enable strict mode :) Apr 04 02:26:44 detect a newline Apr 04 02:27:01 does the onClick work for that also? Apr 04 02:27:58 IHideKeys: do you want to know: A) if the edittext has wrapped to a second line of text (e.g. a long line that overflowed will do this too), B) if the enter key has been pressed explicitly, C) if a newline has been pasted into the text field, D) something else? Apr 04 02:29:39 g00s: bah. too many other violations. sorry, dunno :| Apr 04 02:30:32 groxx lol :D Apr 04 02:31:23 all probably worth fixing eventually, but it's also things like "make sure I have a cache folder", which is a bit tricky to synchronize with "show views from any activity and load from the cache if able" Apr 04 02:33:30 B Apr 04 02:33:36 trying to avoid using a button Apr 04 02:37:57 groxx there is some weird tension between fragments and GA screenviews Apr 04 02:38:18 doesn't seem to really fit that well Apr 04 02:38:31 that does ring a bell, but I don't remember why. my use of it is pretty limited, tbh Apr 04 02:38:44 probably just something I read about how to handle fragments Apr 04 02:39:00 e.g. because of things like drawerlayout leading to a single activity that's conceptually many "screens" Apr 04 02:39:32 yeah - but then cases when several fragments are shown, which is the screen ? Apr 04 02:40:12 I'm going to go with "yes" Apr 04 02:40:21 heh Apr 04 02:40:22 i want to make a notification system after the phone is booted . i want to send the data query from the database from the receiver to the activity. this is how i want to procede . A service with notification inside , start service inside A receiver , and send data from database to the activity. The database will be accessed in the receiver and a data send to the service to filled the content of the notification. is it the right Apr 04 02:43:48 i want to make a notification system after the phone is booted . The content of the notification query from the database . this is how i want to procede . A service with notification inside , start service inside A receiver , and send data from database to the activity. The database will be accessed in the receiver and a data send to the service to filled the content of the notification. is it the right way? Apr 04 02:44:47 hey there, any idea what "Exception raised during rendering: Unable to find the layout for Action Bar." could mean? Android Studio is not rendering any layout previews :/ Apr 04 02:53:41 How do I get a reference to an item in a list? Apr 04 02:53:49 is it LauncherActivity.ListItem? **** ENDING LOGGING AT Sat Apr 04 02:59:58 2015