**** BEGIN LOGGING AT Sun Dec 01 02:59:58 2013 Dec 01 03:20:49 Anyone know why StateMachine and WifiStateMachine would constantly be logged in log cat? Dec 01 04:00:46 Anyone have a nexus 5? Dec 01 04:00:59 i'm sure lots of people do Dec 01 04:01:24 need a new testing phone my atrix is done for Dec 01 04:03:20 I probably would, if Verizon weren't such a bag of dicks Dec 01 04:03:29 fuck verizon Dec 01 04:03:40 that's exactly who i'm leaving Dec 01 04:03:47 i swear it feels like i'm being scammed by them Dec 01 04:04:09 switchin to t-mobile Dec 01 04:04:48 have business account with verizon, and 30 gigs is like 250/month Dec 01 04:05:20 gonna be 70/m now :D Dec 01 04:05:57 I will never ever ever ever go back to t-mobile Dec 01 04:06:04 oh no Dec 01 04:06:06 Worst experience ever with them Dec 01 04:06:07 bad? Dec 01 04:06:21 i have to believe it's better than verizon Dec 01 04:06:45 it's so cute how obsessed USians are with their little carrier spats... until it gets annoying Dec 01 04:07:23 I hate them all. I stay with Verizon because their shit service is better than the others Dec 01 04:07:31 ok point taken, but did I just get dubbed an azn?! score1 Dec 01 04:08:04 t-mobile actually fucked me though... a $400 bill that was the result of their own people explaining things wrong to me, and then they REFUSED to fix it Dec 01 04:08:29 ok so people trouble, service ok though? Dec 01 04:08:48 well, it was like 7 years ago... dunno now Dec 01 04:09:13 but, I switched from AT&T 2 years ago to Verizon, service is definitely better with Verizon Dec 01 04:09:18 well vzn livechat lied to me last night almost cost me $300 and would've been stuck with them another month Dec 01 04:09:32 ya way faster Dec 01 04:09:36 Eeeeeee: anz? Dec 01 04:09:40 er, azn? Dec 01 04:09:44 azn Dec 01 04:09:49 nope :D Dec 01 04:09:55 what is azn? Dec 01 04:10:02 but I'll take that over cajun any day Dec 01 04:10:03 asian Dec 01 04:10:20 what us usian, a us based asian was my guess? lol Dec 01 04:10:34 <_Auron_> I love verizon Dec 01 04:10:46 er, firstly no, I don't think you're asian - you could be, I have no way of knowing... secondly, I didn't imply that at all... thirdly, I do actually live in Asia, but I'm not an Asian :P Dec 01 04:10:57 well i only hate them cause of their price structure Dec 01 04:11:13 hahah i didn't mean anything by it Dec 01 04:11:34 also, "azn", really? because "asian" is a long complicated word to type? Dec 01 04:11:44 habit :P Dec 01 04:11:51 hbt Dec 01 04:12:00 <_Auron_> hobbit Dec 01 04:12:05 HOBBIT Dec 01 04:12:23 so what's an USians? Dec 01 04:12:28 americans? Dec 01 04:12:48 <_Auron_> united states asians I'm assuming Dec 01 04:13:00 that was my guess Dec 01 04:13:24 then i'm labeled as a racist for evening thinking that, hahah jk Dec 01 04:13:36 USians as opposed to other Americans Dec 01 04:13:56 <_Auron_> oh I see, that makes more sense. Dec 01 04:14:07 because american is too long and complicated to type? Dec 01 04:14:26 no, because most Americans don't live in countries where they get screwed by CDMA carriers Dec 01 04:14:32 <_Auron_> no, because US doesn't mean all of america, there is north america and south america Dec 01 04:14:35 j/m Dec 01 04:14:59 I could type "residents of the United States of America", but that *would* be too long and complicated :P Dec 01 04:15:02 man the history lesson going on here tonight Dec 01 04:15:20 <_Auron_> history? this is geography Dec 01 04:15:22 lol. what time is it in Asia? Dec 01 04:16:10 well Dec 01 04:16:18 you have a point there Dec 01 04:16:21 12:15 in the afternoon Dec 01 04:16:40 nice Dec 01 04:25:09 We should start our own carriers Dec 01 04:25:17 with blackjack and hookers Dec 01 04:25:55 i mean they really need one that is just laid back and not only out for the money Dec 01 04:25:57 Leeds: you live in asia? Dec 01 04:26:25 Eeeeeee: impossible Dec 01 04:26:27 "with free tethering because our customers aren't idiots" Dec 01 04:26:31 will be the tagline Dec 01 04:26:33 the infrasructure is too expensive Dec 01 04:26:37 YA Dec 01 04:26:38 DarkSlay3r: I do! Dec 01 04:27:05 I'm pretty sure t-mobile US is aiming for the "laid back" moniker Dec 01 04:27:12 yep Dec 01 04:27:42 i see Dec 01 04:27:42 hey it worked on me, but i'm using mobile internet as main isp Dec 01 04:28:01 Leeds: Looks like we are in the same timezone Dec 01 04:28:04 so unlimited data plan is um yeah important Dec 01 04:28:20 yeah, that's part of the problem - your wired infrastructure is so crappy that people actually start relying on limited-bandwidth mobile connections Dec 01 04:28:28 DarkSlay3r: UTC+8 Dec 01 04:28:46 Yep Dec 01 04:29:05 or GMT+8 as I like to call it, being a supporter of the Empire and all that :) Dec 01 04:29:08 honestly wish i had access to wired connection can't even get 56k where I live. Or maybe they just don't have that anymore? lol Dec 01 04:29:11 Eeeeeee: unlimited until throttled Dec 01 04:29:25 well i can deal hopefully Dec 01 04:29:27 hello I was trying to install google map api yesterday on my ADT. now my ADT is saying "this sdk requires ADT version 22 or above". Dec 01 04:29:28 how throttled? Dec 01 04:29:42 Broadband is supposed to have solved this wired issue Dec 01 04:29:59 It's simple the ISPs know people need internet Dec 01 04:30:01 fibre to the living room, baby - accept nothing less! Dec 01 04:30:14 the closest house is like a couple of acres away from me, i'm in the boonies Dec 01 04:30:19 hahahah Dec 01 04:30:24 that's how they roll in asia eh? Dec 01 04:30:28 yup Dec 01 04:30:40 that is awesome Dec 01 04:31:06 And they will pay a crap ton to get it Dec 01 04:31:06 so the had a meeting Dec 01 04:31:06 lets all serve crappy connections Dec 01 04:32:06 damnit Dec 01 04:32:08 they heard him talking about them and killed his connection Dec 01 04:32:11 i need information from you Dec 01 04:32:17 yep Dec 01 04:32:20 classic Dec 01 04:32:30 http://www.speedtest.net/result/3136388500.png - downloads running a little slow today Dec 01 04:32:42 DO I EVEN WANT TO LOOK Dec 01 04:32:43 The carriers disconnected me because they know that i know their secret Dec 01 04:32:55 ahh that's nothing WHHHAAATTTT Dec 01 04:33:02 is that an M Dec 01 04:33:03 lol Dec 01 04:33:03 ... Dec 01 04:33:13 http://www.speedtest.net/result/3136388500.png Dec 01 04:33:15 for context Dec 01 04:33:39 are you on tmobile now dark? Dec 01 04:34:04 lol leeds said the exact same thing Dec 01 04:34:13 Leeds: Life's good eh Dec 01 04:34:13 No i live in asia Dec 01 04:34:33 I've got a pay as you go $5 a month plan i'm happy with Dec 01 04:34:58 and what does that buy you Dec 01 04:34:59 lol Dec 01 04:35:18 lol Dec 01 04:35:29 shitty internet apparently Dec 01 04:35:44 I pay about the same for my mobile and wired connections - comes to around US$32.50 each... one is for 6GB of fast LTE, other is unlimited fast-as-it-can-go fibre Dec 01 04:36:00 wow Dec 01 04:36:05 really cheap Dec 01 04:36:42 how could I make my own Location test cases? Dec 01 04:37:06 I'd like some location data for when I drive some specific route Dec 01 04:39:21 my tracks is open source, dunno if it tracks driving well though Dec 01 04:39:37 but there's probably a simpler way Dec 01 04:40:49 "the Galaxy Nexus, the Nexus 4, and the Nexus 5 all contain a vulnerability that can allow attackers to interrupt use of the phone. By sending a Nexus phone around 30 flash SMS messages — an attacker can cause the phone to malfunction, frequently restarting or losing its data connection when the messages aren't promptly dismissed." Dec 01 04:41:05 hm, is it possible for a normal person to send a flash sms ? Dec 01 04:41:45 or is that only something your carrier can do ? Dec 01 04:42:13 lol i'm like bout to hit the buy now button and you drop this on me Dec 01 04:46:16 pretty messed up http://www.androidpolice.com/2013/11/29/sms-vulnerability-in-nexus-devices-can-be-exploited-to-force-a-reboot-or-kill-cellular-connectivity/ Dec 01 04:48:10 wonder if an could override them Dec 01 04:48:13 app* Dec 01 04:49:04 lol Dec 01 04:49:36 there is one app mentioned in the article, class0firewall Dec 01 04:49:50 force remote code execution Dec 01 04:50:05 yeah i loled right when i got to it Dec 01 04:50:54 that's a pretty bold theory Dec 01 04:54:27 so wait Dec 01 04:54:50 people can send these to anyone as long as they have their number? Dec 01 04:55:16 can anyone help wht this means "you are installing software that contains unsigned content"? Dec 01 04:55:49 lol pretty self explanatory, where did you download it? Dec 01 04:56:18 Eeeeeee: I am installing it online. it eclipse ADT. Dec 01 04:56:32 yeah, its always said that Dec 01 04:56:35 don't worry Dec 01 04:56:46 g00s: okay thank you. Dec 01 04:57:35 ok so I'm guessing other phones don't act the same way and just pop the sms out of nowhere Dec 01 04:57:56 was thinking how have spammers not abused this yet Dec 01 05:41:58 hi all Dec 01 05:43:19 hi dodobrain Dec 01 05:43:28 why do you want to be dodo? Dec 01 05:47:44 he might be afraid of flying high :-P Dec 01 05:48:59 dodo/train/ Dec 01 05:49:15 or hipster style - be unique since dodo has been gone long time Dec 01 05:57:12 there is no e-payment mechanism in my country, how can I have google play account? Dec 01 06:06:10 shadej: what country? Dec 01 06:06:33 Leeds: Ethiopia Dec 01 06:07:11 a dev account? Dec 01 06:07:19 Eeeeeee: yep Dec 01 06:09:29 don't know what to tell you Dec 01 06:09:29 but I know what I would do in your situation Dec 01 06:09:53 unfortunately, it looks like Ethiopia isn't on the list... so you'd have to have a presence in another country to be able to register Dec 01 06:10:12 https://support.google.com/googleplay/android-developer/answer/136758 is the full list of supported countries Dec 01 06:10:39 oh it's like google local or something? Dec 01 06:10:45 where they send a post card? Dec 01 06:12:06 er, what? Dec 01 06:15:51 wonder if someday we can buy apps with bitcoins :) Dec 01 06:17:12 problem is, you'd buy an app for $2 equivalent, and a week later either you could have bought a car with that much bitcoin, or the developer has declared bankruptcy because he only got 5c work Dec 01 06:17:17 worth Dec 01 06:20:29 Hi, would anyone be willing to help me out with authentication using a google account? Dec 01 06:20:42 I've been searching for a suitable tutorial but have failed to find anything fruitful Dec 01 06:21:42 I'm interested in finding out how from the app I can ask a user to log in with their google account provide a username that is then sent to a web server Dec 01 06:22:05 and which gives it a token? maybe that is then used to get and send data Dec 01 06:28:27 Leeds i remember you don't like anon inner classes :) Dec 01 06:28:40 you would probably find RxJava pretty ugly Dec 01 06:29:05 it will be better with java 8 lambdas Dec 01 06:33:09 n20: Convention is "on" + event type name Dec 01 06:38:00 nobody can help me? Dec 01 07:02:45 Good day Dec 01 07:03:01 is there any way to get the AS 'structure' pane to show things in the order declared? it seems only alphabetical / visibility Dec 01 07:03:10 on a new core i7 this simulator is freezing up Dec 01 07:03:12 is this normal? Dec 01 07:03:42 laggy beyond usability Dec 01 07:04:01 Technodrome maybe try genymotion Dec 01 07:04:29 is it supposed to be that slow? Dec 01 07:04:30 can't be Dec 01 07:05:26 without gpu accel, hacm, and x86 image - yeah, super laggy Dec 01 07:05:36 even on a good machine? Dec 01 07:05:39 yeah Dec 01 07:05:47 well, Technodrome for my i5, it was quite the drag Dec 01 07:06:02 Eclipse and emulator duo were memory hungry Dec 01 07:06:05 oh, you should try it on my Core 2 ;) Dec 01 07:06:16 and kill myself :-P Dec 01 07:06:34 my blood will be on your hands g00s ;-) Dec 01 07:06:57 crazy Dec 01 07:07:07 i didn't know the whole ARM simulator was that slow Dec 01 07:07:10 i'm getting an intel image now Dec 01 07:07:25 try testing on phones if possible for best result Dec 01 07:07:28 i'm using genymotion 4.1.1 image for the nexus s; it takes about 380mb for the player and 150mb for virtualbox Dec 01 07:07:36 i got an s4 here to test Dec 01 07:07:41 really g00s Dec 01 07:07:43 but still i want a nice simulator Dec 01 07:07:49 that is quite neat Dec 01 07:08:58 the iPhone simulator works so perfect on this machine Dec 01 07:09:49 hmm, every seems to praise the apple lot, Dec 01 07:10:09 * Looney gotta learn Obj-C someday for $$$ Dec 01 07:11:08 Looney consider this http://www.businessinsider.com/ios-android-shopping-traffic-2013-11 Dec 01 07:12:17 Looney: what do you do now? Dec 01 07:12:37 perhaps android people know that best way to shop is using PC/laptop Dec 01 07:12:44 a quick question. I put two cascaded fragments in an activity with front on having a videoview. when i play, I can hear sounds but no video. any suggestion? Dec 01 07:12:49 I try to lurk around Technodrome Dec 01 07:12:54 use phones Dec 01 07:13:02 i guess a while back, google had some prototypical simulator in incubation, but it was never released Dec 01 07:13:03 so how will the intel stuff work? will performance heavily improve? Dec 01 07:13:04 borrow from people at work Dec 01 07:13:19 different OS/screen resolution checking Dec 01 07:13:36 thanks Looney Dec 01 07:14:11 you are welcome for whatever reason ;-) stkim1 Dec 01 07:14:47 Technodrome: I would suggest you to try genymotion as g00s says that it is quite a charm Dec 01 07:15:08 which IDE do you use ? ? Dec 01 07:17:25 I am finally in! Dec 01 07:19:55 I am a fresh man at android Dec 01 07:21:59 startup time is forever with this intel image Dec 01 07:22:57 even on genymotion ? ? Dec 01 07:23:13 i haven't got that yet Dec 01 07:23:17 ok Dec 01 07:24:04 do you know how much responsive my laptop became when I quite eclipse + emulator duo Dec 01 07:24:06 literally going on 3 minutes right now though Dec 01 07:28:05 this is a modern core i7 Dec 01 07:28:09 not some old thing Dec 01 07:28:40 what is hcpi or what not ? Dec 01 07:34:12 right now i'm using intel and its not as smooth as when i used it last on my less powerful macbook core 2 duo Dec 01 07:34:15 something ain't right Dec 01 07:45:20 this thing takes a whole GB to just run the emulator Dec 01 07:45:22 crazy Dec 01 07:45:40 anyone know of an easy way to set up authentication for my app using a google account?? Dec 01 07:46:00 not easy, but essentially OAuth2 Dec 01 07:46:02 OAUTH Dec 01 07:46:08 oh damn it you beat me to it Dec 01 07:46:19 not easy. :/ Dec 01 07:46:20 dam. Dec 01 07:46:29 I thought perhaps with it I'd be able tocreate something simple Dec 01 07:46:35 I just need to be able to communicate with the sevrer to fetch and post data Dec 01 07:46:47 and thought it would be mighty simple using a google account Dec 01 07:46:53 know of any good tutorials? Dec 01 07:47:43 Xcode is such a better dev environment Dec 01 07:47:48 as is windows phone stuff Dec 01 07:47:52 perhaps making use of GCM Dec 01 07:48:08 no matter what hardware you have eclipse + this will lag Dec 01 07:48:20 yeah eclipse is a dump hole Dec 01 07:48:45 Technodrome: fortunately, eclipse is not required :3 Dec 01 07:48:56 i'm just getting back into this Dec 01 07:49:08 eclipse is the main java editor so i don't mind putting up with it Dec 01 07:49:18 "the main"? oh please ... Dec 01 07:49:38 every java job i have had, they use eclipse mainly Dec 01 07:49:41 thats what i meant Dec 01 07:49:47 you poor thing :( Dec 01 07:50:22 you have my condolences as well Dec 01 07:50:29 yeah…… Dec 01 07:50:52 eclipse definitely got worse compared to its VisualAge origins Dec 01 07:52:19 how smooth is the intel image for you guys? Dec 01 07:52:27 what are some other ways to make it even faster? Dec 01 07:53:13 use a real device, that's the fastest way to test Dec 01 07:53:25 i'm definitely going to do that Dec 01 07:53:31 and best Dec 01 07:53:38 but i just wanted to have the most optimal pure software solution too Dec 01 07:53:52 though the emulator has certain advantages, but I don't think many people use them, ever Dec 01 07:54:22 well with the intel image now, it's not that bad after you get past the startup Dec 01 07:54:37 and as long as you don't go native ;) Dec 01 07:55:29 i'm not using HAXM Dec 01 07:55:33 should i be using that? Dec 01 07:59:44 * Looney puts on his "to mess with list" fiddling with native code using emulator Dec 01 08:00:03 this is much quicker now with HAXM Dec 01 08:00:30 Looney: it's on my list too ... very very far down Dec 01 08:00:38 hehe Dec 01 08:01:22 the emu is still good for testing a bunch of screen sizes Dec 01 08:01:58 with intel haxm and the intel image, it seems much much more responsive Dec 01 08:02:08 enable gpu accel also Dec 01 08:02:23 what are you testing that needs some awesome performance ? Dec 01 08:02:37 that stuff just makes it usable :) Dec 01 08:02:47 urgh I'm so confused with all this authentication stuff and GCM and whatnot Dec 01 08:03:20 http://developer.android.com/google/play-services/auth.html read this Dec 01 08:03:42 g00s: i been did that Dec 01 08:03:46 GCM is a bit confusing at first, think of it like a "double hop" Dec 01 08:04:05 g00s: i'm just starting out with android, the last android dev i did was in the 1.6 days Dec 01 08:04:18 i just knew that i couldn't use a slow arm emulator at all Dec 01 08:04:25 it was beyond horrible Dec 01 08:05:04 i think google should have did something like genymotion a long time ago Dec 01 08:05:21 a remotely hosted VM? Dec 01 08:05:26 well, at least somedody did :) Dec 01 08:05:51 the vms are copied to your machine; like using the avd manager Dec 01 08:07:00 i should have did this 2 years ago Dec 01 08:07:03 such an idiot Dec 01 08:07:06 i do tons of iOS stuff Dec 01 08:07:14 but Xcode is a walk in the park compared to this Dec 01 08:07:27 dont like ios any more ? Dec 01 08:07:37 i do, but i want to make an app for both Dec 01 08:07:41 + for work Dec 01 08:07:52 gah, this is driving me nuts - Intellijj docs say for the 'structure' pane, if both sort by alpha / visibility are unchecked, the order should be as i declared it Dec 01 08:08:07 but its in some weird order Dec 01 08:08:21 i miss my eclipse structure view Dec 01 08:10:04 Technodrome i only know android and some iOS; but of my buddies that do all 3 platforms; they consider android the most difficult Dec 01 08:10:22 third platform being windows phone? Dec 01 08:10:23 one friend just ported his app to WP in a weekend without knowing shit Dec 01 08:10:24 iOS is easy Dec 01 08:10:32 xcode is quick the simulator is super super quick Dec 01 08:10:36 + its easy to use the real thing Dec 01 08:11:09 WP is quick because you have visual studio, super quick and light DE and its quite similar to android it seems, but i like wp dev, just don't like windows Dec 01 08:12:02 VMs are stupid Dec 01 08:12:06 we should have a simulator Dec 01 08:13:44 yeah i wish. i wonder why they never finished the incubating one Dec 01 08:13:46 bankai_: little known fact is that under the emulator you can get, theoretically, a trace from userspace to kernel and back Dec 01 08:14:01 i'm just saying Dec 01 08:14:02 blah Dec 01 08:14:08 eclipse is just slow, no matter the hardware Dec 01 08:26:18 hello Dec 01 08:29:47 https://gist.github.com/rymate1234/65cdc56bb258a0573814 what am i doing wrong Dec 01 08:30:05 just managed to get my app compiling again after reinstalling OS and this come sup :/ Dec 01 08:30:30 does a custom recovery image count as a custom rom? Dec 01 08:32:22 paulo_: yes Dec 01 08:33:22 rymate1234: apart from "have you cleaned your project recently?" it seems like a pretty obvious error Dec 01 08:36:28 Leeds: yeah i cleaned it Dec 01 08:37:33 ive tried googling it Dec 01 08:37:52 no idea wtf is causing my main class to be missing Dec 01 08:50:03 app work pls Dec 01 08:50:06 :( Dec 01 08:50:42 is the class there or not? Dec 01 08:53:06 lemme check Dec 01 08:53:46 it's in D:\RymateNotes\out\production\Rymate Notes\net\rymate\notes\activities Dec 01 08:54:34 is it in the apk? Dec 01 08:56:25 how do i check Dec 01 08:56:56 i assume i dont just open classes.dex with 7zip Dec 01 09:52:02 I'm setting actionbar title in onCreate method : actionBar.setTitle("Test"); but for a sec it flashes the name of the activity for the title and then it replaces it.. Dec 01 10:01:07 So, SimonVT, if you were to get a callback - which name would be more clear? onFooLoaded(), or fooSubscription()? :) Dec 01 10:31:57 tried to build AOSP for N5 as usual everything works except GSM. any ideas?;p Dec 01 10:39:21 Leeds: no Dec 01 10:39:26 it isn't in the apk Dec 01 11:27:48 developer.android.com is down ? ? Dec 01 11:27:55 or just me :-( Dec 01 11:29:21 http://www.downforeveryoneorjustme.com/developer.android.com Dec 01 11:29:23 want to talk to someone about how best to implement an option in an app Dec 01 11:29:36 specifically which interface mechanism i should choose. Dec 01 11:29:53 ok, so it really looks down :-( Dec 01 11:30:32 nope Dec 01 11:30:35 it's up for me Dec 01 11:31:22 I see It's not just you! http://developer.android.com looks down from here. Dec 01 11:31:49 I don't trust that site :p Dec 01 11:32:06 I can assure you, it's up for atleast me Dec 01 11:32:23 Where do you live, and what's your ISP? Dec 01 11:32:36 it may be down for a specific region or ISP Dec 01 11:33:31 d.android.com might very well be hosten on different servers around the world. Maybe the server(s) in the USA are down, and those in Europe are still pu. Dec 01 11:33:31 `its up for me Dec 01 11:33:35 I'm in Europe :) Dec 01 11:33:44 <- au Dec 01 11:34:00 Australia or Austreich? Dec 01 11:34:20 AU is Australia's top level domain Dec 01 11:34:24 ah Dec 01 11:35:31 aww an aussie and a fellow iineter! Dec 01 11:36:03 * wmealing tips his hat Dec 01 11:36:17 where in au ? Dec 01 11:36:38 Brisbane Dec 01 11:37:02 aww me too, N or S ? Dec 01 11:37:30 Southwest. Dec 01 11:37:39 phew i'm SE Dec 01 11:37:41 And now for something completely different: how do I force hardware acceleration? I have set android:hardwareAccelerated="true" for both the application and the activity, but someview.isHardwareAccelerated() still returns false Dec 01 11:38:09 bankai_: bleach reference ? Dec 01 11:38:46 yep Dec 01 11:39:22 solaris user, i program for red hat.. we can never be friends ;P Dec 01 11:40:01 Peetz0r: target sdk level ? Dec 01 11:40:37 wmealing: solaris admin, thankyou :P Dec 01 11:40:59 Peetz0r: http://developer.android.com/guide/topics/graphics/hardware-accel.html Dec 01 11:41:03 apparently you can't set it on the view level Dec 01 11:41:08 but you can unset it Dec 01 11:41:20 • Dec 01 11:41:21 If you must do this check in your drawing code, use http://developer.android.com/reference/android/graphics/Canvas.html#isHardwareAccelerated%28%29 instead of http://developer.android.com/reference/android/view/View.html#isHardwareAccelerated%28%29 when possible. Dec 01 11:41:28 oh, that was a bad paste. Dec 01 11:41:34 https://gist.github.com/rymate1234/65cdc56bb258a0573814 android studio doesn't appear to be compiling my app classes into the apk Dec 01 11:42:30 probably something dicky in your manifest Dec 01 11:42:39 my target sdk is 19, but the minimum is 9, and my own device is 15 Dec 01 11:42:45 so.. can you see it in the package explorer ? Dec 01 11:43:25 Peetz0r: where / how are you testing ? Dec 01 11:44:01 bankai_: my code is here https://github.com/rymate1234/Rymate-Notes Dec 01 11:44:23 Well, my 'main issue' is that I am using https://github.com/davidschreiber/FancyCoverFlow/ Dec 01 11:44:30 with MapViews from the Google Maps API v2 Dec 01 11:44:52 and the maps don't render properly inside the FancyCoverFlow stuff Dec 01 11:45:03 and the FancyCoverFlow View is software renderen Dec 01 11:45:06 rendered* Dec 01 11:45:27 and I'd like to check if things look/work better wehn hardware rendered Dec 01 11:50:26 oops Dec 01 11:50:28 java: package net.rymate.notes.database does not exist Dec 01 11:50:30 wat Dec 01 11:55:15 I have a DialogFragment that shouldn't be persisted (it shows data that only makes sense for the currently active TCP connection), how do I tell Android not to re-create it next time my Activity starts? Dec 01 11:56:14 Hello everyone.. i'm trying to download the ADT Bundle for Windows but after the download starts it just stop the download after a few seconds, and doesn't restart. I have this issue about for 2 days. Does anyone have this issue? can you, please try to download https://developer.android.com/sdk/index.html and let me know your experience? Dec 01 12:17:57 does SQLite handles duplicates itself or we need to do that manually Dec 01 12:34:28 Hey Guys. I am having a Problem: My Graphical Engine supports multiple rendering-modes (Software, openGL). For that I created one Interface which both classes implement Dec 01 12:35:07 Thing is now hoever, I need to use setContentView() and as "EngineInterface" does not inherit View, that does not work. Unfortunately a Interface can't easily derive from a class. Any clues? Dec 01 12:37:05 I can't just dismiss the DialogFragment in onDestroy, since that isn't even guaranteed to be called. Dec 01 12:41:43 I need another eye on my layout, the bottom TextView/EditText in this layout http://pastebin.com/55Cst7kh are displaying in the wrong place, screeny: http://imagebin.org/279797, what am I missing? the problem seems to stem from the TextView as the EditView positions itself correctly below it Dec 01 12:41:54 Looney: use Private Keys for that :) Dec 01 12:42:11 private keys, hmm Dec 01 12:42:16 gotta check it out Dec 01 12:42:57 ehr, Primary Key Dec 01 12:42:58 oops Dec 01 12:43:29 I have it in my table already Dec 01 12:43:31 hmm Dec 01 12:43:37 No clue? Dec 01 12:43:41 gotta check then Dec 01 12:43:56 oh, then the values of your primary key column are unique by definition. no worries Dec 01 12:44:40 you'll get an sql error whenever you try to insert a row in a table when a Primary key value already exists Dec 01 12:45:22 Disclaimer: I have never used sqlite on Android, I have my experience from working with MySQL and PostgreSQL (and even MSSQL) databases Dec 01 12:50:54 well whether primary key thing or SQLite default behaviour, no duplicates so far Peetz0r Dec 01 12:52:46 it's the primary key Dec 01 12:52:51 not default behaviour Dec 01 12:52:51 ok Dec 01 12:53:53 I happen to have a book from 1988, written for dBase 3, and that pretty much describes the behaviour of any SQL-based database system, even those in use today Dec 01 12:53:57 which is awesome :) Dec 01 12:54:54 good stuff Dec 01 12:55:54 yes. Standards that last for 25+ years :) Dec 01 12:56:01 Hi. Dec 01 12:57:26 How can I tell the FragmentManager that I don't want my Fragment to be saved? Dec 01 12:57:31 hehehe Dec 01 12:57:36 Is there even a way? Dec 01 12:57:51 nm, ignore my layout question, worked around it by aligning to another view Dec 01 12:58:13 fragment.killItWith("Fire").nukeItFrom("Orbit"); Dec 01 12:58:16 :p Dec 01 12:59:09 I can't even do that because onDestroy isn't called like half the time Dec 01 12:59:54 I wasn;t serious. But I wouldn't know the correct answer :( Dec 01 13:00:44 kbwt: my fragment knowledge is umm fragmented :) but can't you remove it in onPause()/onStop() and recreate it in onResume()/onStart() if you still need it? Dec 01 13:01:39 onStop is not called when low on memory Dec 01 13:01:52 and then the user goes back to my app, and the dialog gets restored Dec 01 13:02:04 then try onLowMemory() is Dec 01 13:02:05 kbwt: onLowMemory() is :) Dec 01 13:02:14 :D Dec 01 13:02:26 Is that guaranteed to get called? Dec 01 13:02:29 I wouldn't think so. Dec 01 13:02:41 nothing is guaranteed to be called anyway Dec 01 13:03:03 if I open a shell, su; kill your_pid; then nothing is called Dec 01 13:03:18 So the Fragment lifetime model is broken. Dec 01 13:03:35 yeah, but the same goes for Activities Dec 01 13:03:48 \everything is a little bit broken sometimes Dec 01 13:04:15 s/little bit/hugely/ s/sometimes/almost always/ Dec 01 13:04:37 meh. Usually, everything works just fine (from the user's viewpoint) Dec 01 13:04:42 So I just let the system waste resources on saving and recreating my DialogFragment and close it when it calls the empty constructor Dec 01 13:05:01 yeah, I think so Dec 01 13:05:48 btw: my app has to maintain a TCP connection to interact with a third-party service, and even though I show a persistent notification my service gets killed as soon as I load even a slightly javascript-intensive page in chrome Dec 01 13:06:21 There's no way to recreate the TCP connection, it is stateful. Dec 01 13:06:37 fits in awesome with the whole activity lifetime.... not :( Dec 01 13:07:07 anyone know on why source code imported to eclipse isnt showing any JAVA FILES ? Dec 01 13:07:21 at some point, there's just no way I can hide the flaws of the platform from the user Dec 01 13:10:23 design better. Dec 01 13:11:21 why does your app need to maintain a tcp connection ? Dec 01 13:11:26 kbwt: you are calling Service.startForeground() aren't you? Dec 01 13:11:30 "to interact with a third-party service" Dec 01 13:11:33 i read that Dec 01 13:11:36 but thats not an answer Dec 01 13:11:37 doc_savage: Yes, I am. Dec 01 13:12:02 I checked in dumpsys, it is considered foreground. Dec 01 13:12:27 wmealing: The app's whole purpose is to interact with that third-party service when you don't have access to a desktop computer. Dec 01 13:12:34 anyone know on why source code imported to eclipse isnt showing any JAVA FILES ? Dec 01 13:12:52 let me rephrase the problem Dec 01 13:13:54 is the problem that the connection isnt persistent, or that the user suffers from an non-consistent connection ? Dec 01 13:14:32 and if so, can you man in the middle this connection to deal with a spotty connection or can you design your requirements in such a way that connections are made faster when they fail Dec 01 13:15:10 I can't and don't want to man-in-the-middle that connection because it would be a security flaw. Dec 01 13:15:34 well, you can.. but what you're saying is its not acceptable to your users Dec 01 13:15:45 Right. Dec 01 13:15:51 what is the end data ? Dec 01 13:16:13 ie, what is the third party Dec 01 13:16:21 Let's say it's a MMO server. Dec 01 13:16:28 ok Dec 01 13:16:43 tried to build AOSP for N5 as usual everything works except GSM. any ideas?;p Dec 01 13:16:47 i figured most of them used udp these days anyway Dec 01 13:16:56 * wmealing thinks Dec 01 13:18:02 didnt know there were many MMO that did mobile.. but thats a different issue Dec 01 13:18:43 perhaps this has been suggested before, but a background service is not an option ? Dec 01 13:19:02 does the background service still have that issue with the drop ? Dec 01 13:19:32 I have a background service, using a persistent connection to keep it from getting killed. But it's still getting killed way too easily. Dec 01 13:19:35 kbwt: i'm not being a smart ass here, i'm thinking of ways that you can work around it Dec 01 13:19:50 due to memory ? Dec 01 13:19:52 persistent notification* Dec 01 13:20:12 i'm not following Dec 01 13:20:20 Yeah, something as simple as opening acko.net in chrome is enough to get it killed. Dec 01 13:20:38 i didnt think that was simple to render ;P Dec 01 13:20:51 well, Chrome doesn't do WebGL on Android Dec 01 13:21:17 reddit and imgur are often enough to cause it as well Dec 01 13:21:24 * wmealing nods Dec 01 13:21:49 so when you say persistent notification.. ie, something that reaches into the notification area and display something ? Dec 01 13:22:23 Yeah, that's what you have to do for the process to not be considered expendable. Dec 01 13:22:37 right, startforeground or whatever teh function is. Dec 01 13:22:52 Exactly. Dec 01 13:22:58 * wmealing thinks Dec 01 13:23:26 your service is small, right ? Dec 01 13:23:35 Foregrounding a service doesn't mean it can't get killed, it's just less likely.. You always need to design around the possibility that your app can get killed at any time Dec 01 13:23:37 you're only pulling in what you need to pass to your main app Dec 01 13:23:55 SimonVT: yeah.. thats what i'm thinking too.. but the re-start time of it is unacceptable Dec 01 13:24:01 the tcp connection time Dec 01 13:24:09 It's small compared to the Activities, yes. Dec 01 13:24:27 how small ? Dec 01 13:24:33 Crippling a foreground activity with some random foreground service is more unacceptable Dec 01 13:25:13 i dont know about androids oom killer, but if i was it. i'd kill the largest ones first Dec 01 13:25:16 SimonVT: It's maintaining a connection to a MMO server, killing and attempting to restore it is unacceptable. Dec 01 13:25:48 you know, the other option is simply to not allow them to exit without knowing its going to terminate the connection Dec 01 13:26:16 Hey guys, im trying to make a notification with a set intent, i keep seeing int requestCode, anyone kno what that's about? Dec 01 13:26:20 If I'm not actively playing your mmo I don't care about your connections Dec 01 13:26:51 SimonVT: the use case is that the user is going to a browser to researchs omethign Dec 01 13:26:52 while playing Dec 01 13:26:58 the documentation says nothing other than "requestCode Private request code for the sender" Dec 01 13:26:58 at least, thats what i think Dec 01 13:27:07 which reveals nothing as to what it does for me Dec 01 13:27:36 maybe Dec 01 13:28:12 SimonVT: You care about the connection because you want to get Notifications when someone sends you a chat message. Dec 01 13:28:35 as an mmo user, i'd be ok with proxying that Dec 01 13:29:03 Surely you use a push service for stuff like that, instead of keeping your own connection alive 24/7 Dec 01 13:29:05 Chat goes over the same encrypted TCP connection that is used for everything else. Dec 01 13:29:18 SimonVT: No. I can't change the third-party server. :) Dec 01 13:29:28 and i guess you cant allow two logins, right ? Dec 01 13:29:32 Nope. Dec 01 13:29:48 proxying this whole thing would be so tempting Dec 01 13:30:44 how big is your service in memory kbwt ? Dec 01 13:30:56 Let me check.. Dec 01 13:31:39 i dont know why your service gets chosen to be killed, but i guess mem would be one of the contending reasons Dec 01 13:34:52 It's not the memory usage of the service, it's the memory usage of the entire process Dec 01 13:36:04 so you think it doesn't even try to kill the activities first? Dec 01 13:36:33 thats a good point Dec 01 13:36:55 i remember there being some function calls that can be called when memory is tight Dec 01 13:37:00 you could kill your own activities Dec 01 13:37:47 I guess I could try settings android:process for my service and then using RPC between it and the activities. Dec 01 13:37:52 setting* Dec 01 13:38:58 requires a bit of a rewrite Dec 01 13:39:54 btw, procstats to find out your memory Dec 01 13:40:51 http://developer.android.com/reference/android/content/ComponentCallbacks.html#onLowMemory%28%29 Dec 01 13:40:55 for the low memory dection Dec 01 13:41:02 detection, for when you clean up your activities Dec 01 13:41:19 No, it doesn't kill your activities for you.. Either the entire process or nothing Dec 01 13:41:41 SimonVT: i was thinking he can willingly kill his own activities Dec 01 13:41:43 in low memory Dec 01 13:41:53 to keep the service running Dec 01 13:42:07 crap idea ? Dec 01 13:42:19 I don't have procstats on 4.3, using Galaxy Nexus Dec 01 13:42:41 oh right Dec 01 13:43:45 ps might need to do then Dec 01 13:44:13 Do I need to be rooted for my app's process to show up? Dec 01 13:44:49 i dont think so Dec 01 13:45:34 ah, nvm it does show up now Dec 01 13:46:36 always the way Dec 01 13:47:08 hmm, ps only shows VSIZE and RSS Dec 01 13:48:24 There's always a chance that your process will get killed, you still need to handle that properly.. Then you can try and limit memory usage to, hopefully, make it less likely Dec 01 13:49:04 The funny thing is, it seems to kill all running processes around the time my app dies. Dec 01 13:49:21 There's like a mass-starving in the logcat. Dec 01 13:49:58 all services I mean Dec 01 13:55:07 hey, is there a known bug concerning bitmaps on android 4.4? Dec 01 13:55:16 be more specific Dec 01 13:56:22 could be this line from stacktrace: 12-01 13:57:09.749: D/skia(18938): --- SkImageDecoder::Factory returned null Dec 01 13:56:33 there are no other exceptions Dec 01 13:57:48 Hi there. I'm making a preferences activity and would like to add a confirmation dialog to a checkbox preference. What is the most straightforward way to do this? Dec 01 13:57:55 Bitmaps works fine here Dec 01 13:58:21 SimonVT, was that directed at me? Dec 01 13:58:30 Are you talking about bitmaps? Dec 01 13:58:36 noop Dec 01 13:58:40 Well then Dec 01 13:58:50 yeah... but the timing.. anyway nvm then Dec 01 14:01:55 android.process.media also gets killed very often, but just gets scheduled to restart 5 seconds later Dec 01 14:02:40 and that one seems to be related to usb mounting, I can't imagine it taking tons of memory Dec 01 14:05:29 "Scheduling restart of crashed service com.android.providers.media/.MtpService in 5000ms" then 5 seconds later it dies again and gets scheduled for 20000ms, then 80000ms, etc. Dec 01 14:05:40 so, it backs off Dec 01 14:05:46 is that kind of restart acceptable ? Dec 01 14:05:56 all I did was look at adb logcat | grep media while I browsed a few pages in chrome Dec 01 14:06:17 Well, it would definitely not be ok for my service, but android.process.media seems to be just fine with that Dec 01 14:06:24 now restarting in 320000ms Dec 01 14:07:52 why would it not be ok? Dec 01 14:08:16 ok, thanks. Well I think I have to look into this. Have a nice day! (And thanks for your menudrawer library! It's really nice) Dec 01 14:08:21 to keep getting killed as soon as it starts up, just because chrome is in foreground? Dec 01 14:08:33 well, it makes it available for when the user wants it Dec 01 14:08:40 ie, they wont see the startup Dec 01 14:09:12 Yeah, but other users see you disconnecting and reconnecting all the time. Dec 01 14:10:18 depends if you care about the other users Dec 01 14:10:20 or your users Dec 01 14:10:57 your choices are limited by your third party .. from what i can see Dec 01 14:11:29 there seems to be a few ways around it, none of which are pretty, but might be acceptable to your users.. (i don't know how you can tell) Dec 01 14:11:42 if it was me, i'd just MITM and be done with it Dec 01 14:11:51 I have a custom Dialog, then I dismiss it, I can't open it again Dec 01 14:12:02 its not the most ideal, but I reckon it could work Dec 01 14:12:17 on my samsung s3, it works just fine on the bluestacks emulator Dec 01 14:12:50 who can help me with this problem? Dec 01 14:13:13 not me, but i imagine source code helps. Dec 01 14:13:31 reproduce it in the smallest possible way Dec 01 14:13:36 upload your project go github Dec 01 14:13:44 with the tiny reproducer Dec 01 14:14:00 hm Dec 01 14:14:18 i'll try to debug on my phone Dec 01 14:15:55 or you could use the normal emulator Dec 01 14:15:55 thing Dec 01 14:18:44 Now back to the simple dialog issue. It seems that you cannot call dismiss() in the DialogFragment constructor. Dec 01 14:19:21 So I guess I could set a flag when the empty constructor is called and dismiss() in onAttach or onCreate Dec 01 14:19:40 ugly :( Dec 01 14:24:07 can someone help me figure out how to make a confirm dialog on a CheckboxPreference? Dec 01 14:44:44 wmealing, fixed the not displaying dialog problem Dec 01 14:46:27 good stuff Dec 01 14:48:39 scone, listen for the preference change Dec 01 14:53:36 Calling dismiss in DialogFragment.OnCreateView results in a NullPointerException. I wish the documentation were a bit more explicit about when functions can be called. Dec 01 14:54:08 dismissing something before its initiated is a little odd. Dec 01 14:54:53 Showing something and then dismissing it before it's shown is a little odd Dec 01 15:00:01 I don't want to show it. The FragmentManager just brings it back to life after my application was killed. Dec 01 15:00:44 and onCreateView is called after onAttach, AFAIK Dec 01 15:00:51 onAttach is basically when the dialog gets "shown" Dec 01 15:46:29 Yes, because the fragmentmanager properly restores state Dec 01 15:46:32 Sounds like you don't Dec 01 15:56:43 SimonVT: State that is only valid for the current process/TCP connection. Dec 01 16:01:31 SimonVT: I'm open for anything that doesn't involve proxying the connection. Dec 01 16:06:59 You'll find other apps start off where you left, even if they were killed in the meantime Dec 01 16:14:01 when receiving onSensorChanged in a SensorEventListener, is the SensorEvent instance reused? As in, say there are 4 instances of SensorEvent, each is updated in turn and passed to onSensorChanged, and once the 4th one has been passed, loops back to the beginning and the initial one is set again? Dec 01 16:16:09 https://developer.android.com/reference/android/hardware/SensorEventListener.html#onSensorChanged(android.hardware.SensorEvent) Dec 01 16:16:24 because it seems when i use an ArrayList to store all SensorEvent's received through onSensorChanged. when i just sit and shake the phone, i get about 600+ entries, but, only 5 - 6 unique sets of values, even when using SENSOR_DELAY_FASTEST. Does anyone know if this is what is happening? or is there some other explanation? Dec 01 16:17:29 SimonVT: already read that, this is why i am asking a direct question about it, the note could do with being a little clearer if this is the case Dec 01 16:17:48 "The object may be part of an internal pool and may be reused by the framework." Dec 01 16:18:23 so, yes to my question then? Dec 01 16:20:03 I think it's fairly clear that it might reuse the instance Dec 01 16:20:16 thank you Dec 01 16:21:19 hopefully, a small rewrite from ArrayList to ArrayList (custom class) will fix this. and have the ctor of the SensorEventStore accept a SensorEvent and copy its values Dec 01 16:32:43 SimonVT: how about a grittier question. why? why would sensors be built like that? to only use a set number of event instances? Dec 01 16:45:59 No idea, maybe there's some performance reasons Dec 01 16:55:08 only thing i could think of would be RAM constraints, 3 floats, a long an an int may not take up much room on their own, but if your requesting them at 44khz, that soon adds up to a lot of ram Dec 01 17:03:53 Maybe they just want to leave thrashing the memory up to you Dec 01 17:04:58 Also, that code was probably written 4-5 years ago when phone were a lot less powerful and the GC was a lot worse.. Might not be that relevant today to reuse the event, but no reason to change it now Dec 01 17:12:21 didnt google change how sensor data works? Dec 01 17:12:28 something about sensor package Dec 01 17:30:01 i'm trying to download an app from the play store onto my eclipse device Dec 01 17:30:05 thru the web browser Dec 01 17:30:08 but doesnt work, any ideas ? Dec 01 17:30:17 just hangs on the loading animation Dec 01 17:33:50 anew: by eclipse device, do you mean the emulator? Dec 01 17:34:59 I imported a project X into my workspace. Can I access the resources of X? X is not a library, it is an Android Application. Dec 01 17:37:03 What I mean is that I have an initial project Y which is also an Android Application. So can I access the resources of application X in application Y? Dec 01 17:37:11 Or must X be a libaryr? Dec 01 17:37:15 gsteinert, yes emulator Dec 01 17:38:19 the emulator is woefully underpowered. it may be that it isn't powerful to run the app in any meaningful way Dec 01 17:40:18 tried to build AOSP for N5 as usual everything works except network registration and so, telephony is not functional. any ideas?;p Dec 01 17:40:49 anew: My understanding is that emulator can't use the Play store. Instead, you have to load the app onto some physical device. Then use adb to backup the app from your physical device. Then use adb to load the app into your virtual device. Dec 01 17:41:08 ah ok Dec 01 17:41:17 any way to download the apps other than the playstore ?? Dec 01 17:41:26 then i'll just load it onto the device Dec 01 17:42:21 anew: I think the whole deal with Play store is to help app developers get paid. Also to limit bad people putting viruses into apps. Dec 01 17:42:36 anew: You can get (some) apps from other locations. Such as f-droid.org. Dec 01 17:42:50 will try the...thx Dec 01 18:05:37 any clues why I get the following error msg when I try to install an app in an emulator? Error: Could not access the Package Manager. Is the system running? Dec 01 18:05:51 it happens after I do "avd snapshost load " Dec 01 18:09:50 mdim: I don't know because I've never directly used the ``avd'' command. But I do use ``adb install ...'' to install a package on a running virtual device. Dec 01 18:10:17 mdim: hopefully my alternate approach is sufficient for your needs. Dec 01 18:10:47 and you can use ``adb install ...'' on a physical device too. Dec 01 18:19:40 * _3lionz_ waves Dec 01 18:23:27 <_3lionz_> Just starting out in android development, looking at the "action bar" section of the developer training: https://developer.android.com/training/basics/actionbar/setting-up.html Dec 01 18:24:35 <_3lionz_> When I run the app on my phone (nexus 4, 4.4), it already seems to have an action bar... does this mean I still have to include this section? Dec 01 18:25:46 _3lionz_: Beginning with Android 3.0 (API level 11), the action bar is included in all activities that use the Theme.Holo theme (or one of its descendants), which is the default theme when either the targetSdkVersion or minSdkVersion attribute is set to "11" or greater. Dec 01 18:26:22 If you want an action bar below Android 3.0 you'll need to look at the SUpport Android 2.1 and above section Dec 01 18:26:40 the "Support Android 3.0 and Above Only" section on that page is very short, and basically says "you're done now" Dec 01 18:26:41 it's a choice on your part as to whether the extra work is worth supporting the extra devices Dec 01 18:27:17 <_3lionz_> Ah. So the first section is limiting it to 3.0 and up, other section 2.1 and above. But if I leave it... it could cause issues? Dec 01 18:27:23 <_3lionz_> Or have I misinterpreted. Dec 01 18:32:15 _3lionz_: does your app support versions below 3.0? Dec 01 18:32:35 roadfish: yeah, I use adb to install apps too... it's just that I'd had an issue with snapshots. I've managed to resolve it. Once you load a snapshost, you have to issue "avd stop" and then "avd start", i.e. you need to restart the device Dec 01 18:32:46 with other works, what is the minSdkversion in your AndroidManifest.xml file? Dec 01 18:33:14 words* Dec 01 18:35:50 <_3lionz_> Peetz0r, it was 8, but it says to change to 11 to have 3.0 or over Dec 01 18:35:59 <_3lionz_> target SDK is set as 19. Dec 01 18:55:02 _3lionz_: if you don't want to support versions below 3.0, then you can set minSdkVersion to 11. If you still want (or need) to support older versions, like 2.3, then you'll need to read the other section of the article Dec 01 18:55:51 <_3lionz_> thanks. Dec 01 19:09:58 Is there a CoverFlow-like widget available somewhere that does *not* extend from the deprecated Gallery view? Dec 01 19:13:30 or can anyone help my with my current solution? I'm trying to draw maps in a CoverFlow-like UI. Dec 01 19:13:35 I am now using https://github.com/davidschreiber/FancyCoverFlow/ Dec 01 19:14:15 I have a shared library that works on the 2.3.3 emulator and up, but doesn't load on 2.1 and 2.2. Dec 01 19:14:19 and I have tried both MapView from the Maps Api V2, and now I'm trying GoogleStaticMapView (basically a ImageView) from https://github.com/mitmel/android-staticmap Dec 01 19:14:27 when you guys start a new app, what is the first thing you guys do usually? Dec 01 19:14:27 and the maps just won't render Dec 01 19:14:37 do you guys do full design in an image editor first? Dec 01 19:15:11 The logcat shows: "D/dalvikvm( 393): Trying to load lib 0x40514ca8" Dec 01 19:15:25 and then I get an ANR Dec 01 19:15:44 no errors, just gets stuck there Dec 01 19:16:11 Any idea on how I'd go about debugging this? Dec 01 19:20:21 Peetz0r yuck coverlow Dec 01 19:20:35 totally useless Dec 01 19:21:10 Looks like it's stuck at "0xafd0eb08 in __futex_syscall3 () from obj/local/armeabi/libc.so" Dec 01 19:23:21 * kbwt recompiles with debug symbols Dec 01 19:25:41 :) Dec 01 19:27:21 hmm, that doesn't help Dec 01 19:27:34 "info sharedlibrary" also doesn't list my .so Dec 01 19:30:33 are you trying to override any syscalls in your library? are do you have any preloader functions for ld or for jni in it? Dec 01 19:32:02 no, I only have JNI_OnLoad Dec 01 19:32:25 well, that's a jni loader function. :P Dec 01 19:32:29 Try commenting that out and see Dec 01 19:32:58 (in other words, i don't have any real ideas, but the only thing i can think of that could loop during lib load would be one of those functions) Dec 01 19:34:00 Commented out JNI_OnLoad and JNI_OnUnload, still doesn't return from System.loadLibrary Dec 01 19:34:06 weird. Dec 01 19:34:20 have you tried making another library that's essentially empty? Dec 01 19:34:35 if a library w/ no bound functions (ie just a void foo() or something) in it has the same behavior Dec 01 19:34:41 i'd suspect however your apk is assembled Dec 01 19:34:59 if it does work, you've found someting weird. or at least weird to me. Dec 01 19:35:00 well, it works on 2.3.3+ :s Dec 01 19:35:13 i've seen onload crap itself, but not like this Dec 01 19:35:35 ah Dec 01 19:35:51 well... do you really need to support < 2.3? Dec 01 19:35:59 2.3 is about 25% of the market, I can see supporting that Dec 01 19:36:07 but earlier than that really quickly drops into nothing Dec 01 19:36:09 well, I have a few users on 2.1 and 2.2 Dec 01 19:36:36 and 2.1 is just what's required for the android support libraries Dec 01 19:37:39 Dunno; don't think i ever tried jni code on android that long ago Dec 01 19:37:44 so I'm out of ideas ;/ Dec 01 19:37:49 I'd still try making an empty lib Dec 01 19:37:56 and seeing if it's the loader, period, or something in your library Dec 01 19:38:00 working on it :P Dec 01 19:39:31 <_biff> hi Dec 01 19:39:51 empty lib loads fine Dec 01 19:40:09 <_biff> why do i get for a file xxx cannot be resolved or its not a field Dec 01 19:40:16 <_biff> i imported wav files Dec 01 19:40:27 <_biff> and the reference has error... Dec 01 19:41:26 <_biff> why cant be seen my files? Dec 01 19:42:06 I'll try adding little bits of my library back in Dec 01 19:42:07 <_biff> i made a raw folder for these files...i added to r.java public static Object raw; Dec 01 19:43:03 kbwt: yeah; weird. Good luck Dec 01 19:51:15 Can I get help with flashing a rom here? If not, could I have a reference? Dec 01 19:51:43 #android-root is the right place for that. Dec 01 19:51:48 you're not developing an app. Dec 01 19:52:49 Thanks. Dec 01 20:03:15 This could be the same issue, as I'm also using clang: http://code.google.com/p/android/issues/detail?id=61140 Dec 01 20:04:35 Yeah, that's it. Doesn't look like there are any known workarounds though. Dec 01 20:12:20 * kbwt rebuilds everything with GCC Dec 01 20:14:09 hey, I have a quick question Dec 01 20:14:33 is it about bunnies? Dec 01 20:14:40 for someone with no experience in developing for Android, how hard is creating a simple Android keyboard? Dec 01 20:27:59 Is there a CoverFlow-like widget available somewhere that does *not* extend from the deprecated Gallery view? Dec 01 20:28:01 or can anyone help my with my current solution? I'm trying to draw maps in a CoverFlow-like UI. Dec 01 20:28:06 I am now using https://github.com/davidschreiber/FancyCoverFlow/ Dec 01 20:28:09 and I have tried both MapView from the Maps Api V2, and now I'm trying GoogleStaticMapView (basically a ImageView) from https://github.com/mitmel/android-staticmap Dec 01 20:28:13 and the maps just won't render Dec 01 20:28:38 Swapping the map out for something else works, so the CoverFlow part is working as expected Dec 01 20:44:51 Oww, it still doesn't work with gcc. :( Dec 01 20:45:13 brb in half an hour Dec 01 20:48:49 Hey guys, would this be a good place to ask questions about AOSP, or is this chat more directed to developing android applicaitons Dec 01 20:49:47 jmdetloff: go on Dec 01 20:49:51 You'll have more luck by using one of the mailing lists on s.android.com Dec 01 20:51:46 Do you mean the ones linked here: http://source.android.com/source/community/index.html ? Are there any other forums or lists for discussing AOSP? Notable guides or documentation sites? Dec 01 20:52:16 Yes, those Dec 01 20:52:31 hi people Dec 01 20:53:22 Ok thanks for the info, I'll check those out Dec 01 20:55:30 onr: I wanted to ask about an issue with ViewGroups's invalidateChildFast not creating the correct dirt rect when there is an animation with setFillAfter(true), but I haven't finished filing it yet. If you're interested I can msg you a link once I'm done collecting the details Dec 01 20:56:11 Should I check if a view is already visible before calling setVisibility(VISIBLE). In other words, is calling setVisibility expensive if the visibility had not changed? Dec 01 20:56:40 i just started developing for android (api 19). i have to implement a preference activity, i used preferencefragment for this. the app should have sharedpreferences and two of them should have an onSharedPreferenceChangeListener. i'm confused about this listener. i have a mainactivity and my activity for preferences. in which of the activities do i have to implement the listener? Dec 01 20:57:21 in the fragment Dec 01 20:57:23 It looks like setVisible just sets a flag on the view, and then calls setVisible on its background drawable, which only invalidates if theres been a change Dec 01 20:57:34 So you should be able to call it cheaply without checking Dec 01 20:58:13 thanks jmdetloff Dec 01 21:08:33 * onr bounces Technodrome Dec 01 21:10:46 Hello Dec 01 21:11:29 hey onr Dec 01 21:12:01 onr? Dec 01 21:12:18 my nickname Dec 01 21:12:18 I have a sample project that is generating an R.java file with "public static int" rather than "public static final int" -- does that matter, and why would it not be generating static final? Dec 01 21:20:38 Library projects don't get final id's Dec 01 21:21:36 why is that? Dec 01 21:22:35 the ids are still hardcoded, do they get modified somewhere in the pipeline? Dec 01 21:25:17 Because the id's aren't known until the app that depends on the library is compiled Dec 01 21:26:04 thanks Simon Dec 01 21:27:24 SimonVT: my accelerometer tracking works perfectly now, wrote a small wrapper class that copies out all the values, so if the SensorEvent is reused, it doesnt matter Dec 01 21:28:20 Library projects are compiled into a jar (that doesn't contain R.java) that is then added as a dependency to the project. If the id's were static final they would be inlined. When the app is compiled it will generate an R.java for the library that contains the actual ids.. Otherwise there would be conflicts between the apps ids and the librarys ids Dec 01 21:39:29 These two seem to be the same issue: http://code.google.com/p/android/issues/detail?id=43819 and http://code.google.com/p/android/issues/detail?id=61140 Dec 01 21:41:10 durka42, thanks, got it working. i also missed overriding onResume() and onPause() Dec 01 21:42:49 yeah you have to do that Dec 01 21:48:50 i have a lot of hatred for whoever designed how activities work Dec 01 21:49:41 Using gcc 4.6 fixed my problem Dec 01 21:49:41 Hello. Dec 01 21:49:43 whoever thought that an orientation change required a full destruction and recreation of an activity, rather than onResize or onOrientationChange... deserves a special place in the core of the moon Dec 01 21:50:02 But damn, they really got to resolve this. gcc 4.7, 4.8 and all clang versions are affected. Dec 01 21:50:04 to near Bizzeh Dec 01 21:50:17 Bizzeh, you do know, of course, that recreation can be avoided right? Dec 01 21:50:33 well, it seems to be a gnustl_{static,shared} problem Dec 01 21:50:36 mrhanky: its a cold and eternal solitide in the core of the moon Dec 01 21:50:50 Mavrik: by locking the oreintation? Dec 01 21:50:53 apparently stlport works fine, but I haven't tested it yet Dec 01 21:51:00 Bizzeh, no, by telling Android not to do that. Dec 01 21:51:02 then i would suggest pluto Dec 01 21:51:16 Bizzeh, it's documented in the doc. Dec 01 21:51:16 Mavrik: do tell Dec 01 21:51:18 Re-creating the activity makes it easier to handle different resources for different configurations Dec 01 21:51:28 where? Dec 01 21:51:31 And for the majority of apps, the performance hit is insignificant Dec 01 21:52:23 And everything you have to do to properly handle orientation changes is something you'd have to do anyway, considering there are other reasons an activity might get destroyed Dec 01 21:52:57 Bizzeh, right in the docs: http://developer.android.com/guide/topics/resources/runtime-changes.html Dec 01 21:53:15 ok, one nooby question: how can i set the holo light theme global for my app? Dec 01 21:53:27 i tried 'android:theme="@android:style/Theme.Holo.Light"' in the manifest Dec 01 21:53:30 of course, you'll find out that letting activity recreate is a better choice in most cases :) Dec 01 21:54:47 Mavrik: i ended up binding a service to the activity and moved any logic, stateful information and anything that wasnt "put this on the screen" to that service Dec 01 21:56:49 mrhanky: in what element of the manifest? or ? Dec 01 21:57:00 Dec 01 21:57:16 do i have to do this per activity? Dec 01 21:57:18 Bizzeh, that sounds like a silly idea :) Dec 01 21:57:25 mrhanky: usually not Dec 01 21:57:58 so the xml i added is ok? Dec 01 21:58:15 yeah, I think so Dec 01 21:58:27 but it isn't working? Dec 01 21:59:32 is it possible to use api 14 stuff in an app that is supposed to work with api 11 devices, but just have that feature disabled in pre-14 devices? Dec 01 22:00:12 no, it's still holo dark Peetz0r Dec 01 22:00:34 Bizzeh, yes. Dec 01 22:09:25 is there a good tutorial somewhere for properly using an asynctask to not block the UI? Dec 01 22:09:46 Most little examples are just about using the AsyncTask class, rather than effectively using it to not block the main thread Dec 01 22:10:43 Acidic, that's a strange question Dec 01 22:10:55 why is that Dec 01 22:11:13 if you use it right it never blocks the main thread Dec 01 22:11:17 i.e. never ever call get() Dec 01 22:11:18 until you try and get the result Dec 01 22:11:21 right Dec 01 22:11:31 by definition, anything in "doInBackground" does not run (and thus does not block) in the main thread Dec 01 22:11:33 what is the *right* way to get the result Dec 01 22:11:39 using onPostExecute Dec 01 22:11:51 usually the "result" is doing something, like updating a View Dec 01 22:11:56 Acidic, doing update (or calling back to your activity) in onPostExecute Dec 01 22:12:06 you can declare a final reference to the view and update it in onPostExecute which runs on the main thread Dec 01 22:12:40 do i just make my own setter to set the view to update in the onPostExecute Dec 01 22:13:12 i guess i'm looking for a tutorial that goes over properly using onPostExecute Dec 01 22:13:24 I'm using a little script to build my Android app, it pretty much just does ndk-build -j8 && ant release. Now whenever I change the XML resources, like adding a new id, I get all my resources randomly reshuffled, which results more often than not in an APK that simply crashes. Then I have to 'rm -rf bin gen' and rebuild for the resource ids to be correct again. Dec 01 22:13:52 Is this a known issue, and is there a way to avoid this? Dec 01 22:23:56 Hi Dec 01 22:29:39 is it possible to have two webview on top of each other? Dec 01 22:31:17 I have a weird bug in Eclipse, I have two files mainactivity.xml and signup.xml, whenever I try to change some widget's id in signup.xml, it changes ids in mainactivity.xml too, how I can stop it? Dec 01 22:35:15 I don't suppose any of you guys know how I could debug a system service through Android Studio? For example, I've build AOSP to my device, and I would like to step through the system code that is called by IWindowSession Dec 01 22:37:07 <|0xD34D|> I know how to import system apps and set up breakpoints in those, like SystemUI but I have not messed with actually debugging stuff in the framework Dec 01 22:40:18 I'm not sure what you mean by 'debugging stuff in the framework'? I think importing a system app and setting up breakpoints is pretty much what I want to do Dec 01 22:40:36 Basically, I just want to see what code is run when relayout is called on ViewRootImpl's IWindowSession Dec 01 22:45:54 <|0xD34D|> when I use intellij I can actually step into the framework code, which is where ViewRootImpl is located Dec 01 22:46:58 Yes, but IWindowSession is an IBinder, which calls out to another process. I can step all the way to the relayout call, but I think I have to do something extra to connect to the other process and debug what happens as a result of the relayout call Dec 01 22:48:02 Also, does anyone know of any method to get a view's animation's current transformation matrix? I'd like to be able to find the difference between the translation on the matrix returned by getMatrix, and the translation on the matrix returned by getAnimation().getTransformation() - preferably without applyTransformation being called again on the animation Dec 01 22:49:04 I want to be able to calculate the bounds a view is being drawn in, in its parents coords, taking into consideration any animation transformations - and do it as efficiently as possible of course Dec 01 23:05:27 Is onResume always guaranteed to be called from the UI thread? Dec 01 23:09:34 yeah Dec 01 23:12:25 |0xD34D|: If you could share with me how you import system apps and set up breakpoints in them, in android studio, I'd really appreciate it Dec 01 23:12:41 Or if you know of any relevant resources I could look at Dec 01 23:33:35 Hi guys, if I have a ListFragment and a ListActivity (Standard Master/Detail views), and my ListFragment should show items that have been downloaded, where should the code for the download be? In the Fragment or Activity? Dec 01 23:33:42 Instantly, I think Activity... Dec 01 23:34:56 in a service Dec 01 23:35:27 Android devs, can you help me with this? http://stackoverflow.com/questions/20318629/implementin-login-with-sqlite-in-android Dec 01 23:35:32 I dont want the download to be constant - I want it to either be triggered by the user opening the page, OR, by them pressing a button. Dec 01 23:35:44 Actually Dec 01 23:35:45 The code for performing the download? I'd say Activity over fragment. If its a long download you might want to consider a service Dec 01 23:35:48 In a service isn't a bad idea. Dec 01 23:35:51 Here's the correct link, http://stackoverflow.com/questions/20318629/implementing-login-with-sqlite-in-android Dec 01 23:36:00 Not very sure on how services work... Dec 01 23:36:05 Well Dec 01 23:36:05 I am Dec 01 23:36:08 Just not how to implement them... Dec 01 23:36:28 When I say service, in the design pattern sense Dec 01 23:36:40 Using the service class? Dec 01 23:36:55 it's a separate concern from the presentation Dec 01 23:36:58 Yeah Dec 01 23:37:16 i..e the service will perform the download and store in a local database, and the application itself will pull from the database. Dec 01 23:37:35 or the service will provide the data, ignoring how it does it Dec 01 23:37:42 Yeah Dec 01 23:38:03 Im not too certain on how to implement services though - I've always wanted to, but never had the chance to learn. Dec 01 23:38:44 Where I used to work was adament that we did things the quick way in terms of development (which because of some classes someone made, meant downloading each and every time the app was opened) which made the management very happy. Dec 01 23:39:00 Is there somewhere I could get a decent tutorial and understanding on how to do it? Dec 01 23:39:46 I wasn't saying extend Service, btw. If you're thinking that's what I meant Dec 01 23:39:51 No Dec 01 23:40:05 IntentService seems to do what I want on the Android dev. site. Dec 01 23:40:39 Unless I am misunderstanding? Dec 01 23:42:38 That could work, by the looks of it (I'm still new to android) Dec 01 23:45:25 Hmm, all I need this service to do is download an RSS feed and store the results into a database locally Dec 01 23:48:29 If its an rss feed maybe you do want to refetch it each time you open your activity, won't the feed update regularly? Dec 01 23:51:51 It will Dec 01 23:51:59 Which is why I decided perhaps a survey is better Dec 02 00:06:52 Has anyone managed to install 2 Bluetooth adapters in Windows? Dec 02 00:12:16 Wait Dec 02 00:12:28 Will intent service go run, for example, every hour? Dec 02 00:12:35 Or do I need a different service for that? Dec 02 00:17:11 even when the app isn't running? Dec 02 00:18:42 When the app is open Dec 02 00:18:58 I dont want to destroy users data usage and battery if they dont plan on using my app for the next month haha Dec 02 00:19:14 I want it to run regularly though (user defined with a default should they not specify) Dec 02 00:20:07 ScheduledThreadPoolExecutor? Dec 02 00:20:21 Can it not be done through a PendingIntent? Dec 02 00:21:03 BroadcastReciever with AlarmManager seems to be another recommended method Dec 02 00:34:20 Hi, I'm new to android development and just implemented log in and log out using shared preferences. Everything works fine except that right after log in, my app minimizes such that I end up in the home screen. I have no idea what could be wrong. Can someone help? Dec 02 00:35:45 Dimple, are you sure its not just crashing? Have you checked log cat for a stack trace? Dec 02 00:37:28 jmdetloff: when I open it running pressing the button to switch app, I'm logged in Dec 02 00:37:32 jmdetloff: let me check it now Dec 02 00:39:58 jmdetloff: no, no stacktrace Dec 02 00:40:08 have you got any of it in a try...catch block? Dec 02 00:40:37 I used the LoginActivity that comes with the ide Dec 02 00:40:43 Eclipse? Dec 02 00:40:49 Android Studio Dec 02 00:41:12 How are you storing your preference? Dec 02 00:41:31 sharedPreferences.edit().putString("username", username).commit()? Dec 02 00:41:58 storing them with SharedPreferences.Editor Dec 02 00:42:09 and getting by: sharedPreferences.getString("username", ""); Dec 02 00:42:26 I'd wrap the suspect code in a try catch, drop a breakpoint in catch, and add an uncaught exception breakpoint for good measure Dec 02 00:42:26 I just store a boolean to say the user has logged in Dec 02 00:42:42 Unless finish() is being called by your code, it sounds like some sort of exception is being thrown Dec 02 00:42:48 ^ Thats what I was getting at with the try...catch Dec 02 00:42:57 Just trying to work out which one it could be Dec 02 00:43:25 try { // your code } catch (Exception e) { e.printStackTrace(); } Dec 02 00:43:31 then see what you get in log cat Dec 02 00:43:41 Basically my code is like this, starting in a main activity which checks this boolean, if flase the user is not logged in so it calls the login activity and then directly calls finish on that intent Dec 02 00:43:51 that is the only place I added a finish Dec 02 00:43:54 Ahhh Dec 02 00:43:58 You don't finish() intents Dec 02 00:44:03 finish() ends activities Dec 02 00:44:17 yes I meant it finishes the login activity, sorry Dec 02 00:44:36 you mean main activity? lol Dec 02 00:45:05 no, the log in one, such that once the log in is done, the user can't get there using the back button Dec 02 00:45:24 or at least I hoped that is what it would do Dec 02 00:45:34 So, you load MainActivity. Check if the user is logged in. If they are NOT, you call the LoginActivity, and then end the LoginActivity... Dec 02 00:45:44 Just so I understand, you call finish on the log in activity after login completes, and you never call finish on your main activity? Dec 02 00:45:45 in general, what could cause the app to minimize (how is this even called?) ? Dec 02 00:46:30 In main activity, when a user has not logged in I do: Dec 02 00:46:31 Intent intent = new Intent(this, LoginActivity.class); Dec 02 00:46:31 startActivity(intent); Dec 02 00:46:32 finish(); Dec 02 00:46:40 Use pastebin Dec 02 00:46:57 I don't think you want to finish your main activity after starting the LoginActivity Dec 02 00:47:00 I will next time, this was short Dec 02 00:47:06 Just let the login activity sit on top of the main activity? Dec 02 00:47:07 Doesnt matter Dec 02 00:47:16 ^ I'd disagree Dec 02 00:47:19 ok sorry :) Dec 02 00:47:31 To an extent anyway Dec 02 00:47:37 I'd use a flag on the intent. Dec 02 00:47:42 so, what I wish to finish is the log in activity after log in is done Dec 02 00:47:53 so users can't go back Dec 02 00:47:56 Once you start your log in activity, you want it to be the only activity on the stack Dec 02 00:48:07 and then finish() Dec 02 00:48:19 or perhaps one or the other Dec 02 00:48:30 Either way, I wouldnt leave the other activity open for the user to press back lol Dec 02 00:48:35 I have a TextView that is sort of used as a circular buffer showing up to 1000 lines of text. I'm appending about 1-5 lines per second. When it gets close to about 200 lines, I literally get spammed by GC_CONCURRENT in logcat: https://dpaste.de/3HH6/raw <- multiple of these per second Dec 02 00:48:46 I want: Main -> Login -> Main and finish log in for good Dec 02 00:49:08 It's freezing for so long, my app becomes sluggish. Dec 02 00:49:11 kbwt: GC_CONCURRENT is your garbage collector screaming that it is trying to clear resources. Dec 02 00:49:17 so how do I finish this log in? Dec 02 00:49:22 Anthaas: I know, but all I'm doing is appending text to a TextView Dec 02 00:49:36 kbwt: 1000 lines of the stuff Dec 02 00:50:01 Yeah, how can the TextView not cope with that tiny amount of data? Dec 02 00:50:45 It starts to really slow down at about 200 lines already. Dec 02 00:51:08 Dimple - are you calling finish on your LoginActivity after the login completes? If so, then the Main activity is already finished and no longer in your nav stack, so LoginActivity is closing and you're ending up on the launcher Dec 02 00:53:26 Is TextView so inefficient that my Galaxy Nexus takes ~30ms to GC, multiple times a second? (which is almost twice what AAA games take to render a full frame) Dec 02 00:53:38 jmdetloff: no, I am calling finish in the main activity after starting the login activity, but maybe this does the same behavior? Dec 02 00:53:58 No other apps are running, so it's not under memory pressure. Dec 02 00:54:46 Ok, I got confused by your statement that you want to finish the login activity after login completes. In the code snippet above, you're finishing your MainActivity and Laucnhing your login activity Dec 02 00:54:59 Dimple: Post the code for both activities into pastebin. Dec 02 00:55:04 It sounds like the LoginActivity launches successfully, and you are able to complete the login Dec 02 00:55:11 Anthaas: let me do so Dec 02 00:55:14 yeah, a paste bin would be helpful Dec 02 00:55:23 The exact moment text is appended, it blocks and GCs for ~30ms. Dec 02 00:55:43 What is it doing all that time? It has 1 Editable to GC. Dec 02 00:58:31 jmdetloff: yes the functionality works Dec 02 00:58:32 http://pastebin.com/1pTQAFwH Dec 02 00:59:51 http://pastebin.com/abB3MxDS Dec 02 00:59:54 voila Dec 02 01:00:09 Now it should be easier, sorry for the confusion Dec 02 01:00:33 I thought this was an extra noob question as this is my first app ever and the first thing I actually try to do :) Dec 02 01:02:44 So, does the app only close if you call attemptLogin()? Can you pop a breakpoint in this method and step through? Find out at which point the application is closing? Dec 02 01:02:55 I removed the finish and it still minimizes, and this time does not log in Dec 02 01:03:30 yes, let me try this one Dec 02 01:04:14 does that work when connected to a real device, or I need an emulator? Dec 02 01:04:47 breakpoints will work on your device Dec 02 01:10:42 jmdetloff: ok thx Dec 02 01:14:36 I did so and it ends up in Looper.class and executes quitSafely() ?? Dec 02 01:14:56 after the attemptLogin is finished Dec 02 01:17:08 How does it end up there? Did you step into that class, or are you hitting an exception breakpoint there? Dec 02 01:19:02 I don't even know what that class is Dec 02 01:19:12 let me run it again to see how it goes there Dec 02 01:22:13 Oh. Dec 02 01:22:49 It GCs when I call delete() on the Editable to remove text from the beginning (since I want the TextView to behave as a circular buffer) Dec 02 01:23:15 Is there a better way to do this? Dec 02 01:24:00 jmdetloff: I'm not sure, but there is an if finish on line 238 of the log in acitivy Dec 02 01:24:07 -if Dec 02 01:24:20 ah, lines numbers are off wait Dec 02 01:24:34 jmdetloff: line 220 Dec 02 01:25:20 that would finish the log in and I would end up in the main again, and then another finish is called Dec 02 01:25:40 how do you guys manage "sessions" ? Dec 02 01:26:24 Ah I see Dec 02 01:26:27 man i have a crappy problem. some UI over top of an image, and when u begin zooming the image, i dissapear the overlaying UI. but on one tablet, that area that the overlays used to be, doesn't get updated, so as u zoom the image, those areas just stay blank, actually with a little artifacting from the image, but only at first, and then it just freeze, and even calling invalidate on it doesn't help Dec 02 01:26:50 Well it looks like that finish is being called, and your MainActivity was also finished when you launched the login activity Dec 02 01:26:58 i.e. i show a login activity and get a username and password. the3n authenticate with a backend server.. where/how do i store/manage the "logged in" flag/variables (including username and other data) while the user is busy flipping between activities/fragments ? Dec 02 01:27:04 So after calling that finish, both your activities are finished and the app closes Dec 02 01:27:36 Either do not finish the MainActivity when the launch the LoginActivity, or launch an intent to start a new MainActivity when you finish the LoginActivity Dec 02 01:28:12 jmdetloff: let me try the second option Dec 02 01:28:53 jmdetloff: as when i removed the finish in the main the app still closed and i am not logged in when i start it again Dec 02 01:30:13 If I got 3 tabs in my layout, I also have a navigationdrawer. There's a viewpager so you can swipe between the tabs. However, if I put a mapfragment (api v2) as the 2nd tab (3 tabs in total). I can only swipe to the third tab or swipe to open the menu Dec 02 01:30:26 Can't swipe to go to the first tab. Dec 02 01:34:04 My google-fu is not helping me either. Found a question at SO that it should work. But yeah, it doesn't Dec 02 01:34:17 ViewPager within NavigationDrawer and then a mapfragment within the viewpager Dec 02 01:35:23 jmdetloff: that works but now pressing the back button bring me to the log in screen, arg Dec 02 01:35:47 jmdetloff: also unsure whether to call finish after starting the main activity after login is completed? Dec 02 01:36:21 Dimple, are you finishing the LoginActivity after you launch the MainActivity? Dec 02 01:36:58 jmdetloff: both seem to do exactly the same Dec 02 01:37:25 jmdetloff: I end up login, app doesn't close, but when I press the back button app closes and when I reopen I am logged out Dec 02 01:38:12 I am not sure if I want the app to close or not, but I do not want to it bring me to this log in screen ever unless I press the log out button Dec 02 01:39:51 I don't understand how this finish works Dec 02 01:40:02 I wonder if it's just impossible. meh sucks so much Dec 02 01:40:16 does it finish the activity I am calling this from or the activity I just started? Dec 02 01:40:42 Calling finish finishes the activity it is being called from Dec 02 01:41:06 When you call a method within a class, unless you specify the receiver - receive.methodName() - the received is the class you've called from Dec 02 01:41:58 I guess what I want is finish log in from the main activity Dec 02 01:42:59 You're going to need a reference to the LoginActivity to finish it from the MainActivity, but that is probably not what you want to do. You probably want the LoginActivity to finish itself when it launches the MainActivity. Dec 02 01:43:57 Can anyone help me in this? http://stackoverflow.com/questions/20318629/implementing-login-with-sqlite-in-android?noredirect=1#comment30323251_20318629, I am getting "No Such Table: TABLE_LOGIN" Dec 02 01:44:13 When you say that you reopen you are logged out, do you mean that the SharedPreferences no longer has your "hasLoggedIn" key set to true? Dec 02 01:44:59 jmdetloff: what I believe happens is that the back button brings me back to the login activity, but maybe I am wrong Dec 02 01:45:15 jmdetloff: how could my preference file get erased? this is not possible or? Dec 02 01:46:07 or it gets erased every time the app closes?? Dec 02 01:46:15 this is not what I want obviously Dec 02 01:46:57 I guess I'm just a bit confused as to the behavior you're observing at this point. Could you give me one more step by step rundown of what happens when you try to go through your login process? Dec 02 01:47:28 Yes Dec 02 01:48:43 Log in -> Main (no more app closing at this point) -> I press back button, app closes -> I reopen app and end up in log in screen while I want to be in the main activity already logged in Dec 02 01:49:53 Hmm, that is strange. Are you sure that finish() is being called on your LoginActivity? Dec 02 01:53:34 Yes Dec 02 01:53:43 is this file deleted when the app closes? Dec 02 01:54:53 Your SharedPreferences should not be deleted when the app closes, no. I asked because you said that upon restarting the app you were 'logged out' which i took to mean your SharedPreferences did not have the logged in key. Dec 02 01:55:07 If that were the case, I would assume it was not being written correctly in the first place, not that it was being deleted. Dec 02 01:55:18 I see Dec 02 01:55:31 well you see in log in activity where it is written Dec 02 01:55:36 is anything wrong there? Dec 02 01:55:56 No, it looks right to me, just trying to narrow down what could be going wrong Dec 02 01:57:33 I checked with the debugger and it is written correctly Dec 02 01:58:51 So, if you force stop the application and relaunch it, is it launching the MainActivity or is it opening up the LoginActivity? Dec 02 01:59:44 If it is opening up the LoginActivity, it would seem boolean hasLoggedIn = settings.getBoolean("hasLoggedIn", false); is returning false, which would imply that "hasLoggedIn" isn't being written successfully somehow Dec 02 02:01:37 before I try that something weird: I tried the whole process with the debugger and it works correctly Dec 02 02:01:44 without the debugger it doesn't Dec 02 02:01:47 Oo Dec 02 02:02:20 force closing the application works Dec 02 02:02:37 it's the back button which makes it go back to log in activity Dec 02 02:03:07 Hitting the back button at what point makes the app navigate to the log in activity? Dec 02 02:04:06 after I log in I am sent to the main activity Dec 02 02:04:11 then I press the back button Dec 02 02:04:13 the app closes Dec 02 02:04:30 well it doesn't really close because it is still running Dec 02 02:04:32 it hides Dec 02 02:05:00 when I go in the list of running apps, select it again, it brings me to the log in screen instead of the main acitivity Dec 02 02:05:52 Hmmm that is very strange. And if instead of backing out of the MainActivity you force close the app and restart - then it loads the MainActivity correctly? Dec 02 02:09:16 jmdetloff: it does Dec 02 02:09:29 Could you update your LoginActivity paste bin to reflect the changes you've made? Dec 02 02:09:37 Yes Dec 02 02:11:40 jmdetloff: http://pastebin.com/9AW6Rpib Dec 02 02:12:13 When I press the back button does it finish any activity or just puts the app in the background? Dec 02 02:12:35 When you hit the back button the default implementation is to finish the current activity Dec 02 02:13:13 So it makes sense that hitting the back button on your MainActivity brings you to the launcher. The part that confuses me is why you're ending back in LoginActivity when you reopen the application, since you finished LoginActivity Dec 02 02:14:13 Ok, lets try another thing. Go through all the steps you described up to the point where you back out of the MainActivity and end up on the launcher. Then, drop a breakpoint in the onCreate method of both MainActivity and LoginActivity and reopen the application Dec 02 02:14:42 I'd like to determine if the application is reopening directly to LoginActivity, or if it is opening MainActivity which in turn is launching LoginActivity for some reason Dec 02 02:14:54 ok Dec 02 02:17:25 jmdetloff: it goes to the main activity Dec 02 02:17:30 but as I said before Dec 02 02:17:37 when I run when the debugger it works Dec 02 02:17:47 only when I run without it doesnt't Dec 02 02:17:54 arghs Dec 02 02:18:03 is android dev this painful?? Dec 02 02:19:25 Hmm, that is interesting. So it launches the correct activity when opening back up, but for whatever reason settings.getBoolean("hasLoggedIn", false); is returning false - but only when you're not using the debugger? Dec 02 02:20:04 Could you try logging the value of hasLoggedIn at this point, trying it again without the debugger, and then reading the outputted value to see if what I just said is correct? Dec 02 02:26:45 jmdetloff: how do I do this log in? Dec 02 02:26:56 jmdetloff: can I not just print it in this console somehow? Dec 02 02:28:00 You can use System.out.println(String.valueOf(hasLoggedIn)) Dec 02 02:28:10 you'll see it printed in logcat Dec 02 02:29:08 It'll look something like: "com.example.myapplication I/System.out﹕ false" Dec 02 02:29:30 So you can search on System.out in the android studio log cat console Dec 02 02:35:33 so, when I hit back and end up in log in screen, it does not go to Main before Dec 02 02:35:52 which means the back makes it go back to log in or so? Dec 02 02:37:47 I see, thats pretty strange. I'm not sure what would cause your application to launch the LoginActivity instead of the MainActivity when reopening... Also sorry but I've got to head out to dinner. Sorry you haven't resolved your issue yet. If you'd like, create a stack overflow post including as much info as possible, and if you msg me the link I'll take a peek later tonight Dec 02 02:39:02 thank you for helping Dec 02 02:39:35 jmdetloff: also sometimes it does work, like when I did log in before recompiling then back button Dec 02 02:39:44 No prob, my gmail is the same as my nick, if you end up posting a SO post Dec 02 02:39:51 jmdetloff: works a few times and then the problem comes again Dec 02 02:40:14 When something works 'sometimes' I think it could be a race condition, perhaps things are happening out of order due to asynchronous code Dec 02 02:40:28 Anyways, good luck resolving your issue Dec 02 02:40:42 I decided to do log in like this because I thought most of the time a user doesn't log out, so I do not want log in to be my main acitivity Dec 02 02:40:55 but maybe it is not the right approach Dec 02 02:41:11 No, I agree that you should have a separate activity for your Login UI Dec 02 02:41:51 jmdetloff: separate yes, but I meant that I do not want to start in the log in acitivity always, just trigger it when needed Dec 02 02:42:01 I will try the other way round for now Dec 02 02:42:09 thanks and have a good meal jmdetloff Dec 02 02:44:11 Someone pleas help with this, http://stackoverflow.com/questions/20318629/implementing-login-with-sqlite-in-android **** ENDING LOGGING AT Mon Dec 02 02:59:58 2013