Skip to content

Functional Constructs for Databinding + Kotlin + RxJava

Notifications You must be signed in to change notification settings

rakshakhegde/ObservableFlow

Repository files navigation

ObservableFlow

Pt 1/3

Pt 2/3: Stepper Indicator

Pt 3/3: SugarPreferences

BuddyBuild Minimum API 9 Release rakshakhegde newsletter

Functional Kotlin constructs like map(), filter() and 12 more functions built for Android Data Binding library + RxJava 2 support. Check out the release apk here.

// tiny preview
val source = ObservableField("source")
val dest = source.filter { it.startsWith('s') }
		.map { it.capitalize() }

Please sign up for my newsletter rakshakhegde 📧

Installation

In your global build.gradle file:

allprojects {
	repositories {
		...
		maven { url 'https://jitpack.io' }
	}
}

Include any one of these in your app build.gradle file:

dependencies {
	compile 'com.github.rakshakhegde.ObservableFlow:observableflow:0.1-alpha'
}

(or, if you need RxJava 2 support. This includes the above ObservableFlow core library.)

dependencies {
	compile 'com.github.rakshakhegde.ObservableFlow:rxdatabinding:0.1-alpha'
}

Motivation

I personally find writing code in a functional way just makes my logic neatly encapsulated and broken up into simple parts. So I built these functional constructs as a part of my other projects where I use DataBinding and Kotlin (which is almost always), that kinda emulates RxJava's map and filter which I loved and used a lot, but for ObservableFields. So now I've open sourced it hoping others might find it useful too. Cheers :)

By the way, if you like my work please show some ❤️ and star this project above

GitHub stars GitHub forks GitHub watchers GitHub followers Twitter Follow

Documentation

Here are all the functions and extension functions for ObservableFields. Even though I've attempted to document how to use all the functional statements, the best source of documentation are the tests themselves. Here I'll just show the function prototype and code snippet in pairs. Hopefully meaning of the code becomes immediately apparent.

  • map
     val text: ObservableField<String> = ObservableField("Hello World")
     val textLength = text.map { it.length }
     inline fun <T, R> ObservableField<T>.map(
     	dstObsrv: ObservableField<R> = ObservableField(),
     	crossinline f: (T) -> R
     ): ObservableField<R>
  • vararg map
     val totalLength = map(text1, text2) {
     	text1.get().length + text2.get().length
     }
     inline fun <T> map(vararg sources: Observable, crossinline onChange: () -> T): ObservableField<T>
  • filter
     val evenLengthTextOnly = text.filter { it.length % 2 == 0 }
     inline fun <T> ObservableField<T>.filter(
     	defaultVal: T? = null,
     	crossinline predicate: (T) -> Boolean
     ): ObservableField<T>
  • onPropertyChanged
     text.onPropertyChanged { // block executed for every subsequent property change
     	val newVal = get()
     	println("text changed to $newVal")
     }
     inline fun <T : Observable> T.onPropertyChanged(crossinline listener: T.(propertyId: Int) -> Unit):
     	Observable.OnPropertyChangedCallback
  • vararg onPropertyChanged
     onPropertyChanged(text1, text2) {
     	performAction(text1.get(), text2.get())
     }
     inline fun onPropertyChanged(vararg sources: Observable, crossinline onChange: () -> Unit):
     	List<Observable.OnPropertyChangedCallback>
  • bind
     text.bind { // block executed immediately and for every subsequent property change
     	val newVal = get()
     	println("text changed to $newVal")
     }
     inline fun <T : Observable> T.bind(
     	crossinline listener: T.(Int) -> Unit
     ): Observable.OnPropertyChangedCallback
  • vararg bind
     bind(text1, text2) {
     	println(text1.get() + text2.get())
     }
     inline fun bind(vararg sources: Observable, crossinline onChange: () -> Unit):
     	List<Observable.OnPropertyChangedCallback>
  • observableListOf
     val observableInts = observableListOf(5, 6, 7, 9) // Int type inferred
     fun <T> observableListOf(vararg items: T): ObservableArrayList<T>
  • bindToList
     observableInts.bindToList { // block executed immediately and everytime anything changes in the list
     	forEach { item -> performActionOn(item) }
     }
     inline fun <T, S : ObservableList<T>> S.bindToList(crossinline listener: S.() -> Unit):
     	ObservableList.OnListChangedCallback<S>
  • onListChanged
     observableInts.onListChanged { // block executed everytime anything changes in the list
     	forEach { item -> performActionOn(item) }
     }
     inline fun <T, S : ObservableList<T>> S.onListChanged(crossinline listener: S.() -> Unit):
     	ObservableList.OnListChangedCallback<S>
  • rx
     val nameObservable = ObservableField("First")
     
     nameObservable.rx()
     	.subscribe(onNext)
     fun <T> ObservableField<T>.rx(): Observable<T>
  • toField - share()s the RxJava Observable so that when the last propertyChangedCallback is removed, the Observable is also finished
     val dest = Observable.create<String> { emitter = it }
     		.toField()
     fun <T> Observable<T>.toField(defaultVal: T? = null): RxObservableField<T>
  • rxOnPropertyChanged
     rxOnPropertyChange(src1, src2, src3)
     	.map { src1.get() + src2.get() + src3.get() }
     	.subscribe(onNext)
     fun rxOnPropertyChange(vararg observables: android.databinding.Observable): io.reactivex.Observable<Int>
  • rxBind
     rxBind(src1, src2, src3)
     	.map { src1.get() + src2.get() + src3.get() }
     	.subscribe(onNext)
     fun rxBind(vararg observables: android.databinding.Observable): io.reactivex.Observable<Int>

Miscellaneous

  • There are no extension functions for all the other primitive Observables, like ObservableInt, ObservableFloat, etc. because I didn't find these optimisations worthwhile to write code and tests for. So I just use ObservableField<Int>, ObservableField<Float>, etc even for Android projects and they are auto-boxed, well, automatically.
  • For the RxJava toField() binding, it was initially non-explicitly-writable, i.e., set() function was made useless and deprecated, so as to avoid side-effects, one of guiding principles of Functional Programming. But I once found myself desperately needing to change the value of that ObservableField imperatively. So I have reverted it back to a normal ObservableField whose value can be imperatively mutated anywhere in your code. But just don't go bonkers using this side-effect. Try and use it functionally as much as possible (:

Created & Maintained By

Rakshak R.Hegde (@rakshakhegde) Wanna contribute to this project? Pull requests welcome ❤️

Note: This project is a part of an ongoing series of open-source projects I'm working on. If you liked my work, please consider following me on Twitter @rakshakhegde and Github. Also please share this project with fellow developers. I'm also starting a newsletter of my own called rakshakhegde 📧. It would be great if you could fill in your email address and subscribe to my new project announcements, releases or blog posts. Subscribe here :)

License

Copyright 2017 Rakshak Hegde

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.