Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Trip into the async world @ NYC Kotlin Meetup
Search
Roberto Orgiu
October 16, 2019
Programming
130
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
Trip into the async world @ NYC Kotlin Meetup
These are the slides of my talk from NYC Kotlin Meetup @ Dropbox
Roberto Orgiu
October 16, 2019
More Decks by Roberto Orgiu
See All by Roberto Orgiu
Wellness & Droid
tiwiz
0
140
Behind the curtains
tiwiz
0
89
The Importance of Being Tested
tiwiz
0
450
An Android Dev start to Kotlin MPP
tiwiz
0
210
Fantastic API and where to find them
tiwiz
0
100
Flipping the Koin @ GDG Dev Party
tiwiz
1
79
Flipping the Koin
tiwiz
2
180
Trip into the async world
tiwiz
1
150
GraphQL IRL (Android Makers)
tiwiz
0
160
Other Decks in Programming
See All in Programming
PHP に部分適用が来るぞ!……ところで何それ?おいしいの? #phpcon / phpcon-2026
shogogg
0
430
【やさしく解説 設計編・中級 #1】一つの車に、運転手は一人 ~ある倉庫システムの事例から~
panda728
PRO
0
200
Haskell/Servantを通してWebミドルウェアを捉え直す
pizzacat83
1
630
Built Our Own Background Agent at LayerX #aidevex_findy
layerx
PRO
9
3.8k
エンジニアにデザインハーネスを 〜デザインプロセスを規定するためのハーネス〜 / Design harness from an engineer's perspective
rkaga
2
1.8k
The Past, Present, and Future of Enterprise Java
ivargrimstad
0
440
komatsuna「分散システムにおけるバグ分析手法」
komatsunaqa
0
130
改善しないと、タスクが回らない。 “てんこ盛りポジション” を引き継いだ情シスの、入社3ヶ月の業務改善録
krm963
0
230
地域 SRE コミュニティ最前線 - ホンマでっかSRE勉強会
tk3fftk
0
290
アルゴリズムは何を圧縮しているのか ─ Haskell から育った「圧縮代数」というメンタルモデル
naoya
16
3.7k
【やさしく解説 設計編 #0】DDDのコード、読めるのに分からない人へ
panda728
PRO
2
290
Go言語とトイモデルで学ぶTransformerの気持ち / fukuokago23-transformer
monochromegane
0
150
Featured
See All Featured
Avoiding the “Bad Training, Faster” Trap in the Age of AI
tmiket
0
200
Digital Projects Gone Horribly Wrong (And the UX Pros Who Still Save the Day) - Dean Schuster
uxyall
1
2.1k
Git: the NoSQL Database
bkeepers
PRO
432
67k
16th Malabo Montpellier Forum Presentation
akademiya2063
PRO
0
300
How to build a perfect <img>
jonoalderson
1
5.8k
Navigating the moral maze — ethical principles for Al-driven product design
skipperchong
2
430
Skip the Path - Find Your Career Trail
mkilby
1
170
Getting science done with accelerated Python computing platforms
jacobtomlinson
2
370
The untapped power of vector embeddings
frankvandijk
2
1.8k
Optimizing for Happiness
mojombo
378
71k
Designing for Performance
lara
611
70k
Testing 201, or: Great Expectations
jmmastey
46
8.2k
Transcript
Trip into the async world
Rob
Why RxJava?
Why RxJava? Retrofit
Why RxJava? Retrofit Async ops
Why RxJava? Retrofit Async ops Basic operators
Why RxJava? Retrofit Async ops Basic operators Advanced operators
Why RxJava? Retrofit Async ops Basic operators Advanced operators
Why RxJava? Retrofit Async ops Basic operators Advanced operators
implementation ‘com.squareup.retrofit2:retrofit:2.6.1’
@GET("/path/to/resource") fun networkRequest(): Single<Data> @GET("/path/to/resource") suspend fun networkRequest(): Data
Retrofit.Builder() ... .addCallAdapterFactory(CoroutineCallAdapterFactory()) .addCallAdapterFactory(RxJava2CallAdapterFactory.create()) .build()
CoroutineCallAdapterFactory() RxJava2CallAdapterFactory.create()
@Test fun test() { runBlocking { val response = api.networkRequest()
assertThat(response)... } api.networkRequest() .test() .assertValue { } }
None
class DataViewModel()) : ViewModel() { private val subscriptions = CompositeDisposable()
val output = BehaviorSubject<LCE<Data>>.create() fun loadCurrentStatus() { disposable += store.get(BarCode.empty()) .startWith(LceLoading()) .onErrorReturn({e -> LceError(e)}) .map({data -> mapEmptyData(data)}) .observeOn(AndroidSchedulers.mainThread()) .subscribeOn(Schedulers.io()) .subscribe(output) } override fun onCleared() { disposable.clear() } }
class DataViewModel()) : ViewModel() { private val subscriptions = CompositeDisposable()
val output = BehaviorSubject<LCE<Data>>.create() fun loadCurrentStatus() { disposable += store.get(BarCode.empty()) .startWith(LceLoading()) .onErrorReturn({e -> LceError(e)}) .map({data -> mapEmptyData(data)}) .observeOn(AndroidSchedulers.mainThread()) .subscribeOn(Schedulers.io()) .subscribe(output) } override fun onCleared() { disposable.clear() } } Hidden knowledge
te val subscriptions = CompositeDisposable() utput = BehaviorSubject<LCE<Data>>.create() oadCurrentStatus() {
isposable += store.get(BarCode.empty()) .startWith(LceLoading()) .onErrorReturn({e -> LceError(e)}) .map({data -> mapEmptyData(data)}) .observeOn(AndroidSchedulers.mainThread()) .subscribeOn(Schedulers.io()) .subscribe(output) ide fun onCleared() { isposable.clear() Learning curve
class DataViewModel()) : ViewModel() { val output = MutableLiveData<Data>() fun
loadCurrentStatus() { output.postValue(LceLoading()) viewModelScope.launch { try { val data = store.get(BarCode.empty()).await() if (data.isEmpty()) { output.postValue(LceError(EmptyResultSetException())) } else { output.postValue(LceSuccess(data)) } } catch (e: Exception) { output.postValue(LceError(e)) } } } }
class DataViewModel()) : ViewModel() { val output = MutableLiveData<Data>() fun
loadCurrentStatus() { output.postValue(LceLoading()) viewModelScope.launch { try { val data = store.get(BarCode.empty()).await() if (data.isEmpty()) { output.postValue(LceError(EmptyResultSetException())) } else { output.postValue(LceSuccess(data)) } } catch (e: Exception) { output.postValue(LceError(e)) } } } }
adCurrentStatus() { output.postValue(LceLoading()) viewModelScope.launch { try { val data =
store.get(BarCode.empty()).await() if (data.isEmpty()) { output.postValue(LceError(EmptyResultSetException())) } else { output.postValue(LceSuccess(data)) } } catch (e: Exception) { output.postValue(LceError(e)) } }
map flatmap filter first last
map flatmap filter first last forEach orEmpty mapNotNull
@NotNull public static <T> Observable<Response<T>> from(@NotNull final ApolloCall<T> call) {
checkNotNull(call, "call == null"); return Observable.create(new ObservableOnSubscribe<Response<T>>() { @Override public void subscribe(final ObservableEmitter<Response<T>> emitter) throws Exception { cancelOnObservableDisposed(emitter, call); call.enqueue(new ApolloCall.Callback<T>() { @Override public void onResponse(@NotNull Response<T> response) { if (!emitter.isDisposed()) { emitter.onNext(response); } } @Override public void onFailure(@NotNull ApolloException e) { Exceptions.throwIfFatal(e); if (!emitter.isDisposed()) { emitter.onError(e); } } @Override public void onStatusEvent(@NotNull ApolloCall.StatusEvent event) { if (event == ApolloCall.StatusEvent.COMPLETED && !emitter.isDisposed()) { emitter.onComplete(); } } }); } }); }
@NotNull public static <T> Observable<Response<T>> from(@NotNull final ApolloCall<T> call) {
checkNotNull(call, "call == null"); return Observable.create(new ObservableOnSubscribe<Response<T>>() { @Override public void subscribe(final ObservableEmitter<Response<T>> emitter) throws Exception { cancelOnObservableDisposed(emitter, call); call.enqueue(new ApolloCall.Callback<T>() { @Override public void onResponse(@NotNull Response<T> response) { if (!emitter.isDisposed()) { emitter.onNext(response); } } @Override public void onFailure(@NotNull ApolloException e) { Exceptions.throwIfFatal(e); if (!emitter.isDisposed()) { emitter.onError(e); } } @Override public void onStatusEvent(@NotNull ApolloCall.StatusEvent event) { if (event == ApolloCall.StatusEvent.COMPLETED && !emitter.isDisposed()) { emitter.onComplete(); } } }); } }); } suspend fun <T> ApolloCall<T>.await(): Response<T> = suspendCancellableCoroutine { cont -> enqueue(object : ApolloCall.Callback<T>() { override fun onFailure(e: ApolloException) { cont.resumeWithException(e) } override fun onResponse(response: Response<T>) { cont.resume(response) } }) cont.invokeOnCancellation { cancel() } }
suspend fun <T> ApolloCall<T>.await(): Response<T> = suspendCancellableCoroutine { cont ->
enqueue(object : ApolloCall.Callback<T>() { override fun onFailure(e: ApolloException) { cont.resumeWithException(e) } override fun onResponse(response: Response<T>) { cont.resume(response) } }) cont.invokeOnCancellation { cancel() } }
val job = Job() val scope = CoroutineScope(Dispatchers.IO + job)
fun doThings() { scope.launch { ... } } fun inTheEnd() { job.cancel() }
val job = Job() val scope = CoroutineScope(Dispatchers.IO + job)
fun doThings() { scope.launch { ... } } fun inTheEnd() { job.cancelChildren() scope.coroutineContext.cancelChildren() }
job.cancelChildren() scope.coroutineContext.cancelChildren() public fun Job.cancelChildren(cause: CancellationException? = null) { children.forEach
{ it.cancel(cause) } } public fun CoroutineContext.cancelChildren(cause: CancellationException? = null) { this[Job]?.children?.forEach { it.cancel(cause) } }
None
kotlinx-coroutines-rx2
class RxThing { fun doStuff() : Single<MyData> }
suspend fun doThings() { val my data = rxThing.doStuff().await() }
class CoroutineThing { suspend fun doStuff() : MyData }
fun doStuff() : Single<MyData> = rxSingle { coroutineThing.doStuff() }
fun doStuff() : Single<MyData> = rxSingle(RxCoroutineExceptionHandler) { coroutineThing.doStuff() }
None
open.nytimes.com @nytdev github.com/NYTimes developers.nytimes.com
QA &