**** BEGIN LOGGING AT Sat Feb 13 02:59:58 2016 Feb 13 03:12:34 if I have an AppBarLayout, with Toolbar + TabLayout, can I make both scroll out of the way? examples I see only put app:layout_scrollFlags="scroll|enterAlways|snap" on the Toolbar Feb 13 03:18:18 well, windows got updated to windows relatively painlessly Feb 13 03:19:40 g00s, always abandon Android UI framework as soon as possible in favor of lower-level implementations Feb 13 03:19:58 knapper_tech O.o Feb 13 03:20:22 i dont want to re-invent coordinator layout and nested scrolling ;) Feb 13 03:20:24 I was thinking of using Spoon, but it has no Gradle plugin...and...what does it do that is useful? Feb 13 03:20:47 wth is spoon Feb 13 03:21:07 some jar that runs instrumentation tests and produces output as static html Feb 13 03:29:05 g00s: yes, as long as the tablayout is still a child of appbarlayout. I did it in the first design of an app I was making, dont remember it being very complicated Feb 13 03:29:56 orbyt_ ok thanks Feb 13 03:31:01 yup make the appbar (which contains toolbar+tablayout) collapse with app:layout_scrollFlags="scroll|enterAlways" Feb 13 03:31:21 don't forget to use app:layout_behavior="@string/appbar_scrolling_view_behavior" on the scrolling view Feb 13 03:31:32 and wrap everything with a CoordinatorLayout Feb 13 03:31:57 adq oh, so you put scrollFlags on the AppBar, not the children then ? Feb 13 03:32:34 * g00s tries Feb 13 03:32:52 oh indeed, you have to put them in the children Feb 13 03:33:12 hm, so children + appbar ? Feb 13 03:33:17 so all 3 views .. Feb 13 03:33:24 i'm checking without appbar right now Feb 13 03:33:41 no need for the appbarlayout itself Feb 13 03:33:56 only its children then Feb 13 03:34:10 ok thanks adq Feb 13 03:35:19 my understanding is that Android will ask an app to serialise if it needs to reclaim memory… if so, how can I force this scenario to test how my app behaves? Feb 13 03:47:57 Anyone familiar with Rx mind checking this out: http://stackoverflow.com/questions/35353242/executing-a-network-request-every-x-seconds-how-to-pause-resume Feb 13 03:48:27 Basically im polling an api every x seconds, but needed a way to pause the polling when the fragment went out of view Feb 13 03:48:41 my current solution has me duplicating code Feb 13 04:08:36 erisco: start a web browser with several large background tabs? Feb 13 04:08:50 Those usually just eat your memory away. Feb 13 04:35:45 Anyone here use JRebel? Is it worth using? Feb 13 04:36:18 I'm using a slow laptop for dev now (used to use a desktop) so Gradle builds are painfully slow and Instant Run is super buggy Feb 13 04:42:07 Tried it for Android when it was in beta and have tried it with other Java projects. I dont know what improvements they've made to the Android version but its useful for doing UI stuff Feb 13 04:43:17 jrebel seems like it works ok for small projects Feb 13 04:43:30 $50 is cheap Feb 13 04:44:44 there's also protify, but I've never optimized for spinning disk Feb 13 05:05:51 I hate the boilerplate of using composition in Java. I'm sitting on a case right now where inheritance would save dozens of lines (but it'd be less easy to test). Feb 13 05:07:51 Inheritance is so often the path of least resistance, though. It's like an alcoholic deciding on "just one more beer". Feb 13 05:09:00 I haven't used Kotlin much; I wonder whether their first-class delegation helps a lot on Android. Feb 13 05:14:14 then use it more and decide Feb 13 05:50:45 And that's all from Buzz McKillington. Now back to the studio. Feb 13 06:30:28 I don't use native delegation much, I actually realize I like it less than I thought I would. I initially wrote a proposal very similar to Kotlin's in 1998 so I was very excited about it. Looking at my Kotlin code right now, I've used it 3-4 times top. Feb 13 06:30:40 Not sure if it's because it's less useful or just that I don't think of using it as often as I should Feb 13 06:31:07 (FTR, my proposal from almost 20 years ago: http://www.beust.com/java-delegation.html) Feb 13 06:39:56 CedricBeust: WHat is native delegation? Feb 13 06:40:25 You can say you implement an interface and automatically delegate it to a field inside your class with the right type Feb 13 06:41:29 CedricBeust: Hmm, do you mean if you have a Java interface that declares getX and setX methods, and the Kotlin delegate allows you to have a field called "x" and thus implement the interface? Feb 13 06:41:56 loke``: https://kotlinlang.org/docs/reference/delegation.html Feb 13 06:43:05 CedricBeust: Oh I see. Feb 13 06:43:17 CedricBeust: I can see how that's useful. Feb 13 06:43:46 (it's also funny how so many of these features aren't even needed in a "better" object system like CLOS :-) ) Feb 13 06:43:57 It heavily hinges on the mantra "inheritance is useful, delegation is the only proper way". As it turns out, inheritance is not remotely as bad as a lot of people claim. Feb 13 06:44:10 Well CLOS has an MOP so it's kind of cheating Feb 13 06:44:24 it's also extremely overkill and hardly ever used in practice (both CLOS and its MOP :) ) Feb 13 06:44:40 s/useful/awful Feb 13 06:45:23 CLOS: "Here is a language and you can completely rewrite its semantis". "Thanks but we'll take the language. Thanks, though" Feb 13 06:45:34 CedricBeust: I wouldn't say "hardly ever used". You might not make many metaclasses, but the ones you do make are generally framework style ones, making them quite central. Feb 13 06:46:30 Can't comment, haven't touched CLOS in a solid 20 years Feb 13 06:46:41 I give you this, as a counterexample. A CLOS/MOP-based implementation of Object-to-Couchdb mapping, with transparent caching using memcached: Feb 13 06:46:43 https://github.com/cicakhq/potato/blob/master/src/potato/metaclasses.lisp Feb 13 06:47:32 Actually, this is better demo. It's an object using the features of this metaclass: https://github.com/cicakhq/potato/blob/master/src/potato/channel.lisp#L8 Feb 13 06:48:07 (also includes change-tracking) Feb 13 06:48:23 Now... I've been doing quite a bit of Kotlin recently, and I have been thinking about ow the same thing would be done there. Feb 13 06:49:11 It seems like some of it can be done using the delegate stuff, but as far as I can tell there is no way I can create a method that is called when a field is modified (and has the ability to override it) Feb 13 06:49:13 Am I wrong? Feb 13 06:49:47 How about setters of properties? Feb 13 06:50:51 val age: Int set(newValue: Int) { ... } Feb 13 06:51:02 CedricBeust: Yes of course. But wouldn't I have to declare the delegate of every property manually? CLOS has SLOT-VALUE-USING-CLASS which is a generic function that is called for _every_ slot modification. Thus, to get change tracking working, all I have to do is implement function method once. Feb 13 06:51:30 I'm not sure what you're trying to do since it's already supported in Kotlin Feb 13 06:51:53 Is it? Feb 13 06:52:02 Oh intercept all sets on properties of a given class? Yeah, don't see how you could do that right now Feb 13 06:52:21 Such a mechanism would impact the performance significantly, even for programs that don't use it. Feb 13 06:52:53 But at least the amount of boiletplate needed can be minimised to a single "by XXX", right? Feb 13 06:53:03 I'm still a huge fan of one of the rules that guided C++'s design: You should never pay for features you don't use. Kotlin scores pretty high on that scale Feb 13 06:53:04 XXX can implement all the necessary logic? Feb 13 06:53:37 CedricBeust: "high" as in "no paying"? Or "high" as in "you're paying a lot". Feb 13 06:53:50 I mean that as a positive Feb 13 06:53:59 Kotlin doesn't charge you (much) for features you don't use Feb 13 06:54:46 CedricBeust: For Android that's good, actually. And Android is the only yhing I'm using Kotlin for. I'd much rather use, say, ABCL for Android, but it scroes so much on the opposite end of that scale that it isn't practivcally possible. Feb 13 06:54:59 ABCL? Feb 13 06:55:48 Common Lisp on the JVM. It's very good actually. Anyway, Scala and Clojure suffer from the same problem when it comes to Android. Feb 13 06:57:19 scala doesn't suffer on android Feb 13 06:59:24 pfn: It doesn't? I was taking to a Scala developer who said it does. I don't know Scala so I wouldn't know myself. Feb 13 06:59:41 it doesn't Feb 13 07:00:19 there's no downside to using scala on android except for lack of official support Feb 13 07:00:25 pfn: Would you recommend using Scala for Android development, or are there other issues outisde of the langage or ART itself? Feb 13 07:01:05 I fully recommend scala for android, if that interests you, I find kotlin to be pretty limited Feb 13 07:01:33 pfn: Kotlin _is_ pretty limited, but as a "better Java" it does what it says on the tin. Feb 13 07:01:45 That said, I disagree with a some of their design decisions Feb 13 07:02:21 yes, I agree that it's a good java++, and that's what the masses want. Feb 13 07:02:30 I'm a dynamic typing kind of guy though. Scala seems to be incredibly static :-) Feb 13 07:03:29 That said, dynamic languages on the JVM tends to have performace problems unless the compiler has really good type inference, and even then it's easy to screw it up, causing an innocent change to cause a performance drop. Feb 13 07:03:46 (a static language simply refuses to compile if the type inference fails) Feb 13 07:04:13 Kotlins' limitations are a feature. If you don't need this feature. use another language. Feb 13 07:05:21 loke``, scala has a pretty amazing type system, and I utterly despise working in dynamically typed systems Feb 13 07:05:41 pfn: Sure. I never said you were wrong or that I was right. :-) Feb 13 07:06:05 loke``, but yes, it is absolutely static Feb 13 07:06:19 Both systems has their benefits and drawbacks. Which ones are important to your project depends on a huge number of factors. Feb 13 07:06:57 pfn: I'm hygely facinatied by static languages though. I started learning Haskell but despised the experience. Not because of the staticness, but because of other aspects of the language. Feb 13 07:07:04 Perhaps I should look at Scala. Feb 13 07:07:26 lots of people try to emulate haskell on scala Feb 13 07:07:36 loke`` Ceylon is apparently going to be supported on Android, within the next release if i recall Feb 13 07:07:41 that's basically the entire premise of scalaz Feb 13 07:08:20 pfn: Scalaz and its community seems to be quite painful. Feb 13 07:08:39 I'm actually trying to come up with a pure api for activities and fragments. Feb 13 07:09:01 thats mostly dibblego, he's always been painful. Feb 13 07:09:09 Thanks for the info. Need to go for lunch now. Feb 13 07:09:32 loke`` maybe you'll me using dart soon :P Feb 13 07:10:40 For a more reasonable approach, look at the cats library Feb 13 07:10:49 But plenty of very solid concepts come from scalaz/cats, no doubt Feb 13 07:10:53 if coming from lisp, dart is not attractive Feb 13 07:11:04 crap, the basic CoordinatorLayout / AppBarLayout thing acts weird when you go full screen Feb 13 07:11:52 cats is basically a reimplementation of scalaz, except intended to be modular, and without the political issues Feb 13 07:11:55 so i have one of these cheap iTag BT LE devices and i can't get it to pair. i'm stumped about how to go about debugging the connection Feb 13 07:12:16 suggestions welcome Feb 13 07:12:16 Hi. Feb 13 07:12:26 rigel hmm, in dev options you can turn on HCI logs Feb 13 07:13:40 I have a WebViewClient, and I would like to fire an event in it when onReceiveError() happens (i.e. when it can't open a page) so that the MainActivity can show a button. How can I do that? Feb 13 07:13:52 CedricBeust, I never knew you were infamously suspected for being "the" scala troll, heh Feb 13 07:14:05 uh? Feb 13 07:14:07 lol i saw a nick called cbuestwatch Feb 13 07:14:14 g00s: im not seeing a dev options in settings Feb 13 07:14:40 rigel you do have dev mode enabled right ? Feb 13 07:15:20 that guy is Simon Ochsenreither, not sure why he's so obsessed with me, but he calmed down a lot Feb 13 07:15:38 what is it again, about -> status ? Feb 13 07:15:57 soc claims that it's not him as well Feb 13 07:16:05 * CedricBeust shrugs Feb 13 07:16:20 Cedric u work at google? Feb 13 07:16:29 Used to Feb 13 07:16:32 g00s: now i do, i cant believe i didnt have it enabled Feb 13 07:16:35 https://www.reddit.com/r/scala/comments/454ahd/recent_dottyscala_fud/ Feb 13 07:16:41 b4b i thought you were a bot ! Feb 13 07:21:13 5 second app startup time for nyt, that's pitifully long for something that just reads the news Feb 13 07:21:25 I blame Groovy Feb 13 07:21:44 I mean these guys picked Groovy just so they could have the lambda syntax Feb 13 07:21:57 Talk about putting themeselves (the developers) before their users. Feb 13 07:22:24 "The code looks nice, who cares if it's slow" Feb 13 07:23:41 groovy was a good amount of time Feb 13 07:23:56 speaking of json parsing, yeah, gson is fucking slow Feb 13 07:24:08 and slow is anything that relies on reflection Feb 13 07:24:34 pfn: gson is slow? It was decent last I checked, do you have some data? Feb 13 07:24:38 What is the most idiomatic way to show a button when the web view cannot load a page? Feb 13 07:25:05 Argonaut and circe are amazingly fast on android. stuff that was taking like 5 seconds to parse in gson was like sub second in argonaut Feb 13 07:25:24 I'm a bit skeptical of that claim Feb 13 07:25:27 CedricBeust, several hundred kb of json Feb 13 07:25:58 every reflective call takes roughly 50ms on android, on a modernist devuce Feb 13 07:26:11 modernish Feb 13 07:26:17 That's not data for the claim Feb 13 07:26:33 look ups are slow overall, invocations pretty fast. But yes, reflection is still a no-no on Android overall Feb 13 07:26:33 i don't have anything handy right now Feb 13 07:26:54 I'm still skeptical of some JSON that would take 5 seconds to parse in Gson and subsecond in any other framework (argonaut or other) Feb 13 07:26:54 not at a pc to put good comparisons together Feb 13 07:27:12 If you ever have a github project I can clone and build to run tests, I'd be interested to Feb 13 07:27:28 I recommended gson across our org, happy to revisit that choice if there's data to support it Feb 13 07:27:30 it was private client feeds Feb 13 07:27:48 I'll anonymize it some day and write up some comparos Feb 13 07:28:07 Would be great. Make it available to others too so we can run it on various computers Feb 13 07:28:21 on the jvm, the diff is negligible Feb 13 07:29:14 Yes, I'm interested in the Android case Feb 13 07:30:22 Hey! I am trying to fetch user data from google play game services but have no idea where to start Feb 13 07:30:33 I have integrated it to my game Feb 13 07:30:53 but don't know how to fetch user data like name, emails connected etc Feb 13 07:32:55 I'm actually interested in trying out rust the next time I have a problem that needs solving in jni Feb 13 07:33:18 Java -> jni -> Rust? Feb 13 07:34:53 yeah, building an so directly with rust that's loaded Feb 13 07:36:15 That's a bold strategy, let's see if it will pay off, Cotton. Feb 13 07:39:01 I very rarely use jni, though, so it'll be unlikely any time soon Feb 13 07:45:37 so in genymotion i sometimes get this glitch http://wstaw.org/m/2016/02/13/001.png Feb 13 07:45:53 when expanding notification bar or changing activities Feb 13 07:46:02 someone has encountered similar problem? Feb 13 07:46:26 fedora, radeon 7770 open gpu drivers Feb 13 07:46:50 i mean wtf is this Feb 13 07:52:04 How can I write a replacement NfcAdatper service? Feb 13 07:53:36 *NfcAdapter Feb 13 08:06:59 who here voting trump? Feb 13 08:07:22 hah we'll see if I even voite Feb 13 08:07:35 ever see the south park where they elect a new mascot? Feb 13 08:07:53 Not appropriate for this channel Feb 13 08:56:29 Can anyone tell me what the most idiomatic way is to show a retry button when WebView fails to load a URL? Here is my question on StackOverflow if you want to answer there: http://stackoverflow.com/questions/35377353/what-is-the-most-idiomatic-way-to-show-a-retry-button-when-the-receiveerror-even Feb 13 10:43:14 hello, is there a document mentioning AOSP/Android versions and the linux versions on which they are based? Feb 13 10:55:17 anddam: sadly, that’s not possible – different OEMs use different kernels for different versions Feb 13 10:55:59 While the nexus devices used 3.1.10 on Android 5.1, I’ve seen HUAWEI devices on a 2.6 kernel running 5.0 Feb 13 11:08:44 justjanne: tanks, is there something just for AOSP then? I wasn't able to find any on android.com Feb 13 11:09:03 (or wikipedia that has just a partial description of the relationships) Feb 13 11:09:31 I'm specifically interested in tracing so the different linux versions matter as they implement different tracing facilities Feb 13 11:10:07 I’ll specify api level and kernel version, as known: Feb 13 11:11:38 3 => 2.6.27; 4 => 2.6.29; 5 => 2.6.29; 8 => 2.6.32; 9 => 2.6.35; 11 => 2.6.36; 14 => 3.0.1; 16 => 3.0.31; 17 => 3.4.0; Feb 13 11:11:51 After that, no info was released in the official Release Announcements anymore Feb 13 11:11:58 yay for docs Feb 13 11:12:00 thanks Feb 13 11:12:09 manually taken fromhttps://en.wikipedia.org/wiki/Android_version_history Feb 13 11:16:41 anddam, https://android.googlesource.com/kernel/ Feb 13 11:17:06 adq: that doesn’t tell you which android version used which kernel, though Feb 13 11:18:58 well, it partially does, take android wear kitkat kernel for example: https://android.googlesource.com/kernel/msm/+/android-msm-dory-3.10-kitkat-wear/README Feb 13 11:19:41 adq: and which Android version does that relate to? Feb 13 11:19:42 and some version inside makefile & other config https://android.googlesource.com/kernel/msm/+/android-msm-dory-3.10-kitkat-wear/Kconfig leads to https://android.googlesource.com/kernel/msm/+/android-msm-dory-3.10-kitkat-wear/Makefile (KERNEL_VERSION) etc Feb 13 11:19:45 it's just painful Feb 13 11:19:45 is there any good note taking application for android and ubuntu..i have used evernote earlier when i was using windows..now i have shifted to ubuntu completely and i'm missing only my evernote. Feb 13 11:19:50 justjanne, android wear kitkat Feb 13 11:19:56 oh, now I see it Feb 13 11:20:10 i took an example, but i agree it's just painful Feb 13 11:20:33 something I’d do when continuing work for the wikipedia article, but not today Feb 13 11:20:39 also the codename are obscure for wear, like it's all based on fish names Feb 13 11:20:46 that reminds me, gonna have to check the edit history of some articles Feb 13 11:20:53 and you need to know the relation between manufacturer/OEM and fish codename -__- Feb 13 11:21:04 i cannot imagine what it is for regular android device Feb 13 11:21:17 (and funny i took wear cause it's the least open of all android stuff) Feb 13 11:22:28 the main point is all kernels are seriously outdated on android anyway Feb 13 11:34:55 so I'd have to make a list of platforms and go checking one by one Feb 13 11:35:09 yes. Feb 13 11:35:17 And that’s only for AOSP devices Feb 13 11:35:23 other vendors have their kernels in other places, Feb 13 11:35:27 if available at all Feb 13 11:35:32 painful, as you said, but again I probably won't need the whole map but my choice will be limited to a few targets Feb 13 11:35:40 thanks for the info Feb 13 11:35:53 it might be best to check at runtime Feb 13 11:36:03 or to write an app which sends the kernel version to your server, Feb 13 11:36:06 and make it popular Feb 13 11:36:16 it's for a in-house study, no app deploye Feb 13 11:36:19 deployed Feb 13 11:36:26 no, deploy an app to get the stats Feb 13 11:36:33 like, a flashlight app which phones home Feb 13 11:36:52 but that'd be useful to gather statiscal info Feb 13 11:37:28 which is not my point Feb 13 11:37:42 well, it’d give you a full map version => kernel Feb 13 11:37:46 including all weird vendor bugs Feb 13 11:40:57 is there a particular online course that's considered the de facto, particularly given someone who's already a programmer, but on other languages, but has done android before but wants to learn to do it properly? Feb 13 12:21:53 does someone have experience with https://github.com/serso/android-checkout ? Feb 13 12:43:02 > To keep your public key safe from malicious users and hackers, do not embed it in any code as a literal string. Instead, construct the string at runtime from pieces or use bit manipulation (for example, XOR with some other string) to hide the actual key. The key itself is not secret information, but you do not want to make it easy for a hacker or malicious user to replace the public key with another key. Feb 13 12:43:04 ROFL Feb 13 12:43:23 > You should obfuscate your In-app Billing code so it is difficult for an attacker to reverse engineer security protocols and other application components. At a minimum, we recommend that you run an obfuscation tool like Proguard on your code. Feb 13 12:43:58 Every time I read that, I get the feeling like the people writing that have never even had any experience with designing secure systems Feb 13 12:44:05 Spoiler: Obscurity is not Security. Feb 13 12:44:24 justjanne, unfortunatly theres no other way if everyone can decompile your app :/ Feb 13 12:44:40 danijoo_: Everyone can decompile and modify your app content anyway Feb 13 12:44:54 yes Feb 13 12:45:03 thats what I just said :) Feb 13 12:45:08 the only choices you have are (a) trusting a third party (very few people manage to reverse and modify Google Play Services) Feb 13 12:45:19 or (b) loading content at run time from your server Feb 13 12:45:25 thats why you verify iab products on your own server Feb 13 12:45:27 but even with (b), anyone can modify it anyway. Feb 13 12:46:01 If we’ve learnt anything from piracy on the desktop, it’s that if someone wants to pirate your stuff, they will find a way. Feb 13 12:47:00 no reason to make it easy ;) Feb 13 12:47:07 Also, if I ever try to use an app from you, I’ll at least reverse it and read through every part of it before using it Feb 13 12:47:36 you are not john doe, so its fine Feb 13 12:47:48 hmm Feb 13 12:47:50 that’s the issue: Feb 13 12:48:00 if at least one malicious person ever reverses it, Feb 13 12:48:05 anyone else can get it, too. Feb 13 12:48:07 also have fun with proguarded non java code. it will be a hard read :p Feb 13 12:48:14 Not everyone who pirates games is experienced, Feb 13 12:48:20 I wonder if retrieving the public key from an API and then encrypting it with secure enclave would make this a bit more annoying. Feb 13 12:48:22 but it’s enough for one person to release a crack Feb 13 12:48:31 Mavrik, i thought about that too Feb 13 12:48:40 danijoo_: I’ve started coding as a kid working on Minecraft mods, before MCP was a thing ;P Feb 13 12:48:41 but I think its not really worth the effort Feb 13 12:48:51 The annoying thing is that you need to have some kind of static key anyway. Feb 13 12:48:57 That’s proguarded game code, changing with every release. Feb 13 12:49:00 Since you could use secure enclave to generate the key. Feb 13 12:49:22 Well, that’s the solution most people use. Use the secure enclave to generate a key, transmit public key to server. Feb 13 12:49:29 you can then retroactively revoke it. Feb 13 12:49:36 if someone has the will to go through proguarded XOR methods to receive the key, he will also have the will the extract if from what the server gives you Feb 13 12:49:41 so the pirate can’t get any /future/ content Feb 13 12:50:29 The best working solution is: steganographically mark the content each user receives. When you see pirated content, revoke the access of the user who got the content. Feb 13 12:50:43 now the pirates have to pay at least once every time they try to access anything.. Feb 13 12:50:52 that’s part of the reason why game studios love DLCs Feb 13 12:54:12 not worth doing that in that case :) Feb 13 12:57:21 its not the weak spot anywayss Feb 13 12:57:42 oh nice.. my keyboard starts doing double ss again -_- Feb 13 12:58:02 lol Feb 13 12:58:37 second time I replaced the mechanics.. seems like its time to get a new one. Feb 13 13:08:18 hello guys! Feb 13 13:09:53 Where I work we use samba 4 AD for user authentication. We have thousands of users. We are developing internal android apps and my question is: is there a way for our android apps use our already working samba 4 AD user base for authentication? Feb 13 13:13:48 Hi, what all options do I have to distribute my app for testing among alpha users ? Feb 13 13:14:03 I am using Fabric Beta for now Feb 13 13:14:08 looking for other options Feb 13 13:16:15 suhas_sm, check in your google play developer console of an app in "apk", beta testing / alpha testing and read about it online Feb 13 13:16:54 but I haven't uploaded it on google play yet Feb 13 13:17:00 but might give it a try Feb 13 13:17:37 suhas_sm, https://support.google.com/googleplay/android-developer/answer/3131213 Feb 13 13:17:47 thanks adq :) Feb 13 13:27:50 adq, I will stick to Beta for now. Less Nuances Feb 13 13:34:14 following along the maps example its not clear where the onMapReadyCallback is to be put...seems no matter where I try I get a illegal start of type error Feb 13 13:34:32 * REdOG is obv noob Feb 13 13:44:50 Link? Feb 13 13:45:58 suhas_sm: Google play has added more features to its beta thing Feb 13 13:46:34 suhas_sm: In fabric beta, users have to download another app while in case of google its play store Feb 13 13:47:05 REdOG, your activity or fragment must implement it. Feb 13 13:47:20 I think fabrics beta is way better then google Feb 13 13:47:38 hello Feb 13 13:47:40 more feedback and faster Feb 13 13:47:47 REgOD, you should be getting map reference inside onMapReady Feb 13 13:48:05 Yeah, faster is the key issue here Feb 13 13:48:11 how can i create a secure login system? People say sha1 is not secure anymore Feb 13 13:48:24 Google play says it will be available after hours Feb 13 13:48:41 But I ideally want to release it immediately to my team Feb 13 13:48:52 quick bug fix and release Feb 13 13:49:56 torak: for hashing, bcrypt or scrypt Feb 13 13:50:26 also, think about using session tokens and 2FA Feb 13 13:50:47 If you just use OAuth 2.0, those things are handled by the provider, though, so it’s less work Feb 13 13:51:03 (OAuth is the "login with Facebook/Twitter/Google+/GitHub/GitLab/etc") Feb 13 13:51:18 justjanne: thank you! that is very helpful. :) Feb 13 13:51:38 also, read about salting ;) Feb 13 13:51:53 suhas_sm: ok I think its there correctly but when it is I get "method getMapAsync in class MapFragment cannot be applied to given types method getMapAsync in class MapFragment cannot be applied to given types;" Feb 13 13:53:50 REdOG mapFragment.getMapAsync(this); ? Feb 13 13:54:41 where this is your activity or fragment implementing OnMapReadyCallback Feb 13 13:55:33 yes in the onCreate in main activity Feb 13 13:55:42 Correct Feb 13 13:58:26 where "this" is your activity or fragment implementing OnMapReadyCallback Feb 13 13:58:35 incase you were confused :P Feb 13 13:58:48 still am a little Feb 13 13:59:02 I understand 'this' pretty well Feb 13 13:59:07 heh ok Feb 13 13:59:22 but not quite how to tie together the main and fragment Feb 13 13:59:33 main and fragment ? Feb 13 14:00:24 public class appmain extends fragment implements onmapreadycallback Feb 13 14:01:16 and public class AppActivity extends AppCompat... Feb 13 14:16:11 i got it ! Feb 13 14:21:22 <_maddy> hello Feb 13 14:22:18 hi! Feb 13 14:22:27 how do you fix lint? Feb 13 14:22:54 by doing what lint says Feb 13 14:24:33 it didn't say anythhing really...says lint errors found ... fix lint errors Feb 13 14:24:55 suggests to do what looks like ignoreing them instead Feb 13 14:25:24 fix them or disable lint Feb 13 14:27:13 how do I enumerate theses "thems" Feb 13 14:27:28 <_maddy> I have a problem: my Gradle Build gets stuck on ":app:transformClassesWithDexForDebug" when I select "run app" from android studio Feb 13 14:27:45 all I see is it saying 2 errors 4 warnings but I do not know which lines to fix Feb 13 14:28:46 i see em now Feb 13 14:29:01 lint-results.html Feb 13 15:06:23 WHY IS THIS REFERENCED NOWHERE? o.O Feb 13 15:06:24 http://developer.android.com/tools/data-binding/guide.html Feb 13 15:06:30 Could have saved me so much time... Feb 13 15:07:26 > You can change your data model in a background thread as long as it is not a collection. Data binding will localize each variable / field while evaluating to avoid any concurrency issues. Feb 13 15:07:29 Seriously? Feb 13 15:07:34 Now I’m really angry Feb 13 15:07:46 hm? Feb 13 15:09:41 I could have saved days of work if I had known earlier, but this thing is mentioned nowhere Feb 13 15:10:01 where can i learn about acces tokens? How they work etc? Feb 13 15:29:44 https://www.youtube.com/watch?v=eVD9j36Ke94 Feb 13 16:03:11 hi everyone. i really got stuck and need someone's help. i have Nexus 5 with CyanogenMod 12.1 installed. i created a backup using Online Nandroid backup tool (used tar option for system and data partitions), but when i try to restore it, the phone just boots in a loop into CWM recovery. i successfully flashed CM firmware back and the phone works but i really need all my data back. Feb 13 16:04:02 i tried to restore just /data/system/users/* files from the backup - the phone boots but hangs on asking privacy guard settings for each application Feb 13 16:05:00 so, the question is how to restore the app's data from the nandroid backup? Feb 13 16:05:50 i must say i tried the option in Nandroid Manager which allows to restore user's app and data, without success Feb 13 16:19:15 android-dev044, wrong channel, read topic Feb 13 16:20:04 danijoo: oops. sorry. thanks for pointing Feb 13 16:46:38 is there a forum for code reviews ? Feb 13 17:56:16 suhas_sm: codereview.stackexchange.com Feb 13 17:56:41 thanks, vedu Feb 13 19:03:27 when i'm getting an okhtp result it goes to a new activitybut i dont want that activiy to popup when the app is in the backgorund what's the scoop there? Feb 13 19:03:57 Um, what does that have to do with okhttp? Don't invoke the intent in background. Feb 13 19:11:26 Mavrik well how would i not invoke the intent in the background? when should I invoke it? It seems that normally android is not supposed to surface the views if the app is in the backgorund is there a flag I need to set? Feb 13 19:11:45 Calguy: In some cases, you can simplify such UIs by showing the activity as soon as the download (or whatever) begins. It'll probably have a progress indicator until the download has finished. Not sure whether this applies in your case. Feb 13 19:12:08 Otherwise you'll have to say `paused = true;` in onPause and `paused = false;` in onResume, AFAIR. Feb 13 19:13:00 Calguy, I don't understand what kind of use case warrants poping up a new activity when data loads from the internet at all. Feb 13 19:13:06 What are you doing exactly? Feb 13 19:13:22 TacticalJoke, that's a horrible solution :P Feb 13 19:13:33 Mavrik: Which one? Feb 13 19:13:51 Tracking lifecycle with paused = something, it's not really maintainable :/ Feb 13 19:14:04 Yeah, I agree, but I seem to remember reading about cases where people have had to do this. Feb 13 19:14:16 Though maybe they were mistaken. Feb 13 19:14:35 TacticalJoke: but doesn't it get drawn in onCreate? Feb 13 19:14:45 Calguy: What gets drawn? Feb 13 19:16:43 TacticalJoke, yeah, usually it's better to use LocalBroadcastReceiver / Bus / RxJava observable to communicate from service to shown UI Feb 13 19:16:48 Was less flaky Feb 13 19:17:39 Mavrik: when it UPLOADS something to the internet, it goes to a confirmation page saying the upload succeeded Feb 13 19:18:07 Mavrik: but they don't want that to happen when the app is backgrounded...but it shoudl show hwne it is foregrounded Feb 13 19:19:01 Calguy: The simplest solution is, as above: show the activity *first*, with some kind of progress bar or "uploading" message until the upload has finished. Feb 13 19:19:26 TacticalJoke and then ti goes to a different activity Feb 13 19:19:43 Why a different activity? Just hide the progress bar and show "upload complete" or something. Feb 13 19:19:49 also on some cameras it only does 4;3 an dlooking for a good layout and solution that handleds that Feb 13 19:23:07 *sigh* Feb 13 19:29:53 Calguy: It's kinda hard to help since you seem to be jumping from topic to topic (OkHttp -> starting activities in the background -> cameras). It'd be best for everyone if you described what you're trying to do here. Feb 13 19:32:55 im good Feb 13 19:34:27 crap I missed the valentines sale on Nexus devices Feb 13 19:51:23 TacticalJoke: first I take a pic, then they review it, then they upload it, then there is a confirmation page Feb 13 19:51:33 3 activieis and the confiramtion apage does he uploading Feb 13 19:53:49 RustyShackleford: Which one were you looking at? Feb 13 19:54:05 3 activities that is Feb 13 19:56:49 Show an "uploading" activity (as above) or simply use a notification for the upload confirmation. Problem solved. Feb 13 19:57:59 It doesn't seem wise to pop up an activity after some asynchronous event completes. Feb 13 20:02:07 barq: 6p with matte gold color was 50 bucks off Feb 13 20:02:14 I see newegg and best buy still have that price Feb 13 20:02:34 I can't buy anything without making it a huge dilemma Feb 13 20:02:42 lol Feb 13 20:06:49 I want to add a FAB but the anchor view is in another layout (im using include tag). Is there a way i can reference the id of the anchor view or should i move the code in the same layout as fab ? Feb 13 20:14:01 csst0111, i had this issue few weeks ago -___- include and merge does not understand id out of their own file in the editor Feb 13 20:14:23 csst0111, you can set the anchor programatically or indeed put the layout in the same file :' Feb 13 20:14:37 i did not try to make AS bypass the error but it might work Feb 13 20:18:37 RustyShackleford wouldn't there be a presidents day sale ? Feb 13 20:18:48 maybe? Feb 13 20:18:51 I dunno, its over Feb 13 20:18:58 pres day is monday right ? Feb 13 20:19:50 hi there. I'm new to GCM. Just to understand, should I call at each application startup the InstanceID.getToken ? (I'm looking at RegistrationIntentService from GCM tutorial) I thought if the token was sent to my custom server, I can avoid this call no ? Feb 13 20:19:51 yeah probably some sales monday Feb 13 20:20:19 if the graphite or aluminum ones would go on sale, I'd prefer that Feb 13 20:20:35 I would have gotten gold for the $50 savings though Feb 13 20:24:36 I'm trying to import a 450,000 word text file dictionary list of words into a sqlite database. I don't think it is loading the whole file because I can only find queries up to the word babbie. Any idea why this might be the case? https://gist.github.com/emilabraham/7e306e0fd0126f45d6bb Feb 13 20:24:40 whichever is excess inventory will go on sale , like the pink ones :D Feb 13 20:25:05 honestemu wouldn't use sqlite for that ... Feb 13 20:25:48 g00s: What would you use? Feb 13 20:26:07 i'd serialize the entire dictionary into a big string Feb 13 20:26:08 Use the cloud [tm]. Feb 13 20:26:10 and store it in 1 sharedpreferences Feb 13 20:26:14 key / value pair Feb 13 20:26:18 achuinard trololol Feb 13 20:26:23 anyone here used react native? Feb 13 20:26:38