My Code is Fragmented - Android Development Megathread

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > My Code is Fragmented - Android Development Megathread

«‹›71 »

Thermopyle: Jul 1, 2003; ...the stupid are cocksure while the intelligent are full of doubt. �Bertrand Russell

Volguus posted:

Well, i don't know. The users would use our app to create a thing that then it would be saved into their account online (our web application). That thing needs a name before it can be saved, so the save button pops up a dialog asking for the name and off it goes. Now, one could say that this operation can be called Upload and they'd be kinda right, but to be fair I'm not so sure that Upload is better than Save. At the end of the day is about semantics. The app (the entire solution in general) is targeted at developers not average joe six-pack, so they should most likely know what is going on.

If its any help, a common paradigm is that its just automatically saved with some sort of name like "Untitled".

Google Docs for example.

# ? Dec 23, 2017 00:05

Adbot: ADBOT LOVES YOU

# ? May 14, 2024 00:30

baka kaba: Jul 19, 2003; PLEASE ASK ME, THE SELF-PROFESSED NO #1 PAUL CATTERMOLE FAN IN THE SOMETHING AWFUL S-CLUB 7 MEGATHREAD, TO NAME A SINGLE SONG BY HIS EXCELLENT NU-METAL SIDE PROJECT, SKUA, AND IF I CAN'T PLEASE TELL ME TO
EAT SHIT

Volguus posted:

Right, so an icon for the Save (and not text) is what should be done then when it is on a main menu. Though I agree with the theory that full screen dialogs should get dialog buttons (with text), in practice, even in their page, it simply looks out of place to have one thing in one place (icons, sure main menu but the user doesn't know or care what's main menu or secondary one) and text in another. I know I saw it used in google android apps and it always looked out of place and inconsistent with the rest of the app. But i guess once people get used to it it'll become normal just like the floppy icon did even though it was obsolete when hard-drives appeared and one no longer saved to the floppy.

Well the point is that a full-screen dialog represents an action you're taking - you can select some options or enter some data, but in the end you're going to be making a choice - CONFIRMing or CANCELing that action. The full-screen version moves those buttons up into the title bar, instead of below the dialog contents on the floaty ones, and the CONFIRM button stays as the usual text that states what action you're taking

So if your full-screen dialog is about saving (with various options to set before kicking it off) it would make sense to have that confirm button say SAVE, because that's explicitly what the user is choosing to do. The cancel button cancels the action, backs out of the process. The user has to choose one of those two options

But icons just represent possible, optional actions the user can take in the current context. You don't have to press any of them, and if you do, they might silently perform some action (like saving to a standard location) or they might pop up a dialog with options, where the user is required to decide whether to go ahead or not. The icon is just a thing on the side that you can press, if you like, there could be lots of them and they don't necessarily end what you're doing

So really they do two different things - the full screen dialog really is meant to act like an alert dialog, with its text buttons with explicit actions you have to choose between. It's just laid out like an activity with an action bar because it's full-screen. You might not agree with it or follow it (hell I'm noticing how the Awful post editor should probably follow that pattern...) but there is a logic and consistency to it

baka kaba fucked around with this message at 02:46 on Dec 23, 2017

# ? Dec 23, 2017 02:44

NoDamage: Dec 2, 2000

I'm building a new version of an Android app I originally wrote back in 2012 and just wanted to check if things have changed a lot since then. (I've been mostly in iOS land lately.) In particular, has anything changed with how modern Android apps load data from a local database, display it on screen, and keep it synced if further changes are made to the database?

My original implementation used a ContentProvider as the wrapper around a SQLite database and then used a CursorLoader/CursorAdapter to fetch the data and populate a ListView. All inserts/updates/deletes went through the ContentProvider API which would call ContentResolver.notifyChange() to ensure the ListView got updated when something changed. But now I'm reading that you shouldn't really bother using ContentProviders if you're not trying to expose data to other apps, just use SQLite directly instead. Okay, so if you fetch a bunch of data from SQLite and shove it in a list view. Lots of tutorials and documentation on how to do that.

But then you have a background service running that syncs that data with a server. New rows get added to the database by the background service. How do you (efficiently) tell your ListView to add the corresponding rows?

# ? Jan 1, 2018 05:51

Taffer: Oct 15, 2010

Fundamentally it all works the same, but with more refined tools. If you're using SQL, just go ahead and use Room, the new wrapper library from Google. If you're feeling fancy you can use all the new architecture components, they'll help deal with the awful issues of configuration changes and - relevant to your case - notifying when a data source updates. Use a RecyclerView instead of a ListView (RV is just a way better version of it), create an adapter and tie it in with your Room data and notify it when the data changes.

That's vague info but it's hard to be more detailed with such a broad question.

# ? Jan 1, 2018 06:19

Telephones: Apr 28, 2013

I am using OpenCv with Android to process camera frames...

code:

@Override
public Mat onCameraFrame(CameraBridgeViewBase.CvCameraViewFrame inputFrame) 
{
    Mat frame = inputFrame.rgba();
    mCameraCalculator.processFrame(frame);
    return frame;
}

which sets the frame to a new cropped Mat...

code:

public void processFrame(Mat frame) 
{
    //fill frame outside of square center region
    Mat mask = new Mat(frame.size(), frame.type(), Scalar.all(255));  //create mask
    mask.submat(getCenterSquare(mask)).setTo(Scalar.all(0));        //init mask
    frame.setTo(Scalar.all(0), mask);   //fill

    mask.release()    //this does not solve the problem
}

This causes a memory leak, eventually crashing the application. Additionally, using Android Studio's Android Profiler, I see that the unreleased memory is native. What is going wrong?

# ? Jan 4, 2018 22:59

FAT32 SHAMER: Aug 16, 2012

So we have been building my app to work on the Pixel C and based on my emulators for various sized screens, it looks how I expect it to on vanilla android. Turns out that google has discontinued the Pixel C, so we decided to verify that it works on a Samsung Tab 3, but for some reason everything is too large and takes up more screen than it should. Is there some trick to getting apps to look correct on Samsung? Did I gently caress up by using hardcoded dp and sp? I thought the entire point of using those was to enable the layout to work on all screen sizes?

# ? Jan 24, 2018 16:38

baka kaba: Jul 19, 2003; PLEASE ASK ME, THE SELF-PROFESSED NO #1 PAUL CATTERMOLE FAN IN THE SOMETHING AWFUL S-CLUB 7 MEGATHREAD, TO NAME A SINGLE SONG BY HIS EXCELLENT NU-METAL SIDE PROJECT, SKUA, AND IF I CAN'T PLEASE TELL ME TO
EAT SHIT

It looks like you hosed up by samsunging yourself

These people are talking about an older model but it looks like Samsung was intentionally reporting the wrong density to get bigger icons. Wouldn't be surprised if they're doing it on your model too

You've done right using dp and sp - they're not always perfect, devices go into the closest density category so you get some variation, but they should be pretty consistent. Can't help if computers lie though

# ? Jan 24, 2018 17:00

FAT32 SHAMER: Aug 16, 2012

baka kaba posted:

It looks like you hosed up by samsunging yourself

These people are talking about an older model but it looks like Samsung was intentionally reporting the wrong density to get bigger icons. Wouldn't be surprised if they're doing it on your model too

You've done right using dp and sp - they're not always perfect, devices go into the closest density category so you get some variation, but they should be pretty consistent. Can't help if computers lie though

god drat that sucks

what's a non-lovely tablet that doesnt cost mongobucks?

im beginning to think that if we had just done what I wanted to do and did this app for iOS we would have finished this a long time ago lol

# ? Jan 24, 2018 17:06

Thermopyle: Jul 1, 2003; ...the stupid are cocksure while the intelligent are full of doubt. �Bertrand Russell

FAT32 SHAMER posted:

god drat that sucks

what's a non-lovely tablet that doesnt cost mongobucks?

im beginning to think that if we had just done what I wanted to do and did this app for iOS we would have finished this a long time ago lol

Google doesn't care about Android tablets, so there's not exactly a definitive non-lovely tablet...and what you want is less likely to be a definitive non-lovely tablet and more like a tablet that best represents what your potential users will have. Do you have any idea what your target market is using?

And yes, you probably would've finished an ios version a long time ago!

# ? Jan 24, 2018 17:13

FAT32 SHAMER: Aug 16, 2012

Thermopyle posted:

Google doesn't care about Android tablets, so there's not exactly a definitive non-lovely tablet...and what you want is less likely to be a definitive non-lovely tablet and more like a tablet that best represents what your potential users will have. Do you have any idea what your target market is using?

And yes, you probably would've finished an ios version a long time ago!

It's an internal app so we get to select what tablet to give the consultants that will be using it, so anything goes really

# ? Jan 24, 2018 17:17

Taffer: Oct 15, 2010

Chromebooks? I know it's not strictly a tablet but it fits into the category you're describing. They're cheap too.

# ? Jan 24, 2018 19:03

FAT32 SHAMER: Aug 16, 2012

Taffer posted:

Chromebooks? I know it's not strictly a tablet but it fits into the category you're describing. They're cheap too.

I'm currently working on convincing my boss that the Pixelbook would be the best option since a lot of the consultants are guaranteed going to whine about not having a keyboard, but the bossman feels they dont meet the usecase and also doesnt want to spend $999/consultant

just in case, does anyone have any Pro Tips for fragmenting your app specifically to work around TouchWiz?

# ? Jan 24, 2018 19:50

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

You can always use the Build class to check if it's Samsung or a very specific Samsung model, if you hate your life.

~~Pixel Tablets maybe?~~ edit: lol, discontinued I guess. I'd still like to know about a nice Nexus 7 replacement.

# ? Jan 24, 2018 20:36

LLSix: Jan 20, 2010; The real power behind countless overlords

FAT32 SHAMER posted:

I'm currently working on convincing my boss that the Pixelbook would be the best option since a lot of the consultants are guaranteed going to whine about not having a keyboard, but the bossman feels they dont meet the usecase and also doesnt want to spend $999/consultant

just in case, does anyone have any Pro Tips for fragmenting your app specifically to work around TouchWiz?

Jailbroken Kindle/KindleFires are <$100 each.

# ? Jan 24, 2018 21:50

baka kaba: Jul 19, 2003; PLEASE ASK ME, THE SELF-PROFESSED NO #1 PAUL CATTERMOLE FAN IN THE SOMETHING AWFUL S-CLUB 7 MEGATHREAD, TO NAME A SINGLE SONG BY HIS EXCELLENT NU-METAL SIDE PROJECT, SKUA, AND IF I CAN'T PLEASE TELL ME TO
EAT SHIT

Is there a way to fix this annoying Intent behaviour?

external intent launches activity, say a URL handler
the activity loads, handles the intent ("this URL says show thing A")
the user navigates elsewhere, showing thing B
the activity is put in the background, and gets destroyed, saving the current state (looking at thing B)
the user opens the backgrounded app, which recreates the activity as it was initially created including that URL intent
because the new activity has this intent, it handles it in the same way and shows thing A again

I basically want to consume that initial intent, so I don't handle it again. I also want to be able to handle new intents that look like the old ones (e.g. if the backgrounded app is opened from an external URL intent again)

I can't work out anything that works. If you call setIntent that only sticks around until the activity is destroyed - it still gets recreated with the original intent. You can't do getIntent and add anything like a flag that says "handled this" as an Extra, it doesn't show up in the recreated activity. Can't even relaunch the current activity with startActivity and a 'neutral' intent, because it's not actually creating a new activity so it doesn't stick as the 'original' intent when the activity's recreated

Is there a way round it? It's a huge pain so there must be some technique... right?

# ? Jan 26, 2018 23:11

Tunga: May 7, 2004; Grimey Drawer

Have you tried messing with the intent flags to get it to reuse the same instance of the Activity? I'm not sure if that will help with this or not but it's worth a go.

Otherwise I think you'd have to persist the screen state to local storage (with Room or whatever you want).

# ? Jan 27, 2018 01:31

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

Why exactly does your activity launching need to be stateful? Could you note that you Did The Thing in onSaveInstanceState?

# ? Jan 27, 2018 01:43

baka kaba: Jul 19, 2003; PLEASE ASK ME, THE SELF-PROFESSED NO #1 PAUL CATTERMOLE FAN IN THE SOMETHING AWFUL S-CLUB 7 MEGATHREAD, TO NAME A SINGLE SONG BY HIS EXCELLENT NU-METAL SIDE PROJECT, SKUA, AND IF I CAN'T PLEASE TELL ME TO
EAT SHIT

I had thought of the savedState thing, my problem was if the 'sticky' intent that's handled and marked as ignored was replaced with a new identical one, a fresh event that would get ignored

But that got me thinking, I experimented a bit and worked out that the intent 1 that first creates the activity (using FLAG_ACTIVITY_CLEAR_TOP here) sticks with it, forever, even when it gets restored. If it's backgrounded and destroyed, and you send another intent 2 (say from a share handler) that brings up the activity again, it gets recreated with Intent 1 again, back to that original state, and then Intent 2 comes in through onNewIntent

So it looks like I fixed it by only handling that baked-in intent in onCreate if there's no saved state, meaning it's a fresh start and that intent hasn't been handled. Otherwise you can just ignore it, and any new intent that restored the activity will just come in through onNewIntent as usual

Sorry if I've just worked out some obvious stuff here, I haven't run into this kind of thing before. I guess I understand the beautiful activity lifecycle a little more now

# ? Jan 27, 2018 03:35

SaTaMaS: Apr 18, 2003

I have an app built on iOS using Scenekit which uses inverse kinematics. I'm thinking about porting it to Android, but I'd like to avoid using Unity if possible. Does Android have an IK library outside of Unity?

# ? Feb 4, 2018 03:41

Taffer: Oct 15, 2010

That's a really vague question. Do you need a 3d engine? 2d? What kind of rendering capabilities do you need to go with it?

Short answer, you'll probably have to use unity. You can try to find a small lightweight engine but those tend to be painful.

# ? Feb 4, 2018 04:02

SaTaMaS: Apr 18, 2003

Taffer posted:

That's a really vague question. Do you need a 3d engine? 2d? What kind of rendering capabilities do you need to go with it?

Short answer, you'll probably have to use unity. You can try to find a small lightweight engine but those tend to be painful.

It's in 3D. I'm using OpenGL, and the capabilities are pretty basic

# ? Feb 4, 2018 04:26

reversefungi: Nov 27, 2003; Master of the high hat!

Are there any good Android tutorials that aren't video-based? Seems like Google's Udacity course is the most highly recommended source out there, but it's mainly videos which is a little frustrating to get through and limits the times I'm able to pick up the tutorial. Some sort of free resource would be most kick-rear end, but book recommendations are good too.

# ? Feb 6, 2018 05:46

FAT32 SHAMER: Aug 16, 2012

The Dark Wind posted:

Are there any good Android tutorials that aren't video-based? Seems like Google's Udacity course is the most highly recommended source out there, but it's mainly videos which is a little frustrating to get through and limits the times I'm able to pick up the tutorial. Some sort of free resource would be most kick-rear end, but book recommendations are good too.

I found Big Nerd Ranch's Android book to be very good and using it I was able to build a great final project in my senior year :shrug:

# ? Feb 6, 2018 08:31

baka kaba: Jul 19, 2003; PLEASE ASK ME, THE SELF-PROFESSED NO #1 PAUL CATTERMOLE FAN IN THE SOMETHING AWFUL S-CLUB 7 MEGATHREAD, TO NAME A SINGLE SONG BY HIS EXCELLENT NU-METAL SIDE PROJECT, SKUA, AND IF I CAN'T PLEASE TELL ME TO
EAT SHIT

I dunno about the best, but

Vogella has a good set covering the fundamentals - sometimes they can feel a little formulaic and light on some detail, but that might just be me looking for specific info
Codepath has a getting started guide / list of resources which I'm not familiar with, but I really like that site - whenever I want to know about a thing, their page on it is always really good and clear, better than the official docs most of the time
Mostly more advanced, but Google's codelabs are a good resource for trying out stuff - they're lessons on different concepts with projects you can import and follow along. There are a couple of 'my first app' ones though
The official docs have some intro tutorials and are pretty good for covering a lot of the framework, but they're also a bit all over the place - you can get the same concept explained on different pages, sometimes conflicting slightly. And some of it is really old, like the Jelly Bean UI is on some of the screenshots. I don't know why they don't put some people on it to keep it up to date each year but hey

I'm guessing you already know Java? If not the official tutorials are good. If you want to learn Kotlin (recently added as a supported language and getting pushed heavily) then they have a nice set of exercises you can code and test in the browser

# ? Feb 6, 2018 09:47

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

Be careful with Vogella, I found some blatantly false info on there a couple times.

# ? Feb 6, 2018 15:28

baka kaba: Jul 19, 2003; PLEASE ASK ME, THE SELF-PROFESSED NO #1 PAUL CATTERMOLE FAN IN THE SOMETHING AWFUL S-CLUB 7 MEGATHREAD, TO NAME A SINGLE SONG BY HIS EXCELLENT NU-METAL SIDE PROJECT, SKUA, AND IF I CAN'T PLEASE TELL ME TO
EAT SHIT

Can you remember what it was? I used it years ago and they seemed pretty good, but now it has... I don't know, a vibe like they're churning out articles. Doesn't feel so much like the one guy running his tutorial website anyway

# ? Feb 6, 2018 16:04

Tunga: May 7, 2004; Grimey Drawer

Lars Vogel is obsessed with Eclipse for some reason so I generally avoid his articles these days. Anyone who thinks Eclipse is a good thing to use to develop Android apps in 2018 is not really someone I'm willing to pay any attention to.

# ? Feb 6, 2018 17:24

reversefungi: Nov 27, 2003; Master of the high hat!

Thanks guys, I'll check out those resources! Codepath and the codelabs seem super handy.

Regarding Java, I remember going through some of University of Helsinki's Java course but didn't find it too helpful. I work professionally in C#, so I figure that since the languages are close enough I could just jump into Android and pick up the relevant additional Java knowledge as I go along. If this is a terrible idea feel free to yell at me.

# ? Feb 6, 2018 18:25

FAT32 SHAMER: Aug 16, 2012

The Dark Wind posted:

Thanks guys, I'll check out those resources! Codepath and the codelabs seem super handy.

Regarding Java, I remember going through some of University of Helsinki's Java course but didn't find it too helpful. I work professionally in C#, so I figure that since the languages are close enough I could just jump into Android and pick up the relevant additional Java knowledge as I go along. If this is a terrible idea feel free to yell at me.

I did the opposite and they�re close enough that you�re going to keep your head above water and possibly even do a bunch of laps, worked well enough for me

# ? Feb 6, 2018 18:27

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

baka kaba posted:

Can you remember what it was? I used it years ago and they seemed pretty good, but now it has... I don't know, a vibe like they're churning out articles. Doesn't feel so much like the one guy running his tutorial website anyway

No, sorry, it was years ago.

# ? Feb 6, 2018 21:04

Tunga: May 7, 2004; Grimey Drawer

The Dark Wind posted:

Regarding Java, I remember going through some of University of Helsinki's Java course but didn't find it too helpful. I work professionally in C#, so I figure that since the languages are close enough I could just jump into Android and pick up the relevant additional Java knowledge as I go along. If this is a terrible idea feel free to yell at me.

I converted my career from C# to Android and had no real problems. It's even better now because Kotlin includes a lot of the good ideas from C# so you finally write Android apps without feeling strangely crippled after coming from C#.

# ? Feb 7, 2018 09:28

speng31b: May 8, 2010

anyone have any experience trying to expose Rx-based architecture to a larger group in an org or being the other end of it and being forced to learn? We've set up our stuff with RxSwift/RxJava+Kotlin and it's been really great for building consistent apps quickly across platforms and cutting down on dumb side effects / mutable state issues. We're in a position where we can make what we've been doing the standard for a larger group within the org and enforce it, but I've been doing Rx-heavy apps for awhile now and the concern is that I might have lost perspective / it might not be practical for a larger group, and we need something more "watered down."

# ? Feb 10, 2018 18:38

Danyull: Jan 16, 2011

Any recommendations for the best way to learn OpenGL ES for someone who has little to no experience with graphics programming? I've been working on an app that applies glitch effects to images and some of the effects take upwards of 15 seconds to process, long enough for me to think my app had crashed before I realized it just was taking a while (especially bad since some of the effects are meant to be interactive by moving your finger across the screen). Right now everything works by getting a bitmap of an image, then turning said bitmap into an array of pixels that are acted upon one by one. I've done some reading on other image editing apps where developers have said they've gone from one filter update every 4 seconds to >30 per second by switching to OpenGL from pixel-by-pixel processing, but the more I read about shaders and textures the more confused I get. I've tried searching for resources but almost everything I've found is for game programming which I'm not sure will be helpful for what I'm trying to do.

# ? Feb 21, 2018 08:23

baka kaba: Jul 19, 2003; PLEASE ASK ME, THE SELF-PROFESSED NO #1 PAUL CATTERMOLE FAN IN THE SOMETHING AWFUL S-CLUB 7 MEGATHREAD, TO NAME A SINGLE SONG BY HIS EXCELLENT NU-METAL SIDE PROJECT, SKUA, AND IF I CAN'T PLEASE TELL ME TO
EAT SHIT

I found these helpful when I had to learn it to get around the slowness of Canvas drawing

http://www.learnopengles.com/android-lesson-one-getting-started/
Tutorial for Android covering all the main stuff - old, but still relevant (OpenGL ES 3 works the same, just has more helpful features)

http://blog.db-in.com/all-about-opengl-es-2-x-part-1/
More of a broad overview of how OpenGL ES actually works as a state machine, why you're calling these weird functions and the reason it's organised the way it is. There's some code in this one but it's aimed at iPhone developers, I just found the explanation of the model really helpful

This is fairly basic so it'll get you drawing shapes on a Surface, which is really all I needed to do, but it sounds like you want to get a bit more involved with shaders and the like to make the glitching. I don't really have any resources for that (the game dev thread might), but you can probably experiment with stuff you find online. You'll be writing that in the shader language (GLSL) though

Also a big tip - you need to be able to spot errors in your code, which generally means calling a "was there an error" function, like the one here

Java code:

public void checkGLError(String op) {
    int error;
    while ((error = GLES20.glGetError()) != GLES20.GL_NO_ERROR) {
            Log.e("MyApp", op + ": glError " + error);
    }
}

what that does is loop through the error queue, popping off errors and logging them with the label you passed. You're meant to call it after an operation that could generate an error, so in practice people tend to put it after every GL call (with a conditional flag so you can turn logging on and off easily) so you can see exactly where the problem appears. It kinda sucks and I'm not sure if there's a better way in ES 3 - the docs still list it along with another thing - the Stack Overflow link says you can use OpenGL debugging on Android (haven't tried it) but it might not be available on newer versions?

One other option you can look into (which I don't know anything much about) is Renderscript, which is meant for fast computation, including image stuff:

quote:

Filterscript defines a subset of Renderscript that focuses on image processing operations, such as those that you would typically write with an OpenGL ES fragment shader.

so it's similar to writing shader code (it's C as well), and just focused on that instead of the whole 3D framework. Some of the old live wallpapers were written with Renderscript (like the one with floating leaves where you tap and get ripples)

# ? Feb 21, 2018 18:46

brand engager: Mar 23, 2011

Hey how do I check what the android emulator is getting stuck on through the command line? I've got a jenkins server that is supposed to run one on builds, but it doesn't seem to be booting.

quote:

12:28:03 $ /var/lib/jenkins//platform-tools/adb start-server
12:28:03 * daemon not running; starting now at tcp:5819
12:28:06 * daemon started successfully
12:28:06 $ /var/lib/jenkins//platform-tools/adb start-server
12:28:06 [android] Starting Android emulator
12:28:06 $ /var/lib/jenkins//emulator/emulator -skin 240x320 -ports 5736,5737 -report-console tcp:5810,max=60 -prop persist.sys.language=en -prop persist.sys.country=US -avd hudson_en-US_120_QVGA_android-25_google_apis-armeabi-v7a_10 -no-window -no-audio
12:28:06 emulator: WARNING: encryption is off
12:28:06 emulator: WARNING: Requested adb port (5737) is outside the recommended range [5555,5586]. ADB may not function properly for the emulator. See -help-port for details.
12:28:11 [android] Emulator reported that the console is available on port 5,736
12:28:11 [android] Waiting for emulator to finish booting...
12:28:11 $ /var/lib/jenkins//platform-tools/adb -s emulator-5736 wait-for-device shell getprop init.svc.bootanim
12:28:11 emulator: Cold boot: snapshot doesn't exist

quote:

12:42:49 $ /var/lib/jenkins//platform-tools/adb -s emulator-5736 wait-for-device shell getprop init.svc.bootanim
12:42:50 [android] Emulator reported that the startup process is ''
12:43:06 $ /var/lib/jenkins//platform-tools/adb -s emulator-5736 wait-for-device shell getprop init.svc.bootanim
12:43:09 [android] Emulator reported that the startup process is ''
12:43:24 [android] Timed-out after waiting 900 seconds for emulator
12:43:24 [android] Stopping Android emulator
12:43:24 $ /var/lib/jenkins//platform-tools/adb kill-server
12:43:24 emulator: WARNING: Not saving state: emulator hasn't finished booting.

# ? Mar 3, 2018 20:48

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

While Jenkins awaits, what happens if you run ADB devices on the server yourself as the Jenkins user? My suspicion is that your not standard port is the cause of the problem but I don't do Build Janitoring myself. I also don't know how you get adb to look on a specific port if it's not there.

You've seen https://developer.android.com/studio/run/emulator-commandline.html#startup-options I'm sure?

# ? Mar 5, 2018 06:19

brand engager: Mar 23, 2011

The plugin that does that randomly generates the ports on each job, and does not give an option to set them manually.

Edit: this plugin https://wiki.jenkins.io/display/JENKINS/Android+Emulator+Plugin

brand engager fucked around with this message at 08:01 on Mar 5, 2018

# ? Mar 5, 2018 07:53

brand engager: Mar 23, 2011

I'm running a copy of it manually to see how long it takes to boot. Been 30 minutes so far and it hasn't finished. This instance doesn't have virtualization features so no hardware acceleration.

Edit: Still wasn't done booting after an hour and a half. :shrug:

brand engager fucked around with this message at 18:04 on Mar 5, 2018

# ? Mar 5, 2018 16:36

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

It should take no longer than 5 minutes at the worst, sanity check that you can start ANY emulator on the build system first. Remove all the arguments and make a standard API 26 target or something.

# ? Mar 5, 2018 19:10

Adbot: ADBOT LOVES YOU

# ? May 14, 2024 00:30

brand engager: Mar 23, 2011

Volmarias posted:

It should take no longer than 5 minutes at the worst, sanity check that you can start ANY emulator on the build system first. Remove all the arguments and make a standard API 26 target or something.

I was running this avd

quote:

Name: hudson_en-US_120_QVGA_android-25_google_apis-armeabi-v7a_10
Device: Nexus 6 (Google)
Path: /var/lib/jenkins/.android/avd/hudson_en-US_120_QVGA_android-25_google_apis-armeabi-v7a_10.avd
Target: Google APIs (Google Inc.)
Based on: Android 7.1.1 (Nougat) Tag/ABI: google_apis/armeabi-v7a

with

code:

sudo -u jenkins ./emulator -avd hudson_en-US_120_QVGA_android-25_google_apis-armeabi-v7a_10 -no-window -no-audio -verbose

Is there a way to define a more basic avd to check with?

# ? Mar 5, 2018 19:26

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > My Code is Fragmented - Android Development Megathread

«‹›71 »