**** BEGIN LOGGING AT Thu Mar 09 03:00:03 2017 Mar 09 03:53:29 interesting https://customers.microsoft.com/en-US/story/raygun Mar 09 03:55:17 what is the "tapUp" event that gesture listener suppots? Mar 09 03:55:32 i can't seem to find a description of what the gestures actually are Mar 09 04:22:27 raygun! everything's better with lasers :) still, .NET and C# creeping into *nix world Mar 09 04:32:13 hi does any one have info about DFP I am just a beginner and I want to create test account on it... ist saying me to create adsense account through I do not have any website or any blog not on just beginning i got stuck Mar 09 06:14:07 is there any way to hide the hint (not the actual hint, the overlay) behavior on a TextInputEditText? Mar 09 06:18:00 perhaps unset text Mar 09 06:19:21 yeah thats what im doing, but the overlay hint behavior is still there Mar 09 06:20:23 setHintEnabled on the inputlayout seems to do the trick, with a TextWatcher Mar 09 06:20:25 how about just EditText without wrap Mar 09 06:20:32 :) Mar 09 06:20:49 though its still pinkish (inheriting from layout colors) Mar 09 06:20:55 but i can live with it Mar 09 06:21:23 think its a bad idea to have a scrollview inside a cardview item in a rv squ /: Mar 09 06:22:35 rv? Mar 09 06:23:17 if it makes it uneasy to scroll then probably its bad Mar 09 06:23:37 yeah Mar 09 06:23:45 a horizontal would be ok Mar 09 06:23:55 but i need vertical inside vertical Mar 09 06:24:24 can we think about workaround? Mar 09 06:25:41 ill open a dialog instead me thinks Mar 09 06:27:39 ah Mar 09 06:27:53 I think I've got my mu-recursive fix Mar 09 06:53:40 Does anyone have problem to launch the android emulator from Android Studio? I could only launch it from command line by specifying the LD_LIBRARY_PATH Mar 09 06:56:24 aoriani i heard there is some bug with AS 2.3 Mar 09 06:56:31 but it should work with 2.2 ... Mar 09 06:56:53 always gotta wait for .1, but actually jake argues .3 Mar 09 07:01:28 g00s: yep, the same applies to the support library. Mar 09 07:02:15 I dunno how I could collect some log to open an issue Mar 09 07:03:36 the android emulator ship with all the libs it needs ( Qt, opengl, ...), but I failed to find any shell script that sets the library path Mar 09 07:04:23 aoriani: it's not a bug Mar 09 07:04:39 o/ g00s Mar 09 07:04:50 raouuuuuuuuuuul11 Mar 09 07:05:02 tx: its a feature? Mar 09 07:05:09 sup g00s Mar 09 07:05:24 It's neither :P Mar 09 07:05:29 Your LD path may not be what it expects Mar 09 07:05:32 doesn't mean what it expects is wrong Mar 09 07:05:48 things should work out the box Mar 09 07:05:53 * raoul11 away Mar 09 07:06:07 blame the ever dynamic and differing world of linux distros :P Mar 09 07:06:17 i hear you Mar 09 07:06:36 pick one effin convention nd stick to it Mar 09 07:06:48 aoriani is it what 3dom describes? https://www.reddit.com/r/androiddev/comments/5x6adh/android_studio_23_is_now_available_in_the_stable/ Mar 09 07:06:49 aoriani: you can just edit the script that runs the emulator Mar 09 07:06:57 or add the path to studio.sh Mar 09 07:08:08 raoul11 i'm catching up reading old post https://www.reddit.com/r/androiddev/comments/5xz26c/latest_trends_in_android_dev_scene/ Mar 09 07:08:51 i thought there would be more trends, seems just MVP, rxjava, and kotlin. i still wonder how popular kotlin is though, or whether its just a vocal majority Mar 09 07:09:00 im doing mvp now on my new app Mar 09 07:09:05 mvp + rx Mar 09 07:10:05 lol vocal minority Mar 09 07:10:12 raoul11 how is it going ? Mar 09 07:10:22 could be better /: Mar 09 07:10:27 the rx part mainly Mar 09 07:10:35 but i only just started with it Mar 09 07:10:44 seems like black magic to me so far Mar 09 07:10:46 g00s: stackoverflow statistics may help to answer minority question Mar 09 07:11:00 g00s: I download the full tarball. I was using umake , so I thought that the problem was that. Mar 09 07:11:32 I like kotlin\ Mar 09 07:11:40 although top leaders are weird languages, which means those provide most quesitons Mar 09 07:12:45 and how I don't need to go full board to use it Mar 09 07:15:33 g00s u gonna try kotlin? Mar 09 07:16:15 no plans at the moment ... Mar 09 07:18:30 Hello, where can i find the image file after a successful build? (i've build the source for angler_userdebug) Mar 09 09:30:51 Anyone else having issues with android studio visibility on an external monitor? Mar 09 09:31:48 My notebooks monitor is 4k, but the attached monitor is only full HD, and the font looks horrible and washed out. It's not such a big issue when reading stuff in the browser, but on the monitor it's horrible. Mar 09 09:32:50 linux? Mar 09 09:35:18 Syzygy_, which OS? Mar 09 09:35:35 Works well on macOS when switching between the retina display and non-retina UWD external display Mar 09 09:35:44 windows 10 Mar 09 09:37:29 vga/dvi/hdmi? Mar 09 09:37:38 4k vga? :D Mar 09 09:37:49 the external is fullHD Mar 09 09:38:10 so probably has vga port as well Mar 09 09:38:37 when im connected via hdmi to my shitty metz monitor it looks horrible Mar 09 09:38:54 hdmi, no vga port Mar 09 09:39:08 do you have any non hdmi ports? Mar 09 09:39:12 no Mar 09 09:39:19 well, usb c/thunderbolt 3 Mar 09 09:39:50 try replacing le cable Mar 09 09:40:31 I don't think it's the cable, stuff in my browser looks fine on the external monitor (even though the top seems cut off) Mar 09 09:40:45 the monitor only has hdmi? Mar 09 09:41:16 I think the monitor does have pretty much everthing that was available a few years ago Mar 09 09:41:49 display port, dvi, hdmi, vga Mar 09 09:42:19 even a few USB connectors (but not usb 3.0 or c) Mar 09 09:42:24 donno, id play with other cable types or a test with another monitor Mar 09 09:42:35 some monitors are just shit Mar 09 09:42:41 I would need to get adapters for that. Mar 09 09:42:46 like my current metz one Mar 09 09:47:02 q: in a stream of data i am getting from a rest endpoint i get an array which might contain two types of objects. when i want to render them (together) in a recyclerview, is it possible to inflate a particular item.xml on the relevant viewholder? Mar 09 09:47:58 So depending on the "type" in the list of items returned, return a different layout? Mar 09 09:48:14 in each item Mar 09 09:48:22 since the list might contain both types Mar 09 09:49:36 https://guides.codepath.com/android/Heterogenous-Layouts-inside-RecyclerView Mar 09 09:49:37 k Mar 09 09:50:09 Its indeed possible, using the View Holder Pattern will help greatly Mar 09 09:51:41 getItemViewType this method is useful Mar 09 09:56:49 HI Mar 09 09:58:47 howdy Mar 09 09:59:20 i just get 2 Xslate D10 tablets (for work) and any of them get the my sim cards detected, baseband is unknown in the about this tablet menu and no imei code too ... constructor continue to tell me the sim is bad !! :/ , i tried 3 sims for 3 differents carrier in france ... (all of the them works in another 4g tablet or phone) is it possible that the sims are the problem or just a bad install / corrupted file ...or ? !thanks Mar 09 10:00:25 Oh, well this is Android Development channel, for making Apps and such.. try /join #android Mar 09 10:22:29 hello, i must in function "onDestroy" use android log or i can use the app log class? Mar 09 10:44:15 grauzikas2, what? Mar 09 10:45:52 Log is static and visible everywhere Mar 09 11:02:10 Hey! If i have a protected ListView mListView; and set mListView.setAdapter(this);, will this result in a retain cycle? Mar 09 11:04:41 who is this Mar 09 11:05:36 It's me, your cousin Mar 09 11:05:59 damn cuz I don't trust you to not leak memory Mar 09 11:24:43 The idea is to create an app that offers a new digital service that will record a person's carbon footprint by using so-called life cycle analyzes. They demonstrate how the climate impact of a product or service is. Then the person's impact on the climate as the sum of the different products and services climate impact. At the end of each month, send the app including a report on the Mar 09 11:24:44 month's release to every user. The report also hints at the changes that the user can take to reduce their carbon footprint,what do you think of this app Mar 09 11:31:38 Kevin-Oculus stupider things have made money on the play store. Mar 09 11:32:53 yes its true Mar 09 11:33:02 and this is stupid to Mar 09 11:33:38 its not my idea i was reading on internet about this Mar 09 11:58:07 whats a good android development book? I'm a c++ programmer with a degree and have worked as one for 5 years.. buying "core java for the impatient" to start picking up on java, but would like a recomendation for a android book as well :) Mar 09 11:59:26 effective java Mar 09 12:00:41 android changes so quickly that I don't think its possible to cover it in a book Mar 09 12:01:05 maybe a book for specific technology like Rx Mar 09 12:01:19 bloony: honestly, https://developer.android.com/training/index.html Mar 09 12:01:25 if you _really_ want a book, https://pragprog.com/book/eband4/hello-android Mar 09 12:01:52 and official material design resource from google.com Mar 09 12:02:16 android's training site is way more than enough :) Mar 09 12:02:53 it links to material design page Mar 09 12:03:15 ? Mar 09 12:04:24 Ah right.. thanks.. :) Mar 09 12:06:06 tx: clicking on any link ends up in material.io https://www.google.com/search?q=site%3Adeveloper.android.com+material.google.com Mar 09 12:08:18 I am so confused. Are you referring to the one I linked? Mar 09 12:08:21 I don't have that issue. Mar 09 12:09:53 bloony: why you don't continue with c++? Mar 09 12:38:49 squ: I am.. in my hobby projects.. I got worn out two years ago in my previous work.. applying for a android developer jobb now.. want to learn something new :) we'll se where it leads though.. Mar 09 12:39:50 bloony: worn out, don't you think c++ is just bad language? Mar 09 12:40:21 squ: haha.. well.. it wasn't language fault :p Mar 09 12:40:46 how do you know that Mar 09 12:41:08 Do you have a point you're trying to make, squ? Mar 09 12:41:58 I had back surgery a year before that and got back to work a bit too early.. also other things happening in my life at the time :) Mar 09 12:42:35 Roughy: just met a guy quitting c++ after 5 years, point is obvious for me Mar 09 12:44:16 bloony: try out ios, objective-c and objective-c++ is closer to c++ and ios is much better than android Mar 09 12:53:27 I've been doing c++ stuff for about 15 years, it's <3 Mar 09 13:03:28 <3 C++ :) Mar 09 13:08:34 I have one activity 'Dict' and two fragments 'Definition' and 'Search'. The onOptionsItemSelected callback from Definition is not called. Please help https://codepaste.net/dvn4so Mar 09 13:08:35 * Mar 09 13:13:54 I have one activity 'Dict' and two fragments 'Definition' and 'Search'. The onOptionsItemSelected callback from Definition is not called. Please help https://codepaste.net/dvn4so Mar 09 13:14:00 Xmls: I have one activity 'Dict' and two fragments 'Definition' and 'Search'. The onOptionsItemSelected callback from Definition is not called. Please help https://codepaste.net/dvn4so Mar 09 13:14:11 Xmls: https://paste.ofcode.org/xKYPKZBD3x9aPAu5H7iusu Mar 09 13:23:56 I'm working on an accountmanager implementation. Is there anyway to remove an account type from the device? I've removed all apps associated with that login type, however it is still possible to create an account of that type. Any experience with this? Mar 09 13:31:11 Troffel, were you logged in? Mar 09 13:31:32 I logged in to my account Mar 09 13:31:41 and deleted the account, etc. Mar 09 13:32:28 point being that on one device, the authenticate activity that you supply, crashes (only when called from settings, and only on that device) Mar 09 13:32:52 hence it seems something is cached with the account type. So I'm looking for a way to clear that account type from the device Mar 09 13:33:08 did you try rebooting? :) Mar 09 13:33:18 ye :D Mar 09 13:34:06 cleap app cache -> uninstall app -> restart device -> compile new build -> attempt to create account through settings -> crash Mar 09 13:34:21 if I launch the activity from the app, it works fine. So it has to be something cached Mar 09 13:35:12 if you uninstall the app now, does it still appear in the settings? Mar 09 13:35:15 oh Mar 09 13:35:26 um, maybe it wasn't uninstalled for all users? because that can happen.. Mar 09 13:35:35 use adb to uninstall? Mar 09 13:35:53 haven't tried through uninstall. But I'm the only user on the device Mar 09 13:36:02 that doesn't matter Mar 09 13:36:04 through adb * Mar 09 13:36:22 what other users would that be? sorry if that's a bu Mar 09 13:36:26 basic question * Mar 09 13:36:49 it might still be installed but just not enabled for your user Mar 09 13:37:40 or maybe another test app (using the login) that I had previously installed, has somethign cached Mar 09 13:37:50 I'm not 100% sure how to reproduce that but I've had it happen once or twice Mar 09 13:38:14 ok, ill look through some adb commands to purge a phone for a user type / package name Mar 09 13:38:26 just adb uninstall your.package.name Mar 09 13:39:08 I'm pretty sure the account types in settings should be based on the account-authenticator xml thing Mar 09 13:39:18 yes, but I meant to also do it for the other testapp I had installed previously, to purge in cached content? Mar 09 13:39:21 but it's been years since I tried writing stuff like that Mar 09 13:39:27 it is Mar 09 13:39:53 deleting that app might have passed the "ownership" to the other app I had installed that utilized same login Mar 09 13:40:00 anyway, thanks a bunch Zharf Mar 09 13:40:05 interesting Mar 09 13:40:32 seeing as the first app installed with a given login type, "owns" the authentication process Mar 09 13:53:48 Are IntentServices async by default Mar 09 13:53:59 Or should I spawn a thread in the handling body when doing network operations? Mar 09 13:55:40 Actually... Maybe this isn't the best way to do this. Mar 09 13:55:48 Looks like requests are processed on a separate worker thread Mar 09 13:56:33 Basically, what I'm trying to do is make a little background process that checks for an internet connection, whether by pinging, asking system services, whatever. And when it has that pushing either one or many JSON objects up to a website Mar 09 13:56:42 Would IntentService be right for this? Mar 09 13:59:15 Seems reasonable Mar 09 14:00:19 If you do not intend to queue up multiple of these tasks, you may just want to go with an asynctask or something Mar 09 14:02:11 I'm fairly new to android, so forgive me if my practice is bad. But what I have so far is data slowly being accumulated from the user and placed into shared preferences. Mar 09 14:03:13 On the last pane, when that's done I'm going to serialize the data into JSON and put it into something that isn't shared preferences, whether it be database, plain file, whatever. Then I'm going to have a background process going over that thing and pushing things to the cloud/removing them from disk as and when it can Mar 09 14:03:31 asynctasks just mess things up :) ... Mar 09 14:06:23 Redrield, the thing with intentservices is that they'll run your code in a thread and when the method returns and there's no more intents in queue it will destroy the service... so doing anything asynchronous in there might not be completed Mar 09 14:07:12 intentservices work ok when you can do everything in it synchronously... or at least wait for things to complete before returning from the method Mar 09 14:08:04 if im following mvp and my model fetches, but I need to transform it, should i be doing that in the model or presenter? Mar 09 14:08:10 fetches data* Mar 09 14:08:19 if that's doesn't work for you, use Service instead, but then you'll need to manage your own threada Mar 09 14:08:46 ... sorry gor typos... I hate on sceeen keyboards Mar 09 14:08:52 dont think there're any rules for that KukOrkan Mar 09 14:09:16 the main issue is to seperate the view and model Mar 09 14:09:25 -issue+goal? Mar 09 14:09:25 hey, currently starting to develop with android. Already get in trouble with the developer.android.com website... For example want to play mp4 videos from the website (https://developer.android.com/training/constraint-layout/index.html#add-a-constraint) but if i click play nothing happens. Used firefox,edge,internet explorer... Some suggestions? Mar 09 14:09:45 youtube? Mar 09 14:09:53 raoul11, hmm ok, i thought there was. im new to this pattern Mar 09 14:10:00 i am as well Mar 09 14:10:01 raoul11, thanks Mar 09 14:10:44 id say the model though KukOrkan Mar 09 14:11:20 the way i implement my presenter is that it takes a viewmanager, calls the model and callbacks back to the viewmanager Mar 09 14:11:52 raoul1, yeah thats what im getting at now, having some issues with unit testing my presenter Mar 09 14:12:02 gonna move my data transformation methods to the model Mar 09 14:12:49 what transformation are you doing? Mar 09 14:13:15 is that a rest model btw? Mar 09 14:13:28 KukOrkan: tbh i dont do any ut Mar 09 14:13:31 yet /: Mar 09 14:13:41 basically sorting json data Mar 09 14:13:55 well in this case, data returned from firebase Mar 09 14:14:08 mhm Mar 09 14:14:30 using retrofit or the realtime db? Mar 09 14:15:03 realtime db Mar 09 14:15:06 event listeners Mar 09 14:15:48 k Mar 09 14:30:13 hi Mar 09 14:30:42 i am setting swipeRefresh.setRefreshing(false) but the SwipeRefreshLayout is not hide, anybody can help me ? Mar 09 14:36:34 hey guys Mar 09 14:37:22 anyone ever work with InfoWindowAdapter? is there a way I can treat it or work with it like I work with other adapters? meaning I want to apply different rules to the view of each object depending on the contents of the objects im working with (Markers) Mar 09 14:38:09 i am setting swipeRefresh.setRefreshing(false) but the SwipeRefreshLayout is not hide, anybody can help me ? Mar 09 14:49:46 Odaym, I think the only way is to have the Marker's tag contain enough information for the adapter Mar 09 14:49:55 yea thats what im doing now Mar 09 14:49:57 in the Snippet Mar 09 15:10:48 I've updated to the latest Android Studio -> 2.3 and gradle 3.3 and I'm getting compiling errors for cannot find symbol class NotNull and cannot find symbol class MultiDexApplication Mar 09 15:10:56 anyone else run into AS 2.3 issues? Mar 09 15:12:32 I had to invalidate cache and resync project after updating gradle Mar 09 15:12:48 I've already tried that Mar 09 15:34:03 I fixed my databinding issues by adding apt 'com.android.databinding:compiler:2.3.0' to gradle Mar 09 15:34:36 also had an annotation issue in AS 2.3 with NotNull, but AS resolved it with a static import Mar 09 15:49:37 hi guys Mar 09 15:50:16 anyone knows how can i raise an Retrofit HttpException on unit tests using Mockito? Mar 09 15:51:27 there is a thenThrow Mar 09 15:51:30 I need something when(presenter.callNetwork()).thenThrows(new HttpException(iTNeedsSomeResponseObject) Mar 09 15:52:16 I don't know how to create this retrofit ResponseObject Mar 09 17:06:59 raoul11: omg so sick Mar 09 17:07:41 u2? Mar 09 17:08:24 yeah Mar 09 17:08:26 sux Mar 09 17:09:16 yeah it does Mar 09 17:09:27 my throat is also finished from all the yelling yesterday Mar 09 17:11:47 i have strep Mar 09 17:11:53 accirding to the Dr Mar 09 17:11:55 ewww Mar 09 17:12:03 fevere nd shit? Mar 09 17:13:26 no fever, but I've been in bed all day Mar 09 17:13:47 sux Mar 09 17:14:00 mines bad but not that bad Mar 09 17:53:17 hello dev`s.. i am stuck in a lil problem! in init.d script.. if -e use for if file exists.. what token should i use too check if its not exisit? if !-e ? :D Mar 09 18:16:04 I am trying to get this code working, but I got some error on missing things in setPermissionForM at the if statement, how can I get it to work? : http://stackoverflow.com/a/39699958/3472305 Mar 09 18:19:08 https://developer.android.com/training/permissions/requesting.html Mar 09 18:20:32 anyone here using C with android ndk? Mar 09 18:22:29 If you have a question concerning it that I'm confident I can answer, then yes. Otherwise no. Mar 09 18:23:06 Roughy: I just want to know if I can run C without needing to make wrappers from C++ Mar 09 18:28:10 Ah, unfortunately not qualified to provide any insight on that Mar 09 18:36:40 Murii: the NDK is actually a C API, not C++ Mar 09 18:36:42 So yes Mar 09 18:36:55 I just downloaded it and saw that Mar 09 18:37:02 awesome stuff Mar 09 18:42:51 indeed, it's all java jni Mar 09 18:43:02 and pretty much all ffi bindings always default to C Mar 09 19:04:29 Hey is this an error on the Android Dev site? https://developer.android.com/topic/libraries/data-binding/index.html#method_references Mar 09 19:04:43 Handlers vs MyHandlers Mar 09 19:05:48 probably Mar 09 19:06:27 fun Mar 09 19:06:46 tryign to learn about databinding and they throw me a curveball Mar 09 19:07:04 btw, DataBindingUtil vs Butterknife? Mar 09 19:07:12 did it make ButterKnife obselete? Mar 09 19:09:55 Tricknology: filed a bug and assigned it to one of our tech writers: https://code.google.com/p/android/issues/detail?id=241298 Mar 09 19:10:16 Oh cool :) Mar 09 19:10:41 (there's a "Developer documentation" template in b.android.com if you find other things) Mar 09 19:10:58 alright good to know. Thanks! Mar 09 19:11:26 Is this just for the documentation or bugs in the Android SDK itself? Mar 09 19:12:32 I think I came across a weird one with PercentRelativeLayout, have yet to create a self-contained test case for it though Mar 09 19:19:04 https://techcrunch.com/2017/03/09/new-version-of-google-app-engine-supports-all-programming-languages/ Mar 09 19:19:49 Tricknology: follow http://tools.android.com/filing-bugs for more general Studio/SDK bugs Mar 09 19:20:15 They'll end up in the same place, but that'll point you to the right category Mar 09 19:20:30 thank you Mar 09 19:20:38 There's another category for issues with the platform APIs too, which is I think what a layout issue would be Mar 09 19:20:47 (or maybe support library) Mar 09 19:40:12 g00s, that's actually interesting Mar 09 19:40:53 I generally like appengine Mar 09 19:41:00 easiest turnkey app hosting Mar 09 19:41:09 have never found anything easier Mar 09 19:41:27 docker, etc. require finding a suitable provider and base image to work with Mar 09 19:48:03 isnt stuff like this discouraged because it's not very testable? Mar 09 19:48:03 android:visibility="@{age < 13 ? View.GONE : View.VISIBLE}" Mar 09 19:48:44 yeah, data binding is bad in general Mar 09 19:48:50 it's basically good for non-programmers Mar 09 19:48:54 lol Mar 09 19:49:05 in the idea that "editing xml is easy" and "non programmers can do it" Mar 09 19:49:10 i thought the point was to get rid of the boilerplate findViewById stuff Mar 09 19:49:24 that's one thing that databinding can do for you Mar 09 19:50:51 so I'm looking at this in a MVVM context.. databinding could actually be a good thing here? Mar 09 19:51:04 I dunno mvvm Mar 09 19:51:52 Model View Binder is a better name for it I think Mar 09 19:52:16 or Model Binder View as it better represents the flow Mar 09 20:09:12 So it’s almost like you have a model, a view, then a presenter? Mar 09 20:09:33 That presents the content from the model onto the view? Mar 09 20:13:21 Hey all. I have a weird issue: I'm writing what boils down to a stopwatch. User hits start -> save current time in milliseconds to shared pref. In onCreate I load that value from sharedprefs. The issue I'm having is the time drifts. I started my stopwatch and googles stopwatch(google stopwatch brings up a js stopwatch) at the same time and ~12 hours later the stopwatches are almost 1 min apart Mar 09 20:13:56 They're reading the actual time Mar 09 20:14:39 I've tried System.currentTimeMillis(), System.nanoTime(), and SystemClock.elapsedRealtime() Mar 09 20:15:05 Dagmar, What's reading the actual time? Mar 09 20:15:14 The first one as I understand it Mar 09 20:16:00 The last one is _not_ the one to use Mar 09 20:16:29 That only represents seconds since the system was booted, and not all of *those* seconds are created equally Mar 09 20:16:53 Dagmar, Really? I must have grossly misunderstood what I read. I thought SystemClock was best to use as it isnt affectwed by user timechanges Mar 09 20:16:55 THey're not actual seconds but "seconds" that the kernel has been running Mar 09 20:17:09 Hmm Mar 09 20:17:14 Interesting! Mar 09 20:17:16 There's a whole bunch of ways that latter one deviates from an actual clock Mar 09 20:17:32 You want _epoch time_ which is what the first function gets you Mar 09 20:17:43 Epoch time is not affected by user timechanges Mar 09 20:17:51 I read some article how java uses different methods to get the time for different systems. But I did not really understand the differences. Mar 09 20:18:21 If you want something that won't simply be _wrong_ if the user changes the clock, you're out of luck. There's no real way to track that Mar 09 20:18:44 ...but epoch time sources are *always* supposed to be the number of seconds since Jan 1 etc Mar 09 20:19:01 If you got a wrong value from one, it's on you Mar 09 20:19:02 Dagmar, Right, epoch time. I thought that SystemClock was epoch since boot and was not affected if the user changed the time, as android still have the exact time since boot to use as reference Mar 09 20:19:15 It is, but not all "seconds" are created equally Mar 09 20:19:39 It's subject to kernel accounting, which happens in timeslices _anyway_ and one or to can easily get lost here and there Mar 09 20:19:45 s/to/two/ Mar 09 20:20:07 Dagmar, Interesting, thanks! Mar 09 20:20:22 Does anyone know if you can use vector assets in a menu resource? Mar 09 20:20:39 My app doesn't really need super duper accurate time, but damn is this annoying me! Mar 09 20:20:59 Kernel: Sort of like you can poll the counters for system load accounting, and the ticks will *mostly* add up to 100 per core Mar 09 20:21:03 Dagmar, So System currentmillis is the best way to go? Mar 09 20:21:08 It's the only way to go. Mar 09 20:21:46 Dagmar, Ah. Awesome, I appreciate the help :) Mar 09 20:22:13 Those ticks won't _always_ add up to precisely the value they're supposed to tho', because CPUs are lots more complex than allow that anymore and you can actually consider a signficant deviation from the correct number a sure sign that the system is having issues Mar 09 20:22:24 I'm going to have to do some more testing and see if I'm somehow rounding ms's and over long periods is causing the drift. Mar 09 20:23:05 Are you doing something insane like repeatedly writing to sharedPrefs storage 1,000 times a second? Mar 09 20:23:48 Also, on the subject of time, is there any method of getting pure milliseconds since epoch without taking system time into account? Mar 09 20:24:04 That is to say the user set time Mar 09 20:24:12 Multiplying epoch seconds by 1,000 Mar 09 20:24:14 ...and madness. Mar 09 20:24:38 There *is* a difference but unless you're putting satellites into orbit it's not worth caring Mar 09 20:25:05 with a sgs4 Mar 09 20:25:05 Where do you get that? System.currentMillis returns the user defined time, right? Mar 09 20:25:20 It returns the _time_ Mar 09 20:25:27 What is supposed to be THE time Mar 09 20:25:35 There's no "user defined" to that Mar 09 20:26:01 Any time zone stuff is supposed to take place _outside_ of that Mar 09 20:26:14 I thought users could change it by switching time zones, is it just GMT? Mar 09 20:26:18 Dagmar, I only write to prefs when the user starts, pauses, or resumes. Mar 09 20:26:18 THe system keeps it's own seconds uptime counter because *that* one always moves forward Mar 09 20:26:23 it is UTC Mar 09 20:26:34 Okay, cool Mar 09 20:26:47 konk3r: Changing the system clock because someone changed time zones is some of that Microsoft insanity Mar 09 20:27:05 Dagmar, that hasn't been the case since DOS Mar 09 20:27:15 (and by extension Windows through Win95/ME) Mar 09 20:27:24 If someone was so foolish as to write say, a database server or anything with any kind of time-based locks, all hell breaks loose if time appears to go backwards Mar 09 20:27:30 pfn: Bullshit. They still do it Mar 09 20:27:35 no, they don't Mar 09 20:27:45 It makes me want to stroll into Microsoft offices with a whole box of hand grenades Mar 09 20:27:47 I am doing some math on the time. stuff like now - millis and things of this nature. Mar 09 20:28:26 They were absolutely still storing local time in the RTC as of at _least_ Windows 7 Mar 09 20:28:37 I've just quit caring so I've not looked to see if Win10 still does it Mar 09 20:28:41 It's come up time and time again Mar 09 20:29:01 If someone wants to dual-boot a Windows server and a Linux server, they're just going to have to get used to calling ntpd on boot Mar 09 20:29:11 perhaps you're right, I usually don't care because linux usually has a clockadj thing at bootup Mar 09 20:29:15 if I wanted Mar 09 20:29:53 The last time I tried to seriously beat it into behaving was with XP, and it invariably resulted in EVENTUALLY Windows freaking out and managing to be timezone-offset wrong Mar 09 20:30:04 I gave it a go again with Win7 Mar 09 20:30:18 Less than a month later it fucked up and I just let it go back to doing what it damn well pleases Mar 09 20:30:37 ...and it seems to *always* catch the Windows admins off guard Mar 09 20:31:29 Due to a confluence of joy and puppies as recently as three years ago I had to tell an admin to set their damn hardware clocks properly so their servers could actually rejoin the AD domain Mar 09 20:32:33 Microsoft's excuse when questioned about "why do you keep doing this dumb thing" has pretty much always been "DOS-compatibility" although they stopped even answering questions about it with Win7 Mar 09 20:33:02 Cuz some idiot's ancient copy of Lotus 1-2-3 running in a DOS box will freak out and break when it directly accesses the system clock to get the time Mar 09 20:33:21 well... Microsoft has the most amazing history of backward compatibility Mar 09 20:33:31 far beyond what *anyone* else has ever done Mar 09 20:33:38 Yeah and in this case it's a terribly broken practice Mar 09 20:33:44 * pfn shrugs Mar 09 20:33:57 it doesn't matter as long as Windows is the only thing running on a particular system Mar 09 20:33:57 I wouldn't call it a useful thing to be supporting when the result is the WRONG solution Mar 09 20:34:06 Actually, it'll matter if you reboot at the wrong time Mar 09 20:34:25 ...and can matter if you restore a backup, or crash at a certain time Mar 09 20:34:40 about the only wrong time is system shutdown during dst change Mar 09 20:34:47 Like, all the reasons everyone else stores a wholesome UTC epoch seconds value in the hardware clock Mar 09 20:34:53 and anything internet connected will fix itself relatively shortly Mar 09 20:35:16 Unless it gets it's time server information as a result of joining the AD domain which it can't exactly do if it's several hours ahead of reality Mar 09 20:37:10 I am not a windows admin. I just know these things because they've _all_ wound up as a trouble ticket on *my* desk Mar 09 20:37:17 ...becasue I was the asshole responsible for DNS and NTP Mar 09 20:38:04 It takes real balls to send me a trouble ticket about a time server that's got an antenna lead running up to the roof Mar 09 20:38:31 if that box could *be* wrong, most of the conditions that would lead to it would have flaming computers falling from the sky. Mar 09 20:41:25 bc Mar 09 21:15:05 woop, editing Android Studio source is kinda fun Mar 09 22:45:53 is Jack fully supported yet? Mar 09 22:46:08 trying my hand at Lmbda functions but I think one of my dependencies is not allowing it Mar 09 22:46:16 Error:Could not get unknown property 'classpath' for task ':app:transformJackWithJackForUs_productionDebug' of type com.android.build.gradle.internal.pipeline.TransformTask. Mar 09 23:16:46 that's a gradle problem... Mar 09 23:26:09 damn gradle.. Mar 09 23:28:58 jack sucks, anyway Mar 09 23:29:03 there's no reason to use it over retrolambda Mar 09 23:29:14 well, except for using default methods on android 7+ Mar 09 23:29:38 I think that's the only feature that jack enables vs. retrolambda Mar 09 23:30:28 streams too Mar 09 23:33:58 streams only require lambdas Mar 09 23:34:06 they're a library feature otherwise Mar 09 23:34:19 unlike lambdas and default methods Mar 09 23:34:32 which are language features (and subsequently VM) Mar 09 23:36:23 hmm I'll have a look.. retrolambda you say? Mar 09 23:36:35 yaya internet is down! Mar 09 23:36:44 (retrolambda also supports default methods, but using a compile-time hack) Mar 09 23:59:09 Why would Android be auto focusing on the first EditText in a layout when the activity is started. Mar 09 23:59:18 It's happened on 2 separate occasions in different activities Mar 10 00:00:32 that's pretty typical, something within your view hierarchy is pretty much always focused Mar 10 00:01:25 The EditTexts aren't even the first things defined in the layout though... Mar 10 00:01:36 I have a spinner and a few TextViews before that Mar 10 00:05:21 Is there at least a way I can close the keyboard so that it doesn't obstruct the UI Mar 10 00:33:01 wonder why the 400 concurrent connection limit https://news.ycombinator.com/item?id=13831370 Mar 10 00:54:10 I think I should probably try to use some service like that. Mar 10 00:54:38 g00s: thanks for the links that you post Mar 10 00:54:48 I always learn something Mar 10 01:50:23 Melatonina ;) Mar 10 01:50:54 c# 7 looks nice too https://news.ycombinator.com/item?id=13834511 Mar 10 01:57:22 Hiiiiiii ALLLL !!! Whats Appp :) Mar 10 01:58:38 I got the app working - the ble background and timer(s) etc - bunch of things just had to sit back and digest the requests - look at what I had on hand and cross fingers Mar 10 01:58:49 just wanted to say Thank you Mar 10 02:02:04 so here's anohter question - if you have a code base and you resell it to another client just need to change branding - [1] any issues with this on the playstore ? [2] whats' the best approach version control wise - ie. Fork the codebase or Brnach it or clone and create new repo ? Mar 10 02:22:27 fels: you can probably just use different product flavors for the different clients. can change branding etc per flavor Mar 10 02:22:54 teasp00n: whats a product flavour ? Mar 10 02:23:55 https://developer.android.com/studio/build/build-variants.html is this it ? searched and found this - never knew about it - Mar 10 02:28:54 fels: yeah its part of that Mar 10 02:30:27 thanks teasp00n - I'll try to reseach it and problbly have a bunch of questions :) Mar 10 02:31:37 anyone here work in a software shop - I need to know about agreements that hand over IP - I've been told Work for hire is the term I am looking for Mar 10 02:46:12 Hm... Mar 10 02:46:26 For some reason, the scale on my RelativeLayout is messed up. Mar 10 02:46:52 What I have is a RelativeLayout inside a ScrollView. And when I try to center in parent, the placement is too high Mar 10 02:47:02 either center in parent or center horizontally Mar 10 02:49:06 s/horizontally/vertically Mar 10 02:59:24 Redrield: does the relativelayout match parent for height? do you have a collapsing toolbar when you scroll? Mar 10 02:59:53 Not sure if you've used anko, but I can give the code that I'm using **** ENDING LOGGING AT Fri Mar 10 03:00:03 2017