**** BEGIN LOGGING AT Tue Sep 10 03:00:00 2013 Sep 10 03:00:51 Another dumb question... So I've set up my project based on the Master-Detail template for the main activity. I made a button in my DetailFragment layout and tried to specify an onClick method for it. My problem is, it seems to be invoking that method on the DetailActivity rather than the DetailFragment (which is where the detail logic proper is happening) when I tap Sep 10 03:01:39 What's the proper course of action here? Can I get the onClick invokation to go to the Fragment class instead, or will I have to move logic to the Activity? Sep 10 03:02:52 never use onClick in xml Sep 10 03:03:01 it's always the wrong solution Sep 10 03:03:13 Google really should get rid of it Sep 10 03:03:47 Okay, set up onclick from the Fragment's code. Got it. Sep 10 03:04:44 it's useful for nothing but a simple reactive hello world demo Sep 10 03:15:18 pfn, yay i got it working. I load 52 ImageView's into my LinearLayout, which is inside my HorizontalScrolLView in my first ever app. Starting to get a feel for things. Thanks Sep 10 03:15:49 that sounds memory heavy Sep 10 03:16:14 I probably will use OpenGLES at some point and just draw whats visible on screen, they're small playing card images Sep 10 03:19:17 hello Sep 10 03:20:19 Totally new to developing for android. Using eclipse with the latest sdk. I created a helloworld app, and it says no AVDs are found. I've seen plenty of tutorials on creating one and I think it's pretty self explanatory, but I just wanted to ask if this is normal behavior? Does everything come installed without AVDs, or should I be looking to fix something. Sep 10 03:21:06 i had to create one, it populated the options for me though when i chose from the device drop down Sep 10 03:21:24 ackpacket: might be a good time to consider starting with the new Android Studio Sep 10 03:21:42 no reason to get too entrenched with Eclipse right when it's going out the window Sep 10 03:21:54 BHSPiMonkey: Downloaded it, and would love to, but everything on google's article for android dev deals with eclipse Sep 10 03:22:30 If I could find a *detailed*, and equally importantly *recent* tutorial for android studio, I would be thrilled Sep 10 03:22:44 ackpacket: for what it's worth, the AVD Manager can be found in Android Studio under Tools>Android :) Sep 10 03:23:13 honestly, tutorials are mostly focused on the classes and XML you deal with while making an app Sep 10 03:23:20 and that stays the same regardless of your IDE. Sep 10 03:23:33 Hmm... Sep 10 03:23:44 it doesn't really benefit you much to learn all the Eclipse menus and conventions, in my opinion Sep 10 03:24:03 I dunno. I would feel more comfortable if I put my hands/eyes on a walkthrough that actually uses a-studio Sep 10 03:24:15 if I could find that i'd jump at the switch Sep 10 03:24:22 Doesn't android studio have plenty of bugs as well? Sep 10 03:24:25 (and the projects you start now will be harder to work with in the future) Sep 10 03:24:44 works for me(tm) Sep 10 03:24:48 ackpacket: it's not perfect, but I've been using it without any trouble for a few months Sep 10 03:25:17 and I do believe there are a good amount of visual tutorials already tailored for it... Sep 10 03:26:28 just be cognizant of how invested in Eclipse you let yourself get :) Sep 10 03:27:45 ackpacket: getting back to your question, I think if you poke around the menus you'll find the Android submenu which contains the AVD Manager probably Sep 10 03:28:28 though if you only wish to test on an actual device, you don't need to set up any AVDs anyway Sep 10 03:29:46 lo Sep 10 03:29:48 Well i've had eclipse for a while and for several other languages Sep 10 03:30:00 So that's not a worry of mine Sep 10 03:30:06 it'll always be in my library Sep 10 03:31:46 migrating in the future is easy Sep 10 03:31:58 no reason to rush into android studio Sep 10 03:32:07 other than eclipse sucking Sep 10 03:32:38 my reasoning is that I'd simply rather focus my tool familiarity with what's going to be the standard down the line Sep 10 03:33:49 does android not have java.util.concurrent.LinkedBlockingQueue ?? Sep 10 03:33:54 what is the android equivallent? Sep 10 03:34:17 I'm using a ViewPager and need to update a spinner in one fragment in it when something happens in another fragment. Should I be updating all of the fragment in the view pager by setting the adapter again? or is there a better way? Sep 10 03:34:33 all I need is an array queue Sep 10 03:34:40 I've read a lot and don't quite have a grasp on the best or "right" way to do this... Sep 10 03:35:24 I'm designing a sort of "Call in progress" Activity (like what you might see in a Skype/Hangout/phone call). Should my logic/audio/IO stuff take place in the Activity itself, or in some other class like a Service? Sep 10 03:35:55 (Keeping in mind that the "call" should continue working if the Activity gets pushed to the background, like a regular phone call) Sep 10 03:37:07 or maybe is there a way to kick off a method when a fragment in the view pager comes into view? Sep 10 03:37:29 i changed the extension of a resource, and am having "resource already defined" gradle problems Sep 10 03:38:44 supercell: try a rebuild? Also make sure you renamed all versions of that resource (I accidentally renamed some xhdpi images without renaming the mdpi versions and couldn't build) Sep 10 03:39:26 it wont build because of this, =/ it says Originally defined here. Sep 10 03:39:31 but no indication of where that is lol Sep 10 03:40:16 supercell: restart your IDE? Sep 10 03:40:29 supercell: perhaps an issue with permissions Sep 10 03:40:50 i feel like it has an entry for the .jpg extension and an entry for the .png Sep 10 03:40:58 i just have no idea what hidden file this dumb ide is using =\ Sep 10 03:43:12 ugh ill recreate a file with the .jpg and delete it usign the ide Sep 10 03:43:15 i bet that works Sep 10 03:45:08 ack, still trouble. Gradle: res/drawable-hdpi/myimg.png:0: error: Resource entry myimg is already defined. Sep 10 03:45:41 so, i've created this custom view that displays a list of readings from the accelerometer; i made this mostly so my students wouldn't have to figure it out on their own Sep 10 03:45:48 (i'm a TA, before you all get too alarmed that i'm teaching...) Sep 10 03:47:02 would it be a terrible idea for this class to also get a handle to a file to write out the readings, then somehow close it when the activity in which it's hosted is destroyed? i can appreciate that there might be a problem if there were more than one of these views on screen at a time Sep 10 03:47:38 oh. I didn't realize there was a seperate "Rebuild" project... that did something different than build project... that worked, thanks Sep 10 03:58:10 kevinb: hey thanks for help man, I used those ideas to make live wallpaper hella more effecient Sep 10 03:59:18 cool. Shaders are fun Romain Guy has a nice post about shaders at http://www.curious-creature.org/2012/12/13/android-recipe-2-fun-with-shaders/ if you're looking for more effects for your wallpapers :) Sep 10 03:59:31 thanks Sep 10 03:59:46 blur edge is nice too Sep 10 04:00:16 as far as opening my file handle from a custom view goes, is there a nice place that i can close it from, too? Sep 10 04:00:53 actually, i'm just looking to dump readings in some place that other apps can access; is there a more appropriate way to do that than writing to some random file? Sep 10 04:01:50 I wonder if android supports pipes Sep 10 04:05:55 kevin its nice effect but dont think I need it Sep 10 04:06:30 yeah i haven't found a use for it either Sep 10 04:07:01 yeah not only that but my program adds 100 paths then draw once Sep 10 04:07:08 thats how its so fast compared to old way Sep 10 04:25:51 where is app settings stored at? Sep 10 04:26:00 I broke something and need to kill saved settings Sep 10 04:26:10 oh wait android has that utility Sep 10 04:29:14 It's so nice that Google Play does this alpha testing stuff. Sep 10 04:29:20 No more distributing APKs myself. Sep 10 04:30:52 XXCoder, just delete app data from settings Sep 10 04:31:19 android was too old to have that so I just uninstalled whole app. should kill settings too Sep 10 04:31:27 foroyo Sep 10 04:31:31 huh Sep 10 04:31:43 my computer isnt kickass enough to run more modern android lol Sep 10 04:32:05 Clear data has been there forever Sep 10 04:32:18 lemme look Sep 10 04:32:30 froyo has clear data Sep 10 04:32:41 guess I missed it Sep 10 04:33:35 it has uninstall, clear cache (disabled), clear defaults (disabled) and force stop Sep 10 04:34:17 probably pre-froyo lemme look Sep 10 04:34:27 so, no ideas about how i should close a file i opened from a custom view...? Sep 10 04:34:33 2.1 Sep 10 04:34:46 i'm wondering if i should call some method that closes the file in the hosting activity's onDestroy() method Sep 10 04:35:33 even 2.1 should have clear data Sep 10 04:37:25 ah finally found the bug Sep 10 04:38:15 missed one code on settings change from str to int Sep 10 04:38:26 so it was stikll accessing as string when its int Sep 10 04:44:15 I <3 having typed settings Sep 10 04:45:25 https://github.com/pfn/advanced-keyguard-manager/blob/master/src/com/hanhuy/android/bluetooth/keyguard/Settings.java Sep 10 04:45:50 I never have a problem with wrong settings types when I do that Sep 10 04:46:55 defaults are always correct and consistent, too Sep 10 04:46:58 mines typed too just that it was accessing data as string when it was int lol Sep 10 04:47:05 it was due to me changing data type Sep 10 04:47:48 fianlly tested it on phone. damn it looks REAL good Sep 10 04:47:57 it finally got past alpha status Sep 10 04:48:08 too bad it lost 2 pretty cool effects Sep 10 04:49:33 yeah, that sounds untyped.. Sep 10 04:50:13 crap, in that painful spot where i can't decide on massive refactoring or start over Sep 10 04:50:26 only testing on phone now? I rarely develop so long on an emulator, so many constraints.. Sep 10 04:50:50 Anyone got a Snapdragon 800 device that they can slap me with? Sep 10 04:50:57 I test it when its known stable. its my only phone and I cant afford to upgrade it lol Sep 10 04:51:21 Aw, just found out my testers have to pay to do alpha testing. Sep 10 04:51:24 hmm stringsetting, booleansetting and so on Sep 10 04:51:28 interesting Sep 10 04:51:30 Two steps forward, one step back :| Sep 10 04:52:18 Nilium, yeah, I'd love to give testers free access Sep 10 04:52:28 but Google won't make it so Sep 10 04:52:48 Google seems wholly opposed to the idea of there being any way to give anyone a free copy of a paid app Sep 10 04:52:53 XXCoder, only for the assignments/Initialization Sep 10 04:53:16 of course. out of memory error Sep 10 04:53:17 i'd like it so i change my paid app to free + IAP Sep 10 04:53:28 I'm not sure why its so sensive to that Sep 10 04:53:35 I believe you can change from paid to free but not vice versa? Sep 10 04:53:39 it should hold 2 images just fine. well and one small Sep 10 04:53:44 Nilium, correct Sep 10 04:53:45 free app + paid unlock thing is really nice distribution wise. it makes it easy to include your own non-play-store unlock thing for gifting or alternative payment methods or whatever. it means you only have to maintain one build, rather than maintaining a paid version and a trial version. but obviously it only makes sense if it makes sense to have a free version at all Sep 10 04:53:46 Nilium: well got idea for you Sep 10 04:53:51 make build alpha Sep 10 04:53:57 some.app.name.alpha Sep 10 04:54:01 make that one free Sep 10 04:54:06 Too much effort. Sep 10 04:54:10 that way you dont give up paid status of paid Sep 10 04:54:16 I'll just ask 'em to pay and then refund it manually. Sep 10 04:54:16 er nonalpha app Sep 10 04:54:17 that's pretty terrible on many fronta Sep 10 04:54:28 IAP has the disadvantage that some users feel it's a one time thing, not for all their devices. and the disadvantage that you don't get listed in the "Top Paid" section of the play store, you're just competing in top-free and top-grossing which are more competitive Sep 10 04:54:29 really? it wasnt very hard for me Sep 10 04:54:55 kevinb oh, didn't know about the metrics aspect. weird Sep 10 04:54:58 I know who my testers are, I'll just refund them via wallet, since I don't think that invalidates the purchase status Sep 10 04:55:08 I just made main app library for other 2 projects, free and paid. and changed minium files Sep 10 04:55:15 easy to do, but a data intensive app means you lose it between alpha and release packages Sep 10 04:55:35 node battle and node battle free is actully both same, just 2 files different Sep 10 04:55:40 unless you implement some sort of sync mechanism, which is more work Sep 10 04:55:56 Nilium, it does Sep 10 04:56:04 revoke purchase status Sep 10 04:56:07 Damn it. Sep 10 04:56:13 g00s: there is also a theory that the top-paid apps are ranked partially by uninstall rate, so unlocker apps which take up virtually no space have a kind of undeserved advantage against full paid apps which would be higher on the list of things to uninstall Sep 10 04:56:14 not if you manually pay em Sep 10 04:56:19 not refund Sep 10 04:56:22 I know Ok, I'll ask them to pay then.. uh.. this is a pain. Sep 10 04:56:40 PayPal them a refund... Sep 10 04:56:48 about the only way Sep 10 04:56:57 or gobank if they use that Sep 10 04:57:16 gobank is pretty awesome Sep 10 04:57:34 paypal is just ok Sep 10 04:57:46 maybe if you pay em slightly more than they paid Sep 10 04:58:21 hey is special promo code possible? Sep 10 04:58:27 no Sep 10 04:58:27 do you know why am I receiving this error?  Emulator] Failed to allocate memory Sep 10 04:58:43 So if I click the refund button in Wallet, that makes it so they haven't purchased the app anymore? Sep 10 04:58:49 elisa871, 768mb max on windiw Sep 10 04:58:53 windows Sep 10 04:59:05 Just to make sure we're talking about the same thing Sep 10 04:59:35 Nilium: correct, they will not be able to install future updates and if you're using LVL it will return unlicensed Sep 10 04:59:39 Drat. Sep 10 04:59:50 Ok. I will owe them beers. Sep 10 04:59:56 This seems like an acceptable compromise. Sep 10 05:01:35 yeah Sep 10 05:05:07 pfn 768mb for what? Sep 10 05:05:16 where can I change it pfn? Sep 10 05:05:38 in the emulator settings Sep 10 05:05:45 pfn are you talking about RAM? Sep 10 05:05:58 of course Sep 10 05:15:04 trying to fgure how to do decoderesources so downsample to argb 444 Sep 10 05:15:13 dont need 64 bit Sep 10 05:15:22 what Sep 10 05:15:26 64bit textures? Sep 10 05:15:40 bitmap apparently defaults to that. Sep 10 05:15:56 argb_888 Sep 10 05:16:06 that's 32bit Sep 10 05:16:11 hmm odd Sep 10 05:16:30 8+8+8+8 :P Sep 10 05:16:39 indeed youre right Sep 10 05:16:53 just figuring how to make my app able to load 2 large pictures. Sep 10 05:17:08 3 max Sep 10 05:18:10 "If you use ARGB_8888 then each pixel will be Sep 10 05:18:10 64 bits. ARGB_4444 -- 32 bits, RGB_565 -- 16 bits. " Sep 10 05:18:20 guess that guy was wrong Sep 10 05:18:39 haha Sep 10 05:18:49 each pixel! Sep 10 05:19:21 Photoshop supports 64bit at least Sep 10 05:19:42 XXCoder: heh, someone can't math, apparently Sep 10 05:19:55 apparently. someone else corrected it dowb thread Sep 10 05:19:57 down Sep 10 05:20:25 trying to find if theres way to decoderesources set to argb 444 Sep 10 05:20:46 because apparently theres mem limit Sep 10 05:21:01 how can I make use of SystemClock instead of "private Chronometer mydChronometer;" for start , stop and reset buttons? Sep 10 05:21:07 is it very different? Sep 10 05:21:56 BitmapFactory.Options I guess Sep 10 05:23:12 hey, if onDestroy() isn't called for my activity, is it a problem if i have an open file handle floating around? Sep 10 05:23:58 it's not mission-critical that the file be closed; it's just a sample application for a lab for a class that's not really devoted to writing high-reliability android apps or anything... Sep 10 05:27:14 why wouldn't onDestroy() be called for your method? Sep 10 05:27:55 i'm not sure; i've just heard there are instances where it won't be called for an activity (if it's force-quitted, i imagine) Sep 10 05:28:25 all the cases I'm aware of will end up with the process dead Sep 10 05:28:25 i wonder if it's sufficient just to expect my students to stick myDumpView.close() into their activity's onDestroy() method Sep 10 05:28:27 force quit destroys everything Sep 10 05:28:38 ah, then i won't worry about it, then Sep 10 05:28:41 in which case, it doesn't matter if you haven't closed a file Sep 10 05:29:22 Android might as well as be doing passing it a 9 signal Sep 10 05:29:37 killall -9 woo Sep 10 05:29:45 right, point taken :) Sep 10 05:30:01 diphtherial there is an internal class for journaling if you need it Sep 10 05:30:36 is there a way for views to clean themselves up when their host activity is destroyed, or should i just expect my users to call some destruction method on the custom view? Sep 10 05:30:41 g00s: ah, good to know Sep 10 05:30:57 /frameworks/base/core/java/com/android/internal/util/JournaledFile.java Sep 10 05:31:04 i don't really need to store data with any degree of reliability; it's just a dumpfile of accelerometer readings that will eventually be imported by another application Sep 10 05:31:17 that's good to know for other purposes, though Sep 10 05:31:20 oh, do that in an intentservice Sep 10 05:31:40 bonus: flle i/o off the main thread Sep 10 05:32:07 i agree, but that introduces a lot of complexity to this lab :\ Sep 10 05:32:38 it's just supposed to give them a taste of reading from a hardware sensor; it's not meant to be very complicated Sep 10 05:33:17 they also have a week to complete it, so going from nothing to an understanding of how services work seems ambitious for a class that's not even strictly about android development Sep 10 05:33:42 they're also mechanical engineering students for the most part, with not that much familiarity with java, let alone developing for mobile devices... Sep 10 05:35:28 i agree that it's frustrating that i don't have the time to teach them the best way to do it Sep 10 05:36:02 just curious, is the class an android dev class or a sensors class or what? Sep 10 05:37:05 kevinb have you done anything with persisting large amounts of sensor data (on the phone) ? Sep 10 05:37:21 no Sep 10 05:37:25 large is relative i guess ;) Sep 10 05:37:37 for our other projects, we've used a sqlite database Sep 10 05:37:53 periodically dumping the contents to a server and flushing it when we receive an ack Sep 10 05:38:18 diphtherial: fwiw, I would start up a HandlerThread, and use the SensorManager.registerListener variant which accepts a handler (which is associated with that HandlerThread) Sep 10 05:38:31 that should only be maybe 3 lines of boilerplate to set it up Sep 10 05:38:32 diphtherial that all kinda needs to be done in a service O.o Sep 10 05:38:49 (saving off to a server) Sep 10 05:38:52 g00s: yes, it's done in a service in that case...sorry, heh, different project, not what we're doing in class Sep 10 05:39:11 the constraint with the class is mostly that i don't want to go into services quite yet; i've just introduced them to activities Sep 10 05:41:47 you know, i wonder if its actually easier to teach about services, and then control it via the command line (i think pm does this) Sep 10 05:41:58 i dunno, maybe not Sep 10 05:42:16 that's a good point, since the class is mostly centered around using phones for personal instrumentation Sep 10 05:43:11 i think the complex lifecycle of activities, being stateless as possible, rotation changes, all the callbacks, is more difficult Sep 10 05:43:16 a service starts and stops :) Sep 10 05:43:42 yeah, it gets complicated when you overlap the service binding stuff with the intent stuff Sep 10 05:43:49 or if you use aidl for IPC Sep 10 05:44:41 i agree; services are probably closer to how they already think about programming... Sep 10 05:46:24 diphtherial if i remember, you can send intents to the service to start / stop it , or send info … and use Service.dump() to print out to the logs Sep 10 05:46:32 so you kinda get a command line interface Sep 10 05:46:38 poor mans :) Sep 10 05:48:14 neat :) Sep 10 06:03:41 diphtherial here is a tutorial on sensor logging; shane conder has a few android books http://mobile.tutsplus.com/tutorials/android/android-barometer-logger-acquiring-sensor-data/ Sep 10 06:13:37 how should I start the timerTask? Sep 10 06:20:01 hi all Sep 10 06:23:31 is there an event for coming back from sleep mode? Sep 10 06:35:18 anyone ever had a problem with signature verification failed for an IAP when you know for sure the public key is correct? Sep 10 06:48:56 … does your 64 bit key change? Sep 10 06:59:54 how do I send a String or int from an Activity to a Dialog? Sep 10 07:01:59 Is there any easy way to get an horizontal list view? Sep 10 07:04:18 gavilan2 nooooooooo Sep 10 07:04:59 anybody know how to get an int or string from the activity to a Dialog? Sep 10 07:05:13 O.o Sep 10 07:05:27 why multiple oos? Sep 10 07:05:56 for emphasis :D Sep 10 07:06:28 usability nightmare. people don't like flinging through lists Sep 10 07:06:32 horizontally Sep 10 07:06:52 g00s: That should be my choice :) Sep 10 07:07:00 ;) Sep 10 07:07:29 "Microsoft investors want Ford's Mulally as the next CEO" oh man, wut? Sep 10 07:08:54 better or worse than Elop? Sep 10 07:09:53 hello Sep 10 07:10:02 I need a pro at App programing Sep 10 07:10:27 need something like this app https://play.google.com/store/apps/details?id=eu.chainfire.nomoarpowah&hl=en Sep 10 07:10:42 the app should just start android, no other feature when plug usb Sep 10 07:11:04 I contacted that developer and no answer received Sep 10 07:11:04 Leeds i think MS is a bit different than something like ford / boeing; they are more like a tech conglomerate like samsung. the next ceo will have to have a grip on many different things Sep 10 07:11:07 g00s: So anywah, how do I make an horizontal list if I do need it? Sep 10 07:11:29 gavilan2 no idea. knowing its bas ui, never invested in finding out Sep 10 07:11:33 *bad Sep 10 07:11:58 you could use a viewpager i guess Sep 10 07:12:14 recycle the views yourself, play trickery, etc Sep 10 07:13:55 g00s: Can you show multiple views in a view pager? Sep 10 07:14:33 hm, on the screen at once - not sure. maybe SimonVT knows Sep 10 07:14:58 alex_PP this is cool http://www.theverge.com/2013/9/10/4712962/olympus-om-d-e-m1-micro-four-thirds-camera Sep 10 07:15:07 "WE WERE ABLE TO USE A NEXUS 7 AS A REAL-TIME VIEWFINDER" Sep 10 07:15:12 they shouted ;) Sep 10 07:16:23 view finder? Sep 10 07:16:34 ha Sep 10 07:16:39 i was just looking at this Sep 10 07:16:40 https://play.google.com/store/apps/details?id=eu.chainfire.dslrcontroller&hl=en Sep 10 07:16:47 wondering when it'd be out of beta Sep 10 07:16:48 hey guys, what do you use to drag a view: Sep 10 07:16:57 set translation or setoffset top and bottom? Sep 10 07:17:04 alex_PP when USB host on Android becomes reliable Sep 10 07:17:06 or settop/setbottom? Sep 10 07:17:12 which will be, when hell freezes over, probably Sep 10 07:17:14 so, never? Sep 10 07:17:30 they bothed it on the new Nexus 7 *again*, so Sep 10 07:17:33 *botched Sep 10 07:17:44 :( Sep 10 07:18:00 need a new camera before that anyway Sep 10 07:18:25 g00s, finally got around to taking some more photos! Sep 10 07:18:56 alex_PP oh nice; are they up anywhere ? Sep 10 07:19:03 nope, didn't want to Sep 10 07:19:13 was at a protest Sep 10 07:19:14 are you happy with them ? Sep 10 07:19:26 and cambodia doesn't have the worlds greatest human rights record Sep 10 07:19:28 yeah Sep 10 07:19:31 a few nice ones Sep 10 07:19:40 oh, so these pictures are part of a narrative / story Sep 10 07:19:45 maybe published some day Sep 10 07:19:56 by the great alex_PP Sep 10 07:20:01 ha Sep 10 07:20:10 unlikely, they're going in my private collection Sep 10 07:22:33 alex_PP i read about some photographer at riots in egypt Sep 10 07:23:02 got pictures of guys damaging people's homes … they beat him up and took his camera Sep 10 07:23:10 he got it back, but they forced him to delete the sdcard Sep 10 07:23:16 happens Sep 10 07:23:18 which is , of course, lame Sep 10 07:23:25 ge got back and un-erased it Sep 10 07:23:28 hehe Sep 10 07:23:34 just fat32 :) Sep 10 07:23:51 even 'low level format' i have no idea what this really does in my canon Sep 10 07:23:58 too fast to overwrite with 0s Sep 10 07:24:13 i'm pretty sure that just kills the FAT Sep 10 07:24:52 delete all seems to go through and delete them one by one, missig any you're set as protected Sep 10 07:25:08 still doesn't 0 it out Sep 10 07:25:20 yeah, great trick :) Sep 10 07:25:48 some sandisc cards come with some software to recover stuff Sep 10 07:25:52 pretty good of them Sep 10 07:26:19 :D Sep 10 07:57:55 morning all Sep 10 07:58:08 brb Sep 10 07:58:10 hey Napalm|wrk Sep 10 07:58:31 sorry, just updating my PTV3000 firmware.. wants me to connect to its SSID Sep 10 07:58:35 one mo Sep 10 08:08:24 how can I make use of the home button in android phones ? I don't know how to define it in "activity_main.xml" rather than this http://codepad.org/X63wLiC1 Sep 10 08:10:20 i'm confused; are you talking about the hardware/soft home button on the bottom of the device, or a button in your app that just happens to be called 'Home'? Sep 10 08:15:29 diphtherial: yes , I want to use the hardware home button but I don't know how to write its xml in the activity_main.xml Sep 10 08:17:14 i didnt disconnect? Sep 10 08:17:17 oh wait.. hard line Sep 10 08:17:18 damn Sep 10 08:17:20 lol Sep 10 08:17:23 hey g00s Sep 10 08:17:34 :) Sep 10 08:17:46 update firmware on router ? Sep 10 08:18:23 nope, PTV3000 is the wireless display reciever that works with Nexus devices Sep 10 08:18:31 Hm, I wonder if it's worth bothering about LDPI-specific resources Sep 10 08:18:40 the Miracast and WiDi one Sep 10 08:18:50 Seems like it's almost impossible to find an LDPI device now. Sep 10 08:18:51 Nilium: i dont anymore Sep 10 08:19:06 Nilium: if they have a ldpi device it will scale down the mdpi ones Sep 10 08:19:15 Works for me. Sep 10 08:19:32 I suppose if I hear anything from an LDPI user I'll 1) ban their device and 2) make some LDPI assets and update as needed. Sep 10 08:19:47 In that order. Sep 10 08:21:01 will be interesting to see final iOS 7 bits tomorrow :D Sep 10 08:26:34 weird shit https://code.google.com/p/android/issues/detail?id=15499 Sep 10 08:26:52 * g00s trials and tribulations of using attached databases Sep 10 08:27:01 I honestly don't care that much about iOS 7. I mean I think it looks cool and I'll be happy to use it, but I just want Mac OS 10.9 right now. Sep 10 08:27:11 I must have the multimonitor improvements. Sep 10 08:27:20 And GL 4.1 support. Sep 10 08:27:23 oh yeah, i hear good things about 10.9 ;) Sep 10 08:27:27 Even if I'm ticked off over it only being 4.1 Sep 10 08:27:55 It's like, you were so close, Apple, SO CLOSE, but it's like they consistently try to stay behind on GL support. Sep 10 08:28:08 Nilium you don't by any chance use a vector drawing tool? i have been using inkscape, but they basically said today there is no OTA for a newer mac version Sep 10 08:28:19 The closest I get to vector stuff is paths in Photoshop. Sep 10 08:28:28 So nothing really actually intended for vector work. Sep 10 08:28:32 ok Sep 10 08:31:32 Heya guys, quick question Sep 10 08:31:44 I have a surface holder, a camera preview, all that jazz, it works fine Sep 10 08:31:53 Do you have a lobster? Sep 10 08:32:00 I select the optimal display size, that works too Sep 10 08:32:07 Yes, I have a lobster too. Sep 10 08:32:16 Good. Lobster's crucial. Sep 10 08:32:24 Remember to take it for walks. Sep 10 08:32:36 The problem is - on some resolutions, the data returned via the picture callback doesn't reflect the stuff the preview shows. Sep 10 08:32:50 hello folks Sep 10 08:33:04 This lobster ain't walking anywhere, unless you count the path from my mouth to my stomach walking. Sep 10 08:33:09 In that case, it'll do a helluva walk. Sep 10 08:33:37 Tee_Pee, picture callback gives you a grabbed picture which may or may not be processed (depending on driver and settings) Sep 10 08:33:48 Tee_Pee, also, picture resolution is set separately from preview resolution :) Sep 10 08:34:20 is there a alternative for pthread support under android Sep 10 08:34:37 ( i try to compile asterisk, and it's a bit hard without complete pthread support ) Sep 10 08:35:16 uhm no. Sep 10 08:35:44 No, no, you need the lobster alive. For the sacrifice! Sep 10 08:35:45 Tee_Pee, also, picture resolution is set separately from preview resolution :) Sep 10 08:35:47 Oh god, thank you Sep 10 08:35:49 finally Sep 10 08:35:52 fucking finally Sep 10 08:35:59 Have a lobster Sep 10 08:36:08 God, kids these days and their lack of lobster sacrifice etiquette. Sep 10 08:36:11 Tee_Pee, just make sure to check the fun cases Sep 10 08:36:28 Tee_Pee, when resolutions for pictures have different aspect ratios than previews ;) Sep 10 08:36:33 Oh I do... I've copied the camera preview from the dev samples Sep 10 08:36:34 there are fucked up devices like that out there :P Sep 10 08:36:48 I select the correcta spect ratio preview, everything works on a multitude of devices Sep 10 08:37:04 the only issue that I was having was that on some devices the returned image was vastly different than the preview Sep 10 08:37:38 hurdman, you get bionic supported subset of pthreads, for anything else you just don't have an implementation Sep 10 08:38:25 I don't whether I should laugh or cry, I've spent 3 hours on this yesterday Sep 10 08:38:38 Tee_Pee, you lucky bastard Sep 10 08:38:42 only 3 hours to get camera working. Sep 10 08:38:52 No no no, 3 hours on this particular issue Sep 10 08:39:14 getting the preview to not stretch but rather have a 1:1 ratio, cropped, in the center of the camera content area Sep 10 08:39:26 I don't know... 3-4 days? Sep 10 08:39:55 also, fuck android 4.0 and OpenGL out of memory. Sep 10 08:40:32 :P Sep 10 08:40:55 ics users on the decline, time to say minsdk = 16 ? Sep 10 08:41:04 :D Sep 10 08:41:12 I do 16. Sep 10 08:41:23 I do 16. Sep 10 08:41:46 10, if i remember rightly Sep 10 08:41:50 lucky, I do 8. Sep 10 08:41:55 most of the really broen stuff was sorted by then Sep 10 08:42:06 Because the apps I do are region specific, and I live in the backwaters of nowhere Sep 10 08:42:40 I do v4 Sep 10 08:42:51 That being said Sep 10 08:43:21 http://developer.android.com/about/dashboards/index.html It does seem that it's phasing out globaly. Sep 10 08:43:24 globally * Sep 10 08:43:32 2.2 that is... Sep 10 08:43:41 I should start doing api 10. Sep 10 08:44:18 For alarmmanager if I have broadcastreceiver as inner class... and onReceive method overridden in it.. Sep 10 08:44:30 It creates a Alertbox on alarm. Sep 10 08:44:39 but it dosn't work if the app is not running. Sep 10 08:44:40 Looking at my own app's statistics, api 8 is about 2%, api 10 is the second highest, 17.6% Sep 10 08:44:44 Is it because I have it as my inner class? Sep 10 08:44:59 Tee_Pee yeah, i still have a lot of GB people Sep 10 08:45:04 Mavrik: ok thx Sep 10 08:45:20 Josh-, is it static? Sep 10 08:45:29 Umm no..? Sep 10 08:45:39 I get boatloads of emails asking/telling me to support 2.3 Sep 10 08:46:00 demanding Sep 10 08:46:04 even Sep 10 08:46:05 Josh-, it's probably because you can't just show a dialog froma service without being in foreground Sep 10 08:46:06 http://ideone.com/BmfFhM Sep 10 08:46:06 Josh-, try that Sep 10 08:46:25 Mavrik, Well, toast dosn't work either. Sep 10 08:46:46 but if I make it a distinct class, toast works... I havn't tried alertbox from outside. Sep 10 08:47:08 Josh-: add ; Sep 10 08:47:14 at line 13 Sep 10 08:47:20 nb-ben, that was wrong paste. Sep 10 08:47:22 Give me a second Sep 10 08:47:28 alertboxes never work from the background Sep 10 08:47:42 wehat you can do is have an activity that has the dialog theme Sep 10 08:47:49 http://ideone.com/3EZaex Sep 10 08:47:51 style even Sep 10 08:47:51 that one Sep 10 08:47:58 Ah Sep 10 08:48:17 Tee_Pee, So, I can start an activity using intent from the broadcast receiver? Sep 10 08:48:21 Is that what you want to say? Sep 10 08:48:25 of course Sep 10 08:48:28 Okies Sep 10 08:48:35 Josh-: line 40, forgot } Sep 10 08:48:46 start up an activity that has the Theme.Dialog (iirc) style Sep 10 08:48:54 then do with it as you please :) Sep 10 08:48:59 nb-ben, Haha, I am just pasting half pastes, removing much of irrelephant data, so I may forget stuff like that. Sep 10 08:49:14 Josh-: then how am I supposed to debug it? Sep 10 08:49:43 Dosn't matter anyway, I got my answer from Tee_Pee for now. Sep 10 08:49:45 Josh-: unless you intentionally removed that } at line 40 of the paste, you forgot to close upyour broadcast receiver Sep 10 08:49:59 nb-ben, I just pasted some part of my click event. Sep 10 08:50:04 Upper and lower. Sep 10 08:50:10 Skipped a lot of middle part. Sep 10 08:50:23 Anyway, thanks a lot Tee_Pee Sep 10 08:50:51 no problem Sep 10 08:50:56 Mavrik Sep 10 08:51:12 there seems to be the case that the available preview sizes don't match the available picture sizes Sep 10 08:51:14 is that possible? Sep 10 08:52:03 Tee_Pee, very common actually Sep 10 08:52:15 picture sizes are usually noticably larger than screens :) Sep 10 08:52:21 you know, 8Mpix or whatnot Sep 10 08:52:58 Excuse me while I go cry for a bit more. Sep 10 08:53:28 Tee_Pee, you CAN just grab the preview frame tho ;) Sep 10 08:54:43 every day is school day. Sep 10 09:00:47 I need to make calls to a rest api, what should I use? Sep 10 09:00:55 A bed. Sep 10 09:01:49 let me try that again, I need to make calls to a restful api, does android have componenets specifically for that or what package should I use? Sep 10 09:02:39 there is all kinds of stuff. never tried it, like http://static.springsource.org/spring-android/docs/1.0.x/reference/html/rest-template.html Sep 10 09:02:40 zquad, Android has a HTTP library and a JSON parser Sep 10 09:02:56 zquad, there's tons of 3rd party libraries that make your life easier tho Sep 10 09:16:48 Hey all Sep 10 09:19:14 my home server is acting up so I'm having to run irssi on my laptop Sep 10 09:20:31 now I can see joins, parts and netsplits Sep 10 09:20:35 whoa buddy Sep 10 09:20:48 it'll be fun trying to have a conversation in here Sep 10 09:21:11 hey romainguy, how's life in Nestle's promotions department? ;-) Sep 10 09:29:07 Hello! Can someone explain what means ....="?android......" Sep 10 09:29:25 ? qualifier Sep 10 09:29:45 xgear: I'm not entirely sure, but it seems to query the values you have set in your theme Sep 10 09:31:15 xgear: so if you have @style/RSA.Light.TextAppearanceMedium in your Theme Sep 10 09:31:39 then ?android:TextAppearanceMedium will set return that style Sep 10 09:31:50 sorry Sep 10 09:31:55 then ?android:TextAppearanceMedium will return that style Sep 10 09:32:36 and if i havn't this style in my theme? Sep 10 09:33:28 Hi All Sep 10 09:33:58 xgear: it will take the default from your parent theme Sep 10 09:34:11 hey MickeySoFine Sep 10 09:43:16 Hey there! Has anyone switched from ORMLite to greenDao without data loss? How compatible are their ORMs? How complex would it be to convert the SQLite db so it can be loaded by greenDao? Sep 10 09:47:10 Is there any informed guess about the KitKat release date? Sep 10 09:48:34 I'm asking because my app is currently targeting api 16, uses a whole shebang of external libraries so I'm not that motivated to update to every new api level. Sep 10 09:49:05 However, the "fragment in fragment" feature of api 17 might help me with some scheduled new features. Sep 10 09:49:41 so I'm now wondering if I could spare myself some work by waiting for KitKat and do just one app overhaul instead of two. Sep 10 09:52:16 why would you "update app to new api level"? Sep 10 09:57:03 because the new api level allows fragments in fragments, which makes scheduled new features easier to implement. Sep 10 09:57:20 "fragments in fragments" is also included in the new google support library. Sep 10 10:01:45 how do I set a layouts ontouch ? Sep 10 10:02:26 ok sorry stupid question Sep 10 10:08:13 not sure if I understand the question, but you can either attach a listener to it via code or put a method name in the xml layout (use Lint to make sure you don't misspell the method name) Sep 10 10:10:37 RedNifre, um, that's in support lib anyway Sep 10 10:11:44 yes, but not in the old one I'm using at the moment. So I'd like to do one big update of target api and all libraries, because I think a batch update of everything is less work than updating things separately. Sep 10 10:24:33 if I replace a layouts contents with some new elements that have click listeners and there were previously some elements with click listeners in the previous view, will GC clear those up for me? Sep 10 10:30:29 my ide is telling me it cannot find symbol: import org.apache.http.impl.client.DefaultHttpClient; this is my first app, does android not have this package? Sep 10 10:39:00 hi guys Sep 10 10:44:12 can you use wifi direct/p2p to connect a phone to a normal pc? or maybe a windows pc with wifi? Sep 10 10:47:22 Gaz`: programmatically? Sep 10 10:48:42 hello d00ds :) How would one retrieve the current user location Sep 10 10:49:00 I need the latitude and longitude, I seem to have problems on the different devices Sep 10 10:49:28 I am currently using locationManager = (LocationManager) getSystemService(Context.LOCATION_SERVICE); Sep 10 10:49:31 yes fdans Sep 10 10:49:55 tagrudev: you need a location listener Sep 10 10:50:05 yeh implements LocationListener { Sep 10 10:50:25 then something to request updates, like lm.requestLocationUpdates(LocationManager.GPS_PROVIDER, 5 * 60 * 1000, 1000, this); Sep 10 10:50:27 I do use Sep 10 10:50:33 LocationListener sir Sep 10 10:50:54 the if all is well onLocationChanged is called.. Sep 10 10:51:07 sometimes it doesnt work on my n1 for ages, then suddenly it works awesome Sep 10 10:51:23 tagrudev: can you put your code in a pastebin? Sep 10 10:51:26 sure Sep 10 10:54:07 my ide is telling me it cannot find symbol: import org.apache.http.impl.client.DefaultHttpClient; this is my first app, does android not have this package? I am using android studio 0.2.7 Sep 10 10:55:05 fdans, https://gist.github.com/tagrudev/6507811 Sep 10 10:55:09 check that out Sep 10 10:57:05 lms Sep 10 10:59:01 ohhhh tagrudev Sep 10 10:59:09 you're using getLastKnownLocation Sep 10 10:59:32 it's reported as not being very reliable Sep 10 10:59:41 you're basically not really using the lisetener Sep 10 10:59:41 am I invisible? Sep 10 11:00:05 fdans, suggestions ? Sep 10 11:00:12 yeah Sep 10 11:00:38 that's how vogella does it Sep 10 11:00:43 you need yourLocationManager.requestLocationUpdates() Sep 10 11:01:17 let me try to fix your pastebin Sep 10 11:01:32 ah no it's a gist Sep 10 11:01:43 you can fork and edit it Sep 10 11:01:47 yeah im on it Sep 10 11:03:16 tagrudev, fdans why not look at the android samples that comes w/ the sdk? Sep 10 11:03:52 alexfu, we got this, it's just that the listener is not really being used Sep 10 11:04:28 alright Sep 10 11:05:58 zquad, that package exists Sep 10 11:07:09 so something might be wrong with my ide? Sep 10 11:07:12 alexfu Sep 10 11:07:37 zquad, which are you using? Sep 10 11:07:54 this is my very first android app. I am using android studio 0.2.7 Sep 10 11:08:49 do Build > Rebuild Project. Sep 10 11:10:36 tagrudev: https://gist.github.com/fdansv/6507877 Sep 10 11:11:41 tagrudev: hope that makes sense Sep 10 11:11:50 fdans Sep 10 11:11:53 let me try that ouy Sep 10 11:11:55 out& Sep 10 11:11:56 ty Sep 10 11:13:20 alexfu, no difference I am even gettint package android.app does not exist error Sep 10 11:13:49 zquad, right but what does the output say when you try to do a rebuild. Sep 10 11:14:39 compilation failed see compiler error output for details Sep 10 11:14:48 Gradle: error: package android.app does not exist Sep 10 11:15:15 pfn: network access on a password app?? How do I know you aren't sending all my passwords to terrorists Sep 10 11:16:09 it would be nice if google drive let you get away with not requiring network access Sep 10 11:16:16 but i guess that could potentially lead to antitrust issues Sep 10 11:19:13 I have micromax a116 phone and added to /etc/udev/rules.d/51-android.rules a line for detecting the phone in lsusb as SUBSYSTEM=="usb", ATTRS{idVendor}=="0bb4", MODE="0666", GROUP="plugdev" but it does not show up. it does not show up in adb. I also tried vendor as 0e8d. Sep 10 11:19:45 zquad open terminal/command line and cd into your project dir and run ./gradlew build --stacktrace Sep 10 11:19:51 mms_, is your current user in the plugdev group? Sep 10 11:20:12 you can check with the "groups" command Sep 10 11:20:21 root is user Sep 10 11:20:49 zquad, ok then run gradlew.bat build --stacktrace Sep 10 11:21:01 mms_, you are running your desktop environment as root? Sep 10 11:21:03 sonOfRa: yes root and other users are part of plugdev group I checked Sep 10 11:21:26 sonOfRa: no. I opened root terminal to edit udev rules file Sep 10 11:21:39 oh. Yes, but the user that is running your DE must be in plugdev Sep 10 11:21:47 sonOfRa: I checked my login user as well as root both are in plugdev group Sep 10 11:21:49 you might have to restart the session for it to come into effect, if you added Sep 10 11:22:07 sonOfRa: I did that with udevadm command Sep 10 11:22:25 any way to check for its syntax or any errors ? Sep 10 11:22:48 at times on boot one can see rules file spitting errors but how to check it without having to reboot linux Sep 10 11:23:03 also I do not think there is any syntax error but just to be sure Sep 10 11:23:12 which log files shows all ok Sep 10 11:23:24 anyone know a way to see or catch an intent from adb? including it's extras? Sep 10 11:23:34 like im trying to get the extras from 09-10 07:22:37.047 192-862/? I/ActivityManager: START {act=com.google.glass.action.VIDEOPLAYER flg=0x10000000 cmp=com.google.glass.videoplayer/.WatchActivity (has extras)} from pid 579 Sep 10 11:24:37 mikedg: you might be able to declare your own activity which claims to support that action, then the normal activity selector should open Sep 10 11:24:46 and if you pick your fake activity, you can do whatever you want to the intent Sep 10 11:24:59 i think it specifically asking for that component though Sep 10 11:25:08 so i'm sol without rooting Sep 10 11:25:09 there is also https://play.google.com/store/apps/details?id=uk.co.ashtonbrsc.android.intentintercept&hl=en Sep 10 11:27:56 actually I think i might be wrong about the component name Sep 10 11:28:48 also note I am running via vbox Sep 10 11:54:44 I'm trying to run the emulator on 64 bit ubuntu server (in a VM) and it says: "SDL init failure, reason is: No available video device" Sep 10 11:54:48 What can I do? Sep 10 11:55:00 actually in vbox in usb section I added filter for active device for it then vbox installed some driver Sep 10 11:55:13 now it shows up fine in debian guest Sep 10 11:55:24 other than give up and go to wwdc Sep 10 11:55:34 ok bye guys Sep 10 11:55:38 anthony_: Do other 3D things work as the same user? glxgears, etc.? Sep 10 11:56:15 where is glxgears? Sep 10 11:57:37 (I don't know what that is) Sep 10 12:00:17 anthony_: It's in mesa-utils. It's just a programme that draws a 3D gears animation as fast as it can. I suggested it because it's a very simple 3D programme that will make sure you have proper access to the video device. Sep 10 12:01:00 glxgears says: "Error: couldn't open display :0.0" Sep 10 12:01:25 when I close and reopen my shell, it says: "Error: couldn't open display (null)" Sep 10 12:01:56 Are you even running X? :P Sep 10 12:02:48 Ohh... "Ubuntu server" - so no. :P Sep 10 12:02:51 I don't know Sep 10 12:02:52 lol Sep 10 12:02:55 :( Sep 10 12:02:59 yea ubuntu server isn't going to run X Sep 10 12:03:24 so basically, there is no hope at all? Sep 10 12:03:26 and no 3D things without X server :P Sep 10 12:03:38 you need a graphical environment to run the emulator anthony_ Sep 10 12:03:54 at least some form of Xvfb Sep 10 12:03:56 you can run a headless server Sep 10 12:04:03 I want a headless server Sep 10 12:04:10 then install one :P Sep 10 12:04:26 but installing another OS isn't possible right now (only phone internet) Sep 10 12:04:46 guess I'll try to install it on my mac Sep 10 12:04:51 anthony_: Perhaps you want to use NX forwarding on VNC. Sep 10 12:04:57 on -> or Sep 10 12:04:59 Macs come with X11 Sep 10 12:05:36 well, you can start X11.app on Mac, tell it to accept IP connections, and set DISPLAY on the server apropriately...:D Sep 10 12:05:43 There's a -no-window option, if you want to run it headless. http://developer.android.com/tools/help/emulator.html#startup-options Sep 10 12:06:02 how would I set display? Sep 10 12:06:18 the ip is something like 10.0.x.x Sep 10 12:06:35 anthony_: DISPLAY=10.0.x.x:0 Sep 10 12:06:37 export DISPLAY=10.0.x.x:0.0 Sep 10 12:06:45 or DISPLAY=10.0.x.x:0.0 Sep 10 12:07:04 If you're connecting with OpenSSH, OpenSSH can take care of X forwarding fo ryou. Sep 10 12:08:15 Is X11.app the same thing as xquartz.app? Sep 10 12:08:52 probably Sep 10 12:08:54 I don't use macs Sep 10 12:09:37 hmm, so glxgears just sits there not outputting anything now Sep 10 12:09:52 does x authenticate with ./.ssh? Sep 10 12:10:17 No, SSH authenticates with ~/.ssh Sep 10 12:10:24 X authenticates with xauth Sep 10 12:11:26 ah crap, you're trying to run accelerated. Just install the emulator on your Mac Sep 10 12:12:22 Doesn't XGL do acceleration? Sep 10 12:12:57 xlq: GLX does indirect, which most people don't want Sep 10 12:13:09 Hmm. Sep 10 12:13:31 indirect is slow and was used mainly by CAD/CAM/CAE tools Sep 10 12:14:09 "No protocol specified Error: couldn't open display 10.0.1.24:0.0" Sep 10 12:15:12 http://dyhr.com/2009/09/05/how-to-enable-x11-forwarding-with-ssh-on-mac-os-x-leopard/ Sep 10 12:15:52 does webkit in android supports plugins only from adobe flash? what is the purpose of "AOSP/development/samples/BrowserPlugin" then? Sep 10 12:17:03 :) very close. it flashes on my screen Sep 10 12:17:58 well I give up Sep 10 12:18:01 thanks for you help Sep 10 12:18:03 your* Sep 10 12:18:40 Anyone knows why downloading the NDK from http://dl.google.com/android/ndk/android-ndk-r9-linux-x86_64.tar.bz2 takes forever? Sep 10 12:18:49 i.e. 7-8 hours Sep 10 12:18:56 Any mirror? Sep 10 12:18:59 not CDN ? Sep 10 12:20:00 anthony_: And do you get any errors in your terminal? (Have you tried -verbose?) Sep 10 12:20:50 yeah, xiq error libGL.so: cannot open shared object file: No such file or directory Sep 10 12:20:50 Failed to load libGL.so Sep 10 12:21:21 oh oops Sep 10 12:21:27 it was missing a comma in the ports Sep 10 12:22:07 Well, no, it means you're missing libGL.so :P Sep 10 12:22:19 yeah, but it works with the comma Sep 10 12:22:33 only problem now is to get jenkins to run it through x Sep 10 12:22:37 :o Sep 10 12:24:33 Freaking Google use a CDN for http://dl.google.com/ after switching it from C++ to Go Sep 10 12:24:55 They claim the new system is must better after it was written in Go, bollocks! Sep 10 12:25:12 They should've written it in Stoople Stop! Sep 10 12:25:28 4 hours to download 400 MB file! Sep 10 12:25:41 that has nothing to do with the language it was written in Sep 10 12:26:30 might not even be related to google Sep 10 12:26:53 networking 101 = you're fucked if there's any leg you don't control ;) Sep 10 12:27:50 We have a thing or two to learn from millipedes. Sep 10 12:28:22 I have the fastest internet connection in the world! Sep 10 12:28:34 for home consumers Sep 10 12:28:44 the problem is with Google's servers Sep 10 12:28:51 I do not like to consume homes. They taste very brick-ey. Sep 10 12:29:31 or you might just be living in the middle of nowhere Sep 10 12:30:01 cff: your connection is "fastest" only to first router/switch Sep 10 12:30:28 hell, even your home router/endpoint could be fucking you without lubrication (search for "bufferbloat") Sep 10 12:30:40 raw bandwidth means *nothing* Sep 10 12:30:42 ive recently got fiber to the home, but ive not noticed a major boost on most sites Sep 10 12:30:48 the servers i'm accessing cant keep up :P Sep 10 12:31:06 hey Sep 10 12:31:07 only major improvement ive noticed is uploading apk files to my site Sep 10 12:31:16 from minutes to seconds Sep 10 12:31:17 guys, how i can extract a full firmware .pac file from my phone? Sep 10 12:35:00 bbl Sep 10 12:35:17 got it xlq, replaced emulator with a shell script that sets DISPLAY and then runs the real emulator Sep 10 12:35:24 lol Sep 10 12:44:03 7's the key number here. Think about it. 7-Elevens. 7 dwarves. 7, man, that's the number. 7 chipmunks twirlin' on a branch, eatin' lots of sunflowers on my uncle's ranch. You know that old children's tale from the sea. It's like you're dreamin' about Gorgonzola cheese when it's clearly Brie time, baby. Step into my office. Sep 10 12:45:21 nseidm1 must be compiling. Sep 10 12:52:12 so, at the end of the day - fuck working with android cameras Sep 10 12:52:16 fuck it to hell. Sep 10 12:52:51 and you haven't even tried recording a video yet. Sep 10 12:52:52 :P Sep 10 12:53:09 and hopefully I won't have to. Sep 10 12:54:21 What I've got is an activity with a navigator bar on top (re-created it manually instead of the action bar so I can have better control of where the camera preview goes), controls on the bottom, and a square camera preview centered vertically in the area between the two Sep 10 12:55:04 getting to that point + the imageview showing the correct image after capturing is quite enough for me. Sep 10 12:56:20 switching to the front camera is giving me trouble on a nexus tablet though Sep 10 13:07:04 Hey. Question: I've followed this guide to create a custom list view: http://goo.gl/BTjFGM however it only explains how to create the view, not how to actually incorporate it into my own app. Any suggestions on where to look? Sep 10 13:09:07 Put the ListView in a layout file you inflate in your app's activity or fragment. Sep 10 13:10:00 Well the entire list view is in a separate project file, so I suppose simply importing that project will give me access to it? Sep 10 13:12:26 the entire listview is a class Sep 10 13:12:55 with likely a layout file for the child views Sep 10 13:13:03 and likely a container for the listview itself Sep 10 13:13:51 I'm not sure what that means in terms of what I have to do. Sep 10 13:19:35 you can move the ListView class into your application Sep 10 13:19:43 also move the related layout files Sep 10 13:20:12 So basically move all of the created files into my own project? Sep 10 13:22:10 not all the created files Sep 10 13:22:12 just the ListView Sep 10 13:23:45 Well the example I did had 3 files and 2 layout files. Sep 10 13:27:28 do you know what the 3 files do? Sep 10 13:28:14 One is a custom adapter Sep 10 13:28:44 The other is the custom list view. Sep 10 13:50:34 when you're using a CAB is there any way to get at the view of the list item that is being selected from onItemCheckedStateChanged? Sep 10 13:53:42 cab? Sep 10 13:55:21 contextual action bar Sep 10 13:55:24 I figured it out though Sep 10 13:56:30 as soon as I asked it I realized what to search in google to get it :/ Sep 10 14:02:01 i'm trying to flash my android phone but i can't get in boot mode Sep 10 14:02:27 some help? Sep 10 14:02:43 hey guys Sep 10 14:04:05 can a device with a dp of 353 still have a resolution of 1280 by 800? Sep 10 14:04:41 that's a wierd question :) Sep 10 14:05:01 yes, because a device with resolution of 353 can have canvas size of 1280x800 if it's of apropriate physical size Sep 10 14:05:09 mikedg, that's why I also made it open source Sep 10 14:05:19 damn Sep 10 14:05:26 ok so im kinda really stumped here Sep 10 14:05:34 so people think oh its open source it must be safe Sep 10 14:05:42 while you move to the bahamas Sep 10 14:05:47 i have to seperate my apks based on the resolution of the device Sep 10 14:05:49 not the density Sep 10 14:05:57 clever clever Sep 10 14:06:11 javaDon: DPI is resolution. The amount of pixels in horizontal or vertical direction is related to physical size of the device Sep 10 14:06:13 javaDon: why Sep 10 14:06:22 mikedg, I suppose I could be nefarious ;-) Sep 10 14:06:23 javaDon: so filter for physical size Sep 10 14:06:31 how? Sep 10 14:07:03 check the manifest attributes - you have small, large, xlarge etc. Sep 10 14:07:09 yes Sep 10 14:07:15 but then again, I have password managers for other platforms which people use in large numbers, and I haven't moved to the Bahamas yet Sep 10 14:07:18 but i have just one problem with that Sep 10 14:07:34 2 of my resolutions fall under the same category Sep 10 14:07:37 then you need a better business plan Sep 10 14:07:49 p_l, filtering by "sw" attributes is probably better tho Sep 10 14:07:52 javaDon: then you need a better design ;) Sep 10 14:07:53 I probably ha e 100k active users for chromeipass and passifox combined Sep 10 14:08:19 Mavrik: I'd prefer more properly scalable, responsive designs ;) Sep 10 14:08:23 passthefox! Sep 10 14:08:39 (partially because my phone has DPI that I tell it to have at the moment, per app) Sep 10 14:08:56 and with those, I could totally be stealing creds with no one the wiser Sep 10 14:09:02 p_l, yeah, that's why using sw parameters is better Sep 10 14:09:09 p_l, or you get wierd results in some cases Sep 10 14:09:21 or maybe you just use it to browse porn websites Sep 10 14:09:28 Is there somebody present who is an expert on UP vs BACK navigation behaviour? Sep 10 14:09:31 (like Note being set as a "large" device rewarding you with lots of pissed off users in ratings :P ) Sep 10 14:09:39 There's 3 sw's that are most useful, 533, 600, 720 Sep 10 14:09:43 533 is for Kindle Sep 10 14:10:09 the ipad has an equivalent sw of 666 Sep 10 14:10:12 coincidence? Sep 10 14:10:14 sw? Sep 10 14:10:36 The thing is that I thought I understood the difference, but I don't understand GMails navigation. Sep 10 14:10:47 RedNifre: yes, up is like climbing a ladder, back is like twerking Sep 10 14:10:59 RedNifre: what confuses you about gmails navigation Sep 10 14:11:01 it could be wrong Sep 10 14:11:02 I thought it was like this: Up goes up in the app's hierachy, back goes back to where you were before. Sep 10 14:11:10 that is generally it Sep 10 14:11:19 except in exception circumstances Sep 10 14:11:53 javaDon, http://developer.android.com/guide/practices/screens_support.html#DeclaringTabletLayouts Sep 10 14:11:57 GMail behaves like this: You write a mail, you get a notification that somebody sent you a mail, you open that new mail, then you press back but you aren't back in your compose mail activity, you are in your inbox. Sep 10 14:12:12 Google argues that this is right because usually back leads to the inbox. Sep 10 14:12:18 I find that argument completely appaling Sep 10 14:12:23 as do i Sep 10 14:12:45 ignore the argument for that it makes no sense Sep 10 14:12:58 ill piss on that argument Sep 10 14:13:52 Rollin round, sittin on dubs. Sep 10 14:13:52 Countin the urs, Sep 10 14:13:52 Was high on shrubs. Sep 10 14:13:54 Coolin in my Escalade. Sep 10 14:13:55 Man I'm paid, I got it made! Sep 10 14:13:58 Take me to your special place. Sep 10 14:14:00 Close your eyes, Sep 10 14:14:02 Show me your face. Sep 10 14:14:04 I'm gonna piss on it! Sep 10 14:14:27 good ole Dave Chapelle said it best Sep 10 14:14:35 so what sw sould i use fo say a 480 by 800 ? Sep 10 14:14:55 you wouldn't use sw to target that Sep 10 14:14:59 xhdpi /hdpi / mdpi/ ldpi Sep 10 14:14:59 no diff just according to its resolution Sep 10 14:16:03 just stick most of your stuff in xhdpi, tweaks for hdpi and xxhdpi as needed, and accommodate 7 in tabs with sw600dp, 10 in tabs with sw720dp, and Kindle with sw533dp Sep 10 14:16:28 javaDon, all of those have different dp sizes Sep 10 14:16:34 sorry i goot bumped off Sep 10 14:16:35 don't forget landscape!!! Sep 10 14:16:41 stop looking at display resolution, it's a meaningless parameter Sep 10 14:16:42 only prtrait Sep 10 14:16:47 this is a game not an app Sep 10 14:16:55 relevant number is width/height in dp and density. Sep 10 14:17:12 ok i understand that Sep 10 14:17:34 so there isn't a good way of dealing with resources if you have them resolution based Sep 10 14:17:46 because you have tablets with 480x800 screens and small phones with 1080x1920 Sep 10 14:18:08 we dont have support for those sizes Sep 10 14:18:16 thats way to big Sep 10 14:18:21 you dont support anything at this point Sep 10 14:18:24 big in what way? Sep 10 14:18:35 1280 by 800 is the max Sep 10 14:18:41 why? Sep 10 14:18:51 dunno that what the designer did Sep 10 14:18:51 mikedg ok, so in conclusion we understand back and up, but the GMail team doesn't. Sep 10 14:19:16 RedNifre: some people believe strongly that back and up should have special behaviors in some circumstances Sep 10 14:19:22 it's confusing as shit though Sep 10 14:19:32 and i dont think tha case you mentioned is a good argument Sep 10 14:20:24 I agree that back can have special functions, e.g. in a web browser. Sep 10 14:20:46 But when I read that incoming mail, pressed back and was in the inbox I panicked because I thought I had lost my half composed mail. Sep 10 14:21:07 Luckily it goes to "drafts" in that case, but still: if I press back I want to go back, not up :/ Sep 10 14:21:22 A similar thing is screen size and screen dpi Sep 10 14:21:52 I thought that it would be best if the devices were honest about those, so I can make my app render in a matching way. Sep 10 14:22:31 Then the galaxy tab comes around, pretends that it has a small screen and hdpi instead of a large screen with mdpi and Google says "That's actually really good, that way everything looks bigger!" m( Sep 10 14:22:41 I find that to be completely insane :( Sep 10 14:23:07 complex specifications hit real world problems. Real world problems make short work of specifications and their intent. Sep 10 14:23:28 isnt the sw tag for your layouts Sep 10 14:23:31 not the manifest? Sep 10 14:23:33 Everything is corrupted, and design always loses to practicality. The wise person develops a simple design and enforces constraints so that it works reasonably while being something everyone can understand Sep 10 14:23:56 having like 5 different metrics as in android layouts guarantees most people won't understand some part of it and therefore nothing ever quite looks right. Sep 10 14:24:45 for instance cairo, a rendering library for 2d, has similar rich array of metrics but they discovered when used in high-dpi context that the easiest way to retain backwards compatible is actually to redefine pixel Sep 10 14:25:03 javaDon, for manifest as well Sep 10 14:25:11 right Sep 10 14:25:13 1 pixel is now 2 pixels if the DPI is doubled. That way everything using pixels still looks mostly right, and pixel thus becomes yet another weird metric with some assumed DPI value that must be converted. Sep 10 14:25:40 actually 1dp can be 1, 1.5, 2 or 3 pixels currently ;) Sep 10 14:25:50 the same thing happened to HTML at some point. That is how things go. Only the fact that things look fairly reasonable with existing software seems to matter. Sep 10 14:26:24 I don't care about pixels. If I make it look right on an honest tablet it will look too big on a galaxy tab. Sep 10 14:26:39 And if I don't notice that the galaxy tab is lying my app will look too small on an honest tab. Sep 10 14:27:08 the only thing that matters is that the icons and texts have certain visual size in human eyes and that they can be laid out reasonably on the available space. Sep 10 14:27:17 I suspect that Samsung just noticed that there were no apps optimized for tablets yet so they cheated and told the app they were running on a phone-like device. Sep 10 14:27:39 What I find inapproprietate is that Google jumps in and says that this is a good idea. Sep 10 14:27:44 I suspect android doesn't have any unit designating "the size in fovea after projection by eye" Sep 10 14:27:54 im still confused how to seperate this Sep 10 14:28:01 but given that devices are held at different distances, we should have one of those Sep 10 14:28:42 How about the square root of the screen size? Sep 10 14:28:45 if dp has nothing to do with screen resolution i dunnow how to seperate it Sep 10 14:29:14 A tablet that is 2x the size of a phone is held farther away than a phone so you should render everything about 1.5x? Sep 10 14:29:29 dp maps to cm, if I understand it correctly. Sep 10 14:29:37 It lacks the "how far away is the screen" aspect. Sep 10 14:29:50 RedNifre so what would a dp be for 480 px? Sep 10 14:30:04 hi! im trying to implement this tutorial http://processors.wiki.ti.com/index.php/Android-Adding_SystemService , all good and service become visible in adb shell service list, but when im trying to get it from a test app(as show at the end of tutorial) i've got an exception at line "om.setValue()" but exception points to null, is there any other tutorials like this available or what is my mistake may be? Sep 10 14:30:10 RedNifre: dp approximately maps to cm. It depends on screen class, apparently. Sep 10 14:30:33 you multiply the 480 px by the dpi and a constant (don't know the constant by heart, but it must be defined somewhere) Sep 10 14:30:35 if it's hdpi, it has an assumed DPI regardless what the actual DPI is. This is probably because assets are constructed at particular DPI and it's advantageous if you can use them without resizing. Sep 10 14:31:04 exactly alankila Sep 10 14:31:07 DPI = Resolution, XxY in px = screen size Sep 10 14:31:11 ...but the assumed DPI should be close to the actual dpi Sep 10 14:31:12 that is how ive setup the app Sep 10 14:31:23 i have evrything at there exact size Sep 10 14:31:31 pu i dunno what folders to send to what phones Sep 10 14:31:34 RedNifre: more or less. Sep 10 14:31:37 but Sep 10 14:31:39 lol Sep 10 14:31:43 RedNifre: btw, which Galaxy Tab? The first one? Sep 10 14:31:56 I think so, I haven't paid attention to the later ones. Sep 10 14:32:06 RedNifre: the first one was before tablet sizes were provided at all Sep 10 14:32:10 Given that apps no longer look horrible on tablets I assume the newer ones are honest. Sep 10 14:32:14 anyway if I had had a hand in this crap, I'd probably have specced px as the fundamental unit and simply stated that 10 px makes a comfortable-sized text for user to read and hidden every notion about the actual pixel size in layouts. :-/ Sep 10 14:32:25 nah, dp is great Sep 10 14:32:36 the only real confusion is for noobs Sep 10 14:33:02 nah, dp doesn't take distance to screen into account Sep 10 14:33:06 RedNifre: first Galaxy Tab didn't lie, but it didn't implement Tablet UI, because it was 2.2 device (later 2.3), while Tablet UI and new sizes/dpi came with 3.0 Sep 10 14:33:20 RedNifre, that only matters for tvs and computer monitors Sep 10 14:33:23 what we need are fovea pixels :) Sep 10 14:33:40 arcangle pixels Sep 10 14:33:44 no, you hold tablets farther away than phones Sep 10 14:33:48 yes, arcangle pixels Sep 10 14:34:02 RedNifre: anyway the projection generally works such that if you have distance z, and dimension d, you normalize objects to distance 1 by calculating d/z. That is, you divide by distance. So object twice as far has half the size, but if you double the size it physically has it would appear unchanged Sep 10 14:34:11 RedNifre, you hold them both at touching distance, so phones and tablets are about the same Sep 10 14:34:18 touch laptops would also be the same as well Sep 10 14:35:03 RedNifre: this is for planar projection which seems to match the model produced by human eyes and maintained by the brain. We imagine ray shooting from eye, intersecting the projection plane at some point, and then continuing until it meets the actual object Sep 10 14:35:21 the object's image on the projection plane remains unchanged if dimensions and distances are scaled by same factors. Sep 10 14:35:48 I guess that taking the square root of the screen size is a good approximation, what do you think? Sep 10 14:36:05 RedNifre, and yes, tablets have been honest since android 3.2 Sep 10 14:36:25 RedNifre, only android <3.2 had that problem because there were only loosely defined small, large, xlarge buckets Sep 10 14:36:46 all this talk about eye-related models made me think of how bad "retina" name is ;> Sep 10 14:37:07 as of android 3.2, apps no longer need to declare that they work within those arbitrary buckets Sep 10 14:37:19 but with different dp sized screens, which is perfect... until a device fucks up the dp ratio Sep 10 14:37:22 it is a clever name in sense that steve jobs tried to imply that it's the DPI to end all DPIs Sep 10 14:37:31 help please Sep 10 14:37:38 alankila, indeed, except now it's getting surpassed, heh Sep 10 14:37:42 alankila: unless you happen to have spent significant part of a year learning about eyes Sep 10 14:37:51 y am i so lost Sep 10 14:37:54 but really there is no DPI value that is right, as viewing distance must be included in the calculation. Sep 10 14:38:23 moar pixels moar better Sep 10 14:38:39 yea y the f is dpi aven used Sep 10 14:38:41 but if you said it's "retina class" you would presumably argue something like 1 px is always the smallest discernible line in the eye because users view the device at a distance appropriate for the DPI and therefore you can just do everything with px as fundamental unit Sep 10 14:38:43 and dp Sep 10 14:38:44 and sw Sep 10 14:38:48 javaDon, huh Sep 10 14:38:48 and bs Sep 10 14:38:51 dip is not dpi Sep 10 14:38:56 just frustrated sorry Sep 10 14:39:09 that is my point why I think only px should be used and why it should be defined as something that occurs in the fovea Sep 10 14:39:11 more pixels needs more butter Sep 10 14:39:34 more pixels more butter!! Sep 10 14:39:51 so, how about square root: You have a nexus 4 and the perfect size for your icon is 0.5 inch... which is a quarter of sqrt(4 inch)... so on a nexus 10 it would be a quarter of sqrt(10 inch) or about maybe 0.8 inch? Sep 10 14:40:08 if the user has unusually poor eyesight or abnormal view distance, a simple scale factor could handle changing the display to user's preference while not making things any more complex -- you just get screens that appear to have different number of pixels fit on them, but 1 px is still the smallest discernible thing Sep 10 14:40:16 i really dont see how to dectare screen size support for the different dp Sep 10 14:40:19 it makes no sence Sep 10 14:40:23 RedNifre, on nexus 10, it should still be .5" Sep 10 14:40:33 RedNifre, why do you want to scale it up Sep 10 14:40:41 you're holding it at roughly the same distance Sep 10 14:40:45 if a dp of 300 can have a resolution of 1300 how the hell does that make any sence Sep 10 14:40:47 and touching it in a similar way Sep 10 14:41:02 javaDon, because pixels don't make sense when you take dpi into account Sep 10 14:41:07 and yes, that makes total sense Sep 10 14:41:16 300dp is always like 4" Sep 10 14:41:20 pfn because i hold the tablet farther away than the phone Sep 10 14:41:26 RedNifre, wait 'til you hit your 40s Sep 10 14:41:32 then you'll hold your phone farther away than your tablet Sep 10 14:41:33 ;-) Sep 10 14:42:25 ok so we realize that dp is not going to work to seperate screen sizes Sep 10 14:42:29 i dont get it Sep 10 14:42:32 huh? Sep 10 14:42:35 i thought Sep 10 14:42:37 dp not going to work? it works fine Sep 10 14:42:43 xhdpi ment bigger Sep 10 14:42:45 it doesn't work for tv's and computer monitors Sep 10 14:42:48 and ldpi ment smaller Sep 10 14:42:51 no, it means denser pixels Sep 10 14:42:53 not bigger or smaller Sep 10 14:43:06 1" = 72 pixels vs. 1" = 144 pixels Sep 10 14:43:08 for example Sep 10 14:43:13 javaDon: I do think that dp value roughly correlates to the screen's physical size Sep 10 14:43:14 so how do you seperate by bigger or smaller? Sep 10 14:43:46 javaDon, either the -large/-xlarge/-small discriminators if you're on android <3.2 Sep 10 14:43:50 for instance a screen with 100 dp width would be approximately twice as wide as screen with only 50 dp. Sep 10 14:44:02 or -sw600dp, -sw720dp if you want 7" vs. 10" tablets Sep 10 14:44:10 or -sw533dp, as nseidm1 mentioned for the kindle fire Sep 10 14:44:12 I say 'roughly' because dp has an approximate relation to physical size. Sep 10 14:44:25 ...and we haven't even started to talk about the size of your fingers that don't scale with screen size :) Sep 10 14:44:31 Hello guys! I got a problem. Do you know why my OnItemSelectedListener is not called on my spinner? Code is Here: http://pastebin.com/raw.php?i=C3zELv2c Sep 10 14:44:32 my min sdk is 8 Sep 10 14:44:44 RedNifre, that's exactly why dp is an appropriate unit of measure Sep 10 14:44:59 RedNifre, because the primary goal is to scale things so that they are roughly touchable regardless of density and screen sizes Sep 10 14:44:59 ...for things that you tap. Sep 10 14:45:06 dp is reasonable for drawing touch regions and measuring touch sizes, for sure. Sep 10 14:45:46 but application made for device held at 30 cm distance as opposed to 20 cm distance should probably think about scaling all dp measurements that relate to text and images by factor of 1.5. Sep 10 14:45:54 so in that sense dp is not a great choice. Sep 10 14:46:00 30cm and 20cm is not a big deal Sep 10 14:46:12 anything up to about 60" really isn't a big deal Sep 10 14:46:16 well it is exactly 50 % deal Sep 10 14:46:34 you can't seriously be contemplating the relative merits of modifying dp based on how far away from the screen you are :/ Sep 10 14:46:36 if you're a design perfectionist, sure Sep 10 14:46:45 Kegsay, that's what I think... Sep 10 14:46:46 increase distance by 50 %. If you want same readability/size-in-eye, multiply sizes by 1.5 too Sep 10 14:46:55 Kegsay: no. Sep 10 14:46:59 alankila, I don't want same readability or size-in-eye Sep 10 14:47:05 Kegsay: but I'm pointing out that dp is fundamentally a compromise for ignoring this matter. Sep 10 14:47:06 particularly when I'm not using the smallest font-size possible Sep 10 14:47:18 this matter is pretty much irrelevant on touch-devices Sep 10 14:47:20 dp works for touch, but things you look at need arcangle pixels which you can guess based on the screen size? Sep 10 14:47:26 when we're talking larger distances, it becomes much more relevant Sep 10 14:47:34 projection screens, tvs, computer monitors Sep 10 14:47:56 hmm Sep 10 14:48:09 but at touch distances, it really doesn't matter Sep 10 14:48:10 does anyone know why Hiearchy Viewer is showing me 0ms for draw/layout/measure? Sep 10 14:48:53 we support droid 2.2 and up Sep 10 14:49:01 soo... i cant use sw tags Sep 10 14:49:25 Sure you can. Just don't support <3.2 tablets Sep 10 14:49:28 why do you think that matters? Sep 10 14:49:35 Your layout/ would just be phones Sep 10 14:49:48 my problem isnty on tablets Sep 10 14:50:06 my problem is the scree size difference between normal screens Sep 10 14:50:48 an phone screen with over 1280 im guess will come up as largeScreens correct? Sep 10 14:51:13 I don't think anyone changes layout based on phone screen sizes Sep 10 14:51:21 javaDon, not necessarily, no Sep 10 14:51:35 sgs4 is 1920x1080 and it is a regular screen Sep 10 14:51:42 javaDon: large has to do with the physical size, not the resolution Sep 10 14:51:57 fuck me then Sep 10 14:52:00 javaDon, use -large, -xlarge for <3.2 Sep 10 14:52:02 are you deaf? Sep 10 14:52:07 or detect it in code Sep 10 14:52:17 i cnat detect in code Sep 10 14:52:18 by calculating density and screen size Sep 10 14:52:28 im haveing 4 seperate apks Sep 10 14:52:32 what are you trying to do, anyway Sep 10 14:52:42 4 apks? awesome, heh Sep 10 14:52:46 and yeah, F all tablets <3.2 Sep 10 14:52:53 so that fixes your problem Sep 10 14:53:04 all the 0.1 % of users are going to be so pissed Sep 10 14:53:14 no thers still small resolutioned phones Sep 10 14:53:35 then use the -small and -xsmall whatever discriminators Sep 10 14:53:43 really, this isn't hard... why are you making it difficult? Sep 10 14:53:52 i mean theres got to be a way to define what apk gos were Sep 10 14:53:58 ... Sep 10 14:54:01 i used libgdx framework Sep 10 14:54:05 obviously, you're not looking for help Sep 10 14:54:07 * pfn wanders off Sep 10 14:54:17 why does one make multiple APKs for the different sizes. What is the reason? Sep 10 14:54:23 pfn of corse im looking for help Sep 10 14:54:25 alankila, assets too big Sep 10 14:54:40 I see. Only can afford one size, and can't afford realtime resize? Sep 10 14:55:09 yes alankilla Sep 10 14:55:36 since haveing alll assets in one apk will cause the apk to go over a hunders Sep 10 14:55:52 even haveing 2 or 3 will also cause it to be to bg Sep 10 14:55:56 so 4 is a must Sep 10 14:56:17 javaDon, really, why can't you just use manifest discriminators to send to the appropriate devices? Sep 10 14:56:26 you have all the tools you need, what is the problem? Sep 10 14:56:35 Weird... when I run my test project the app lacks all classes from external libraries. Hm.... Sep 10 14:56:55 i dont know how Sep 10 14:57:42 pfn i dont know ho im supposed to send my apk thats was built for a resolution of 1280 by 800 get sent to an ldpi res 1280 by 800 Sep 10 14:57:52 I'm not sure I understood the problem. javaDon, does your app have so many images that providing all dpis is impossible? Sep 10 14:57:59 javaDon, you don't build for RESOLUTION Sep 10 14:58:04 i know Sep 10 14:58:15 if you keep asking that, you are going to be ignored then Sep 10 14:58:32 not of the actual layout design was done through android layouts Sep 10 14:59:01 RedNifre yes Sep 10 14:59:05 it does Sep 10 14:59:24 Hm, my error actually states "make sure it is in the apk"... how do I run a test project in such a way that it makes sure all libraries are in the apk? Sep 10 14:59:29 i didnt set it up through android xml Sep 10 15:00:18 I need help!!!!.... I'm trying and searching for hours now and my spinner listener is not called! Sep 10 15:00:23 http://pastebin.com/raw.php?i=C3zELv2c Sep 10 15:00:30 javaDon if you have so many images you could include them as svg and render them on the device in the appropriate dpi if you are really dedicated. Or simply use xxhdpi and don't care about how it looks on ldpi Sep 10 15:01:06 rendering svg is pretty slow still probably Sep 10 15:02:06 it's far easier to simply rescale, and if you did decoder-based rescaling by integer factors it should still be pretty fast Sep 10 15:02:51 no svg guys Sep 10 15:03:00 this is a game Sep 10 15:03:05 performance matters Sep 10 15:03:11 g00s I meant rendering them once, store the pngs and use them from then on. Sep 10 15:05:30 tell me if this makes sence to you guys 1 sec Sep 10 15:06:03 wheres my tick always box on replacing a debug apps signed with different key Sep 10 15:06:04 grrr Sep 10 15:06:11 at least we get a dialog now Sep 10 15:06:20 thanks to the ADT tool devs Sep 10 15:06:34 xhdpi Sep 10 15:06:35 jumpythefrog.com Sep 10 15:06:39 whoops Sep 10 15:06:45 my bad Sep 10 15:06:51 gotta refresh Sep 10 15:08:33 https://dl.dropboxusercontent.com/u/96964806/platform-6%202.png Sep 10 15:08:36 ldpi^ Sep 10 15:08:45 https://dl.dropboxusercontent.com/u/96964806/platform-6.png Sep 10 15:08:49 xhdpi^ Sep 10 15:08:57 tell me if that makes sence Sep 10 15:08:59 please Sep 10 15:10:20 ? Sep 10 15:11:30 It usually does, but if your apk gets too large you might want to cut the smaller dpis Sep 10 15:12:03 ok so this is why i need the 4 apks RedNifre Sep 10 15:12:19 so Sep 10 15:12:28 dpi is the screen resolution Sep 10 15:12:32 javaDon: the image sizes are correct for the dp scale Sep 10 15:12:42 great Sep 10 15:13:17 no, dpi is the screen density Sep 10 15:13:17 not possible, but you could publish it without graphics and have it load the correct ones on first start. Sep 10 15:13:19 not resolution Sep 10 15:13:42 RedNifere Sep 10 15:13:49 your images are badly compressed, you can save space there. Sep 10 15:13:51 it would also have to load the correct code on start Sep 10 15:13:57 i know Sep 10 15:14:03 there is a relationship to resolution, but screen dpi does not dictate resolution Sep 10 15:14:06 i compressed them already Sep 10 15:14:11 just an example Sep 10 15:14:16 javaDon: ((dpi = dots per inch) == (ppi = pixels per inch)) Sep 10 15:14:24 hello i have a problem with the download manager. i pass it a good url but it automatically send me a broadcast that it failed and the reason is DownloadManager.ERROR_UNKNOWN Sep 10 15:14:31 how can i get the error code that it got ? Sep 10 15:14:52 k Sep 10 15:15:47 but going with that formula our main backgroind images are set to ill give anothere sample Sep 10 15:17:13 https://dl.dropboxusercontent.com/u/96964806/sky1.1-sequence.png Sep 10 15:17:18 xhdpi^ Sep 10 15:17:32 https://dl.dropboxusercontent.com/u/96964806/sky1.1-sequence%202.png Sep 10 15:17:37 ldpi^ Sep 10 15:18:02 if your saying that Sep 10 15:18:24 a phone thats 1280 by 800 can be ldpi Sep 10 15:18:37 it would be getting the wrong assets Sep 10 15:19:00 I want to use actionbarsherlock with maven. do I have to have actionbarsherlock in my eclipse workspace? Sep 10 15:20:55 because a 300 by 384 on a 1280 by 800 would be little tiney ity bitty Sep 10 15:24:00 javaDon, 1280x800 will never be ldpi on a phone, unless it's a 20" phone Sep 10 15:26:32 are there any 4k android devices yet ? Sep 10 15:28:31 Hey, I'm getting on error "unable to resolve target android-8" even though my manifest targets sdk version 15 Sep 10 15:29:18 deadmund, that's what you have in project.properties Sep 10 15:29:37 ok thanks Sep 10 15:30:36 pfn so this brings me back to my original delema though a 1280 by 800 tab can be ldpi Sep 10 15:30:36 got a problem with a nullpointer on a view, after like 20min not using the phone/app it returns a nullpointer Sep 10 15:30:37 that fixed it btw Sep 10 15:30:48 javaDon, so? Sep 10 15:30:58 is it a problem what can be solved in the onpause / onresume? Sep 10 15:31:29 so if i say that the apk with that ldpi image is for ldpi devices the 1280 by 800 tab will get the wrong images Sep 10 15:31:55 the 1280 by 800 ldpi tab that is Sep 10 15:32:58 javaDon, no, you put assets into ldpi/hdpi/mdpi etc. to maintain a consistent size Sep 10 15:33:02 on-screen Sep 10 15:33:18 javaDon, if you want a full-bleed image, then put images into -large, -xlarge, etc. and scale as appropriate Sep 10 15:34:56 full-bleed? Sep 10 15:35:05 edge-to-edge Sep 10 15:35:18 Is there a Stringbuilder in Android-Java which replaces like %s to a String, %d to a dec, ...? Sep 10 15:35:47 yes but most of these images are made so that there is no scale applied to them Sep 10 15:35:47 misterli: String.format Sep 10 15:36:07 javaDon: drawable-nodpi Sep 10 15:36:17 ty Sep 10 15:36:18 Heel Sep 10 15:36:43 clever will not work i did not do the layouts with android xml Sep 10 15:36:50 i used libgdx framework Sep 10 15:37:02 MessageFormat.format can be a bit better than String.format, depending on the formatting needs. Sep 10 15:37:08 hi guys Sep 10 15:37:12 does anyone remember the issue i had yesterday/ Sep 10 15:37:54 i'm having some trouble adding Facebook android sdk to my android project Sep 10 15:38:14 janmayan77, what's the trouble Sep 10 15:38:32 i get an error R cannot be resolved to a variable Sep 10 15:38:45 this happens as soon as i add the Facebook sdk library to the project Sep 10 15:38:59 does anyone have an idea how i could solve this problem ? Sep 10 15:39:00 thanks Sep 10 15:39:01 clean your project Sep 10 15:39:09 hmm i never had that issue as far as i remember have you tried cleaning your project Sep 10 15:39:09 i tried that several times Sep 10 15:39:22 disable autobuild, clean -> completly rebuild -> enable autobuild Sep 10 15:39:37 let me try that Sep 10 15:40:37 janmayan77: there shoud be another problem. there is a problem at your or facebook sdk's resources that prevents the gen folder created Sep 10 15:41:08 look at your logs Sep 10 15:41:42 Is there anyone used maven with an android project? Sep 10 15:42:31 i tried that. i get the same errors. Sep 10 15:42:47 in the problems tab i get several errors like: R cannot be resolved to a variable Sep 10 15:43:41 That error means it can't auto-generate the stuff from your resources - not that this tells you why Sep 10 15:43:56 if you're in eclipse there should be an error console view which shows you Sep 10 15:44:17 janmayan77: look at the res folder if there is a red mark on the icon Sep 10 15:45:23 there's no red mark in the res folder Sep 10 15:45:39 only on the files in the src folder Sep 10 15:45:53 i get errors whenever i try to reference R Sep 10 15:47:08 i notice that there's no R.java inside the gen folder Sep 10 15:47:32 yeah that is why you are receiving that error Sep 10 15:47:34 Hey guys, this is meant to be returning a list of facebook profiles in a listview. I've set the cap to I think 10 profiles each request. It sort of worked for a couple of time when I was at home, but whenever I'd try a couple times more or for some other maybe more common names, the activity would remain blank. I'm pretty sure it has a memory leak somehwere. Now I'm testing the app outside usin Sep 10 15:47:34 g 3G and it's just not returning anything. Can someone spot the issue/memory leak/ know of a way of implementing it better so it's not slow as tits? http://pastebin.com/kgm0uDFw Sep 10 15:48:32 also something i notice… if i try to use text autocompletion in eclipse, e.g. R.layout. Sep 10 15:48:47 ithe ice seems to suggest layouts from inside the Facebook library Sep 10 15:49:56 ide* Sep 10 15:50:34 has anyone experienced something similar in the past ? Sep 10 15:50:51 i've been checking google for hours trying to find a solution Sep 10 15:51:42 janmayan77: did you import R from another package by mistake? Sep 10 15:53:14 no Sep 10 15:53:35 anyone? :) Sep 10 15:53:47 the project runs well once i remove the Facebook library Sep 10 15:56:00 I knew some shit was gonna screw up when importing the Facebook library so I made sure I took an hour to do following each step as carefully as possible and it worked lol *phew* Sep 10 15:56:00 with gson, is there a way to have it log which field actually failed with things like "Expected STRING but was BEGIN_OBJECT" ? Sep 10 15:56:28 but yeah, can anyone help? Sep 10 16:01:42 just incase anyone has missed the question Sep 10 16:01:47 Hey guys, this is meant to be returning a list of facebook profiles in a listview. I've set the cap to I think 10 profiles each request. It sort of worked for a couple of time when I was at home, but whenever I'd try a couple times more or for some other maybe more common names, the activity would remain blank. I'm pretty sure it has a memory leak somehwere. Now I'm testing the app outside usin Sep 10 16:01:47 g 3G and it's just not returning anything. Can someone spot the issue/memory leak/ know of a way of implementing it better so it's not slow as tits? http://pastebin.com/kgm0uDFw Sep 10 16:03:31 janmayan77: my only thoughts are you're adding the facebook library wrong (and screwing up the paths for stuff), or it's the wrong version of the library and it's generating resource errors internally. Having not used it, I don't have any idea beyond that. Sep 10 16:04:49 dragorn, could you by any chance help ? Sep 10 16:05:46 CocoStorm: nope, as I just said to janmayan77, I've never used the facebook library Sep 10 16:06:15 dragorn, ah but it's literally just a memory leak question but for sure no worries :) Sep 10 16:07:44 alright. i'll try including the library code directly into my project Sep 10 16:07:48 thanks Sep 10 16:10:58 anyone else? Sep 10 16:22:05 does IAB v3 no longer support making unmanaged products at all? Sep 10 16:22:46 having an issue consuming this unmanaged consumable i made Sep 10 16:24:01 Coco: What is RequestAsyncTask? Sep 10 16:24:47 facebook stuff? Sep 10 16:25:39 nseidm1, yeah, I've just posted my question on Stackoverflow. I guess I've explained a little better there http://stackoverflow.com/questions/18724221/trying-to-query-for-facebook-profiles-and-sort-them-into-a-listview-android Sep 10 16:25:42 after you update you're bitmaps array you're not calling notifyDataSetChanged Sep 10 16:26:06 also your bitmaps array will be finished populating at an undetermined amount of time in the future Sep 10 16:26:42 nseidm1, at an undetermined amount of time? Sep 10 16:26:46 you are executing the bitmap retrieval using anonymous threads that have no synchronization Sep 10 16:29:34 trying to see what notifyDataSetChanged is Sep 10 16:29:39 I've never heard of it :) Sep 10 16:29:47 without it listview is useless Sep 10 16:29:53 Oh really? Sep 10 16:29:56 yes Sep 10 16:30:14 also what od you mean by the threads have no synchronization? Sep 10 16:30:28 your threads can complete any any time Sep 10 16:30:35 they can also be blocked by network comm issues Sep 10 16:30:46 there is no guarantee in the order in which they will complete Sep 10 16:31:16 nseidm1, I have used thread.join() should that not do the trick? Sep 10 16:31:24 ug Sep 10 16:31:27 heh Sep 10 16:31:31 use a SingleThreadPoolExecutor Sep 10 16:31:47 Executors.newSingleThreadExecutor() Sep 10 16:33:02 nseidm1, so notifyDataSetChanged updates the views? If i clicked back and onResume or onCreate was called would that not update the views? Sep 10 16:33:39 may or may not Sep 10 16:33:51 setAdapter will update the views Sep 10 16:33:59 but that's hella innefficient Sep 10 16:35:33 Okay so two issues are notifyDataSetChanged and SingleThreadPoolExecutor. Do they explain why the things is being tempremental and slow? Sep 10 16:35:45 I'm guessing they will sort the memory leaks Sep 10 16:36:55 how is it possible when i close the app and i reopen it after like 20 min it returns a nullpointerexception on the getview().findviewbyid etc... Sep 10 16:37:10 has it something to do with the lifecyle? Sep 10 16:37:13 c Sep 10 16:37:14 dominee: because getView is null Sep 10 16:37:34 the fragment's view needs to be recreated Sep 10 16:37:47 in which method? Sep 10 16:37:48 as in getView hasn't been called yet Sep 10 16:37:58 rather onCreateView() Sep 10 16:38:19 hmmm\ Sep 10 16:38:42 as you said, has everything to do with lifecycle Sep 10 16:39:08 I am creating an separate package in order to create a background service. In my package I cannot import android.app; am I doing something wrong? Sep 10 16:39:46 what is android.app? Sep 10 16:39:54 tried a couple of things yes, but nothing helps for so far Sep 10 16:40:15 I cannot import package android.app Sep 10 16:40:16 Dominee: ensure you call getView in onActivityCreated() or later Sep 10 16:40:25 zquad: because android.app is a private package Sep 10 16:40:37 why do you want to use it Sep 10 16:40:41 just remove the import Sep 10 16:40:53 getview is called in the ontask after httpRequest Sep 10 16:41:32 Dominee: so, that clearly doesn't relate to lifecycle. Ensure that getView() is called after, and only after, onCreateView() has been called Sep 10 16:41:52 ok Sep 10 16:42:46 nseidm1: want to use android.app.IntentService Sep 10 16:43:11 thanks i'm gonna take a look at it Sep 10 16:43:51 brb Sep 10 16:44:00 zquad: just clean up your imports Sep 10 16:45:20 cleanup? I am using android studio 0.2.7 Sep 10 16:45:25 what do you mean Sep 10 16:45:47 I don't use Android Studio, I won't be of any help Sep 10 16:45:52 BUMP Sep 10 16:52:39 Hi all, I have received a project which contains a drupal website. This website uses a mysql database. Now I have to make an android (+ ios in the future) app that uses the data from the mysql database, is this possible? Sep 10 16:52:50 where would be a good place to set the behavior of a ToggleButton if it's in a ListView item's layout? Sep 10 16:55:10 tnzr, possible inside your adapter Sep 10 16:55:19 FrancescoV: ok.. im going to try in getView() Sep 10 16:55:21 thanks Sep 10 17:03:01 The iPhone5c Sep 10 17:03:22 for cheetos Sep 10 17:03:24 I'm calling it Sep 10 17:04:00 if its not dorritos, im not interested Sep 10 17:09:05 mikedg: the c stands for crap edition ;) Sep 10 17:10:25 the C stands for China Sep 10 17:10:36 is there any way to upload an apk to Play and not have it automatically update for, even if a user has auto-update enabled in Play Store ? Sep 10 17:10:38 Leeds: ooh Sep 10 17:10:46 I mean, not really - but it does Sep 10 17:10:50 it's going to be hard to break through 90-something percent for apple Sep 10 17:11:07 line was insane at the apple store here Sep 10 17:11:28 literally insane? ;) Sep 10 17:11:58 literally, since they were overnight in rain :D Sep 10 17:12:00 Hi, how can I add a SurfaceView to my layout when my app is already running? My layout.xml is just an empty table to which I would like to add several SurfaceViews depending on a variable. Sep 10 17:12:31 i think 'C' stands for concatenation or column :D Sep 10 17:12:38 wow. insane AND literally insane lol Sep 10 17:12:56 Siri, I need some tweets Sep 10 17:13:14 siri now comes with support for twerking Sep 10 17:13:16 quick i need quick tweets about lady gaga Sep 10 17:13:32 Siri, buy me a replacement for you Sep 10 17:13:43 OMG thats such an android dialog, Apple stole dialogs from Android Sep 10 17:14:13 I have long stopped caring about who stole what on styles Sep 10 17:14:18 as long as styles work for me Sep 10 17:14:29 I think ios 7 does a great job at making content, not chrome, pop Sep 10 17:14:54 Apple blows dogs for quarters man! Sep 10 17:24:16 green iphone for me Sep 10 17:24:26 Are there any projects on github that replace the default list of sharing apps? Sep 10 17:24:41 replace? Sep 10 17:25:27 well I have the dialog up on screen mikedg, but when I select facebook the text and subject dont carry over like they do on email and twitter Sep 10 17:26:07 its been suggested that I make a custom one that knows how to pass properly to facebook Sep 10 17:27:46 mikedg: sure, you get 0xfffffe iphone. it's green! Sep 10 17:30:25 hi guys, is it possible to compile android yourself and use the compiled drivers/module for example from my current xperia z? Sep 10 17:31:14 the excessive amount of background services that drain my battery is ridiculous especially when you look at the fact that most of them only provide functionality that i dont care about at all, for example statistics and bullcrap for google or sony Sep 10 17:32:10 or do you guys think its better to just "root the device" and hack away? Sep 10 17:34:46 i guess what i'm asking is how doable it would be to compile your own compatible version of android for a device (and strip garbage) Sep 10 17:34:58 dunno Sep 10 17:35:04 Stealthy: are you developing an app which needs maximum performance? Sep 10 17:35:24 no Leeds just trying to get the best out of my device Sep 10 17:35:30 because otherwise you probably want to ask in #android-root Sep 10 17:35:54 oooh thought it was more allround here ^^ thanks i'll check there as well Sep 10 17:36:28 Stealthy: see where the channel topic says "Root/custom ROMs, head to #android-root" and "Android application development"? Sep 10 17:37:11 didnt notice, dont read them usually Sep 10 17:37:29 oh just read more about iphone. they actually went back to colors? wow. lol Sep 10 17:37:32 I've got smartphone clients that i want connected thourgh some hotspot, creating a LAN - then run a co-op stress puzzle game real time Sep 10 17:37:40 not that it'll get me to buy one. Sep 10 17:37:42 Stealthy: not a great thing to announce Sep 10 17:37:47 much a like nintendo DS set up Sep 10 17:38:25 anyone into this thing? creating local servers, and using it for handling local data Sep 10 17:40:19 are there any easy ways to deal with the facebook no text error? Sep 10 17:42:32 MickeySoFine: I spent a lot of time on that Sep 10 17:43:15 i think the gradle sighningConfigs should have , in addtion to storeFile(), configFile() or something - who puts their password in the build? pretty much everyone would put it in properties file; but gradle.properties i check in. Sep 10 17:43:51 easy enough to load another properties file i guess; but kludgy Sep 10 17:44:17 MickeySoFine: you have to override startActivity, that's the ONLY way Sep 10 17:51:10 anyone using Chris Banes Pulltorefresh lib? Sep 10 17:52:13 i just put a refresh button in the actionbar Sep 10 17:52:15 :D Sep 10 17:52:17 i use Mike Batmans pulltorefresh lib Sep 10 17:52:54 mikeeeeeeeeeeeeeeeeeeeeeee Sep 10 17:53:07 owwww https://i.chzbgr.com/maxW500/7782598144/hE2F89F7D/ Sep 10 17:53:16 g00s, thats a slick idea,however I need the pull to refresh mechanism Sep 10 17:53:25 that gotta hurt Sep 10 17:53:36 trouble is I cant get the click listener to work Sep 10 17:54:07 since offtopicness is less of a problem than originally implied by Leeds: Sep 10 17:54:07 http://images.flabber.net/files/GTA-V---officiele-map---lijkt-op-ninja-turtle.gif Sep 10 17:54:11 this made me lol a bit as well Sep 10 17:54:46 nseidm1, any suggestions of code samples or tutorials I could follow? Sep 10 17:54:49 lol Sep 10 17:55:14 MickeySoFine: you should override startActivity and inspect the intent Sep 10 17:55:26 G+, Facebook, and all other apps expect different things in the extras Sep 10 17:55:49 it's a bit of a mess to share to all of them Sep 10 17:56:08 facebook wants extra_subject, and extra_text Sep 10 17:56:18 G+ wants PlusShare.EXTRA_CONTENT_URL Sep 10 17:56:29 and every other app wants an EXTRA_STREAM Sep 10 17:56:36 it's ridiculous Sep 10 17:56:44 so much for a unified sharing mechanic Sep 10 17:57:02 mikedg does that lib allow both up / down pull to refresh? Sep 10 17:57:11 up? Sep 10 17:57:13 mike batman, are you trolling me? Sep 10 17:57:15 what kind of sorcery is that Sep 10 17:57:26 holy batman, robin Sep 10 17:57:26 chris banes works like that, mode.BOTH Sep 10 17:59:23 am i still connected - anyone say yes please Sep 10 17:59:29 can but he do it in slowmo? Sep 10 17:59:42 I think you got disconned bro, try /reconnect Sep 10 17:59:49 thanks im back Sep 10 18:00:33 i didnt see you reconnect though Sep 10 18:00:37 the problem with the other pull to refresh libs is that you drag down to refresh..i want my new results(older entries) to come at the bottom of the list Sep 10 18:01:00 brx_: then why even bother with a pull? Sep 10 18:01:05 just load stuff at the bottom Sep 10 18:01:16 and then scrollTo()? Sep 10 18:04:29 if i load stuff at the bottom then they still have to scroll back to the top of the list in order to refresh/add more results...and then they scroll back down the list to read them. seems counter intuitive Sep 10 18:08:04 damn it works when i modify his sample app,but not inmy own app Sep 10 18:08:18 which is a kick in the balls Sep 10 18:11:07 just load when they are scrolling down Sep 10 18:11:10 make the list never ending Sep 10 18:12:21 okay got question Sep 10 18:12:26 if I create bitmap Sep 10 18:12:44 then make BitmapShader out of it Sep 10 18:12:58 can I then discard the bitmap or do shader still need it? Sep 10 18:13:25 the shader will still need it Sep 10 18:13:32 so you can lose references to it Sep 10 18:13:36 the shader will keep a reference Sep 10 18:13:40 just don't recycle the bitmap Sep 10 18:13:44 hmm ok Sep 10 18:14:19 I still hit crash issue with out of memory :( its flaw is that it requires background sized bitmap as well as 2 other images (sources) Sep 10 18:14:38 ? Sep 10 18:15:24 java.lang.OutOfMemoryError Sep 10 18:15:24 error Sep 10 18:17:01 have anyone ever notice that airplane mode does not blocks connections on android emulator? Sep 10 18:17:57 mikedg, can you link me to the pulltorefresh lib you use please Sep 10 18:18:05 i made it up Sep 10 18:18:23 sorry Sep 10 18:18:33 mikedg: welcome to "jokes don't carry to non-native speakers" Sep 10 18:18:45 I'm assuming... Sep 10 18:18:59 ye im from zimbabwe me Sep 10 18:20:52 https://github.com/chrisbanes/Android-PullToRefresh Sep 10 18:20:57 http://stackoverflow.com/questions/18723265/using-showcaseview-with-the-v7-appcompat-actionbar Sep 10 18:21:47 Got a newbie question for you all. I know user apps are (or at least can be) developed and debugged via ADT/Eclipse. But how are system apps developed & debugged? For example, say Cyanogenmod wants to develop a special system app. What do they use? Sep 10 18:22:02 nseidm1 im already using that, problem is .setOnItemClickListener() doesnt seem to register for me Sep 10 18:22:17 andrew76: same, mostly Sep 10 18:22:23 nseidm1, how do I override startactivity? Sep 10 18:22:42 @Override in the activity Sep 10 18:23:00 I think the better question is why you're overriding startactivity, though Sep 10 18:23:30 <|0xD34D|> andrew76: I load the system app into intellij idea and then set breakpoints and attach to the debugger Sep 10 18:24:04 well RyanM nseidm1 says the best way to deal with facebook missing text is to override the startActivity method, however, I dont really know how that helps? Sep 10 18:24:08 RyanM: because G+, Facebook, and all other app require different intent extras to share data properly Sep 10 18:24:23 oh, I don't know anything about that. Sep 10 18:24:40 nseidm1, how do I detect which intent was selected? Sep 10 18:24:41 I'd make a helper method, personally Sep 10 18:24:44 MickeySoFine: I explained how that helps above Sep 10 18:25:13 yeah I get that part but how do I tell which intent is selected? Sep 10 18:25:17 MickeySoFine: intent.getComponent().getPackageName() Sep 10 18:26:05 and that's the point of the override, the user selects what app they want to share to, you determine which app was selected, and you populate the extras as required by the app Sep 10 18:27:00 do you have any code snippets that show this? Sep 10 18:27:18 So, I need to retain some data on screen rotation, but I only care about this data when a certain Fragment is present in the UI. When the Fragment is not in the UI, the data should be available for garbage collection. The FragmentRetainInstance demo only shows adding a retained Fragment that is there during the entirety of the app's lifecycle. Any suggestions? Sep 10 18:28:41 MickeySoFine: for example this will detect G+ being selected: if (component.getPackageName().contains("com.google.android.apps.plus")) Sep 10 18:28:55 equals, contains, whatever Sep 10 18:28:56 Leeds: Is there a way to tell ADT to install/run something as a system app instead of a user app? Sep 10 18:28:59 get the package name form the Play Store Sep 10 18:30:33 hey nseidm1 Sep 10 18:30:42 heyyyyyyy mannnnnn Sep 10 18:30:43 ive come on duty now Sep 10 18:30:47 you can retire Sep 10 18:30:48 :D Sep 10 18:30:51 heh Sep 10 18:30:57 lol, I need to, stuff to do here Sep 10 18:31:07 nseidm1, its the steps to getting component that I dont seem to understand Sep 10 18:31:14 meh, must make a purchase of my app so I can give it a rating Sep 10 18:31:25 intent.getComponent() Sep 10 18:31:42 pfn: sucks, I know Sep 10 18:32:06 it always crashes at res = BitmapFactory.decodeFile(filename); when not enough memory Sep 10 18:32:20 XXCoder: well, isn't it obvious what the issue is :-) Sep 10 18:32:42 it is. problem is I can't figure how to do what I want to do and avoid oom Sep 10 18:32:58 decode a piece of it, or decode it in lower resolution, or use largeheap option Sep 10 18:33:01 XXCoder: inSampleSize, use it Sep 10 18:33:17 start with a tool like hat or MAT and figure out where all your memory is going Sep 10 18:33:26 https://lh3.googleusercontent.com/-uCS6nHsvOrE/Ui8hWcfLZkI/AAAAAAAAxbA/fM7JxoFJ_-Q/w421-h650-no/BTv7nsKIMAAR9S2.jpg Sep 10 18:33:33 how do I figure how to calculate it to fit exactly size Sep 10 18:33:42 XXCoder: how big are your bitmaps? Sep 10 18:33:49 XXCoder: as in width and height Sep 10 18:33:57 one matches live wallpaper size Sep 10 18:34:19 2 texture bitmaps as large as picture taken or smaller (depends on user select) Sep 10 18:34:30 XXCoder: there's a tutorial on android developer site about how you can measure the dimensions of a file, and how you can set options to decode it while shrunk by some integer factor. Sep 10 18:34:51 alan thanks ill look into it Sep 10 18:35:07 also if you actually have textures, or plan to upload the files to GPU, do it as soon as you can so you can free the android-side bitmap. Also you can consider decoding in 16-bit color rather than 32-bit color Sep 10 18:35:40 alankila: 16 bit images should be the last solution Sep 10 18:35:48 also note if you have hardware accel enabled most devices require <= 2048x2048 res for the hardware renderer Sep 10 18:36:07 I'm sure romain could speak a bunch toward that topic :-) Sep 10 18:36:17 this wouldn't cause OOMs Sep 10 18:36:24 you'd get a blank image and a warning in the logs :) Sep 10 18:36:37 I dont use gpu, just use basic paint stuff Sep 10 18:36:43 Your order was cancelled because you can't buy from yourself. Sep 10 18:36:44 really? Sep 10 18:36:49 I never knew... Sep 10 18:36:56 XXCoder: the Android framework uses the GPU on your behalf Sep 10 18:36:56 * pfn needs some astroturf love Sep 10 18:36:58 I struggled with OOMs once too and my solution was to decode my images in 512x512px fragments and make all such fragments releaseable which weren't in screen at that very moment Sep 10 18:37:27 oh. how do I offload bitmaps to gpu and can paint whatever? Sep 10 18:37:48 unfortuately each frame any point can be painted on Sep 10 18:37:50 forget about the GPU for now Sep 10 18:38:01 just check where your memory is going Sep 10 18:38:03 so I need entire wallpaper. as well as source 2 images Sep 10 18:38:06 and look at how big your bitmaps are Sep 10 18:38:27 lemme make it output sizes Sep 10 18:39:22 how the hell do I change the credit card google music all access uses? Sep 10 18:39:35 I can't remove the card from wallet, and I can't set it in the play music app... Sep 10 18:39:38 pfn: i had to do that Sep 10 18:39:46 i think theres a subscription tab in wallet Sep 10 18:39:51 it was a pain in the ass to find Sep 10 18:39:53 ah, I see Sep 10 18:39:55 Hi, using android studio, is there any way to test my app easier than running my phone emulator? Like running the app directly on windows. Sep 10 18:39:56 lemme see if i can find it again Sep 10 18:40:01 I just clicked cancel from web and it took me to the subscription page Sep 10 18:40:01 pfn: you need a phone with the primary google account being something other than your dev account Sep 10 18:40:16 nseidm1, yeah, I can do that later Sep 10 18:41:00 man, why would I ever upload music with all access... Sep 10 18:41:07 damn thing has every single thing I ever want to listen to Sep 10 18:41:07 i hate that i have a terabyte of free storage Sep 10 18:41:12 you gotta try subscriptions, by far the best addition ever! Sep 10 18:41:16 yet still have to pay for my 80gigs to keep it Sep 10 18:43:07 mikedg, i got it working with android:descendantFocusability="blocksDescendants" on the root layout of the item (im sure ive been here before:s) Sep 10 18:46:33 odd system.out isnt outputting anywhere Sep 10 18:47:19 why not use LogCat? Sep 10 18:47:30 diehard java devs in the house Sep 10 18:47:37 XXCoder, that happens to me in eclipse when i need to select the device in DDMS tab,not sure if thats any help to you Sep 10 18:47:40 java dev, yuck :p Sep 10 18:47:50 thanks Sep 10 18:48:35 System.out and System.err are indeed not captured in android Sep 10 18:48:37 nseidm1, I've overriden the startActivity method but I only see my choose intent being passed to it, how do I get the one thats be selected by the chooser? Sep 10 18:48:44 alankila, no, it goes into logcat Sep 10 18:48:55 alankila, at least actual S.o. and S.e Sep 10 18:49:05 avoid components that have the package name of your app Sep 10 18:49:06 Hmh. Okay, I guess fprintf(stderr, etc. doesn't do it Sep 10 18:49:16 alankila, perhaps not writing directly into fd1 and 2 Sep 10 18:49:27 would figure zygote would fdup it, guess not Sep 10 18:49:32 there is likely some java-level construct then Sep 10 18:49:51 Facebook's package name, for example, is com.facebook.katana Sep 10 18:51:58 alankila, yes, it's likely that System.setOut/setErr are used to redirect Sep 10 18:53:21 Hi! From a MainActivity, when I enter the "gps-activity", goes back to MainActivity and then in to the gps-activity again, and back to main and do so for e.g. 4 times, from the MainActivity, I then need to press the back-button 4 times to leave the app. So, I assume that every time I enter gps-activity, a new instance/session is created instead of using the existing. I assume I need to set some intent flag for this? Sep 10 18:54:00 launchMode="singleTop" Sep 10 18:54:07 hmm, a potential user brought up a good point, how does GoogleAccountCredentials protect against request forgery? Sep 10 18:54:16 i.e. a request coming from a malicious application or attacker? Sep 10 18:54:23 nseidm1, did you create a custom chooser to get the right intent? Sep 10 18:54:33 nseidm1, that is set in the manifest for the activity? Sep 10 18:54:43 meck0: yes Sep 10 18:55:06 MickeySoFine: use the standard ShareActionProvider Sep 10 18:55:47 nseidm1, thx will try it! Sep 10 18:57:54 Log.w(drawme.getWidth() + " drawme " + drawme.getHeight(), " "); Sep 10 18:57:56 no effect Sep 10 18:58:26 XXCoder: as in you're having Logcat issues? Sep 10 18:58:32 that is a funny Log line you have there. Perhaps you should back away from the keyboard, slowly Sep 10 18:59:38 XXCoder: palm to forehead yet? Sep 10 18:59:48 anyway it should print something I guess unless there's some nasty clever thing that looks at your actual message and decides it's all whitespace and skips it Sep 10 18:59:50 not yet unfortunately Sep 10 19:00:04 you're printing nothing with that log tag Sep 10 19:00:06 <|0xD34D|> interesting log tag Sep 10 19:00:08 " " Sep 10 19:00:35 <|0xD34D|> that's the message Sep 10 19:00:41 ahh so its tag, message? Sep 10 19:00:48 thought it was other way Sep 10 19:00:49 <|0xD34D|> ahh hah Sep 10 19:05:50 nseidm1, it works, thx a lot :) Sep 10 19:06:37 does anyone know if RenderScript lets you write multiple output values in one pass, or if it's just like a fragment shader (the program runs once per output fragment)? Sep 10 19:08:40 knew it Sep 10 19:08:50 adb server messed up. had to kill it Sep 10 19:09:54 okay, main screen size is 960x800. image 1 285x181, image 2 is 50x40 Sep 10 19:14:22 on phone with larger images screen wallpaper is 884x800, image 1 285x181, image 2 2560x1920. Sep 10 19:14:44 image 2 2560x1920 is approaching potential issues Sep 10 19:14:59 yeah I guessed so. its a chance to fail or not. Sep 10 19:15:47 I'm not worried about 6 built in bitmaps. all are small. its user loadable thats problem Sep 10 19:16:10 current load code is res = BitmapFactory.decodeFile(filename); Sep 10 19:16:26 I wonder if I can get any money for my old samsung galaxy tab 10.1 Sep 10 19:16:38 you could pfn. its really very nice tablet. Sep 10 19:16:46 send it to me if you dont wanna bother sell lol Sep 10 19:17:01 if only it were nice... Sep 10 19:17:04 can't do jack squat with it Sep 10 19:17:18 yeah last update f'ed it up Sep 10 19:17:24 root and install latest cm for it Sep 10 19:18:07 on cm it's slow as piss Sep 10 19:18:17 really? hm Sep 10 19:18:48 and we all know that piss is quite a slow substance Sep 10 19:19:05 :-) :-) :-) Sep 10 19:19:08 lol Sep 10 19:21:08 okay I added BitmapFactory.Option Sep 10 19:21:16 but not sure what to use to scale it down Sep 10 19:21:32 http://developer.android.com/training/displaying-bitmaps/index.html Sep 10 19:21:44 http://developer.android.com/training/displaying-bitmaps/load-bitmap.html Sep 10 19:22:53 * pfn turfs Sep 10 19:25:08 pf I checked criaglist. average seem to be around 200 Sep 10 19:25:17 interesting Sep 10 19:25:28 yeah its pretty good if youre seller since its bit old Sep 10 19:25:43 $250 is closer to avg Sep 10 19:26:05 $200 for the galaxy tab 2 around here, no one even bothers with the 1 Sep 10 19:26:23 lol interesting Sep 10 19:26:31 I wouldn't bother with a samsung tab no matter what Sep 10 19:26:45 I use hp tablet Sep 10 19:26:59 one that had webos lol Sep 10 19:27:21 public static int calculateInSampleSize( Sep 10 19:27:22 ... excellent, that would be useful. Sep 10 19:31:34 how do I handle multiple project libraries that use the same support library in Gradle? Sep 10 19:31:55 how would I clear the text from an EditText? I tried .setText(""), but that didn't work Sep 10 19:32:17 Chais, that should work Sep 10 19:35:36 Chais: try putting a Log statement to see if your code is actually being executed. Sep 10 19:36:42 how? Sep 10 19:38:06 what part of that are you confused about? Sep 10 19:39:07 nvm got it. didn't know how to write to the lgo Sep 10 19:39:09 *log Sep 10 19:43:17 ok appearently it's not executed... Sep 10 19:48:01 If I see one more stack overflow answer that says "Saving state on rotation is easy, just android:configChanges="orientation|screenSize"!" I'm going to scream. Sep 10 19:48:37 okay code is supposely workable now. just need to tweak stuff so it knows screen size before loading anything. Sep 10 19:50:28 build, build, buildddddddddd Sep 10 19:50:51 RyanM: Why's that? Sep 10 19:52:12 xlq: Because it's a terrible idea if you're doing anything that has resource qualifiers other than density, screen bucket, or smallest-width Sep 10 19:54:15 grr.. suddenly started getting sqlite database errors when updating app to newer version and adding a new database file. Here is my DataBaseHelper: http://pastebin.com/JLv3DCvE Sep 10 19:54:56 I get errors such as database disk image is malformed (code 11) the first time I open the app after update Sep 10 19:55:20 the second time I run the updated app, there are no errors Sep 10 19:55:22 any ideas? Sep 10 19:55:39 I increment the private static int DB_VERSION by 1 for every update I do Sep 10 19:58:06 did you lose data? Sep 10 19:58:49 hmm.. nah Sep 10 19:58:56 use an else clause for your if statement Sep 10 19:59:02 that doesnt fix your main problem Sep 10 19:59:06 but it makes it easier to digest Sep 10 19:59:13 the app only reads from the database Sep 10 19:59:22 What is the best way to let a user select an address in a mapview? Click or long Click? Sep 10 19:59:59 i gotta run Sep 10 20:00:00 sorry Sep 10 20:00:03 crap Sep 10 20:00:08 thanks though Sep 10 20:00:31 wonder why this just started happening recently Sep 10 20:00:37 worked fine previous versions Sep 10 20:02:50 what does "on tree page 2 cell 46: invalid page number 50" mean? Sep 10 20:03:44 the table has 3 columns and 600 rows Sep 10 20:25:35 hi Sep 10 20:26:55 im trying to insert values on my sqlite database but on android 2.3- I have a syntax error when trying to insert NULL Sep 10 20:29:11 just to confirm before i register /pay for my developer account....... users on the android market can't see the google account /gmail address i use to publish my apps can they Sep 10 20:30:06 occ there is a thing that says 'email developer' on your apps page Sep 10 20:30:11 its just plain mailto: Sep 10 20:30:44 the developer email can be different from the publisher account email Sep 10 20:30:47 ok thing is i have my own domain name and support@mydomainname.com Sep 10 20:30:59 so i want to have my own domain as my contact Sep 10 20:31:08 you actually can set the email developer for each app individually Sep 10 20:31:31 and the publishing account i am registering with -i have all my other s=stuff like banks and webhost on there -i dont want any chance of spam going there Sep 10 20:31:33 i see Sep 10 20:33:30 hi -- can you use RenderScript to write multiple "out" values in one pass, or is it like GLES2 fragment shaders where the program is run once for each output? Sep 10 20:33:59 I'm doing some computer vision algorithms and want to do integral image/summed-area table style algorithms Sep 10 20:34:51 It's easy on the cpu, straightforward in opencl, and not possible in fragment shaders... Sep 10 20:35:08 wondering if RS gives me another option Sep 10 20:37:18 from what I've read RS looks like the same model as fragment shaders, so it's not useful for what I want to do -- would love some confirmation of that. Sep 10 20:37:45 can anyone confrim that sqlite on android 2.3- does not support multiple insert in the same sql request but it's supported on 4.x ? Sep 10 20:39:09 ok, found the confirmation : multiple insert was introduced on sqlite v3.7.11 which is used on jelly bean Sep 10 20:41:28 i wish it was easier to ship your own sqlite, the situation sucks - they disabled extensions so you can't even add user defined functions Sep 10 20:41:43 nm multiple versions of sqlite with their own weirdness Sep 10 20:42:13 you can compile your own sqlite with the ndk, good luck getting it to work with the supplied java bindings though ;) Sep 10 20:43:17 write your own jdbc bindings! Sep 10 20:43:30 thats why the H2 guy gave up, the jdbc stuff sucks on android Sep 10 20:44:00 all the classes are there, aren't they Sep 10 20:44:04 just need to implement the drivers Sep 10 20:44:26 java.sql and javax.sql is all present on android Sep 10 20:46:44 Hmm.. looks like this fellow have the same issue as me: http://stackoverflow.com/questions/14231485/android-jelly-bean-database-corrupted-on-upgrade Sep 10 20:52:23 Quacked i remember reading something about corrupt databases and 3.7.0 WAL; is this a device where the oem modified the default journal_mode ? Sep 10 20:52:57 if journal_mode=DELETE it wouldnt be the issue Sep 10 20:53:06 if =WAL you may have to investigate Sep 10 20:53:46 I have no idea about what you just said Sep 10 20:53:48 :) Sep 10 20:54:09 the device I currently test on is a I9505 Samsung galaxy s4 Sep 10 20:54:13 sqlite is pretty simple in terms of API though. It shouldn't be *too* difficult to provide some bare bones execution capability... Sep 10 20:54:13 stock rom Sep 10 20:54:32 but I also tested on my wife's galaxy s2 with cyanogenmod Sep 10 20:54:42 * alankila thinks the android crap built on top of sqlite is borderline useless. Everything so clunky. Sep 10 20:54:54 It used to work just fine, and I haven't changed code concering the database Sep 10 20:55:40 As long as I remembered to increment DB_VERSION int if I used a new database file Sep 10 20:55:53 My database helper code is here btw: http://pastebin.com/JLv3DCvE Sep 10 20:56:11 there was some nasty ass bug where if you persisted WAL as a db and opened it up first read only it got fucked Sep 10 20:56:22 persisted the settings Sep 10 20:56:44 and this is totally something an OEM could do, maybe by changing framework SQLiteDatabaseHelper.onConfigure Sep 10 20:56:47 It works fine if clean install or if cache is cleaned from the app Sep 10 20:57:55 well anyhow, copy the corrupted db to your machine and see if WAL is enabled. Sep 10 20:58:04 if you enable it in the pragma, it stays on Sep 10 20:58:10 you are shipping a binary database apparently Sep 10 20:58:15 is this really supported behavior Sep 10 20:58:15 alankila: I am Sep 10 20:58:36 ooh,. ctate said it was OK; although he used to day it wasn't Sep 10 20:58:45 i had to call him out on that :D Sep 10 20:59:12 well that is of course an improvement to the previous situation Sep 10 20:59:24 even if you did your best it could take a while to set the database up Sep 10 20:59:43 I ship a binary database with every new app version by copying a new .db file to asset Sep 10 20:59:49 p Sep 10 20:59:53 whops Sep 10 20:59:56 and increment db version so it will overwrite the previous Sep 10 21:00:00 Quacked: if I had to guess the problem is that the databas file is already open Sep 10 21:00:07 now you replace it, sqlite gets confused, you have corrupted db Sep 10 21:00:16 alankila: yeah, thats a good guess, but I believe I close it with code Sep 10 21:00:43 I don't see how you'd close the database handle Sep 10 21:00:46 if(!mDataBaseExist) { this.getReadableDatabase(); this.close(); Sep 10 21:01:13 don't know what public synchronized void close() does though Sep 10 21:01:44 Hmh. Fine. I'm not sure if that works though. Might be better to just forget about using this helper altogether and manage the whole thing yourself. Sep 10 21:01:59 You're not closing it in onUpgrade or copyDatabase Sep 10 21:02:16 Quacked out of curiosity, what is the nature & form of the data you supply with the app ? Sep 10 21:02:43 there's also the possibility of using unix trickery if sqlite does everything in a single file Sep 10 21:02:57 SimonVT: I tried if(myDataBase != null) myDataBase.close(); before line 70 within copyDataBase() Sep 10 21:02:57 you can create a file under new name and then replace the dentry to new inode by renaming the file on top of the actual database Sep 10 21:03:00 didnt change anything Sep 10 21:03:08 once you reopen the database you should get the new copy Sep 10 21:03:19 g00s: couple sentences coupled with names Sep 10 21:03:39 O.o Sep 10 21:03:40 Could just copy it on launch if app is updated Sep 10 21:03:45 SimonVT: should that have been sufficient? Sep 10 21:04:01 Dunno Sep 10 21:04:06 I don't bundle databases Sep 10 21:04:15 I think "copy it on launch" is what Im trying to do :) Sep 10 21:04:20 And I recommend ignoring most of these android helper classes. They seem to be worse than useless. Sep 10 21:04:27 if (newVersion > oldVersion) ... Sep 10 21:04:38 You're trying to do "copy after the framework has had a chance to mess with the db file" Sep 10 21:04:48 just directly use sqlite and store version in a table, when you manage it all yourself there can be no problem from a framework whose assumptions you are violating Sep 10 21:04:56 alankila: I don't have the time to look into database code again when theres probably some minor change needed :/ Sep 10 21:05:00 I'm suggesing copying it before that Sep 10 21:05:14 Before your SQLiteOpenHelper is touched Sep 10 21:05:38 Quacked could you elaborate on the data? maybe you can get around shipping sqlite all together Sep 10 21:05:45 SimonVT: hmm Sep 10 21:06:30 g00s: I'm hesitating, haha Sep 10 21:06:51 so much work if you suggest I should do it differently :P Sep 10 21:07:00 depending on how data is queried it might just as well be some flat file thing. My first paid app to a customer actually used ObjectOuptutStream instead of SQLite and I'm not proud of that, but I was pressed for time and confused about all the noise around the database and contentproviders and crap, I didn't understand how much of this shit I should use Sep 10 21:07:21 anyways, the app just retrieves random jokes when hitting a button Sep 10 21:07:23 i wonder what google will do when sqlite 4 comes out; 3 and 4 are kinda divergent branches. 4 will not be compat with 3 Sep 10 21:07:32 g00s: never update? Sep 10 21:07:38 but for every joke theres a name as well that can be used to retrieve jokes that have that name Sep 10 21:07:39 alankila busniess as usual Sep 10 21:08:04 Quacked this seems kinda flatly organized; not db worthy :P Sep 10 21:08:15 to ship anyhow Sep 10 21:08:32 I put everything in an excel file and just put it into a .db Sep 10 21:09:01 how do you query this? Sep 10 21:09:06 fts ? Sep 10 21:10:34 just select * from db where name = 'some name' Sep 10 21:10:52 would a key / value store be better ? Sep 10 21:11:02 name as the key ? Sep 10 21:11:20 there are several jokes per name Sep 10 21:11:53 I don't understand why this just started happening now Sep 10 21:12:14 meh; even using sqlite, i would just supply the csv and drop the old table, load a new one Sep 10 21:12:48 if its 10k jokes, maybe disable indexes until the load is finished. under 1000 all you need is a bulk insert Sep 10 21:12:58 in one trasaction Sep 10 21:13:08 yeah Sep 10 21:13:17 maybe I wouldve if I was to start over Sep 10 21:13:28 but I want to figure out why this is failing Sep 10 21:15:27 from 1.01-1.02 version of my app, where I put in just a few more jokes, there are no errors.. but from 1.02 - 1.03 it crashes the first time Sep 10 21:16:33 does anyone know whether it is legal to distribute images to my clients Sep 10 21:16:38 emulator or device images Sep 10 21:16:45 once I've built them with AOSP Sep 10 21:17:19 sleepster: provide the source to linux kernel, I guess that's the only thing license obligates from you Sep 10 21:17:26 If you build it you can Sep 10 21:17:36 As long as you don't bundle gapps or stuff like that Sep 10 21:17:47 ah I see. so I modify the AOSP Sep 10 21:17:51 but only the android platform Sep 10 21:17:53 not the kernel Sep 10 21:18:04 I believe that is protected under Apache Sep 10 21:18:37 I am surprised Google doesn't provide this licensing information anywhere :) I looked all over developer.android Sep 10 21:19:07 I suspect the licensing information is provided on top of every copyrighted file. That's probably more than 1 % of AOSP altogether Sep 10 21:19:33 yeah that's true Sep 10 21:19:43 so just review every one of them and you're good Sep 10 21:19:44 And on http://s.android.com/source/licenses.html Sep 10 21:20:30 afaik to be part of android open source project, it is required that the file be licensed under the apache license used Sep 10 21:20:35 but the kernel is an exception Sep 10 21:20:54 then there's bunch of subprojects that are also included which carry their own licenses Sep 10 21:21:36 can't have font rendering without freetype, nor www without webkit, and so on and so on. Sep 10 21:22:23 surely going from 545 rows in the database to 550 shouldnt matter, hehe Sep 10 21:22:23 yep Sep 10 21:23:04 hello boys Sep 10 21:23:56 g00s: in the shipped database files journal_mode is set to delete Sep 10 21:23:59 is that wrong? Sep 10 21:24:07 no, that is good Sep 10 21:25:39 reading over the iphone 5 'distilled' from ars; seems pretty boring HW wise. faster processor, dual mode flash, etc. Sep 10 21:25:45 hmm..page count differs from the two versions Sep 10 21:26:02 page count is 48 on one and 49 on the other Sep 10 21:27:52 maybe thats the problem? Sep 10 21:28:22 I am almost certain we already diagnosed the problem. the database is open already, and you replace it by a combination of truncation + write Sep 10 21:28:35 this confuses the open sqlite handle. Sep 10 21:29:05 I have to close it outside of the helper? Sep 10 21:29:09 I know you tried to close it but I suspect it just doesn't close it or doesn't close all open handles to it Sep 10 21:29:47 I don't know what is required to use that helper correctl Sep 10 21:30:01 but why would it happen only for the latest change of db and not the previous versions? Sep 10 21:30:22 I don't know anything about sqlite implementation so I can't answer that question. Sep 10 21:31:02 a likely candidate is that metadata structures are located in different places on the two files Sep 10 21:31:21 its not Sep 10 21:31:57 I see size for the problematic version is 50176 while the others were 49152 and less Sep 10 21:32:25 Hi, in a listviev with buttons is better to implement onClick listener in parent activity or in the adapter? Sep 10 21:33:48 m8: I'd say the general rule is to implement the click listener closest to the actual action Sep 10 21:33:58 wonder if theres a weird size limit somewhere Sep 10 21:34:19 alankila, ?! Sep 10 21:34:27 that is, there's more context in the listview and its related objects than in the activity for figuring out what the click meant, therefore the listview is probably better place. Sep 10 21:34:33 nah, thats not it.. it fails from 50-56 as well Sep 10 21:35:20 alankila, i've to extend listview? Sep 10 21:35:29 m8: I usually put my onClickListeners in the holder Sep 10 21:35:46 nseidm1, the holder in the adapter? Sep 10 21:36:06 if the items have buttons, yea Sep 10 21:37:15 nseidm1, it's logic to put listener in the parent ListActivity, create an holder in adapter, tag the convertView with the adapter, in onClickButton in the activity get the parent and the holder for modify the content of the listitem? Sep 10 21:38:05 I just keep a position reference in the onclicklistener that I keep in the holder Sep 10 21:38:13 so theres a click listener for each item position Sep 10 21:38:19 then I update the listener with the position Sep 10 21:40:11 gotta sleep Sep 10 21:40:13 thanks for helping Sep 10 21:41:32 nseidm1, my flow is useless? Sep 10 21:43:55 if it doesn't work, yes Sep 10 21:44:01 it works Sep 10 21:44:10 then that's all that matters :-P Sep 10 21:44:20 but cleaner is better :O Sep 10 21:48:39 not an android question, but since many smart folk are here :) why use a web-based RSS aggregator service, when a mobile app could be written do directly fetch feeds? Just so the feeds be fetched over a browser from elsewhere if necessary? Sep 10 21:49:35 kbs would it be more of a multi-device / syncing thing ? Sep 10 21:50:39 g00s: true - but could the function of the service (such as it is) could be just to sync some opaque data (say, an encrypted OPML file) between clients? Sep 10 21:50:39 im trying to pay my android publisher fee but it says technical error try again later Sep 10 21:50:40 :( Sep 10 21:51:25 occ this is just the beginning ;) Sep 10 21:52:02 lol Sep 10 21:52:21 well if anyone here works for google can you please check up on this Sep 10 21:52:37 it is at play.google.com/apps/publish/signup Sep 10 21:52:41 lol, support ? :P Sep 10 21:52:43 <|0xD34D|> the beginning of an epic journey Sep 10 21:52:56 <|0xD34D|> make sure to pack for an extended journey Sep 10 21:53:04 oh yeh ill email their support. good idea.. if i can find it Sep 10 21:57:29 wow Sep 10 21:57:40 you fixed it. thanks Sep 10 22:17:57 do i need to setup merchant account to sell apps at all? or it is only for in app stuff like buying gold with real money in a game Sep 10 22:18:46 Does anyone know if, on device rotation, it is possible for any messages to be processed on the uithread while a Fragment with setRetainInstance(true) is detached? Sep 10 22:27:19 lol, I think I've only ever had one question answered here. I think it's mostly because my questions are about obscure details. Sep 10 22:29:39 messages? Sep 10 22:29:48 you mean will the handler run? Sep 10 22:30:17 then yes, an instance of that fragment is active, so any references and anything you send to it will be executed Sep 10 22:32:13 pfn: In particular, I'm wondering if a message posted to the UI thread can be handled while getActivity() will return null Sep 10 22:32:23 very specifically on rotation. Sep 10 22:32:25 posted from/to where? Sep 10 22:32:49 posted from a background thread to a Handler associated with the UI thread's Looper Sep 10 22:32:51 Fragment { Handler handler.post ... } then yes, it will continue to run if the fragment is detached while it is retained Sep 10 22:33:29 bleh. null-checks it is. Sep 10 22:33:33 "a Handler" still doesn't mean anything Sep 10 22:33:48 but if the fragment is retained, and something references that fragment, then yes, it runs Sep 10 22:33:54 It's a subclass of handler that overrides handleMessage Sep 10 22:34:16 how is it related to the fragment? Sep 10 22:34:40 Actually, let's pretend it's an AsyncTask Sep 10 22:35:07 since that's almost exactly what it acts like Sep 10 22:35:41 once started/queued the asynctask will always run unless explicitly canceled, cooperatively Sep 10 22:35:56 so in that case, I'm wondering whether onPostExecute can happen while a retained fragment (that is *not* recreated) does not have a valid activity reference. Sep 10 22:36:05 pfn: I know it will run Sep 10 22:36:12 it's a question of when Sep 10 22:36:13 why wouldn't it happen Sep 10 22:36:17 that's part of the contract Sep 10 22:36:30 part of what contract? Sep 10 22:36:37 the handler is live, there's nothing delaying its execution or delivery of the message Sep 10 22:36:47 the handler knows nothing about the fragment's lifecycle Sep 10 22:36:49 sure there is. Sep 10 22:36:50 why would messages get delayed Sep 10 22:37:03 it can't run if there's something running on the UI thread, for instance. Sep 10 22:37:20 messages are processed on the ui thread Sep 10 22:37:21 that's it Sep 10 22:37:32 nothing "runs" unless you're intentionally blocking the UI thread Sep 10 22:37:47 when you run stuff in the UI thread, you are effectively the current message in the queue getting popped off Sep 10 22:37:51 yes Sep 10 22:38:53 but your message will inevitably run, unless you're doing behaving poorly and blocking the ui thread Sep 10 22:38:59 so what you're saying is that control is returned to the UI thread's looper between onDetach and onAttach Sep 10 22:39:13 huh? Sep 10 22:39:19 the UI thread's looper is the only looper running Sep 10 22:39:28 on the ui thread Sep 10 22:39:34 yes. Sep 10 22:39:45 what does control have to do with this? Sep 10 22:39:58 I believe you are leaving out some vital context behind what exactly is occuring here Sep 10 22:40:02 seeing as I can't read minds Sep 10 22:40:07 * pfn wanders off Sep 10 22:40:27 the issue is not IF it will run Sep 10 22:40:28 you keep insisting that it will run Sep 10 22:40:43 it will run when its position in the queue is at the front Sep 10 22:40:51 just like every other message that gets posted to the message queue Sep 10 22:41:09 which is obvious Sep 10 22:41:09 (unless it's cancelled, of course) Sep 10 22:41:09 yes Sep 10 22:41:26 messages don't get reordered in the queue Sep 10 22:41:29 that's how loopers work Sep 10 22:41:35 I know how they work Sep 10 22:41:52 I need to know about the implementation of retained fragments Sep 10 22:41:55 I keep answering your questions, you ask when, I tell you when Sep 10 22:42:04 the implementation of retained fragments has nothing to do with handlers or message queues Sep 10 22:42:24 It runs on the UI thread, does it not? Sep 10 22:42:33 "it"? Sep 10 22:42:35 what is "it" Sep 10 22:42:43 the fragment retention code Sep 10 22:42:48 what retention code Sep 10 22:42:55 an instance does not "run" Sep 10 22:43:21 what is "the fragment retention code" Sep 10 22:43:22 the code that recreates the fragment and calls onDetach and onAttach Sep 10 22:44:06 the lifecycle methods are messages in the queue, yes Sep 10 22:44:19 right Sep 10 22:44:30 when making a merchant account to sell apps do i have to specify VAT id? i am mainly student not making much. so dont pay taxes yet Sep 10 22:44:49 so what I'm asking is whether onDetach and onAttach are called by 1 or 2+ of those messages Sep 10 22:45:17 why does it matter? Sep 10 22:45:28 if it's called by 1 message, then no other code runs on the UI thread between them Sep 10 22:45:45 if it's 2+, then code can run on the UI thread, and that code needs to check if the Fragment is attached. Sep 10 22:46:42 do i need to do this? To register for VAT, you need to apply to HM Revenue & Customs (HMRC). If HMRC approves your application, they will send you a VAT registration number and certificate. Sep 10 22:46:57 occ, if it says you need to, you need to Sep 10 22:47:00 if it doesn't, then you don't Sep 10 22:47:16 RyanM, what code are you posting that might come in between? Sep 10 22:47:38 and for the most part, if your fragment is getting detached, you have to assume something else is going on in the meantime Sep 10 22:47:39 pfn: code that updates the view based on a network response Sep 10 22:47:50 particularly if you're going through a rotation Sep 10 22:48:30 yeah, I only care about rotation. I have other means of stopping requests in other situations. But if the device is rotating, I want the onPostExecute to happen. Sep 10 22:48:50 there should be a buttload of messages flying through the queue during rotation Sep 10 22:49:02 alright, sounds like I should null-check the getActvity() return value Sep 10 22:54:21 hm, anyone have a strategy for using sqlite ATTACH wrt SQLiteOpenHelper? wondering if i should have one Helper for the main db; before any statement ATTACHEs to another db something invokes a separate child db SQLiteOpenHelper getReadableDatabase() and discards the connection so schema update code can fire for the to-be-attached db; that seems cheesy though Sep 10 22:54:51 yuck Sep 10 22:57:42 what do i put as merchant information: card statement name Sep 10 22:57:47 do i just make something up Sep 10 22:58:35 oh its for credit card statements Sep 10 22:59:12 ill put something really embarassing sounding then Sep 10 23:08:56 how can I get logcat from my emulated device and my actual device at same time? Sep 10 23:09:24 or in least switch between em on eclipse Sep 10 23:11:15 XXCoder: click the devices in the DDMS perspective in eclipse Sep 10 23:11:23 okay looking Sep 10 23:13:59 thanks! Sep 10 23:14:30 no problem Sep 10 23:14:36 yay the rescale works Sep 10 23:14:59 it uses lot less memory Sep 10 23:15:14 lemme try TWO pictures custom. couldnt do it before Sep 10 23:16:28 OOM error Sep 10 23:16:56 but definitely better Sep 10 23:17:17 2560x1920 to 1280x960 Sep 10 23:20:40 I have EditTextPreference with android:inputType="numberDecimal", how to limit to up 2 digits after .? Sep 10 23:21:22 for EditText it could be possible with TextWatcher, but for EditTextPreference? Sep 10 23:27:49 one strange thing Sep 10 23:28:07 I notice cm10.1 dont have a way to change live wallpaper setting without setting it again Sep 10 23:29:35 that's how it usually is Sep 10 23:29:45 This might seem obvious, but do you people know why there is no support for third party lockscreens? Is it only to avoid people getting locked out? I am asking because on my phone the pattern/PIN stuff comes up after the usual lock screen. Sep 10 23:30:02 strange. froyo I can just long click and select "change setting" Sep 10 23:30:39 hmm it does present a problem for my live wallpaper. theres no way to make it repaint since preview would reset and set reset flag to false Sep 10 23:31:08 since froyo opens dialog directly its no problem but in cm10.1... Sep 10 23:33:09 Hi, how i can force the actionbar to show the icons even if i have a hardware menu button? Sep 10 23:34:34 Justin_T, showAsAction="always" Sep 10 23:34:48 overflow items will never have the overflow icon with hardware menu buttons Sep 10 23:35:04 i have my items with showasaction="always" Sep 10 23:35:07 stupid action bar should always show the overflow icon regardless of menu button presence Sep 10 23:35:16 * XXCoder ponders how to reset paint since settings dont work same way as froyo... Sep 10 23:35:21 it should be clickable/unclickable in the presence of hardware buttons Sep 10 23:35:31 yes, like the gmail app for example Sep 10 23:35:41 or whatsapp ? Sep 10 23:35:46 ? Sep 10 23:36:09 * pfn doesn't have any phones with a hardware button to know Sep 10 23:36:14 I want to add a search button for my app Sep 10 23:39:19 my phone has HW buttons ! Sep 10 23:39:22 \o/ Sep 10 23:39:59 g00s, your phone isn't something to be proud of.... :p Sep 10 23:40:06 :( Sep 10 23:42:08 Whatsapp uses ActionBarSherlock for their Action Bar Sep 10 23:42:10 I see Sep 10 23:42:50 pfn: you want an unclickable overflow when there's a menu button? that's like the worst of two evils Sep 10 23:43:14 JakeWharton, yeah, but it brings visibility to what is otherwise missing Sep 10 23:43:25 because users have no indication that a menu action is available Sep 10 23:43:56 why not just show menu icon to show menu option as being there? Sep 10 23:44:03 then why not make it clickable? Sep 10 23:44:05 XXCoder, that's basically what I said Sep 10 23:44:25 JakeWharton, it can be made clickable, but since they want to encourage the menu button use... might as well leave it unclickable Sep 10 23:44:26 jake yeah. clickless menu icon meh Sep 10 23:45:16 but how do I know to use the menu button from an unclickable three dots on the screen? Sep 10 23:45:21 looks like I have to make picture reset with settings change. bah. Sep 10 23:45:35 JakeWharton, it's simply a visual indicator that the menu button has available actions Sep 10 23:45:51 while still retaining the visual style of devices with non-physical buttons Sep 10 23:46:15 i understand what it is, but it conveys nothing by being a useless button on screen Sep 10 23:46:18 JakeWharton, especially since google has taken that design ethic more than just on android itself Sep 10 23:46:22 it's not a useless button Sep 10 23:46:25 it's a visual indicator Sep 10 23:46:33 that's like saying the battery status icon is useless because it's unclickable Sep 10 23:46:34 with no indication how to use it Sep 10 23:46:43 it's not useless. it conveys infromation Sep 10 23:46:56 the overflow icon conveys information Sep 10 23:46:59 "there is a menu action here" Sep 10 23:47:11 but how do I know to press the menu button for it? Sep 10 23:47:13 how you arrive at the menu is device dependent Sep 10 23:47:16 they are 4+" apart Sep 10 23:47:20 don't look alike Sep 10 23:47:30 actually, if you squint, they look somewhat similar Sep 10 23:47:31 :p Sep 10 23:47:35 at least on old android devices Sep 10 23:48:22 JakeWharton, particularly, there is no better solution Sep 10 23:48:33 make it clickable! Sep 10 23:48:37 fuck the menu button Sep 10 23:48:49 at least that one has logic Sep 10 23:48:50 I suppose you could make the icon clickable, but then, tell all the device mfrs to stop making useless hardware menu buttons Sep 10 23:49:06 since I don't want that redundancy Sep 10 23:49:21 is there a way to set the text label on a switch to appear to the right of the switch instead of to the left? Sep 10 23:49:21 why? my epic I have 3 ways to hit menu button Sep 10 23:49:22 I suppose a hacked up keyboard layout could be uploaded to device to remap it Sep 10 23:49:33 don't buy a phone with one, you'll never see it Sep 10 23:49:34 on screen, physical capactor button on screen, and keyboard button Sep 10 23:49:47 switches don't have labels... do they? Sep 10 23:49:50 wait what? why the hell would you make a button present, but unclickable? Sep 10 23:49:55 it's not a button Sep 10 23:49:59 it's a visual indicator Sep 10 23:50:02 pfn: I mean the text attribute Sep 10 23:50:03 that conveys fucking information Sep 10 23:50:05 :p Sep 10 23:50:15 to visually indicate that some other button that shouldn't even exist can be clicked Sep 10 23:50:25 I can do it with a textview but.... Sep 10 23:50:29 what does it convey? Sep 10 23:50:29 i think that is just about the only way to make the menu button situation worse than it is now Sep 10 23:50:49 99% of users never know they can press the menu button on their phones Sep 10 23:51:00 it's as useless as that stupid search button was Sep 10 23:51:06 samsung won't get rid of the menu button Sep 10 23:51:17 samsung thinks it's their branding or something Sep 10 23:51:19 pfn: it was slightly less useless Sep 10 23:51:20 heh I have 2 of that. :P capactor button and keyboard button Sep 10 23:51:23 fuck samsung Sep 10 23:51:34 Prime326, in my 2 years of droid1 ownership, I never pressed the search button except for andchat Sep 10 23:51:35 but yes, they're pretty married to it and they're selling a lot of devices Sep 10 23:51:46 pfn: not even by accident ? Sep 10 23:51:51 I pressed search button 3 times. all by mistake Sep 10 23:51:54 maybe by accident and thinking wtf happened Sep 10 23:51:59 but otherwise, completely useless Sep 10 23:52:02 well that's why it's not longer there Sep 10 23:52:11 pfn: most apps didn't even use it at all Sep 10 23:52:13 well, the menu button has the same sitation Sep 10 23:52:15 i used to use long-search for voice search to say "Navigate to whole foods" Sep 10 23:52:32 pfn: at least the menu button sort of made sense Sep 10 23:52:38 but then i became a big boy and learned how to get there by myself Sep 10 23:52:44 the problem is people don't know when they can press the menu button Sep 10 23:52:48 plenty of apps used it for options and stuff Sep 10 23:52:51 google went and addressed exactly this situation Sep 10 23:53:01 except to go and punish device manufacturers that continue to put in a menu button Sep 10 23:53:04 yeah they got rid of it Sep 10 23:53:11 their users are still completely befuddled as to when menus are available Sep 10 23:53:16 pfn: so disable its backlight when there's no menu items or something. not putting a useless action item you can't click Sep 10 23:53:30 it's not an action item Sep 10 23:53:32 it's a visual indicator Sep 10 23:53:46 and disabling the backlight is completely non-obvious Sep 10 23:53:48 Samsung's menu button does serve a useful purpose. It disables the even more irritating software menu button that takes up the entire nav area. Sep 10 23:53:54 when the lights on these devices come on and off with no rhyme or reason Sep 10 23:54:14 i don't know why i'm arguing this awful idea. it will never happen, and the amount of effort i've put into this conversation is more than anyone should every thing about this shit ever Sep 10 23:54:38 I prefer software buttons, but having the one software menu button is awful. Sep 10 23:54:42 but anyways does anyone know how to set the text that appears with a switch widget to appear to the right of the switch? Sep 10 23:54:54 JakeWharton: did https://github.com/square/picasso/issues/110 ever make it into 2.0 ? Sep 10 23:54:56 I just find it sad that google purposefully makes the menu button on their number one device oem hard to use Sep 10 23:55:17 pfn: which device is that? Sep 10 23:55:17 pfn: they do? Sep 10 23:55:25 Prime326, sgs3, sgs4, etc. Sep 10 23:55:42 pfn: ah I don't have those so I wouldn't know Sep 10 23:56:01 google should do more to make the menu button useless so that samsung finally ditches it and android can grow up more Sep 10 23:56:37 kevinb: samsung is not going to ditch their distictions. samsung needs oddities like touchwiz to hide android so that if they have to they can switch to tizen Sep 10 23:56:45 even on my own apps, when I use an sgs3/4 I wonder when the hell I can press the menu button Sep 10 23:56:48 done Sep 10 23:57:14 Im using sherlock Sep 10 23:57:16 I usually resort to putting a textview at the bottom of the screen that says, "press menu for more options" Sep 10 23:57:20 do all most currrent android devices with a camera use V4L2 underneath the Camera API? is this accessible via the NDK? Sep 10 23:57:33 i wonder if we'd feel differently if google had provided better guidelines on how it should've been used, and even enforced them maybe. But then they'd be apple Sep 10 23:58:44 i hope google locks things down more with 5.0 Sep 10 23:59:24 terminal IDE is good app? Sep 10 23:59:34 sure, why not Sep 10 23:59:36 funkbox: no. not yet Sep 11 00:00:48 there isn't backdoor in the app? or logging of other activities on the device? Sep 11 00:01:35 Prime326: there's CTS but it's not really good Sep 11 00:03:11 Anyone who knows how to format an S2 that's been encrypted? Can't access clockworkmod or recovery. Sep 11 00:03:48 ashleymoore: #android-root Sep 11 00:04:04 alright, thanks Sep 11 00:05:10 hm, well i was going to say minSdk=16 or something for an existing app, but i still have quite a few GB users :( so i might use appcompat. but reading through that, I don't seem much discussion on how it is different than the full impl provided by the framework. Sep 11 00:05:20 g00s: how many? Sep 11 00:05:34 funkbox 1/3 about Sep 11 00:05:42 appcompat-v7 works fine Sep 11 00:05:43 a little less Sep 11 00:05:43 we're cutting support in the next major release for <14 with about 15% GB Sep 11 00:05:50 the only thing appcompat-v7 is missing is a Theme.AppCompat.Dialog Sep 11 00:06:28 but the 2.3 menu looks so much better on appcompat vs. abs Sep 11 00:07:03 hm, ok. it has the CAB then ? Sep 11 00:07:18 yes Sep 11 00:07:19 i was hoping to forget about all of this :| Sep 11 00:07:26 ok, thanks Sep 11 00:07:32 I told you that a long time ago Sep 11 00:07:40 you can see that it's there in the apidoc as well Sep 11 00:08:23 ok, so my dialogs might look a little weird then without Theme.Appcompat.Dialog. Sep 11 00:08:47 ok Sep 11 00:08:47 you have many dialog activities? Sep 11 00:08:56 oh, i have no dialog activities Sep 11 00:09:01 regular dialogs Sep 11 00:09:11 dunno what regular dialogs use or if they look right Sep 11 00:09:13 * pfn shrugs Sep 11 00:09:49 i think just AlertDialog Sep 11 00:09:58 they probably look fine Sep 11 00:10:11 it's 2.3, who cares if it looks like ass for those users Sep 11 00:12:12 hi Sep 11 00:12:13 help Sep 11 00:12:23 tried flashing cyanogen-mod, failed... Sep 11 00:12:30 it got stuck on boot-load Sep 11 00:12:47 and now when I try to restore the backup I made.... Sep 11 00:12:59 xenaxon: #android-root Sep 11 00:12:59 it gives me error: "Error: cannot mount /system" Sep 11 00:13:28 funkbox: any idea? Sep 11 00:13:33 try #android or #cyanogenmod this server Sep 11 00:13:54 we need a bot Sep 11 00:13:57 xenaxon: read the topic Sep 11 00:13:58 i need help adding a custom layout to actionbarsherlock. - the custom layout view does appear however, there is no highlight when selected and no hint text when long pressed. Any ideas on how I can fix this. Sep 11 00:14:08 haha Sep 11 00:14:12 huh, i missed this one http://developer.android.com/reference/android/support/v4/content/FileProvider.html Sep 11 00:14:32 okay I'm getting this weird effect when painting 1% transparacy texture over and over. it does not build up to 100% of that texture as I expected Sep 11 00:15:11 legov: what topic? Sep 11 00:15:19 can I make it "round up" so it can actually get there? Sep 11 00:15:47 less than 20 transpacy has strange issues whule higher than that it gets to 100% proper texture Sep 11 00:16:42 xenaxon: /topic #android-dev Sep 11 00:17:13 g00s, that's pretty neat Sep 11 00:17:28 too bad there isn't an arbitrary Streams provider that could be used Sep 11 00:18:34 pfn: share a fifo? :) Sep 11 00:18:38 (no idea if that would actually work) Sep 11 00:20:48 sometimes I think I want to send input streams to a MediaPlayer and stuff, and an arbitrary Stream content provider would make it nice to just write data from the output of stuff like libavcodec Sep 11 00:21:21 is it possible to make the height of a custom actionbar layout view use the full height of the actionbar? As of now I have a slight padding above and below the view. Sep 11 00:21:38 Is there some magical way to get rid of the announcement about play store policy changes? Sep 11 00:21:50 'Cause I've clicked that X about 50 times and it's still showing up every time. Sep 11 00:22:04 At this point, I am actively questioning my sanity. Sep 11 00:23:09 have to click the X 65536 times Sep 11 00:23:28 that way they know you're really done with it Sep 11 00:24:19 i wonder if its possible to use NIO Pipe class for IPC Sep 11 00:25:31 wtf Sep 11 00:25:42 stackoverflow does not have perment logoff Sep 11 00:25:53 I just log off then click login and it auto logins Sep 11 00:26:21 XXCoder: that's kinda the point of OAuth... Sep 11 00:26:30 I do not want it to be automated Sep 11 00:26:46 then get an OAuth provider that doesn't do that? Sep 11 00:27:02 I didnt install anything Sep 11 00:27:09 just on stack website Sep 11 00:27:51 and when you click login, how do you login? Sep 11 00:27:57 well it logins Sep 11 00:28:07 not even showing username and password Sep 11 00:29:14 How did you sign up? Sep 11 00:29:34 used openid registered Sep 11 00:30:14 http://meta.stackoverflow.com/a/190757 Sep 11 00:30:45 Gah, even clicking the X in chrome instead of safari won't get rid of it Sep 11 00:30:57 * Nilium scratches his head Sep 11 00:31:55 I am concerned because theres no way to log off and prevent anyone else from logging in as me Sep 11 00:32:01 very bad security Sep 11 00:32:11 XXCoder: did you see my link? Sep 11 00:32:15 yes Sep 11 00:32:41 re-reading to be sure I didnt miss something Sep 11 00:33:04 well its yahoo then stackoverflow Sep 11 00:33:08 I dont use yahoo Sep 11 00:33:28 I logged off directly at stackoverflow and when I try login it dont ask me anything Sep 11 00:33:47 ah "Correct, if I remove my cookies of openid.stackexchange.com then I have to enter my password, otherwise I don't have to. Thanks for your answer! " Sep 11 00:36:08 finally saved password so I can access it anywhere Sep 11 00:49:56 but...even if you delete the cookie someone can still log in as you... Sep 11 00:50:08 nope, I tried. I cant Sep 11 00:50:31 you need to log out of yahoo or whatever too. Sep 11 00:50:31 alright Sep 11 00:50:54 How can I grab a uri to a video thumbnail from mediastore? Sep 11 00:51:00 cookie does get regenerated at login unfortunately so I cant login at stackexchange anywhere but my pc Sep 11 00:51:08 I hate enforced remember login Sep 11 00:51:25 im a bit confused as to what i have to pass into MediaStore.Video.Thumbnails.getContentUri() Sep 11 00:51:51 MediaStore.Video.Thumbnails.getContentUri(String volumeName) what exactly is the volume name?... Sep 11 00:53:43 is this gibberish or? http://stackoverflow.com/questions/18731035/strange-effect-with-transparencies-on-android Sep 11 00:54:27 Pyuruku, I would guess /sdcard ? Sep 11 00:54:59 alex_PP: how would i specify the video im trying to get the thumbnail uri for?... Sep 11 00:56:08 that method gets the tumbnail table Sep 11 00:57:27 you would open a cursor to the uri it gives you, then look for the thumbnial you wnat in there Sep 11 00:57:43 and use that data to load the thumbnail with getThumbnail Sep 11 00:59:36 XXCoder, pictures? Sep 11 01:00:54 Inbox (24) Sep 11 01:00:56 gargh Sep 11 01:06:46 My MainActivity that extends AndroidApplication and implements IActivityRequestHandler passes itself as an advert requestHandler (also a requestHandler for in app payments). Why would my requestHandler be null the second time my app opens just after closing it? Sep 11 01:07:45 How do I add string files to my build.xml file so that whenever it changes that ant will build in the string changes? Sep 11 01:12:49 huh Sep 11 01:14:02 alex_PP: sorry was having dinner Sep 11 01:14:07 a second Sep 11 01:15:15 https://dl.dropboxusercontent.com/u/53432577/device-2013-09-10-181442.png Sep 11 01:21:42 any idea alex_PP? Sep 11 01:23:35 XXCoder, it's not enforced, log out of your openid provider and it won't be remembered Sep 11 01:23:56 how? s far I know only log out is at stackoverflow Sep 11 01:24:07 how do you log into stackoverflow? Sep 11 01:24:09 using gmail? Sep 11 01:24:42 directly using openexchange Sep 11 01:24:52 then log out of openxchange Sep 11 01:25:05 ok lemme look a sec Sep 11 01:28:07 pfn when I look there Sep 11 01:28:13 I am not logged in Sep 11 01:28:24 I'm still logged in at stack overflow Sep 11 01:29:26 login then logout then Sep 11 01:31:54 no change Sep 11 01:32:14 * pfn shrugs Sep 11 01:32:20 I guess stackoverflow is busted Sep 11 01:32:26 guess so Sep 11 01:42:24 hope someone can figure how to do "ceiling" on transparacy paint lo Sep 11 01:53:24 Anyone else working on an OpenGL ES 3 application? Sep 11 01:56:39 im trying to copy bootanimation to my galaxy s3 but it says permission denied and my device is rooted Sep 11 01:57:54 XXCoder, sorry was having a discussion Sep 11 01:57:59 np man Sep 11 01:58:12 pic is https://dl.dropboxusercontent.com/u/53432577/device-2013-09-10-181442.png so dont need srollup Sep 11 01:58:12 not sure what's going on really Sep 11 01:58:29 my guess is its rounding down to 0 if changes less that certain point Sep 11 01:58:54 which bit os the artefact you're looking at Sep 11 01:59:02 all Sep 11 01:59:12 second image is grass Sep 11 01:59:22 should be normal green blades and stuff Sep 11 01:59:26 first image is space Sep 11 01:59:37 its painting transparency level 1 over and over Sep 11 01:59:45 but its not fading in grass all way in Sep 11 02:01:00 easy for you to set test Sep 11 02:01:23 paint any picture on black at transparency level 1 repeatly Sep 11 02:01:25 i'm not really sure what you're trying to do Sep 11 02:01:30 cross fade two bitmaps? Sep 11 02:01:34 oh its wallpaper effect Sep 11 02:01:49 its not working properly at lower transparency. 20+ works fine Sep 11 02:04:56 hmmm Sep 11 02:04:59 sorry, no idea Sep 11 02:05:04 no problem. Sep 11 02:05:14 should not be too hard to reproduce Sep 11 02:05:33 just paint paintshader at transparent level 1 of one texture to other Sep 11 02:06:22 repeatly till it dont change anymore. result should be that second one is same as first but it does not Sep 11 02:06:55 all other live wallpaper stuff is external to this basic issue :) Sep 11 02:07:23 you're drawing the same bitmap at alpha=1 over and over again without clearing? Sep 11 02:07:52 well yeah livewallpaper its not all over at once, however. Sep 11 02:08:00 then what you're seeing is expected Sep 11 02:08:18 there's just not enough precision in RGB888 to do this Sep 11 02:08:36 ahh hmm is there to "round up" so minium change is 1 not 0? Sep 11 02:09:40 if you set alpha to 0 then nothing is drawn Sep 11 02:09:52 correct Sep 11 02:10:01 what I mean is color pixel wise Sep 11 02:10:26 given that blending math is done with integers and tricky shifts I'm not surprised this happens Sep 11 02:10:30 just don't do this :) Sep 11 02:10:48 well it works fine above 20. maybe I can set that as minium Sep 11 02:10:58 but then I like stuble fade stuff Sep 11 02:11:06 a subtle fade is fine Sep 11 02:11:14 just don't draw over and over with low alpha values Sep 11 02:11:32 yeah. unfortunately my livewallpaper requires that Sep 11 02:11:46 I'm sure you can find another way Sep 11 02:11:51 basically its 2 images, stuff moves around and pushes one way or other Sep 11 02:11:59 effect depends on alpha draw Sep 11 02:13:33 XXCoder: so you can't just draw the two images with a progressively increasing alpha on the second one? Sep 11 02:13:47 they dont draw to each other but to third image Sep 11 02:14:19 what? Sep 11 02:14:38 its bit diffult to explain Sep 11 02:19:13 dunno maybe I have to go back to cpu expensive per pixel Sep 11 02:41:48 I'm trying to test an app on my droid mini, but Eclipse doesn't seem to be finding the phone Sep 11 02:42:08 Im on windows 7, I believe I have the drivers installed correctly Sep 11 02:46:48 residentbiscuit: does adb devices report it? Sep 11 02:46:58 if not, make sure your phone is in USB debugging mode Sep 11 02:50:01 It is in usb debugging mode. Not sure how to check adb Sep 11 02:55:22 adb devices shows no devices **** ENDING LOGGING AT Wed Sep 11 02:59:59 2013