We have an existing Android Avocado Facts application created in an older Kotlin Thursdays post, but for any Android application to be seriously useful, we will have to upgrade our facts to be more than just hard-coded. Feel free to follow along with our Github repo. In this post, we will set up an Async Task like a pro!

References

Async Tasks in Android

Async Tasks are a lightweight threading construct given by Android OS. Being an Android-specific concept, it is not created by Java or Kotlin. Async Tasks makes most common task of doing some heavy-weight operation off main thread easy and simple.

The Android OS is able to refresh the UI 64 frames per second — to achieve this, the UI Thread, or the Main Thread, needs to be free as much as possible. Long running tasks like communicating with APIs or databases could lock up your UI thread if it is not done in a background thread. One way to achieve this is by off-loading any longer running tasks to other threads. AsyncTasks is very powerful construct provided by OS to help achieve this.[more details]






Creating an AsyncTask in Android

An AsyncTask has three generic parameter types:

AsyncTask<Params, Progress, Result> 
- params: the type of the parameters sent to the task upon execution
- progress: the type of the progress units published during the background computation
- result: the type of the result of the background computation

We can actually think about implementing our Async tasks in the order these parameter types are defined!

Step 1:

We must analyze if any runtime parameters need to be sent for execution of the task. In our example to generate a new fact, we only need to invoke the function. Hence, the param type is Void.

Step 2:

Do we need to know about the progress percentage of this task?

In our async task, there is no concept of progress hence the progress type is also Void. One example where this is crucial will be downloading a big file.

Step 3:

What is the result of task? The type we designate here also tells us how will we can update our UI with the result. In particular, we want to show the user new fact which is of type String.

All these steps gives us our signature

AsyncTask<Void, Void, String>() {

Step 4:

Start by extending the abstract class and implementing minimum

override fun doInBackground(vararg params: Void?): String {

Step 5:

Any ui update if need will need.

Shortcut tip: ctrl + o (override method shortcut in android studio) will help do method setup
override fun onPostExecute(result: String?) {

Here we need to set result string as Textview’s text value. We have method in activity that will update it for us. So to call it we need to pass that to task so it can call it. Changing class to have

private class NewAvocadoFactTask(val activity: MainActivity)

Okay! Now just call is left

Step 6:

In onClick method call

moarButton.setOnClickListener {
NewAvocadoFactTask(this).execute();
}

Now for the MOST important step: you are leaking your activity!

How?? What??

AsyncTask is a threading construct. AsyncTask started a new thread and it doesn’t know your app may or may not have been killed by the time we have our response. In other words, AsyncTask runs independent of the main app in the background thread. So if the app is closed first without first ending the task, that task could still be running. After the response is returned, AsyncTask is responsible for returning the result to the main thread itself as a callback.

Is passing the actual activity val activity: MainActivity really safe? Should we try to call something that could be garbage collected?? Maybe we should use a wrapper called WeakReference.

var reference: WeakReference<MainActivity> = WeakReference(activity)

What is WeakReference? Glad you asked. It is a wrapper type that would be ok if the object it was keeping track of was garbage collected. Incase that happens it will indicate you the object you wanted reference for is no longer present.

reference.get() will yield a null. So we make our UI update call with elvis operator

reference.get()?.updateAvocadoFact(it)



Gotcha(s)

  1. Always pass any context in AsyncTask with WeakReference Wrapper
var reference: WeakReference<MainActivity> = WeakReference(activity)

2. Use OnProgressUpdate() and OnPostExecute() to show something on UI

override fun onPostExecute(result: String?) {
Log.v(tag,”onPostExecute is running on ${Thread.currentThread().name}”)
  super.onPostExecute(result)
result?.let {
// if reference.get is not null execute the function
reference.get()?.updateAvocadoFact(it)
}
}

Logging will help you see which functions are executed on Main thread since UI elements can only be touched by Main thread.

3. Always create a new Task object for a execute.

moarButton.setOnClickListener {
NewAvocadoFactTask(this).execute()
}

From threading rules of Async Task

The task can be executed only once (an exception will be thrown if a second execution is attempted.)



You can see the complete Github repo here:

We’ve set up our Async Task, but we actually need the request for the task to connect to a database so we can grab our Avocado Facts. In the next post, we will set up a database in Firebase and get our Async Task connected! See you next week.

Views: 214

Happy 10th year, JCertif!

Notes

Welcome to Codetown!

Codetown is a social network. It's got blogs, forums, groups, personal pages and more! You might think of Codetown as a funky camper van with lots of compartments for your stuff and a great multimedia system, too! Best of all, Codetown has room for all of your friends.

When you create a profile for yourself you get a personal page automatically. That's where you can be creative and do your own thing. People who want to get to know you will click on your name or picture and…
Continue

Created by Michael Levin Dec 18, 2008 at 6:56pm. Last updated by Michael Levin May 4, 2018.

Looking for Jobs or Staff?

Check out the Codetown Jobs group.

 

Enjoy the site? Support Codetown with your donation.



InfoQ Reading List

Presentation: Accelerating LLM-Driven Developer Productivity at Zoox

Amit Navindgi discusses the systematic shift at Zoox from fragmented documentation to an AI-driven ecosystem. He explains how they built "Cortex," a secure platform integrating RAG, multi-modal LLMs, and contributor-friendly agent APIs. He shares practical strategies for driving adoption through AI champions and hackathons, emphasizing the move from deterministic workflows to autonomous agents.

By Amit Navindgi

Moonrepo Releases Moon v2.0 with WASM Plugin Toolchains and Overhauled CLI

Moonrepo has released moon v2.0, its first major update since v1, featuring a plugin-based toolchain system and support for multiple configuration formats including JSON and TOML. The CLI has been restructured, enhancing task inheritance and Docker integration. Notable changes include a shift in architecture and improvements to VCS support.

By Daniel Curtis

Scaling Social Systems in Software Organizations

Fast-scaling teams must rebuild trust and psychological safety as their social systems expand. Intentional, redundant communication across multiple formats can keep everyone aligned. Cross-team rituals, buddy systems, and rotating facilitators can reduce silos by building bridges between teams. Leaders accelerate this by modeling the vulnerability they want to see.

By Ben Linders

Pinterest Engineers Eliminate CPU Zombies to Resolve Production Bottlenecks

Pinterest identified and resolved CPU starvation issues that affected machine learning training jobs on its Kubernetes-based platform, PinCompute. The engineers traced the problem to an unused Amazon ECS agent, which caused memory cgroup leaks. By disabling the agent, they stabilised performance. This case illustrates the importance of understanding system defaults for effective troubleshooting.

By Mark Silvester

Anthropic Traces Six Weeks of Claude Code Quality Complaints to Three Overlapping Product Changes

Anthropic published a postmortem tracing six weeks of Claude Code quality complaints to three overlapping product-layer changes: a reasoning effort downgrade, a caching bug that progressively erased the model's own thinking, and a system prompt verbosity limit that caused a 3% quality drop. The API and model weights were unaffected. All issues were resolved April 20.

By Steef-Jan Wiggers

© 2026   Created by Michael Levin.   Powered by

Badges  |  Report an Issue  |  Terms of Service