**** BEGIN LOGGING AT Tue Dec 29 02:59:58 2015 Dec 29 03:18:07 yeah pfn Dec 29 03:34:01 Do I need to use breaks in my Switch statement if the statement is an intent to a new Activity? Is it bad practice to just not use any breaks in this switch? Dec 29 03:34:47 I can't think of any reason why the break would be needed after every case Dec 29 03:41:57 The break stops the execution at that case. If you dont use a break, all the cases would run afaik. Dec 29 03:45:56 pfn: it is? Dec 29 03:46:16 liuwenhao, it is always bad practice to omit break Dec 29 03:46:39 liuwenhao: if you use return then I wouldn't bother Dec 29 03:46:53 and yes, return yields unreachable code Dec 29 03:46:54 but it is bad practice to have multi-return statements Dec 29 03:46:57 as a static error Dec 29 03:47:54 orbyt_: my reasoning was that launching a new activity would kill the code execution anyways. But I guess there is no downside to having the break (other than a bit more text in the switch) Dec 29 03:49:08 liuwenhao: i dont think it instantly kills the execution Dec 29 03:49:41 pfn: there is probably an annotating you can use to get rid of that Dec 29 03:49:48 annotation* Dec 29 03:50:15 orbyt_: that's good to know, will save me some debugging in the future :) Dec 29 03:51:13 frack :/ Dec 29 04:16:13 i have the typical CoordinatorLayout + AppBar + TabLayout + ViewPager setup, but in dismay to see the AppBar scrolling when the ViewPager's content doesn't have enough material to scroll; is there a way to fix that? i only want stuff scrolling if it can't fit on the screen Dec 29 04:20:51 capella-s3 you need a new router ! Dec 29 04:21:33 capella-s3 http://thewirecutter.com/reviews/best-wi-fi-router/ Dec 29 04:22:16 g00s_ I just changed the default channel for both bands, noticing a new neighbor ssid... so far no disconnecting Dec 29 04:25:46 * g00s_ wonders if LTE-U will screw everybody up Dec 29 04:26:06 yah, thats, um, interesting tech Dec 29 04:37:35 hm, for some reason my viewpager doesn't show its fragments once in a while ... wonder if its one of those things solved by posting a runnable Dec 29 04:41:20 because android Dec 29 04:42:24 eh i wonder ... i'm setting up my viewpager in fragment onCreate, maybe thats not a good place Dec 29 04:43:11 oh crap, does PagerAdapter handle child fragment properly Dec 29 04:43:46 i mean, nested fragment Dec 29 04:45:25 pass child fragment manager to fragment pager adapter Dec 29 04:46:45 as for scrolling, update the behavior Dec 29 04:48:08 pfn ah thanks, i passed the wrong fragmentManager ... saved me tons of time Dec 29 04:52:27 better answer: quit using nested fragments Dec 29 04:57:11 so far i haven't run into any problems with nested fragments - turns out that Fragment.requestPermissions thing is broken for ALL fragments Dec 29 04:57:54 as well as DialogFragments, etc Dec 29 04:58:29 but i also realize, i'm probably skating on thin ice - so i don't try to be overlay fancy either :D Dec 29 05:01:21 is it, that's interesting Dec 29 05:28:50 guys im dynamically adding textviews in a recyclerview and with ~10 or so added textviews per item its causing some decent lag. Any tips for how to do something like this without the lag? Dec 29 05:35:09 this thing is pretty cool http://www.sqltabs.com ... always amazed what people build with electron Dec 29 05:35:42 nift charts from queries, rendered with d3.js .. Dec 29 06:35:41 CedricBeust_ pretty interesting! https://news.ycombinator.com/item?id=10803775 Dec 29 06:47:52 http://arstechnica.com/information-technology/2015/12/the-app-ocalypse-can-web-standards-make-mobile-apps-obsolete/ Dec 29 06:53:57 g00s_: nope. I used to think that, was really big into Meteor.js. Web based apps arent yet performant enough on mobile. Dec 29 06:54:49 i think its inevitable, pretty sure google wants to go in this direction too Dec 29 06:55:46 g00s_: ! Dec 29 06:55:54 hey thepoosh ! Dec 29 06:55:59 sup man? Dec 29 06:56:28 g00s_: 5-10 years, maybe. Actually I really like what microsoft is trying to do with cross platform apps Dec 29 06:56:48 orbyt_ closer to 5 years :D Dec 29 06:56:54 thepoosh no much, by you ? Dec 29 06:57:05 ok i guess Dec 29 06:57:25 lol every time i read androidweekly now, i think, " i wonder what thepoosh thinks about this" Dec 29 06:57:33 this week I thought Dec 29 06:57:37 "meh" Dec 29 06:57:42 i thought so haha Dec 29 06:58:08 saw this yesterday: http://inthecheesefactory.com/aca/The%20Android%20Cheese%20Sheet%20rev%204.pdf Dec 29 06:58:15 has some issues but overall it's ok Dec 29 06:58:41 yeah saw that, forgot which issues but yeah - some Dec 29 06:59:13 g00s_: Yeah saw that, I don't understand why people get excited about that Dec 29 06:59:24 the singletone pattern is wrong Dec 29 06:59:29 CedricBeust_ i understand why people get excited :D Dec 29 06:59:38 Why? Dec 29 06:59:53 many developers want to see Java 8 classpath Dec 29 06:59:59 thepoosh: The singleton pattern is just fine Dec 29 07:00:14 What does this commit have to do with Java 8? Dec 29 07:00:24 stuff like nio.2 and whatever else is missing / incompatible Dec 29 07:00:47 CedricBeust_: it's bad for multithreading Dec 29 07:00:50 It's just Google replacing Harmony little by little and using software that's not encumbered by license concenrs Dec 29 07:00:54 thepoosh: It's not. Dec 29 07:00:56 needs to have a double lock Dec 29 07:01:13 thepoosh: Only if you implement your singletons with statics. Don't do that. Nothing wrong with singleton. Dec 29 07:01:15 and I would prefer working with something like Dagger's @singletone Dec 29 07:01:30 CedricBeust_: I'm not saying singletone pattern is bad Dec 29 07:01:32 thepoosh: Exactly. The problem is with statics, not with singletons. Dec 29 07:01:44 I was talking about what's in the doc Dec 29 07:01:58 thepoosh: http://beust.com/weblog/2011/03/10/rehabilitating-the-singleton-pattern/ Dec 29 07:02:42 CedricBeust_: this is what's in the doc: http://imgur.com/AP80eOW Dec 29 07:03:10 That's just bad code, written in times when we didn't know better. Dec 29 07:03:18 agreed Dec 29 07:03:52 when you use statics it should look more likethis: https://www.reddit.com/r/androiddev/comments/3yf4z4/great_android_cheat_sheet/cyd4kr2 Dec 29 07:04:22 Back in the days, I probably encouraged the Android team to write code like this. Times have changed. Singleton is a fine and unavoidable pattern, just learn to implement it propery Dec 29 07:04:35 and this is another way to implement with statics: https://www.reddit.com/r/androiddev/comments/3yf4z4/great_android_cheat_sheet/cye3as0 Dec 29 07:05:16 CedricBeust_: most of the times, singeltones have too much information stored in memory Dec 29 07:05:22 IMO Dec 29 07:05:47 Nothing wrong with that. The complications come when the singletons is mutable. If it's immutable, store as much data in it as you want, doesn't matter. Dec 29 07:06:19 If it's mutable, lock judiciously and as fine grained as possible Dec 29 07:21:21 http://pastebin.com/t8iNkgyW anyone know why the categories are being split into individual characters instead of the whole strings? Dec 29 07:23:11 for some reason my animated values aren't animated, but jump directly to the target value. https://pastee.org/fbnbh Dec 29 07:24:50 hang on. might be because of the | char being a control char in regex according to stackoverflow Dec 29 07:27:29 the interesting part is just in the onCreate method.. Dec 29 07:28:24 did you try to increase the duration or put it elsewhere than onCreate? Dec 29 07:28:36 it might start and finish before your activity is fully visible Dec 29 07:31:10 adq: it gets retriggered on every page change.. Dec 29 07:31:19 adq: so the animations should run every second.. Dec 29 07:31:37 adq: i might move the construction and starting to onResume though. Dec 29 07:32:35 adq: and i set the duration to 1000 ms for precisely that reason [to be able to clearly see if it's running or not] Dec 29 07:33:24 maybe it's because you trigger both: Dec 29 07:33:24 color_activate_animator.start(); Dec 29 07:33:25 color_deactivate_animator.start(); Dec 29 07:33:40 i don't see any delay between them Dec 29 07:34:06 they are applied to different bullets.. next_bullet_index and previous_bullet_index should apply to different bullets.. Dec 29 07:34:16 maybe i'm going about this the wrong way alltogether though :) Dec 29 07:34:23 then i have no clue Dec 29 07:34:30 maybe i should explain what i'm trying to achieve.. Dec 29 07:35:12 the screen has a view pager that displays different pages [three altogether]. below the view pager there is some content that is not moved.. Dec 29 07:35:21 and three bullets that indicate which page is shown Dec 29 07:35:22 I thought I read somewhere that I could have AS convert any VectorDrawable files to PNG files at build time. Is this possible? Dec 29 07:39:41 surf2b1 yea Dec 29 07:39:51 the requirement is that the bullets fade to the active and inactive color.. Dec 29 07:41:19 g00s_, can I just drop a vectordrawable in res/drawable and it will generate PNGs? Dec 29 07:41:57 surf2b1 should put it in res/drawable-anydpi-v21 Dec 29 07:42:09 adq: thanks for taking a look though.. appreciated.. Dec 29 07:42:30 crap, my PreferenceFrafmentCompat isn't showing up in a viewpager :| Dec 29 07:42:55 is there something special about this thing ... Dec 29 07:44:42 Where do I wire up the UI elements of a fragment, onCreateView or onViewCreated? Dec 29 07:45:28 g00s_, so I dropped a vector drawable in /res/drawable-anydpi-v21 but in the ouput APK I don't see any generated PNG files. Dec 29 07:45:43 your minSdk has to be < 21 Dec 29 07:45:59 and using gradle plugin 1.5.+ Dec 29 07:46:16 g00s_, and ughh... I tried the support preference lib. It didn't seem worth it to me for backporting material design. Dec 29 07:46:59 Nick007: wire up as in registering listeners, etc? Dec 29 07:47:23 g00s_, minSdk is 14 and gradle is 1.5.0 Dec 29 07:47:27 final RecyclerView getListView() lol Dec 29 07:47:39 surf2b1 hmmm Dec 29 07:50:30 Where do I wire up the UI elements of a fragment, onCreateView or onViewCreated? Dec 29 07:50:47 oh sorry fps, I didn't see that Dec 29 07:50:59 yes, as in setting listeners Dec 29 07:51:34 surf2b1 hm PreferenceFragmentCompat gave everything ridiculous jumbo sized font Dec 29 07:51:41 i swear Dec 29 07:51:58 it's so ugly. You need to use the 14+ lib too I believe Dec 29 07:53:04 g00s_, I decided to hold off on using the pref lib after reading this answer: http://stackoverflow.com/a/32108439/1048340 Dec 29 07:55:27 surf2b1 omg Dec 29 07:55:41 almost like they never tested it :D Dec 29 07:56:05 i guess ian thinks his google+ post constitutes valid documentation too Dec 29 07:56:27 Who's Ian? The next Chris Banes? :P Dec 29 07:57:51 surf2b1 https://plus.google.com/+AndroidDevelopers/posts/9kZ3SsXdT2T :D Dec 29 07:58:28 yeah this is not worth it Dec 29 08:02:57 g00s_, the vectordrawable to png at build time worked when I moved it from drawable-anydpi-v21 to drawable. The vector is placed in anydpi-v21 at build time. :) Dec 29 08:11:57 O.o Dec 29 08:12:05 anyone here? Where do I wire up the UI elements of a fragment, onCreateView or onViewCreated... by wiring I mean setting listeners and stuff Dec 29 08:12:23 surf2b1 https://commonsware.com/blog/ Dec 29 08:13:31 Nick007, It *should* work in both places but I prefer onViewCreated Dec 29 08:16:23 surf2b1, BNR taught me to do it in onCreateView, Android Developer Guide says to do it in onViewCreated Dec 29 08:16:39 surf2b1, I guess I'll follow you and prefer onViewCreated Dec 29 08:17:58 surf2b1, I'll inflate the view in onCreateView and wire up the stuff in onViewCreated... is that practice fine? Dec 29 08:18:09 Nick007, yep Dec 29 08:18:42 surf2b1, thanks a lot Dec 29 08:53:03 hi to all! how can i record phone call programmatically ? Dec 29 08:54:01 you can't, sorry Dec 29 08:54:23 but i see a lot of apps in google play with same functional Dec 29 08:54:32 oh? which ones? Dec 29 08:55:27 storkme, https://play.google.com/store/apps/details?id=com.smartcoach.callrecorder01 this one for example Dec 29 08:56:25 I guess they just use the phones microphone Dec 29 08:58:10 in that case i take it back i have no idea how it works Dec 29 08:59:17 decompile the apk and check how they do it :p Dec 29 09:01:09 probably they can capture it at hardware level but i doubt it will work with the latest devices Dec 29 09:01:46 maybe more than 1 app can listen to the microphone Dec 29 09:39:07 It works on S5 at least! Dec 29 09:43:11 is there a view that i can use that has two states and uses two drawables for those two states and which animates between those drawables on state change? Dec 29 09:45:04 transitiondrawable doesn't let me query the state, so i have to maintain it myself Dec 29 09:45:25 if i use a valueanimator for the color of a shape in a shapedrawable i have to maintain the state myself, too Dec 29 09:49:32 "state" meaning that i have a number of these things and i need to fade in one, and fade out the last one, which is a bit of a pain Dec 29 09:50:09 As4xk, what works ? Dec 29 09:50:40 Recording conversations. It was discussed earlier :p Dec 29 09:51:21 As4xk, what is S5 ? Dec 29 09:51:28 fps, what about ViewSwitcher Dec 29 09:52:02 webus: A phone from Samsung Dec 29 09:52:44 As4xk, yes on my Lenovo A606 also it works as build in tool. But i want same functional on other my phone... Dec 29 09:55:07 any good pattern for login screen? once I started LoginActivity first, check if user autorized and proceed to main screen Dec 29 09:55:19 Too many method references: 66455; max is 65536. -_- Dec 29 09:55:26 so close :p Dec 29 09:55:49 another time I've started from MainActivity and opened login screen if necessary Dec 29 09:56:10 but none of these are ideal Dec 29 09:56:42 maybe there is a 3rd option Dec 29 09:56:57 vigilancer, what about starting MainActivity and overlaying it with a fullscreen dialogFragment if not logged in Dec 29 09:57:16 danijoo: checking it out. thanks Dec 29 09:58:03 danijoo: and dialog fragment will hold autorization logic? Dec 29 09:58:18 vigilancer, yep Dec 29 09:58:26 i used that once Dec 29 09:58:59 switched to the activity method you mentioned though Dec 29 09:59:05 because of LoLCycle of Fragments :p\ Dec 29 09:59:06 danijoo: there will be mess with activity result callbacks. I need autorize via FB/TW also Dec 29 09:59:46 danijoo: I thought about something like conditional setContentView() Dec 29 09:59:58 danijoo: interesting. looks like a fit. especially if i find a way to specify the duration :) Dec 29 10:00:12 danijoo: i suppose it's part of the animation one sets? hmm hmm hmm Dec 29 10:00:14 vigilancer, or just 2 fragments in mainactivity and a transition between them Dec 29 10:00:25 fps, no idea about animation sorry :/ Dec 29 10:00:51 webus: I think it is different on device from device. I tried the same in 2007. On most devices back then, the audio was routed directly from the radio modem to the speaker. i.e. the application process was seperated from the audio on hardware level. Still some applications was able to record the audio on some devices by recording audio using the normal mic. This worked because the audio was echoed inside the phone. Dec 29 10:01:41 This was in 2007 though. So situation moght not be the same now. I do not know how it works on S5, i just know it works (a friend of mine records all conversations) Dec 29 10:43:24 As4xk, thanks! Dec 29 11:59:25 Hi. I use the ACTION_OPEN_DOCUMENT_TREE intent to ask for an uri for which i can read/write to a directory on external sd card. "Normally" i run startActivityForResult(...), but now i want to do it in a PreferenceFragment (which is not an activity). How can i get the result from the ACTION_OPEN_DOCUMENT_TREE when i'm not in an activity? Dec 29 12:01:40 As4xk, forward it from the activity to the fragment Dec 29 12:09:17 pfn: are you around? Dec 29 12:21:39 In my PreferenceFragment's onCreate I have "addPreferencesFromResource(R.xml.preferences);" In preference.xml i have (which is a programatically class extending DialogPreference). How do i forward the result from the ACTION_OPEN_DOCUMENT_TREE to PathListPreference? There is noe reference to PathListPreference from my PreferenceFragment Dec 29 12:21:41 danijoo Dec 29 12:24:54 As4xk, there are methods to get a referece Dec 29 12:25:03 I guess my new question is: How do i reference "variables" in preference.xml programatically? Dec 29 12:25:34 danijoo: do you have some experience with sbt? Dec 29 12:25:44 and protify? Dec 29 12:27:10 thepoosh, SBT it's for Scala right ? Scala it's not good for Android... Better Kotlin Dec 29 12:27:23 As4xk, you cant. you have to do it promatically Dec 29 12:27:26 webus: thanks, talking about a build system Dec 29 12:27:39 and I program in Java Dec 29 12:27:41 thepoosh, why not Gradle ? Dec 29 12:27:42 thepoosh, yes but I dont use really use it anymore Dec 29 12:28:01 webus: too slow for multidexed apps Dec 29 12:28:13 webus, sbt allows to use stuff like protify which is not available with gradle. also its faster Dec 29 12:28:22 danijoo: I'm trying to understand the integration guid and it seems I misunderstood the part with the build variants Dec 29 12:28:29 care to help a fellow understand? Dec 29 12:28:37 thepoosh, danijoo you use SBT with Java ? Dec 29 12:28:41 cons is you have to deal with the moods of protifies maintianer :p Dec 29 12:28:56 I cannot confirm nor deny at the moment Dec 29 12:29:06 webus: https://github.com/pfn/protify Dec 29 12:29:14 danijoo: So i need to make the PreferenceScreen programmatically? Dec 29 12:29:16 webus, did it once. right now, no Dec 29 12:29:31 As4xk, parts of it Dec 29 12:29:42 the callback and stuff like that As4xk Dec 29 12:31:16 danijoo: do you remember the variants/flavours configuration stage? Dec 29 12:32:53 danijoo: Sorry, I do not understand. How do I make part of it programmatically? Dec 29 12:33:01 thepoosh, i always did that through the gradle-build plugin Dec 29 12:33:27 that happens automatically from there? Dec 29 12:33:34 As4xk, find the inflated preference in the fragment and modify it with java code when you get the activity result Dec 29 12:33:45 I already defined all the buildtypes and flavours in build.gradle Dec 29 12:33:53 thepoosh, it grabs the flavors/build variant from the build.gradle file, yeah Dec 29 12:34:02 frick on a stick! Dec 29 12:34:04 at least last time i used it Dec 29 12:38:31 hmmm Dec 29 12:38:43 it claims install is an invalid key Dec 29 12:38:45 :S Dec 29 12:39:54 Thanks danijoo :) Dec 29 12:40:41 did not know about the findPreference() method in PreferenceFragment Dec 29 12:42:26 thats why you should read the docs about the classes you are using 8-) Dec 29 12:47:18 danijoo: totally not understanding this Dec 29 12:47:36 pfn: can you upload a sample project or something for the idiot commonfolk like me? Dec 29 12:48:59 thepoosh, are your buildtypes recognized? Dec 29 12:49:08 I think so Dec 29 12:49:15 check the generated sbt file Dec 29 12:49:31 hmm Dec 29 12:49:35 1 min Dec 29 12:49:50 if yes, you have to switch to the variant you want to use in the sbt console Dec 29 12:49:57 and then do the usual android:install Dec 29 12:58:46 danijoo: I missed a step Dec 29 12:58:48 :S Dec 29 12:58:51 works now Dec 29 12:59:49 good to hear :) Dec 29 13:00:07 well, I was too quick to be happe Dec 29 13:00:09 *happy Dec 29 13:00:19 now I have: protifySettings cannot be applied to libraryProject Dec 29 13:01:41 trying again Dec 29 13:01:42 :/ Dec 29 13:01:55 this is not as trivial as pfn is presenting it to be Dec 29 13:01:57 IMO Dec 29 13:02:05 indeed. but dont tell him :p Dec 29 13:02:23 seems like tagging him was a mistake I'm gonna pay for... Dec 29 13:02:27 :P Dec 29 13:09:05 anybody using ServiceStack for backend communication Dec 29 13:25:20 thepoosh, how is it not trivial, the error is clear, protifySettings goes on app not libraries Dec 29 13:25:25 and fuck you danijoo Dec 29 13:25:39 I love you too pfn :) Dec 29 13:26:23 pfn: I understood that Dec 29 13:26:34 now trying to understand why it's till not building Dec 29 13:27:02 flavors part is a little confusing I grant Dec 29 13:27:14 nah, that was fixed with the auto generation Dec 29 13:27:31 what part isn't building Dec 29 13:27:32 I'm stuck on having jar definitions side by side with build.gradle Dec 29 13:27:40 Error: more than one library with package name 'android.support.v7.appcompat' Dec 29 13:28:37 do I need to manually exclude those in my app:build.gradle Dec 29 13:28:37 ? Dec 29 13:29:57 grep appcompat 00-gradle-generated.sbt Dec 29 13:30:09 I think i got it Dec 29 13:30:22 had duplication in library projects and gradle Dec 29 13:30:26 as part of the transfer Dec 29 13:30:38 you can exclude them in a local.sbt to avoid impacting your gradle builds Dec 29 13:31:37 pfn: http://pastie.org/private/wabdrrzdwjnmqt40cyv26q Dec 29 13:32:20 even so, being used transitively across projects shouldn't cause that Dec 29 13:33:44 thepoosh, pastebin the build. command and full log? Dec 29 13:35:04 danijoo, I was never an asshole to you until you became completely ungrateful Dec 29 13:35:43 pfn: are you talking about the protify:install output? Dec 29 13:35:49 or the initial sbt build? Dec 29 13:36:02 whatever is failing Dec 29 13:36:07 https://gist.github.com/yishai-glide/c405737b64f6e1e609fa Dec 29 13:36:14 but everything is better Dec 29 13:36:48 if the initial succeeds, then the latter should not fail Dec 29 13:38:46 pfn, no need to apologize. Everyone has his vices :) Dec 29 13:39:40 thepoosh, last GlideApplication/android:rGenerator Dec 29 13:40:31 although, unpacking the appcompat aar multiple times like that is wrong... I wonder why Dec 29 13:41:11 Maybe that's a bug, Dec 29 13:41:16 * pfn ponders Dec 29 13:41:35 pfn: https://gist.github.com/yishai-glide/27ef8733a4f01e75d261 Dec 29 13:44:59 thepoosh, pastebin the 00-gradle-generated.sbt too? Dec 29 13:46:00 pfn: refresh Dec 29 13:47:52 pfn: will be back later today Dec 29 13:48:00 birthday party for my daughter Dec 29 13:48:10 thepoosh, oh, do you have a project.properties? Dec 29 13:48:12 remove it Dec 29 13:54:17 meh, this is probably a bug :( Dec 29 13:59:27 well, more of an unsupported configuration, still a pain Dec 29 14:03:16 hi is there any possible way to change the defaoult dir in whitch you keep the adv images ? Dec 29 14:05:18 what means adv? Dec 29 14:05:46 ^ N3cr0m4nc3r Dec 29 14:13:07 hi all Dec 29 14:13:22 how i can give an variable to an another class? Dec 29 14:15:28 o_O? Dec 29 14:16:17 How can i check if an Uri is a local file or not? Dec 29 14:16:21 ##java snooky Dec 29 14:17:36 As4xk, by its format. Dec 29 14:18:09 if it starts with a / or with file:// its a local file for example Dec 29 14:18:40 there's probably a way to query the file system Dec 29 14:19:04 new File(someURI).exists() Dec 29 14:19:24 theres also URLUtil.isFileUrl Dec 29 14:19:55 I don't think that's true. When i ask the user to choose a directory (and i choose a local file) i get "content://com.android.externalstorage.documents/tree/....." as a result Dec 29 14:20:22 i think content:// is always local Dec 29 14:20:26 http://nopaste.linux-dev.org/?911357 Dec 29 14:20:51 line 41, regid is null but on line 70 toast write me the regid Dec 29 14:22:23 The isFileUrl takes a String not Uri. The "new File(someURI).exists()" in a try catch looks like a nice solution though. Thanks! Dec 29 14:23:12 snooky, its an async task Dec 29 14:23:31 when line 41 is finished, it doesnt mean the task is finished and the result is already there Dec 29 14:23:41 no Dec 29 14:24:02 yes Dec 29 14:24:06 i need the return from getRegId in an variable Dec 29 14:24:17 you cant do that with an async task Dec 29 14:24:28 so what must i mean? Dec 29 14:24:29 you need a callback or something like that Dec 29 14:24:49 and wait for the async task to finish Dec 29 14:25:04 i dont know -.- Dec 29 14:25:31 snooky, you have to wait until onpostExecute is executed. You dont have access to the regId until that moment Dec 29 14:25:45 this will be after onCreate has already finished. At an undefined time in the future Dec 29 14:26:13 but, when toast me write the redid to my phone so the execute must be finish, or not? Dec 29 14:26:21 try to move line 42 to 46 in onPostExecute Dec 29 14:26:46 snooky, yes its finished when the toast is shown. But the toast is shown after the code you wrote in onCreate Dec 29 14:27:00 thats how threads work.. Dec 29 14:27:22 so i must it make after the oncreate? Dec 29 14:27:37 you must make it to run after your onPostExecute Dec 29 14:28:17 try what I said Dec 29 14:28:30 move line 42 to 46 to line 70 Dec 29 14:29:30 ok Dec 29 14:29:36 <_3mpty> q. about data binding library, I'm toying with mvvm like implementation and I'm not sure how interact with context based things from view model, should I pass it and store it in view model or do it via some kind of fancy interface? Dec 29 14:29:55 and then i can store the regid into my database and can send notification from my webbrowser to the application Dec 29 14:32:28 no, content: url are not guaranteed to be anything Dec 29 14:32:39 just that they're served by a content provider Dec 29 14:32:58 and how i can split it to many files Dec 29 14:33:29 As4xk, you can't check arbitrary url for locality Dec 29 14:49:29 pfn: What do you mean? Dec 29 14:49:50 you can't determine if any content: url is local Dec 29 14:51:50 Hmm. So when i use the ACTION_OPEN_DOCUMENT_TREE intent to ask the user for a directory (since access to sd card is denied), then i cannot check if it is a local directory or not? Dec 29 14:52:50 Not practically. Dec 29 14:52:55 You also probably shouldn't :) Dec 29 14:55:15 My end goal is to set an FileObserver on that directory, and the FileObserver require a String path. Is this possible at all then? Dec 29 14:55:56 to do what exactly Dec 29 14:56:14 and no, using saf will not ever result in a local file path Dec 29 14:56:26 Get notified when a new file appears in that directory insted of checking for new files at fixed intervalls Dec 29 14:57:17 write your own directory browser Dec 29 14:57:31 and, of course you need external storage permission Dec 29 14:58:46 how would you expect to do this without permission Dec 29 14:59:54 I think now that I have misunderstood something. I thought that I was not allowed to access the SD-card without using saf? Dec 29 15:00:46 "write your own directory browser" feels like reinventing another wheel :/ Dec 29 15:00:49 irrelevant, how would you add a file observer without permission Dec 29 15:01:08 and it's not Dec 29 15:01:22 since there is no file directory picker Dec 29 15:02:34 what kind of permission are you talking about? App permissions on the phone? In that case, why would i not have that? Dec 29 15:04:09 My thought was. Use saf to let the user choose a directory. Get path for selected directory by doing so-far-unknown-magic on the returned Uri. Use FileObserver to sign up for change notifications, ..., profitt Dec 29 15:06:32 I do not care if the app must ask for all possible permisisons (at install point, still running android 5) as the app is primarily for me. So i trust it not to do bas stuff Dec 29 15:07:08 Was "Stay Awake" removed from the developer options for 5.x? I'm not seeing it on my 6p Dec 29 15:07:36 bass stuff, the best stuff Dec 29 15:11:46 pfn: Am I making any sense? :p Dec 29 15:15:21 any ideas why a grid view's item will not correctly set a background color Dec 29 15:15:55 i'm using: Dec 29 15:16:14 As4xk, no you're not, you said you don't have sdcard permission Dec 29 15:16:40 View view = GetRecycledViewFromAdapter(); view.setBackgroundColor(R.color.MyColorDefinition); Dec 29 15:16:56 the color in Android Studio correctly gets rendered when i open the colors.xml file Dec 29 15:17:11 but when i run it on real hardware it comes out blackish Dec 29 15:17:18 nothing at all like the color i've specified Dec 29 15:17:40 and even if i set different colors to different items, they all come out the same blackish way Dec 29 15:19:01 pfn: Ahh. When i said "(since access to sd card is denied)" it was because I thought that I was not allowed to access the SD-card without using saf (after Android ~4.4) Dec 29 15:20:35 But now i feel like this assumtion is? Can I access internal and external sd card in the "good old way" as long as I have the permissions set for the app? also for Android 5 and 6? Dec 29 15:20:47 is wrong* Dec 29 15:23:31 https://www.dropbox.com/s/zcgf90kg240v0wm/device-2015-12-29-172239.png?dl=0 Dec 29 15:23:40 top two should be a light blue Dec 29 15:23:53 instead ALL that ain't default are that blackish color Dec 29 15:24:41 As4xk, yes, wrong Dec 29 15:27:55 Nice! Then it's back to the drawing board. Thanks, and sorry for the confusion :) Dec 29 15:28:30 any suggestions as to what might be wrong Dec 29 15:28:38 i'm just getting into android Dec 29 15:31:05 Ravenheart, getResources().getColor(R.color...) Dec 29 15:31:32 type system failure strikes again Dec 29 15:31:45 setBackground accept an int Dec 29 15:32:14 hi everyone Dec 29 15:32:50 hmm pfn what should i use as a Theme Dec 29 15:33:02 as the one with only an int is obsolete Dec 29 15:33:11 can i crawl website Using Selendroid or Appium ? Dec 29 15:34:06 Ravenheart, ignore the deprecation and use it anyway Dec 29 15:34:30 and use setBackgroundColor Dec 29 15:34:34 Iresf: wat? Those are testing tools Dec 29 15:36:03 yiati : how to crawl website in android ? i know jsoup but jsoup can not handle event Dec 29 15:36:40 Iresf: crawling websites has nothing to do with Android Dec 29 15:37:00 You can write it in straight up java Dec 29 15:37:09 then use it however you want Dec 29 15:39:21 yiati : i dont know how tell you about my problem because my english is not good Dec 29 15:39:43 stupid question once again, suppose the color is Colors.TRANSPARENT Dec 29 15:40:18 (i actually need the default color) Dec 29 15:42:23 well an if did the trick Dec 29 15:42:31 wonder what the best way to do it is Dec 29 15:43:48 Is there no establshed naming convention for android app code? Dec 29 15:44:27 i have used selenium for getting html source of my portal on college's website but i can not convert my code that i have written in java to android Dec 29 15:45:06 w0bni Same as regular Java except fields are usually called mFoo Dec 29 15:45:27 If you agree to follow that madness of course :) Dec 29 15:45:58 CedricBeust_: How about activity classes, for example? ListContactsActivity, ContactsActivity? Dec 29 15:46:10 Name them like regular Java classes Dec 29 15:46:25 follow standard java naming conventions Dec 29 15:46:28 how to login to a website and get html source and handle event (javascript method ) in android Dec 29 15:46:30 ? Dec 29 15:46:37 Thank you, guys :) Dec 29 15:46:55 use the m member crap if you like, I don't Dec 29 15:47:09 different org have different policy around that Dec 29 15:47:14 I've had orgm Dec 29 15:47:48 orgs reject code with m prefix on review and vice versa Dec 29 15:48:09 Iresf: You're not going to find any great solution to that approach. Anything would be very hacky. Dec 29 15:52:18 yiati : i can login to website using Jsoup and i can get html source in android programming but i cant handle event (javascript method) Dec 29 15:52:25 Iresf: Also it may not be legal, and your school could end up going after you if they catch you scraping Dec 29 15:52:46 check their robots.txt Dec 29 15:53:36 hi all Dec 29 15:53:47 erveryone here hows can helped me in german? Dec 29 15:56:03 you'd be faster if you ask in english Dec 29 15:56:31 i want split my main.java in two three files Dec 29 15:56:46 but than i can dont use the fuction in the main.java Dec 29 15:57:29 http://nopaste.linux-dev.org/?911450 Dec 29 15:57:33 thats my code Dec 29 15:57:52 snooky gets told to learn java every time hes in here Dec 29 15:58:01 and always comes back without java knowledge :( Dec 29 15:58:13 the nerve of some people Dec 29 15:58:42 This is the second time i am onto learning android , first time i learned simple things like activities , . Do you think i should learn it with some buddy ? Dec 29 15:58:57 jackhum: Sure, never hurts Dec 29 15:59:20 yiati: on scale of 1-10 where do you grade yourself in android developement ? Dec 29 15:59:32 20 Dec 29 15:59:38 :D Dec 29 15:59:39 :D Dec 29 15:59:47 21. Dec 29 15:59:57 20 is not legal age yet. Dec 29 16:00:15 depends on your state :p Dec 29 16:00:19 è qui in Italia Dec 29 16:00:25 und in deutschland Dec 29 16:00:32 i just want to know what is correct way to learn android , i have an idea for app , its simple just like tinder Dec 29 16:00:35 jk I'm in texas :D Dec 29 16:00:41 i am thinking of using parse as backend Dec 29 16:01:12 Shia LaBeouf has the best advice on learning android Dec 29 16:01:16 dont make something like tinder please :) Dec 29 16:01:24 yiati: yeah , that was nice advice Dec 29 16:01:30 unless it's tinder for Android programmers Dec 29 16:01:36 yiati: just do it Dec 29 16:01:45 Tennis: no i am not Dec 29 16:01:51 we can all critique the acceleration of the swipe animation Dec 29 16:02:08 i like object oriented as long as i am not the object Dec 29 16:02:19 my application need to store data of users Dec 29 16:02:42 so , research told me , i need to use parse , unless i can write my own backend Dec 29 16:02:57 I'm implementing MultiAutoCompleteTextView.Tokenizer for autocomplete the words and adding a whitespace to the end of the word, anyway if the user adds a punctuation symbol after the autocompleted word I'd like to behave like other text does, autoremoving the whitespace before the punctuation symbol, is this behavior from android, the keyboard or what? Dec 29 16:03:00 i dont know what parse is but build a db Dec 29 16:03:01 "it's simple, just like tinder" Dec 29 16:03:06 right, as if tinder is simple Dec 29 16:03:13 Tennis: parse is baas , Dec 29 16:03:24 I'm sure they've blown millions on their mobile apps already Dec 29 16:03:25 Tennis: its currently owned by facebook Dec 29 16:03:39 yiati : is possible rewrite selenium library for only android ? Dec 29 16:03:44 mmm , don't put any money in facebooks pocket Dec 29 16:03:46 like "hi pfn ," becomes "hi pfn," Dec 29 16:04:01 Tennis: lol , so what should i do? Dec 29 16:04:03 * Tennis quits whatsapp ;) Dec 29 16:04:15 nice Tennis Dec 29 16:04:15 Tennis, yeah. better put it into google :P Dec 29 16:04:30 or Dec 29 16:04:34 use an open source DB hosted on an a years worht of free AWs Dec 29 16:04:51 lol yeah amazons not much better i suppose Dec 29 16:05:10 Tennis: so you think i should go for app engine or aws? Dec 29 16:05:11 Iresf: It's already only for android Dec 29 16:05:12 self host if you have the skills to manage your DB and backend Dec 29 16:05:26 app engine does not have good tutorials for beginners Dec 29 16:05:27 it will take you dev time but is definitly the cheapest solution on long run Dec 29 16:05:48 appengine is simple, and tutorials are overrated Dec 29 16:05:53 Don't worry about optimizing early though Dec 29 16:06:01 any of them can work Dec 29 16:06:03 just do it Dec 29 16:06:04 pfn: you must be really good in this stuff Dec 29 16:06:12 yiati : but we can not do anything using selendroid like selenium on windows or linux Dec 29 16:06:16 yiati: lol , just do it Dec 29 16:06:25 jackhum, you will best serve yourself by learning to read manuals and reference material Dec 29 16:06:26 Appengine and AWS have both free plans / are free in the first year to try them out Dec 29 16:06:37 ^ Dec 29 16:06:39 but if you get big... Dec 29 16:06:40 appengine is completely free, forever, below a certain traffic threshold Dec 29 16:06:49 aws has a free 1 year micro instance Dec 29 16:06:53 Iresf: Such as Dec 29 16:07:16 i was reading an article which says snapchat uses app engine Dec 29 16:07:40 then i read about something called endpoints on gae tutorials , and swooosshh , i was dead Dec 29 16:08:20 some knowledge in java servlets is nice-to-have before starting app engine Dec 29 16:08:27 but not really required if you follow the tutorials Dec 29 16:08:36 then , someone suggested me to go for parse in earlier stage , that way i can learn android first . and later when my application gain users , in that time i can write my own backend Dec 29 16:09:04 was that someone gave me wrong suggestions Dec 29 16:09:13 yiati : i can login to a website and i can get html source automatically Dec 29 16:09:21 right now on 1-10 i am 1 on android developement Dec 29 16:10:25 yiati : i can login to a website and i can get html source automatically but i can not create app with selendroid to do that Dec 29 16:10:46 Iresf, I think the only way to execute js in an android app is through a webview. Not what you want I think Dec 29 16:10:59 you cant just call a js method from your java code Dec 29 16:11:27 (as far as i know) Dec 29 16:11:30 Iresf: What method are you using to login to a website and get html source on windows/linux that wouldn't work on Android? I doubt that you couldn't use the same method on Android with a couple tweaks. If it's the method you suggested before "is" possible on any platform including Android, but it's still hacky/possibly not legal on any platform Dec 29 16:13:21 yiati : why is it hacky ? Dec 29 16:14:14 hmm, FAB doesn't animate elevation on press by default? Dec 29 16:14:32 Iresf: because you are not coding around any maintained API, but depending on ui/html state Dec 29 16:14:37 runtime Dec 29 16:15:09 i must read your answer and understand them : yiati Dec 29 16:19:28 yiati : if i installed my app written using selendroid to login to webstie without running server on host , does it work ? Dec 29 16:21:04 Did you try? Dec 29 16:22:04 yiati : no Dec 29 16:22:31 but i know it needs to server on host Dec 29 16:22:43 is it true ? Dec 29 16:24:56 yiati : do you think it does not need to run server on host after it installed on device ? Dec 29 16:28:05 Iresf: I would suggest actually trying to accomplish what you want, and asking for help on anything you cannot find by researching Dec 29 16:39:57 Is Xposed development chat allowed here? I'm trying to read a file belonging to another app from within initZygote, but it always throws an EACCES permission denied error. Of course, I can always read it using su, but I feel like there must be a way to avoid doing that with the ability to hook everything Dec 29 16:41:27 whats the easiest path to develop a native plugin for android? as in what dev environment is easiest to setup? Android Studio? Eclipse? Dec 29 16:41:39 preferably written in cpp Dec 29 16:43:14 andr3wmac: Android Studio supports NDK development now to my knowledge Dec 29 16:43:23 sweet :D I'll try that route Dec 29 16:51:02 andr3wmac: Why do you want to write a native app? Dec 29 16:51:13 plugin Dec 29 16:51:20 thing Dec 29 16:51:33 What do you call an android plug-in? Dec 29 16:51:46 sounds like the setup for a bad joke :P Dec 29 16:52:01 I need to access some android specific things from Unity Dec 29 16:52:19 like launching a service to run in the background and interfacing with it Dec 29 16:52:32 That part will work fine in Java Dec 29 16:52:46 well I'm a cpp programmer by trade so I'm far more comfortable in cpp Dec 29 16:52:51 java is yuck :P Dec 29 16:53:00 Yeah I suspected so. You're setting yourself up for some hard times Dec 29 16:53:01 but I'll use it if I have to Dec 29 16:53:07 or if it'll save me a lot of time or something Dec 29 16:53:12 You know C++, Java will be easy to learn Dec 29 16:53:21 Yes it will save you headaches to default to Java Dec 29 16:53:38 okay. I'll just do that then. that can be done via android studio I imagine? Dec 29 16:53:44 or should I get eclipse? Dec 29 16:53:52 Android Studio Dec 29 16:54:25 Since you work with Unity, you might end up having to do some native parts but I recommend not going there until you know more about the platform and stick to Java until then Dec 29 16:55:45 basically we're running a download in Unity and if you hit the home button and go do something else obviously the unity Activity is paused and the downloading stops Dec 29 16:56:12 currently utilizing C#'s HTTPRequest class to do the downloading Dec 29 16:56:31 so we need an android plugin to launch a download service and report back Dec 29 16:57:02 does this sound doable from java? Dec 29 16:58:34 andr3wmac: it is doable, you can do it in C++ if you want to share the code between multiple platforms Dec 29 16:58:52 naw, I'll need to a roll a different one for iOS anyway Dec 29 16:59:05 so if I won't hit any major roadblocks with java I'll just go that way Dec 29 16:59:10 path of least resistance and all that Dec 29 17:06:17 andr3wmac Yes you launch background services in Java, it's straightforward Dec 29 17:06:33 and it's exactly for the kind of task you are doing, e.g. background downloads Dec 29 17:20:36 anyone use PreferenceFragmentCompat ? only way i could get this to not look like rat's ass was using PreferenceThemeOverlay.v14.Material Dec 29 17:20:59 still missing dividers, but thats minor Dec 29 17:53:30 Hello there ^^ I'd need a help with a code [App Inventor environment]. So if someone have some time to help please PM me Dec 29 17:56:21 Unidentified don't think anyone here uses AI Dec 29 17:56:48 well i had to try :) Dec 29 17:57:55 i'm so confused about storage, I'm trying with file, but it's just not working.. then with something called Tiny DB and it worked a bit but when i want to build other stuff on it, just doesn't work Dec 29 17:58:14 i'm just losing my mind Dec 29 18:00:40 pfn: nay news? Dec 29 18:01:25 thepoosh, yeah, that's an unsupported configuration :-( sorry, I don't have a good solution at the moment; i.e. project A depends on project B, C and D with B, C and D each depending on the same library Dec 29 18:02:48 so because there is more than one project that has appcompat as a dependency I can't build? Dec 29 18:03:34 thepoosh, yeah, if it were a diamond graph with the tail end pointing to appcompat, it would work Dec 29 18:03:46 hmmmmm, no can do Dec 29 18:03:49 mmmkay Dec 29 18:03:56 if anything changes let me know Dec 29 18:04:06 yeah, I'll see if I can fix it, but I don't know of a good solution at the moment Dec 29 18:04:36 ugh, marshmallow dialer sucks, it enters touchtones too quickly for conference call systems :( Dec 29 18:06:19 i have the OPO so still on L Dec 29 18:17:16 so weird, i had a few PreferenceFragments i moved over to PreferenceFragmentCompat - some show up OK, the others are totally blank :| yes, i overrode onCreatePreferences Dec 29 18:20:41 wow, i think it doesn't like my preferences.xml file Dec 29 18:21:15 thepoosh, good thing, I've reproduced the issue, bad news, I don't have a solution :p thanks for trying though Dec 29 18:21:34 oh noez Dec 29 18:21:38 goold luck Dec 29 18:23:35 https://github.com/pfn/android-sdk-plugin/tree/master/sbt-test/android-sdk-plugin/multiproject-same-dependencies Dec 29 18:45:19 i've just started learning java.. specifically for android.. (android studio).. i'm just fiddling around trying to learn it.. but i've encounted an error.. and was hoping you guys could help me... my code: https://codeshare.io/Pexic + the error i'm getting.. Dec 29 18:46:57 Sinorai, try it with the change I made ;) Dec 29 18:47:02 (line 9) Dec 29 18:47:54 in java you need to declare a variable before you can use it. Dec 29 18:47:56 cool thanks... ehm could you explain why said soultion could work ? where did i go wrong ? Dec 29 18:47:59 ahhh ok.. Dec 29 18:48:32 this means ¨this class has a TextView names test¨ Dec 29 18:48:43 then methods can use that variable after you set it in onCreate Dec 29 18:49:00 ahh ok.. Dec 29 18:49:35 you could also do that in the method. then the variable is only available in the method itself Dec 29 18:49:41 like this.. Dec 29 18:50:35 changeText would not be able to ¨see¨ test but the code setText in onCreate would work Dec 29 18:51:31 cool thanks.. it works now ^^ Dec 29 18:51:49 yw Dec 29 18:53:10 is there a + karma here ? Dec 29 18:53:23 nope Dec 29 18:54:14 oh.. ok..anyways.. thanks... any tips for a noob like me? Dec 29 18:57:30 learn java first :p Dec 29 18:58:44 is there an "easy" way to do so ? Dec 29 18:58:51 nope Dec 29 18:58:56 nope :) Dec 29 18:59:06 eclipse it is then... Dec 29 18:59:22 eclipse is the extra hard way thoug :p Dec 29 19:00:51 ahh so there IS an less harder way... care to share ? Dec 29 19:00:57 a* Dec 29 19:00:58 intelij idea Dec 29 19:01:02 since android studio bases on it Dec 29 19:02:13 anyone been using AS 2.0 on mac os x? Dec 29 19:02:28 is it stable? worth the plunge? talk to me people Dec 29 19:03:15 * frankdrey is still on 1.5.1 Dec 29 19:03:26 at work im still on 1.4 Dec 29 19:03:36 lexton if it was stable, it would be in the stable channel ;) Dec 29 19:03:36 I'm still on 1.4 Dec 29 19:04:06 yes I understand but I just wanted to get an idea of how unstable it is especially on mac os x el capitan Dec 29 19:04:31 so I'm sorry that I misled you Dec 29 19:05:17 maybe I'll get drunk on new year's eve and install it on my mac pro Dec 29 19:05:22 Anybody have suggestions for an "android board" computer? Not a "fully built" android device... yet something like the motherboard from one supplier, and an LCD from another? Dec 29 19:06:09 knob there is the dragonboard, and various o-droids Dec 29 19:08:05 g00s, checking those out... thank you! Dec 29 19:08:05 dragonboard looks interesting Dec 29 19:11:00 thepoosh still there ? Dec 29 19:13:10 g00s: Dec 29 19:14:46 thepoosh you mentioned, using nested fragments - are you using them in a View Pager ? that is, are the pages nested fragments Dec 29 19:15:11 some with a pager and some without Dec 29 19:16:01 lexton: Just do it. You can install is side by side with the released version Dec 29 19:16:01 having problems where only the first fragment show up. The titles are all there, but all the pages are only the first fragment Dec 29 19:16:30 I'm gonna go full nike and just do it Dec 29 19:16:30 oh well, maybe i have to init the viewpager in onViewCreated instead of onCreateView, maybe some weird timing thing Dec 29 19:16:54 if you don’t have a current project, sure, go for it. I’ve you’ve got something for work, though, you might want to be more careful Dec 29 19:17:37 I'm using my macbook pro for my work project, so the mac pro can be the guinea pig system Dec 29 19:17:46 I'm itching to try AS 2.0 Dec 29 19:19:58 so, what's the big advantage of an IntentService over a normal thread? Dec 29 19:20:05 if i kill the application... the service gets killed as well Dec 29 19:20:20 is it just that if my app goes backgrounded... it'll guarantee that the service will keep running? Dec 29 19:22:41 IntentService is more independent of the ui Dec 29 19:22:58 isn't a normal thread also independent of the main ui thread? Dec 29 19:23:15 i think you can even flag a thread with a priority "backgrounded" Dec 29 19:23:54 basically i have some very heavy code that takes a while and runs in background (at whatever speed) and i'm trying to figure out if an intentservice is really giving me benefits over a thread Dec 29 19:24:03 i thought it would allow me to kill the process and keep the thread running Dec 29 19:24:19 but the service seems to go down if i kill the application Dec 29 19:24:32 but will keep running if the application is backgrouded Dec 29 19:25:31 There are overlaps in the use case. but in general IntentServices are for longer running tasks like downloading a file Dec 29 19:25:53 while you would use an async task to quickly get the data for your listview from a database Dec 29 19:26:08 right... Dec 29 19:26:41 i was just downloading off a normal thread... do you know of a use case where the intentservice will win over a thread? Dec 29 19:26:53 im trying to understand what the "win" is Dec 29 19:27:14 to justify moving to a service Dec 29 19:27:50 the win is independence of the activity Dec 29 19:28:20 or more concrete Dec 29 19:28:23 of any lifecycle Dec 29 19:28:44 it doesnt look like it... if i kill the activity the service goes down Dec 29 19:28:50 no Dec 29 19:28:55 if you kill the apps process it goes down Dec 29 19:29:02 ok Dec 29 19:29:17 thats not the same as killing activity Dec 29 19:29:18 so if i was to use a thread and i moved to a different acitivty... my thread would die? Dec 29 19:29:24 probably Dec 29 19:29:38 thats kinda weird since the thread doesn't seem related to the activity Dec 29 19:29:48 at the very least, you’d no longer have a connection to the thread Dec 29 19:29:57 i’m not sure if it would die, though Dec 29 19:30:03 No, it wouldn't die. Dec 29 19:30:09 it usually doesnt Dec 29 19:30:31 so... Dec 29 19:30:37 no, I was wrong about that. However, you wouldn’t have any connection to the thread, so you’d be unable to do anything with it Dec 29 19:30:49 from what you guys are saying, they're about equivalent Dec 29 19:30:53 yeah. theres no point in using an asynctask in that way Dec 29 19:31:02 i dont mind not having connection to the thread... it's justa task that can go on its own Dec 29 19:31:07 they are usually used as async helpers to load data FOR the activity Dec 29 19:31:16 the IntentService is probably better integrated with the Android system than a plain thread Dec 29 19:31:34 "probably" Dec 29 19:32:32 just do it with an asynctask. once you see how much they suck, youll never run into that trap again :p Dec 29 19:32:44 any reason why they suck? Dec 29 19:33:04 i dont actually use asynctask, i use thread with runnable but i think they;re about the same? Dec 29 19:33:04 for one, in most versions of Android currently in use, it’s a single queue Dec 29 19:33:09 they dont play well with the rest of the framework Dec 29 19:33:16 like lifecycle, orientation change.. Dec 29 19:33:36 define "don't play well"? Dec 29 19:33:53 is the framework randomly gonna kill some threads or something? Dec 29 19:34:58 http://simonvt.net/2014/04/17/asynctask-is-bad-and-you-should-feel-bad/ Dec 29 19:35:13 http://www.onsandroid.com/2011/12/difference-between-android.html here's a nice table Dec 29 19:35:44 thats defining pretty well on what i mean with ¨dont play well¨ Dec 29 19:36:19 It won't randomly kill threads. When the app is "in the background", if it's running a service then the process is less likely to be killed than if it isn't running a service. Dec 29 19:36:24 It's mentioned on . Dec 29 19:36:32 "don't play well" == "most people use it wrong" Dec 29 19:36:36 I dont agree with that table because thread and asynctask is pretty much the same Dec 29 19:37:09 asynctask is some kind of wrapper arround a thread Dec 29 19:37:15 Why, when defining a variable as the output of a class's method, do we use () after the class name and what would go in the parenthesis? Dec 29 19:37:19 thats what the doc of asynctask seems ot be saying too Dec 29 19:37:42 (i.e. String str = Class().getString() ) Dec 29 19:37:45 i reckon they should have just built all this in the thread interface... Dec 29 19:37:46 noahmg123: https://docs.oracle.com/javase/tutorial/ Dec 29 19:37:47 DrBenway: AsyncTask is just a wrapper around ThreadPoolExecutor (which managers Threads) and Handler (for getting back to the main thread). Dec 29 19:37:49 sounds overly complicated Dec 29 19:38:16 I never use it, personally. I'd rather use ThreadPoolExecutor. Dec 29 19:38:19 you could use threads too Dec 29 19:38:19 4 different ways of spawning background process and no one is really sure what they're for Dec 29 19:38:32 but this will result in a lot of boilerplate if you want to update views etc Dec 29 19:38:43 SimonVT: I have been trying to google it but I cannot find the specfically where it is. I'll look in the link you gave me though. Dec 29 19:38:58 i was using thread and moved some of the code to an intentservice and then was surprised that they behaved just about the same. the only thing i havent managed to test is "what happen to my process when the system is overloaded by other apps" Dec 29 19:39:00 DrBenway: The crowds who don't know about this stuff drown out those who do. The web is full of blog posts with bad advice. Dec 29 19:39:17 its just hard to explain in general Dec 29 19:39:22 SimonVT's blog (posted above) isn't bad; I mean blogs in general. Dec 29 19:39:30 name what you want to do and we can discuss the pros/cons of different methods to do it Dec 29 19:39:35 “4 different ways of spawning background process and no one is really sure what they're for” That’s android for ya Dec 29 19:41:20 so anyway to close this discussion as a "yes" "no" answer... i have this independent long running task which i want to execute backgrounded. it doesn't signal anyone about it finishing and would hope it'll stay running for as long as possible even if the app goes backgrounded. I might queue up multiple of those tasks as the application run and would prefer having them queued instead of paralelle since they use a lot of resources. Intent Dec 29 19:41:20 Service sounds like the right thing to use? Dec 29 19:41:22 noahmg123: It's basic java (/oop).. If you know neither, you should at least spend a few hours reading up Dec 29 19:42:30 SimonVT: I know both fairly well, but I guess I missed the answer. I am looking into it. Dec 29 19:42:53 DrBenway: How important is it that this code be compatible with regular, real Java? Dec 29 19:43:01 not at all Dec 29 19:43:04 it's android only Dec 29 19:43:33 then an IntentService is probably slightly better Dec 29 19:43:39 Uhuh... Dec 29 19:43:41 alright done... Dec 29 19:43:43 thanks Dec 29 19:43:49 most of the setup for threading will have been done for you Dec 29 19:44:03 SimonVT: OK, I think I found the answer. Sorry for my noob question. Dec 29 19:44:07 of course, if you already have code written for all of this, that might be better :p Dec 29 19:44:51 well i ported my Thread code to IntentService in very little time Dec 29 19:45:15 as long that you have atomic tasks, it's really just a container for the task Dec 29 19:45:53 pretty much Dec 29 20:00:03 if i'm using preference-v14 fragment in a viewpager, should i be using support-v13 FragmentPageAdapter ? Dec 29 20:01:48 greets all Dec 29 20:02:30 g00s, i think you should use native fragments at all :s Dec 29 20:02:33 thats a negative Dec 29 20:02:46 danijoo_ lol should or shouldn't ? Dec 29 20:02:53 shouldt sorry :D Dec 29 20:02:58 i have a Bluetooth/BLE scanning problem that I am looking for some help with before I open an official bug Dec 29 20:03:06 anyone game to talk about it? Dec 29 20:03:41 danijoo_ i'm going flipping mad; I have a FragmentPageAdapter with 3 fragments, but only the first one show up - even though the titles for the others are OK Dec 29 20:04:44 g00s, have you tried a FragmentStatePagerAdapter? Dec 29 20:05:14 swooby not yet, i just copied the adapter from cheese squared Dec 29 20:05:17 maybe i'll try that Dec 29 20:05:51 is it kosher to paste code in here? (first time in this channel) Dec 29 20:06:03 swooby BLE is fun. whenever i do BLE scan, my wifi disconnects and i get a (!) symbol in status bar Dec 29 20:06:39 I'm running in to something similar to https://code.google.com/p/android/issues/detail?id=67272 Dec 29 20:07:16 it is never kosher to paste code Dec 29 20:07:18 in ANY channel Dec 29 20:07:22 g00s, sounds super strange.. Dec 29 20:07:23 could someone tell me the benefits of using fragments over normal activities? Dec 29 20:07:41 Keheira, same idea as single-page-applications Dec 29 20:07:42 danijoo_ yeah, i have a few BLE apps on my device, they all do the same thing Dec 29 20:07:44 on the web Dec 29 20:08:21 hmm, thepoosh left Dec 29 20:08:21 swooby thats an old bug - supposedly it was fixed in 4.4 ? Dec 29 20:08:31 "supposedly" Dec 29 20:08:34 so activities is the single page and fragments are like...cms? Dec 29 20:08:35 4.4.4 to be exact Dec 29 20:09:06 i work around hundreds of active BLE/iBeacon devices, and I have to factory reset my stock Nexus 5 every month or two Dec 29 20:09:31 lol Dec 29 20:10:12 keheira, fragments [although they can be UI-less] are reusable elements that can be imbedded in different layouts Dec 29 20:10:31 swooby EddyStone off to a good start then :D Dec 29 20:10:38 def Dec 29 20:11:18 i can imagine a CES/GoogleIO conference w/ hundreds of IoT devices causing DoS to all Androids in range Dec 29 20:11:56 thanks! Dec 29 20:12:06 oh i see they marked it as Obsolete, so surely it must be fixed Dec 29 20:12:37 g00s, obsolete == ignored :D Dec 29 20:12:57 obsolete means no longer occurs on current version Dec 29 20:13:00 keheira, you can think of fragments as a complex type of view, but they are distinctly different Dec 29 20:13:09 or assumed to be fixed by massive changes and too lazy to revalidate Dec 29 20:13:11 submit again Dec 29 20:13:28 danijoo_ yeah, like 'we changed a bunch of stuff and don't know what works and what doesn't any more can you guys retest everything and file new bugs :D' Dec 29 20:13:38 e.g. change from bluez bluetooth stack or vice versa Dec 29 20:13:39 On the new stock dialer, when you tap a contact to call a circle reveals the call screen. How can I do that easily with an activity transition? Dec 29 20:14:03 ViewAnimationUtils.createCircularReveal Dec 29 20:14:07 yeah, i'm going to open a new android bug soon, but thought i would check here to see if anyone AOS could help Dec 29 20:14:11 doesn't work well for an activity transition, though Dec 29 20:14:36 swooby good luck, hope it doesn't get merged into the obsolete one :D :D :D Dec 29 20:14:40 man, my 2014 motox feels so small now that I've been using my 6p regularly Dec 29 20:15:38 swooby so it's ok if I only use fragments for list? cause I don't know they don't seem too useful otherwise Dec 29 20:15:39 noahmg123, what is "that"? show a contact circle, or open the call screne? Dec 29 20:15:57 swooby: use the animation. Dec 29 20:16:00 sorry for the somewhat begineer question I just started getting serious about development Dec 29 20:16:19 * extor wonders whether it's a handicap to have a small screen size if one is learning to write apps for the first time or whether a 4.5" would be sufficient ¯\_(ツ)_/¯ Dec 29 20:16:33 extor, 4.5" isn't a small screen Dec 29 20:17:09 pfn, I thought it was the smallest size around? And 7" and 8" ones are the larger sizes for tablet phones Dec 29 20:17:13 pfn: thx. still looking into it. the tutorials I have found have not been very helpful. I guess I am going to have to work it out myself. Dec 29 20:17:22 extor, lolwut Dec 29 20:17:34 noahmg123, activity transitions typically don't use reveals Dec 29 20:17:36 keheira, my opinion is that every activity should be broken up in to fragments and that dedicated activity only layouts should be avoided Dec 29 20:17:57 noahmg123, you could probably write such a scene, but it's kinda ugly Dec 29 20:18:02 extor, you can use an emulator Dec 29 20:18:17 extor, there are no 7 and 8" phones Dec 29 20:18:17 swooby, why go with the overhead of fragments + boilerplate of fragments where its not needed Dec 29 20:18:39 Swooby thanks for letting me pick your mind. I guess I'll work on using fragments a bit more Dec 29 20:18:43 Yeah, I'm not very keen on Fragments. So much more complexity. Dec 29 20:18:52 truckcrash, so yes? Dec 29 20:19:05 they are useful. but you dont need them *everywhere* Dec 29 20:19:05 danijoo_, that's dangerously close to a "design patterns" question Dec 29 20:19:09 pfn: OK, well here's the deal. I will have an activity with a list of sections, each section then has some subsections. When you tap a section it will show the subsections. I was thinking about doing this in another activity, but what do you think? Dec 29 20:19:10 pfn, ok so maybe I was thinking 5.5" phone tabs Dec 29 20:19:35 truckcrash, so is this phone too small to use as a tester for apps? http://www.phonearena.com/phones/ZTE-Maven_id9485 Dec 29 20:19:56 extor, it's better to test on a small phone than big Dec 29 20:20:09 and that's not small Dec 29 20:20:10 And why is that pfn Dec 29 20:20:10 I cant find a good replacement for my Nexus 5 -_- the 5x is so disappointing.. Dec 29 20:20:17 extor, nope, and pfn is somewhat correct. Ideally you can test on both big and small Dec 29 20:20:21 you should use fragments when you can't say with absolute 100% certainty that you won't want to reuse your code in a different layout Dec 29 20:20:29 pfn: ^^ Dec 29 20:20:34 but if you use an emulator, such as the one with android studio, you cna try any size you want Dec 29 20:20:35 * extor is relieved Dec 29 20:20:52 swooby, I can still copy/paste my activity code once i need to Dec 29 20:20:54 swooby: That does raise the question of whether the extra complexity of Fragments is justified, though. Dec 29 20:20:58 to create a fragment Dec 29 20:21:03 its a matter of 5 minutes of work Dec 29 20:21:07 copy/paste is bad and not maintainable Dec 29 20:21:09 (in most cases) Dec 29 20:21:11 Is android studio the best dev environment out there, and perhaps the only one Dec 29 20:21:23 i just put all the stuff that would be in activities in my fragments, so i have mostly empty activities Dec 29 20:21:26 extor: Yes. Dec 29 20:21:28 extor, yes and the only one dedicated to android Dec 29 20:21:35 you should also strive to never have duplicate code Dec 29 20:21:38 One IDE to rule them all! Dec 29 20:21:48 * extor wonders what the current language is for apps, is it a derivative of java like before or something related to C# now Dec 29 20:22:02 TacticalJoke computer status ? Dec 29 20:22:13 extor, java Dec 29 20:22:15 g00s: I still have the old laptop. :s Dec 29 20:22:19 but not the newest version of it Dec 29 20:22:25 danijoo_, any fancier ones that are out there which support multiple platforms? Dec 29 20:22:33 <_abc_> Hello. I have very bad news. The latest studio 1.5 update broke adb on linux which adb depends on a glibc version which is not in any mainstream distribution. Specifically I run wheezy and it has GLIBC_2.13 ; the updated adb requires GLIBC_2.14 . I know in theory how to fix it, but I would like to know if clicking on automatic updates in Android Studio is as unsafe as allowing auto updates in MS Windows Dec 29 20:22:35 extor, kotlin, scala... Dec 29 20:22:40 <_abc_> . Dec 29 20:22:47 extor, basically everything that runs in the jvm can be used Dec 29 20:23:03 Hmm I wonder where it was that I heard that C# was becoming the language now. Maybe it was chatter about xamarin studio Dec 29 20:23:10 pfn: id you see my message above? Dec 29 20:23:29 extor, there are frameworks that allow you to do that Dec 29 20:23:35 but its not the recommended way Dec 29 20:23:54 And I suppose google itself has no plans to convert from java to C# Dec 29 20:24:31 extor, its not that easy to do even if they had plans Dec 29 20:24:36 extor, i would say that is a 100% safe statement to make Dec 29 20:24:38 but Im pretty sure they dont Dec 29 20:24:51 I guess there's still a hot battle raging between java and .NET over devices Dec 29 20:25:25 TacticalJoke danijoo_ extor did you guys see that thing where they checked in a bunch of openjdb classes Dec 29 20:25:27 extor, i don't see any serious percentage of devices using .NET Dec 29 20:25:35 google would rather develop their own language than to use something MS invented Dec 29 20:26:02 Good thing the learning curve will be not so steep for me Dec 29 20:26:20 g00s, they did? Dec 29 20:26:22 if i was a betting man, i'd bet that google will introduce web framework for mobile apps to replace the native stuff, but over time Dec 29 20:26:29 Apple itself invented a variant of objective C did they not Dec 29 20:26:43 extor, its not that easy in the android world Dec 29 20:26:46 g00s, so androidJS perhaps? Dec 29 20:26:51 fragmentation is one reason for that Dec 29 20:26:55 apple likes their native crap, they will stay native. i can see MS adopting more web stuff for web apps also, since they are losing so badly Dec 29 20:27:02 danijoo_, because of all the existin hardware wired for JVM? Dec 29 20:27:05 so, no takers on helping to figure out why BLE scanning stops working (hangs and locks up phone) after scanning ~2000 unique BLE/iBeacon devices? Dec 29 20:27:22 extor, you cant just throw a version with another language out Dec 29 20:27:28 because devs wont code for both platforms Dec 29 20:27:39 and you cant update all android phones to the new one because of fragmentation Dec 29 20:27:46 extor i think maybe html5 / dart related Dec 29 20:28:21 So the java used to code android is vanilla java and not some variant I suppose? Just wondering if the existing mountains of tutorials out there for Java will be usable Dec 29 20:28:26 if they would introduce another (optional) language right now, it would take another year or 2 until there are enough devices to care about it Dec 29 20:28:27 the Chrome Dev Summit 2015 keynote - i think that clearly explains where google is going Dec 29 20:28:38 but most android devs watch android dev summit :D Dec 29 20:28:53 <_abc_> How does one downgrade adb in Android Studio 1.5 please? Dec 29 20:29:04 then again, it could still be a big a/b test and not even google knows :D Dec 29 20:29:12 fetch an older sdk zip maybe _abc_ Dec 29 20:29:38 <_abc_> adq: I just finished a fresh install and stupidly upgraded when asked. Dec 29 20:29:44 <_abc_> I am not playing that game again. Dec 29 20:29:48 by the way, ms updates are not really "unsafe", it's even the contrary, they were signing their update years ago when evilgrade would defeat most of update system of many apps and OS Dec 29 20:29:54 Why would you want to downgrade adb? _abc_ Dec 29 20:30:29 <_abc_> Because some dweeb released it depending on glibc_2.14 and only glibc_2.13 is in whezy Dec 29 20:30:32 <_abc_> *wheezy Dec 29 20:30:33 why would you not update your glibc? Dec 29 20:30:47 <_abc_> Updating glibc means basically updating the entire system Dec 29 20:30:52 wheezy is pretty old, isnt it? Dec 29 20:30:53 <_abc_> Aka reinstall Dec 29 20:30:57 1 or 2 versions behind Dec 29 20:31:01 <_abc_> It is oldtstable since September Dec 29 20:31:07 <_abc_> Not 'behind', 'stable' Dec 29 20:31:25 <_abc_> Also systemd-less which is important for other things I do with it Dec 29 20:31:43 * danijoo_ loves SystemD Dec 29 20:32:09 <_abc_> In general sticking with bleeding edge libs is something Oracle used to do, and also Cadence etc with their systems, guaranteeing they would only run on Fedora or official paid Red Hat Dec 29 20:32:21 <_abc_> bleeding edge and or special release numbers Dec 29 20:32:33 <_abc_> and pinning their binaries to them too, breaking relinking at runtime Dec 29 20:36:22 <_abc_> What part of the studio contains adb? SDK tools? Dec 29 20:36:31 <_abc_> ? Dec 29 20:36:44 i think so Dec 29 20:37:35 omg, i used tablayout.setTabsFromViewPager instead of tablayout.setupWithViewPager :| Dec 29 20:37:44 android studio 1.5.1 on mac, builds are extremely slow .... any suggestions on how to speed them up? Dec 29 20:38:03 _abc_: I think it’s platform-tools Dec 29 20:38:14 I don’t see an obvious way to downgrade it, yet Dec 29 20:38:36 <_abc_> I would like to f*** the people who do this with a Sequoia tree. Dec 29 20:38:56 <_abc_> Also where is the procedure to back out of an upgrade? Dec 29 20:38:57 nownot well, first is slow. make sure you're using gradle 2.9 or 2.10 Dec 29 20:39:00 <_abc_> Even Windows has hat Dec 29 20:39:00 (also some of the tools have moved around over time, I think, but I’m not sure adb has, and if it has it may have been quite a long time ago) Dec 29 20:39:02 <_abc_> *that Dec 29 20:40:21 you use android studio and not stuff from the cli? Dec 29 20:40:28 g00s : ohhh that might be it, thanks, changing now Dec 29 20:41:43 I don’t use AS for anything real, just manage the sdk and tools with the ‘android’ command. I guess AS manages them independently, I don’t know a lot about it. Dec 29 20:42:17 yeah i still use the standalone sdk manager Dec 29 20:43:26 can you use the standalone sdk manager with AS? I thought that when I installed AS it downloaded a whole new copy of the sdk for its own use - maybe I missed the step of pointing it to where my sdk install already was. Dec 29 20:48:38 duboisj sure you can, you have to abort that 'd/l all the things' wizard Dec 29 20:49:01 Ah, I see. Good to know. Dec 29 20:49:13 basically just click close button. then it will say "re run when you restart or something like that Dec 29 20:49:18 just say no Dec 29 20:50:18 TacticalJoke https://www.reddit.com/r/androiddev/comments/3yoxqc/using_android_studio_with_multiple_applicationsa/ Dec 29 20:50:21 maybe get 16G ram :D Dec 29 20:51:12 <_abc_> umm remote running android studio over ssh -X gets me a black window with nothing in it... Dec 29 20:51:25 <_abc_> Wtf. It used to work. There's something about swing I forget? Anyone? Dec 29 20:51:43 _abc_ god that has to be horrible UX Dec 29 20:51:49 <_abc_> hm? Dec 29 20:51:56 <_abc_> No it's on a lan it is fast Dec 29 20:51:59 <_abc_> but very black ;) Dec 29 20:52:02 you dawg, we put lags in your lag so you can wait while you wait Dec 29 20:52:18 <_abc_> The bottleneck is not the gui, it's the java sloth... Dec 29 20:52:34 <_abc_> Also delegating rendering to a second machine is supposed to make it faster Dec 29 20:52:39 _abc_ lol you are saying network latency < swing hahaha Dec 29 20:52:47 <_abc_> precisely. Dec 29 20:52:53 well, thats why java never took off on the client :D Dec 29 20:53:03 <_abc_> well it CRAWLED. harhar Dec 29 20:54:21 you used to be able to do this easy with old rosylyn, because it was based on http microservices on top of nancy Dec 29 20:54:36 so have tools on server, running rosylyn, and then IDe on desktop Dec 29 20:54:58 they moved away from that though, was kinda flaky Dec 29 20:55:11 _abc_: there's something about naming your root window LG3D that was important last time I needed to get an IDEA-based IDE running on linux Dec 29 20:55:36 Hi! I have a question regarding Android Auto: Is it only possible to implement apps to use the media service or messaging or is it possible that I write an app that for instance controls some other stuff in the car and integrates to the android auto ui (e.g. as an additional button in the bottom bar)? Dec 29 20:55:41 <_abc_> LG3D sounds like a nsa code for xxl fried at mc'ds Dec 29 20:56:06 <_abc_> Aerobit: wait, what is LG3D in this context? Dec 29 20:56:26 _abc_: I have no idea, but doing "wmname LG3D" has solved weird java GUI issues for me Dec 29 20:56:39 one of those magical solutions Dec 29 20:56:50 <_abc_> You realize LG3D is a monitor by LG right? Dec 29 20:57:00 <_abc_> Some developer forgot the forceps in the patient? Dec 29 20:57:23 * _abc_ shudders Dec 29 20:57:26 generally, doing remote X11 for most anything these days works like ass Dec 29 20:58:27 eh, java has taken off quite well on the client Dec 29 20:58:34 "never took off" Dec 29 20:58:39 yet here we are in the age of java on the client Dec 29 20:59:33 lots of Java Web Start around Dec 29 20:59:35 _abc_: we made a dedicated build machine for Android that’s newer than our others, running Debian 8. libc is new enough for the latest AS according to release notes, although we don’t run AS on there. Dec 29 20:59:46 p_l noooo say it isn't so Dec 29 20:59:54 dunno, haven't seen much jnlp around anymore Dec 29 20:59:57 not for a long while Dec 29 21:00:16 I’m pretty new to Android, but it wouldn’t surprise me if you had a fair amount of trouble running new tools on Wheezy. Dec 29 21:00:48 _abc_: I don't think it's anything to do with an LG monitor name, more spoofing WM names for Java Dec 29 21:00:52 You might have to consider a separate machine, unless you want to reinstall the older AS you had working Dec 29 21:00:52 _abc_: http://awesome.naquadah.org/wiki/Problems_with_Java Dec 29 21:00:59 <_abc_> duboisj: I know. It's up to glibc_2.17 or so at least, 14 is required Dec 29 21:01:14 <_abc_> Aerobit: wow. Thanks, reading Dec 29 21:01:39 <_abc_> nawuadah? Big stargate fans there. Dec 29 21:02:21 <_abc_> Now I started netbeans + mobility pack on the same setup, I get 3/4 of the window painted... wtf are these java people doing to my unix?! Dec 29 21:02:26 <_abc_> *naquadah Dec 29 21:05:21 hi Dec 29 21:05:36 I created a "selector" drawable for overflow icon Dec 29 21:05:42 and i used there "dp" Dec 29 21:05:55 but one item covers the whole space... why? Dec 29 21:05:56 wow the new weather underground app looks really nice Dec 29 21:12:05 g00s: the weather underground was a terrorist group Dec 29 21:13:57 actually i just deleted it. screenshots looked much nicer than reality Dec 29 21:14:52 if you had a phone for dev right now, what would you buy? Dec 29 21:15:05 (not the 6p) Dec 29 21:17:51 im thinking about the Moto X play right now.. Dec 29 21:18:14 dont buy motorola Dec 29 21:18:17 NEVER Dec 29 21:18:26 there are no replacement parts Dec 29 21:19:00 if this is your only reason its not something I care about. Dec 29 21:19:10 yeah Motorola was good until they got bought by Lenovo Dec 29 21:19:24 * danijoo_ works on his lenovo laptop Dec 29 21:19:24 LD Dec 29 21:19:26 :D Dec 29 21:20:55 I used to drink at a bar owned by a former WU member Dec 29 21:20:58 The Night Cafe Dec 29 21:22:03 I wish there would be more phones with a fingerprint sensor available right now Dec 29 21:22:09 besides the new nexus´ Dec 29 21:23:13 maybe I should wait a few more months Dec 29 21:27:26 danijoo_: I think the HTC A9 has fingerprint sensor Dec 29 21:27:59 shmooz, i never buy a htc again Dec 29 21:28:30 they dropped software support for my last htc after it was out for 8 months Dec 29 21:36:28 AlmogBaku, there are no replacement parts for any phones these days Dec 29 21:36:46 samsung s6, htc a9, nexus 5x and 6p Dec 29 21:36:49 all have finger print sensors Dec 29 21:37:02 I set this drawable as overflow icon https://gist.github.com/AlmogBaku/179a8bac63df32881902 Dec 29 21:37:13 but why does the first circle fill the whole space?! Dec 29 21:38:24 drawables always fill the space their in Dec 29 21:38:27 they're Dec 29 21:39:11 but i defined the size Dec 29 21:39:55 doesn't matter iirc Dec 29 21:40:21 how can fix that? Dec 29 21:40:34 i want to use xml overflow icon Dec 29 21:40:37 define the space you’re putting it into Dec 29 21:40:52 if you want to use that, why not just define an xml menu? Dec 29 21:41:00 Note: The shape scales to the size of the container View proportionate to the dimensions defined here, by default. When you use the shape in an ImageView, you can restrict scaling by setting the android:scaleType to "center". Dec 29 21:42:37 huh? Dec 29 21:42:42 xml menu? Dec 29 21:44:29 a menu resource Dec 29 21:44:54 http://developer.android.com/guide/topics/ui/menus.html Dec 29 21:47:44 and what's wrong with the new nexus phones, they're all great Dec 29 21:47:57 their price in europe. Dec 29 21:48:10 then import them from somewhere cheaper Dec 29 21:48:26 Nexus 5 is still good Dec 29 21:49:32 shmooz, it is. but mine begins to fall into its pieces :p Dec 29 21:49:41 no fingerprint sensor on nexus 5 either Dec 29 21:49:54 ^ also this Dec 29 21:50:00 although, why would you want the fingerprint sensor... it's not all that useful Dec 29 21:50:15 not many apps take advantage of it Dec 29 21:50:17 playing arround with it Dec 29 21:50:29 not for actual personal use atm. Dec 29 21:50:38 so get the 5X and stop whining Dec 29 21:50:47 the sensor becomes a lot more useful if you’ve got Marshmallow, cause you have the actual Android apis for it Dec 29 21:50:58 the LG V10 also has a fingerprint sensor Dec 29 21:51:09 marshmallow or not, the usage is still pretty niche Dec 29 21:51:50 only because they’ve just started shipping. Usage on iOS is much more widespread, cause they’ve had standard access for a while Dec 29 21:51:55 yeah next they'll do iris scanner Dec 29 21:52:03 I still like moto's active display much more than imprint for unlock Dec 29 21:52:08 all these lock gimicks Dec 29 21:52:12 eh, not that widespread Dec 29 21:52:42 it's useful for confirming purchases for purchases and money sending apps Dec 29 21:52:44 and they expect everyone to follow their bandwagon which apple built ;) Dec 29 21:52:57 it’s useful for password vaults too Dec 29 21:53:06 it's useful for unlocking the phone for using android pay Dec 29 21:53:14 most people don't use password vaults Dec 29 21:53:26 useful for collecting a database of fingerprints Dec 29 21:53:28 they’re starting to. and among those who do, it’s quite useful Dec 29 21:53:53 s73v3r, yes, I unlock my password vault using a fingerprint :P Dec 29 21:54:14 that's like my only use-case for fingerprint right now, outside of unlocking the phone Dec 29 21:54:42 i wonder if marshmallow can let you assign per-fingerprint permissions Dec 29 21:54:50 no Dec 29 21:54:56 like you could let other people unlock the device, but not purchase things, for instance Dec 29 21:55:15 that's just normal multi-user Dec 29 21:55:18 i think you can save your wallet with the fingerprint Dec 29 21:55:31 so they could not purchase because they cant authorize there Dec 29 21:55:35 can it match who's fingerprint is who's and restore their homescreen and apps ? Dec 29 21:55:42 shmooz, no Dec 29 21:55:49 so you can have multiple people using it Dec 29 21:55:49 you pick the user you want to unlock from the lock screen Dec 29 21:55:52 then you can use fingerprint Dec 29 21:55:53 that would be really useful Dec 29 21:56:17 phones arent shared usually Dec 29 21:56:35 tablets? Dec 29 21:56:43 well they are already half way there, you can create multiple accounts Dec 29 21:56:46 yeah there it might be useful Dec 29 21:57:22 would be cool if your wife picks up the tablet and sees her homescreen, and when you unlock it, its yours. yeah Dec 29 21:57:30 but yeah, you can create multi-account, and then add individual fingerprints there Dec 29 21:57:39 but you can't imprint an account right away Dec 29 21:57:43 you have to pick the account first Dec 29 21:57:44 then imprint Dec 29 21:58:00 now that I said it, they will go implement it Dec 29 21:58:16 because it's too cool, like everything I say ;) Dec 29 21:58:25 for whom may be interested https://gist.github.com/AlmogBaku/179a8bac63df32881902 Dec 29 21:58:30 that's my solution Dec 29 21:59:01 AlmogBaku, did you know theres an overflow button already built in you can just use? ^^ Dec 29 21:59:11 yes Dec 29 21:59:13 by defining a submenu and/or showAsAction=never on your items Dec 29 21:59:14 hardcoded png Dec 29 21:59:15 :| Dec 29 21:59:21 AlmogBaku, what appearance are you going for? and why are you trying to avoid platform convention for this? Dec 29 21:59:31 I just wanna change its color Dec 29 21:59:33 there is absolutely no reason to change the overflow icon Dec 29 21:59:36 AlmogBaku, then tint it Dec 29 21:59:46 don't like this tint thing Dec 29 21:59:51 lol ^^ Dec 29 21:59:54 too much code for freakin' UI Dec 29 21:59:55 but you want to change its color... Dec 29 21:59:56 lol indeed Dec 29 22:00:06 and you can tint via attribute for toolbar icons Dec 29 22:00:16 only from v23 Dec 29 22:00:25 android M :| Dec 29 22:00:28 if they had backported tint, you’d have a point Dec 29 22:00:29 for appcompat toolbar icons without v21 Dec 29 22:00:39 arrrghhhhhh how? Dec 29 22:00:40 support lib has tint too Dec 29 22:01:00 since support lib version 23.1 Dec 29 22:01:07 revision* Dec 29 22:01:27 and too much code? DrawableCompat.setTint(DrawableCompat.wrap(button.getDrawable.mutate()), tintColor) Dec 29 22:01:30 one line isn't much at all Dec 29 22:02:11 and btw i need to support v14 Dec 29 22:02:28 nevermind, i already built it Dec 29 22:02:34 this is compatible back to 7 i think Dec 29 22:02:35 why back to v14? Dec 29 22:02:50 do you have a significant amount of paying customers still going back that far? Dec 29 22:03:03 v14 is not that old Dec 29 22:03:08 its android 4 Dec 29 22:03:11 yes it is Dec 29 22:03:35 v14 has < 0.1% market share Dec 29 22:03:52 anyway, how can I tint the overflow icon with xml? Dec 29 22:03:56 because nearly all devices received at least 15 Dec 29 22:03:59 more specifically, among your users, how many are still below 17? Dec 29 22:04:12 you don’t do it with XML, you do it in code Dec 29 22:04:35 hello Dec 29 22:04:46 you know, i've been told that "if it works, don't touch it :D" Dec 29 22:04:49 i have a really stupid problem Dec 29 22:05:03 so i guess i'll keep this stupid vector thing Dec 29 22:05:08 m-dar, chances are the answer is stupid too then :p Dec 29 22:05:14 takes me already too much time Dec 29 22:06:05 it's not just that, but only the galaxy nexus got api14 Dec 29 22:06:15 i auto-added app indexing to my android studio project. I then deleted the auto added code and the library dependency from my gradle file. expect now when I build with gradle a gps version metadata gets added to my manifest. why? Dec 29 22:06:17 and those were all updated to at least api16 Dec 29 22:07:37 meh, all my testing devices are lollipop and newer, so annoying Dec 29 22:08:25 Flash old system images Dec 29 22:08:35 I have an sgs3 with a broken screen on 4.4 and my broken galaxy nexus that doesn't charge on 4.3 Dec 29 22:08:44 nexus 9 doesn't go under 5.0 Dec 29 22:08:48 and 6p doesn't go under 6.0 Dec 29 22:12:29 and the moto x can't be downgraded Dec 29 22:35:06 <_abc_> Can someone please explain how to downgrade from platform tools 23.1 to 23.0 ? Dec 29 22:35:50 <_abc_> I have Android SDK Manager open, I see SDK tools only 24.4.1 and platform tools only 23.1 but build-tools go back to way back when (obsolete) Dec 29 22:36:01 <_abc_> What the hell is going on? There is no way to downgrade at all? Dec 29 22:36:06 on the android, openjdk subject: (not sure if this site is a reliable source) http://venturebeat.com/2015/12/29/google-confirms-next-android-version-wont-use-oracles-proprietary-java-apis/ Dec 29 22:37:01 "Google confirmed to VentureBeat that Android N will rely solely on OpenJDK." Dec 29 22:37:46 \o/ Dec 29 22:38:26 bitkiller oh lol, i just read that on HN Dec 29 22:41:45 _abc_, platform tools don't affect build Dec 29 22:41:50 _abc_, so you don't downgrade those Dec 29 22:45:08 bitkiller thats the best news i've heard ... in a long time ;) Dec 29 22:45:13 * g00s suddenly feels much better Dec 29 22:45:55 huh. Cool Dec 29 22:46:06 how close is OpenJDK to Oracle's? Dec 29 22:46:23 openjdk is the opensourcing of the oracle jdk Dec 29 22:46:24 the important thing is, will this mean that we get Java 8, 9, 10, and beyond? Dec 29 22:46:47 I wonder if that means aosp will be made gpl Dec 29 22:46:59 or if google has paid oracle an obscene sum for a license Dec 29 22:47:05 truckcrash hopefully close enough some projects (netty) dont fret about supporting android Dec 29 22:47:14 or rxjava 2, etc Dec 29 22:47:51 anyway, further refutes the common FUD that "scala is a dead end for android" Dec 29 22:51:24 <_abc_> pfn: adb is part of platform tools and I have to d/g it to match the glibc version. It worked before the unwanted update, i.e. this morning. Dec 29 22:51:34 It might not be a dead end, but kotlin's in the fast lane Dec 29 22:51:53 <_abc_> pfn: So I downgrade those, because I will not reinstall an OS just to match a dependency someone decided to hard code into adb Dec 29 22:52:25 <_abc_> Again: how does one downgrade the adb version used inside studio, part of platform tools Dec 29 22:52:27 _abc_, it's a 32bit glibc, who cares, there's nothing to reinstall Dec 29 22:52:30 you don't downgrade Dec 29 22:52:42 <_abc_> pfn: it requires glibc_2.14 I have .13 Dec 29 22:52:53 then upgrade your userspace Dec 29 22:52:55 <_abc_> So adb does not work Dec 29 22:52:57 _abc_ time to reinstall linux ! Dec 29 22:53:12 debian wheezy, seriously ? Dec 29 22:53:16 <_abc_> It's not possible to upgrade "userspace" glibc upgrade means upgrade of the entire os Dec 29 22:53:28 <_abc_> yes it's called 'oldstable' now. Since September. Dec 29 22:53:32 <_abc_> Tough. Dec 29 22:54:02 debian users have to be tough enough to handles these things, indeed Dec 29 22:54:06 or use ubuntu Dec 29 22:54:07 <_abc_> There are exactly 4 versions to support, is it that hard to NOT mess up important dependencies? Dec 29 22:54:32 <_abc_> why the frick is adb not built on the target system anyway? That WOULD work. Dec 29 22:54:44 <_abc_> Automatically too upon installing studio Dec 29 22:56:10 _abc_, no, it's completely possible to upgrade 32bit glibc Dec 29 22:56:27 <_abc_> Also updates should check local system deps and NOT install if there is a problem. Dec 29 22:56:35 <_abc_> pfn: I tried to and does not work. Dec 29 22:57:10 _abc_ this happened to me also, i had to update from OS X 10.7 to 10.10 when they shipped adb using new glibc or whatever Dec 29 22:57:16 oh wait, adb is 64bit now? Dec 29 22:57:27 <_abc_> Good morning Dec 29 22:57:27 i did file that bug by the way, to not install if the system doesnt meet requirements Dec 29 22:57:36 /home/pfnguyen/opt/android-sdk-linux/platform-tools/adb: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU Dec 29 22:57:39 interesting Dec 29 22:57:44 when did that happen Dec 29 22:57:51 <_abc_> DURING THIS UPDATE Dec 29 22:57:53 <_abc_> GOOD MORNING Dec 29 22:59:09 <_abc_> Basically the update makes studio unusable. Just like that. Upgrading an OS to suit a tool upgrade is deranged, and must be announced ahead of time. Dec 29 23:01:20 Should I use a different adapter for each of my RecyclerViews? Or just make one monsterous adapter that can handle anything I throw at it? Dec 29 23:01:42 liuwenhao, I tend to use just one Dec 29 23:02:01 liuwenhao, my adapter makes three different kinds of views Dec 29 23:02:24 I guess it depends though... Dec 29 23:02:28 liuwenhao: DO NOT MAKE ONE MONSTER ADAPTER Dec 29 23:02:30 Debian wheezy has been oldstable since April, incidentally; and Studio's system requirements do say "GNU C Library (glibc) 2.15 or later" and "Tested on Ubuntu® 14.04" (though there's probably no way to see how long they've been like that). Dec 29 23:03:15 liuwenhao, everything for me is drawn from the same data source and model though, so s73v3r might be right...it depends on the situation Dec 29 23:03:22 <_abc_> fizzie: see release notes for previous version, which worked fine Dec 29 23:03:39 <_abc_> fizzie: Sorry about April/September mixup Dec 29 23:03:49 Mine will be taken from different data sources and different models. So it probably makes more sense to split it up Dec 29 23:03:50 if the adapter is doing the same thing, then feel free to reuse it. but do not shoehorn several screens worth of crap into one adapter Dec 29 23:03:56 <_abc_> fizzie: also they actually require .14 not .15 Dec 29 23:04:08 s73v3r: Thanks Dec 29 23:05:07 <_abc_> adb output: /home/user/android-sdk-linux/platform-tools/adb: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.14' not found (required by /home/user/android-sdk-linux/platform-tools/adb) Dec 29 23:05:16 liuwenhao: http://hannesdorfmann.com/android/adapter-delegates/ Dec 29 23:05:32 <_abc_> later: /home/user/android-sdk-linux/platform-tools/adb: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.15' not found (required by /home/user/android-sdk-linux/platform-tools/adb) Dec 29 23:05:48 <_abc_> So they went from .13 to .14 to .15 in 3 successive versions Dec 29 23:05:56 <_abc_> latest published glibc is .17 Dec 29 23:06:06 <_abc_> If you ask me, they are going too fast. Way too fast. Dec 29 23:07:23 liuwenhao are you trying to do something like cwac-merge ? Dec 29 23:07:47 _abc_: Current stable version of glibc is 2.22, and glibc 2.15 dates to 2011. I kind of disagree there. Dec 29 23:08:07 debian vs modern binaries is always going to be a struggle Dec 29 23:08:16 They also "require" .15 on the system requirements page, and according to archive.org they've "required" 2.15 like that since at least April 8th this year, even if it might've worked without before. Dec 29 23:09:14 I'm looking at that library g00s, that's what I was thinking of doing. But from a code readability standpoint it seems like it would be much better to just split it up (since i only need 3 adapters or so) Dec 29 23:16:51 <_abc_> fizzie: okay I need to check again. Some web resources are stale... damn people who do not put dates on things. Dec 29 23:17:23 <_abc_> fizzie: My install is from one month ago or so, maybe two. Dec 29 23:17:41 <_abc_> fizzie: and it was an online install so no surprizes wrt stale packages Dec 29 23:27:36 <_abc_> fizzie: checking: I had 23.0.0 build tools installed (August 2015) Dec 29 23:28:14 <_abc_> 23.0.1 dep on libc_2.14 and 23.0.2 (current) .15 Dec 29 23:35:49 you must be on a real old system Dec 29 23:41:18 <_abc_> Not old, OLDSTABLE Dec 29 23:41:40 _abc_: get your ssh android studio working? Dec 29 23:41:41 <_abc_> I found an old installer which has access to the previous versions. Google just hid them in devious ways. I HATE that. Dec 29 23:41:50 <_abc_> Aerobit: yes, ssh -XY does it Dec 29 23:41:55 nice Dec 29 23:42:07 <_abc_> Aerobit: Should be in a faq somewhere was getting a bit annoyed. Dec 29 23:42:27 <_abc_> Not ssh -X but ssh -XY Dec 29 23:46:20 <_abc_> Just for laughs, the exact url for locating the packages is in the Android SDK manager log window, specifically, in my case, now, https://dl.google.com/android/repository/repository-11.xml Dec 29 23:46:47 <_abc_> In which file one finds all the zips of all archived versions. Dec 29 23:49:38 I am trying to uniquely identify android devices without requiring a login Dec 29 23:49:50 is hashing Settings.Secure.ANDROID_ID good enough? Dec 29 23:50:46 w0bni, generally considered not good practice; if you're talking to a web service, just make a random token and store it in your app prefs Dec 29 23:50:51 and use that as your consistent identifier Dec 29 23:51:40 dragorn: sounds like a good idea, thanks! Dec 29 23:53:23 hm Dec 29 23:55:12 <_abc_> And only platform tools are 23.1.0 ?! Dec 29 23:55:26 w0bni, so long as you can tolerate the user wiping data on the app you're in good shape, and that doesn't happen too often Dec 29 23:55:44 dragorn: I am a bit worried about that part Dec 29 23:56:28 currently tempted to use a combination of Secure.ANDROID_ID and other data I might find, and falling back to the app prefs solution if I don't get those Dec 30 00:01:45 http://venturebeat.com/2015/12/29/google-confirms-next-android-version-wont-use-oracles-proprietary-java-apis/ Dec 30 00:01:53 Google confirms next Android version wont use Oracles proprietary Java APIs Dec 30 00:22:51 it said won't use /proprietary/ Dec 30 00:23:05 but it /will/ use openjdk Dec 30 00:23:13 according to that link Dec 30 00:23:18 pretty shitty headline Dec 30 00:23:29 so what constitutes a proprietary java api? Dec 30 00:23:36 or does that mean that by switching away from harmony, it's no longer "proprietary" Dec 30 00:23:53 and using openjdk is /not/ "proprietary" Dec 30 00:24:28 yeah i read that headline and thought exactly the opposite of what the article said Dec 30 00:31:18 I need to add a thing to beg for reviews in my apps Dec 30 00:38:45 any way to get status bar height without density dependant calculations? Dec 30 00:38:57 something like ?android:attr/ Dec 30 00:44:27 no Dec 30 00:44:37 at least I don't remember any Dec 30 00:44:48 ok. how to make transparent status bar and place toolbar below status bar Dec 30 00:45:21 maybe I don't need specific size. more specific it's about toolbar position Dec 30 00:46:14 now I can have transparent statusbar or toolbar below status bar but not both Dec 30 00:46:43 I've always done it in code Dec 30 00:46:49 or maybe you can try fitsSystemWindows Dec 30 00:47:18 fitsSystemWindows ignores windowTranslucentStatus=true Dec 30 00:47:41 sadly Dec 30 00:49:04 I don't recall that being the case Dec 30 00:49:24 but like I said, I've always done that in code Dec 30 00:49:28 since I use iota in qicr Dec 30 00:50:55 https://github.com/pfn/qicr/blob/master/src/main/scala/com/hanhuy/android/irc/MainActivity.scala#L97-L213 Dec 30 00:50:58 heh, and it's not quite pretty Dec 30 00:51:37 wow Dec 30 00:53:17 but all my translucent status bar and systemUI stuff is handled automatically Dec 30 00:53:20 so that's nice Dec 30 00:55:35 I'm trying to avoid programmatically creating views if it is not for the sake of incapsulation Dec 30 00:55:46 but maybe it's the only away in this case Dec 30 00:56:23 you don't have to do it programmatically, you can set margins in code after inflation Dec 30 00:57:27 just find your toolbar view, calculate status bar height, and if v19+ add a statusbarheight top margin Dec 30 00:58:10 I do translucent status bar and systemUI Dec 30 00:58:14 so it's a bit more complex Dec 30 00:58:27 and qicr is really just an experiment for my iota library Dec 30 01:00:45 yeah, I meant translucent status bar. I can have toolbar below status bar but than later become solid color Dec 30 01:02:09 ah, fk, lets calculate it Dec 30 01:07:55 anyone know of any android app tutorials written for experienced winapi/.net developers? Dec 30 01:08:33 i have a msft buddy who's got decades of experience and is just looking for a high-level map of the concept differences, and doesn't really wanna wade through all the "this is how you create a project" Dec 30 01:08:37 experienced developer should be ok with java ;) Dec 30 01:09:06 couldn't resist sorry) Dec 30 01:09:22 sarbs if he has decades of experience, shouldn't it be easy to map out a new domain for study ? Dec 30 01:09:29 vigilancer: yeah, the java's not a prob - more of the APIs, Activity/Service/etc lifecycle components, those kinda differences Dec 30 01:09:40 d.android.com is fine for that Dec 30 01:09:47 guides section Dec 30 01:10:05 g00s: yeah, i told him he'd blaze thru the tutorials quick enough, i was just wondering if there even were any "tutorials" for more experienced ppl Dec 30 01:11:19 sarbs: The Busy Coder's Guide to Android Development Dec 30 01:11:29 i disliked that, too wordy Dec 30 01:11:42 if he wants no fluff, thats a bad choice :D Dec 30 01:12:07 g00s: just read through to get concepts Dec 30 01:13:22 what other choice do we have? Android Internals was quite interesting Dec 30 01:14:11 sarbs he's in for an adjustment coming from MSDN docs :D :D :D Dec 30 01:14:36 g00s: haha no kidding Dec 30 01:15:04 although, where there *is* documentation, i've found the android docs pretty good compared to many Dec 30 01:15:16 the Guides are pretty good Dec 30 01:15:21 javadoc - not so Dec 30 01:15:39 true Dec 30 01:15:58 but he'll love the ability to look at the source and figure stuff out lol Dec 30 01:16:07 hope so hehe Dec 30 01:16:11 he's constantly bitching to me about how closed msft is even internally Dec 30 01:16:17 oh, he'll have to do it alot Dec 30 01:17:11 the best is d.android.com Dec 30 01:17:13 the reference manual Dec 30 01:17:45 sarbs it might be good to have him 'avoid' the compat stuff too, in the beginning Dec 30 01:17:57 good call Dec 30 01:17:59 pretend minSdk=21 ;) Dec 30 01:18:14 i'd almost forgotten about compat stuff Dec 30 01:21:12 the straight top-to-bottom reference manual is too good Dec 30 01:23:39 <_abc_> Okay, I now consider it sabotage. adb in the current (and only) platform_tools for linux is hard linked against glibc_2.14 or glibc_2.15 (version pinned). These versions are not available in ANY mainstream distribution as is, only in some obscure obsolete experimental ones. Dec 30 01:23:49 <_abc_> Who the frick is doing these things? Dec 30 01:24:14 <_abc_> 2.13 is mainstream in wheezy, 2.19 - 2.22 are available in jessie Dec 30 01:24:37 <_abc_> in between 2.13 and 2.19 are all experimental ones, including the required 2.14 and 2.15 (exactly these) Dec 30 01:24:52 <_abc_> relinking is not possible because the developers pinned the eglibc version Dec 30 01:25:12 <_abc_> Just EXACTLY what is going on here? Sabotaging linux developers? Dec 30 01:26:13 <_abc_> even the experimental version from end of 2014 only had 2.17, no trace of 2.14 2.15 Dec 30 01:27:44 _abc_: how recent is that platform-tools? Dec 30 01:27:51 _abc_ ... it is what it is, you have to get over this :D Dec 30 01:27:53 <_abc_> November 2015 Dec 30 01:28:06 <_abc_> g00s: no kidding. I will get over it when the fat lady sings Dec 30 01:28:10 adb is built as part of the aosp build, which packages its own dependencies entirely Dec 30 01:28:29 <_abc_> Last time I had such libc fun was with oracle's java boys who packaged something pinned to a glibc version only available in official paid RH Dec 30 01:28:50 <_abc_> sarbs: So it was built on an experimental stream machine Dec 30 01:28:54 <_abc_> NOT A GOOD IDEA Dec 30 01:29:01 _abc_, what are you talking about? Dec 30 01:29:05 _abc_, it is absolutely current Dec 30 01:29:08 <_abc_> Read the backlog. Dec 30 01:29:12 _abc_, ubuntu 14.04 has it, for example Dec 30 01:29:22 so no, it's not experimental Dec 30 01:29:43 <_abc_> glibc_2.14 and 2.15 are and were in the experimental stream with debian with subverions going to .90 and above each Dec 30 01:29:47 wouldn’t the people who declare if something is experemental be the ones who released it? Dec 30 01:29:52 <_abc_> You be the judge of *why* Dec 30 01:29:57 _abc_, rather, 2.19 works fine Dec 30 01:30:46 adb does /not/ pin against glibc 2.15 Dec 30 01:30:47 period Dec 30 01:30:48 <_abc_> pfn: I installed 2.21-6 and tried to run adb with LD_LIBRARY_PATH= set properly, did not want it. It wants exactly libc_2.14 or .15 and says so Dec 30 01:30:54 nope Dec 30 01:30:57 that's not how it works Dec 30 01:31:06 <_abc_> You don't tell me what I read on my terminal please. Dec 30 01:31:08 I have 2.19 on ubuntu lts and it runs adb perfectly fine Dec 30 01:31:09 <_abc_> I tell you that. Dec 30 01:31:16 LD_LIBRARY_PATH isn't set properly Dec 30 01:31:18 <_abc_> What version of adb. md5sum please Dec 30 01:31:21 but keep convincing yourself that it's correct Dec 30 01:31:27 <_abc_> Bullshit I use it all the time with other stuff Dec 30 01:31:39 pfn: Quit being an ass Dec 30 01:31:54 no more help Dec 30 01:31:56 help is closed up for today Dec 30 01:32:04 <_abc_> 0ce53c7848c0b06c90cea313c74084cc android-sdk-linux/platform-tools/adb Dec 30 01:32:05 you weren’t helping to start with Dec 30 01:32:16 <_abc_> say md5sum or find someone else to troll. Dec 30 01:32:26 nope, s73v3r got in the way Dec 30 01:32:29 he stopped me from helping Dec 30 01:32:29 thank him Dec 30 01:32:34 I'm done Dec 30 01:32:53 pfn: you weren’t doing anything to help in the first place Dec 30 01:32:59 the only one to blame is yourself Dec 30 01:33:02 <_abc_> size 2152530 bytes. Anyone? Dec 30 01:33:06 I was doing something to help Dec 30 01:33:09 but you got in the way Dec 30 01:33:23 your troll behavior has stopped me, reflect upon your behavior and come agan Dec 30 01:33:25 again Dec 30 01:33:25 bye Dec 30 01:33:29 clearly you weren’t, otherwise you wouldn’t have stopped Dec 30 01:33:42 <_abc_> Also there are about 50 messages in forums in google about this exact problem. Dec 30 01:33:48 of course I was, I pointed out to him that I have glibc 2.19 on ubuntu 14.04 and it works fine Dec 30 01:34:04 and telling you to stop being an ass is not troling Dec 30 01:34:13 and that is not helping Dec 30 01:34:36 yes, it is, it's pointing out that he';s clearly wrong about it being version pinned to 2.14 and 2.15 Dec 30 01:34:49 no, it isn’t. it’s saying, “Works for me”, which is not help Dec 30 01:34:51 but since you want to insist that you're so right, you go and help him Dec 30 01:35:20 <_abc_> http://5.12.254.190:8881/mypaste/9e6bd57ce5a0dc07bddb exact error message Dec 30 01:35:26 or you could stop being an ass Dec 30 01:35:36 <_abc_> note supplying other glibc versions did not help Dec 30 01:35:36 people don’t treat you like this when you ask for hlep Dec 30 01:35:43 I don't ask for help Dec 30 01:35:48 <_abc_> Will you get a room please? Dec 30 01:35:52 yes, you do Dec 30 01:36:18 nope Dec 30 01:36:20 keep dreaming Dec 30 01:36:26 <_abc_> plonk pfn Dec 30 01:36:50 _abc_, that simply indicates that the required min-version symbol is not present Dec 30 01:36:59 <_abc_> Seriously, a glibc with devel status and subversion going to 90 should not be used for anything Dec 30 01:37:08 you're very wrong here Dec 30 01:37:23 <_abc_> 2.14.90 ?! Dec 30 01:43:15 much like Cartman and Kyle conersation D Dec 30 01:43:33 *conversation Dec 30 01:43:40 <_abc_> The hockey stick took a long time to show up... Dec 30 01:44:37 don't get it Dec 30 01:45:24 <_abc_> Some discussions in South Park are settled with a hockey stick Dec 30 01:45:30 If you have a Fragment (parent) with a Toolbar, and 2 nested Fragments - and you would like the options menu from the nested fragments to show up in the paren't toolbar - is that possible ? Dec 30 01:45:56 what happens when you try to create an options menu in the nested fragment? Dec 30 01:46:34 s73v3r i would expect it to go to the activity's action bar Dec 30 01:46:53 and you have a separate toolbar from that? Dec 30 01:47:45 s73v3r i have an activity, with DrawerLayout. it has actionbar. But the Left Drawer - has a fragment, with nested fragments Dec 30 01:48:04 yeah i want the menus to be in the drawer's parent fragment, not in the activity's AB Dec 30 01:48:54 I might try it to see, but I’m thinking you’re right, and it would show up in the actionbar. Mainly based on google’s anemic support of nested fragments so far Dec 30 01:50:05 i'll try this http://stackoverflow.com/questions/29164263/add-new-option-menu-from-child-fragment-of-fragmenttabhost-inside-outer-fragment Dec 30 01:50:15 s73v3r one of those good ideas, poor implementations Dec 30 01:50:39 sadly i think that’s your best bet Dec 30 01:52:01 vigilancer, how go your adventures with sbt? or have you stopped using protify Dec 30 02:00:51 pfn: I'm using sbt occasionally , via plugin tho. Pure sbt builds require to deal with scala from time to time, not much time saver after all) Dec 30 02:01:18 Wonder if I should put the Google announcement in the topic, SimonVT thoughts? Dec 30 02:02:49 i dunno. it hasn’t really affected anything yet Dec 30 02:02:58 What announcment Dec 30 02:04:17 openjdk? Dec 30 02:05:48 "Toolbar#inflateMenu : Inflate an XML menu resource into this toolbar. Existing items in the menu will not be modified or removed." -- crap, so how does one reset the menu Dec 30 02:06:53 There's even Swing in that commit, for crying out loud. They can't possibly think of implementing that. Dec 30 02:07:10 lol what about JavaFX Dec 30 02:07:28 material design is going to be replaced with Swing Dec 30 02:18:05 s73v3r now they can port AS to Android :D :D Dec 30 02:39:30 vigilancer, I never expect you to do pure sbt builds, I mostly expect via. android-gradle-build Dec 30 02:39:37 <_abc_> fwiw I got glibc_2.15 from slackware 14.0 and installed it in a non system path, am running avd with LD_LIBRARY_PATH= method and it works Dec 30 02:39:50 <_abc_> Thanks for the patience. Dec 30 02:39:57 vigilancer, as for dealing with scala, it's much more pleasant than java :p Dec 30 02:40:43 path to the dark side and all Dec 30 02:40:54 <_abc_> Next gripe: I have a SSE2 only capable athlon on this headless box, but the kvm for x86 wants SSE3 and won't run at all, accelerated or not, without. Is this reasonable? Is there a way to run on x86 device in the emulator without acceleration? Dec 30 02:41:46 change one line of code and the build take 1min 30 seconds ... upgraded my gradle today, change the vram, thoughts / suggestions? running 1.5.1 on mac and testing directly on device Dec 30 02:42:42 gradle is slow, use instant run Dec 30 02:42:50 ... if you can Dec 30 02:42:55 <_abc_> ... Dec 30 02:43:05 * _abc_ hates "instant" used in this context Dec 30 02:43:20 if you want a working instant run, use protify Dec 30 02:43:30 "Faster than the default speed" run Dec 30 02:43:35 is that only available on 2.0? if not link on how to get this setup? Dec 30 02:44:16 nownot: Is your app available on github by any chance? Dec 30 02:44:23 nope Dec 30 02:44:29 Ok Dec 30 02:44:55 I'm working on a build system, I was curious to try it and see if its incremental build is better than Gradle's Dec 30 02:45:10 CedricBeust_, if you're looking for testing apps, should start with basics like plaid, viewserver+sample, u2020 etx Dec 30 02:45:25 u2020 already building but writing these other ones down, thanks Dec 30 02:45:31 its pretty frustrating ..... Dec 30 02:45:42 CedricBeust_, oh android Topeka Dec 30 02:45:48 CedricBeust_ i guess also iosched app Dec 30 02:46:40 CedricBeust_, forkhub/android is good to build as well, heh Dec 30 02:46:52 since that's the only decent github mobile app Dec 30 02:47:00 and it's not on play yet Dec 30 02:47:18 CedricBeust_, u2020 builds? all flavors or just a canned config? Dec 30 02:47:26 All flavors Dec 30 02:48:09 https://github.com/cbeust/u2020/blob/build-with-kobalt/kobalt/src/Build.kt Dec 30 02:48:37 hot Dec 30 02:48:47 Doesn't have all the bells and whistles but the bigger part is done (resource merging, manifest merging, variants, retrolambda, predex and dex, etc...) Dec 30 02:49:20 I also build all my client apps with sbt, but they're all closed source, of course Dec 30 02:49:22 and kobalt's own incremental engine on top of that Dec 30 02:50:01 Take a look at the bottom of this file: https://github.com/cbeust/kobalt/blob/master/kobalt-incremental-tasks.md Dec 30 02:50:44 checksum is good but heavily disk intensive Dec 30 02:50:56 Yeah, it's not the final plan but it works ok for now Dec 30 02:50:57 I had to move away from checksum. based incremental tasks Dec 30 02:51:09 because it is incredibly slow, even on ssd Dec 30 02:51:10 I'll offer alternatives soon, like timestamps Dec 30 02:51:23 but for now, checksum is very effective, even on large projects Dec 30 02:51:32 and it's task agnostic: applies even to tasks that are not file based Dec 30 02:53:32 my checksum gated aapt task was taking like 6 seconds to process checksum while just letting aapt run only took 3 seconds Dec 30 02:53:35 pfn: you asked about sbt adventures. It'll be nice to have control on instrument or at least understanding of internals. Guess I'm just too slow for scala) Dec 30 02:53:36 as an example Dec 30 02:53:58 FYI, created an issue with your suggestions, feel free to add to it: https://github.com/cbeust/kobalt/issues/89 Dec 30 02:54:19 pfn: Interesting about aapt, will keep that in mind Dec 30 02:54:36 as for scala/java comparison.. kotlin my new hope for now) Dec 30 02:54:50 I'm not overly aggressive for the Android build, I'm assuming there's already a lot of intelligence in the com.android.builder code so I'm more likely to always run them and hope that Google is being smart about that Dec 30 02:56:34 I actually switch to timestamp checks for predexing for example, since calculating checksums for all these jars seems silly Dec 30 02:57:43 pfn: Obviously, I'd love to get contributions from you for kobalt-android since you've written so much code using com.android.builder already, but I understand you prefer Scala over Kotlin :) Dec 30 02:59:33 I'm C++ coder And Working with Qt/C++ App, my App now to read IMEI, My Code http://pastebin.com/z9SS1UR1 What I'm asking is java code is right ? Dec 30 02:59:58 from line 2 to 20. **** ENDING LOGGING AT Wed Dec 30 02:59:59 2015