DEV Community: Felix

Caching in WebFlux - GraphQL context, using Caffeine

Felix — Fri, 19 Dec 2025 19:08:39 +0000

By no means do I know if this is the best way, or even a suggested way, but it's the way that I got it to work. Happy to share. 🤙

We needed to cache expensive and recursive results in a GraphQL context, using Spring Boot and WebFlux. One could argue that we should refactor the schema to ensure that it does not happen, but we wanted to keep that as flexible as possible so 🔩 that.

Only using the classic config with the famous and reliable @Cashable() annotations simply did not work. If I understand it correctly it just caches the Producer, and since it's executed every time it's subscribed to, the end result is just the same as not caching at all. Booger me.

Well how do you do it then? I ran into a few roadblocks along the way. Using futures to cache the results kinda worked. But I quickly noticed that this did not cache recursive calls in the same query, on a repeated nested schema mapped value, since the producers are already spun up before the first value is cached. Second call is perfectly cached however. But this still led to a few thousand (extreme edge case) calls for no reason at all.

So you have to make it atomic, this makes it possible to cache the "mid flight" result of a producer, and route the other producers there once you have your lil' result. Great.

So what happened when I decided to test it? 💩 hit the 🪭 instantly, because it was executing the cache at the same time that the Mono's where resolving (on same thread), leading to recursive update exception. It was solved by forcing it to be async by subscribing to it, and then having a fallback for empty Monos so that you're not stuck in a forever deadlock.

Anyway, here's the config and imlp:

@Configuration
public class CacheConfig {

    @Bean
    public AsyncCache<UUID, FooObject> fooCache() {
        return Caffeine.newBuilder()
                .expireAfterWrite(15, TimeUnit.MINUTES)
                .maximumSize(10_000)
                .buildAsync();
    }

    @Bean
    public AsyncCache<String, BarObject> barCache() {
        return Caffeine.newBuilder()
                .expireAfterWrite(60, TimeUnit.MINUTES)
                .maximumSize(5000)
                .buildAsync();
}

the values being arbitrary, use whatever you'd like.


public class CacheUtil {

    public static <K, V> Mono<V> cacheMono(AsyncCache<K, V> cache, K key, Supplier<Mono<V>> loader) {

        CompletableFuture<V> future = cache.get(key, (k, executor) -> {
            CompletableFuture<V> f = new CompletableFuture<>();

            // subscribe on a scheduler to guarantee async execution
            loader.get()
                    .subscribeOn(Schedulers.boundedElastic()) // forces async for same millisecond edge cases
                    .subscribe(
                            f::complete,
                            ex -> {
                                f.completeExceptionally(ex);
                                CompletableFuture.runAsync(() -> cache.asMap().remove(k, f)); // don't save the exceptions
                            },
                            () -> f.complete(null)); // stops deadlock on Mono.empty()

            return f;
        });

        return Mono.fromFuture(future);
    }

    public static <K, T> Flux<T> cacheFlux(AsyncCache<K, List<T>> cache, K key, Supplier<Flux<T>> loader) {
        return cacheMono(cache, key, () -> loader.get().collectList()).flatMapMany(Flux::fromIterable);
    }
}

and finally the actual use of one, I have not tried making into aspects because I'm lazy, and it's Friday evening. So it's just wrapper. Anyway, it's fairly simple:


    @Autowired
    private final AsyncCache<UUID, FooObject> fooCache;

    public Mono<FooObject> getFooOrSomething(UUID fooId) {
        return CacheUtil.cachedMono(fooCache, fooId, () -> fooService
                .findFoo(fooId));
    }

you can do all sorts of flatMapping, zipping or whatever in there, it's the final result of the publisher that is actually cached, no matter what that might be.

I know that there's no GraphQL code directly in here, but I still mention it because that's what ultimately created the need for me. 📈

Merry Christmas folk 🎄

Rename java packages on all branches: script.

Felix — Thu, 10 Aug 2023 09:30:40 +0000

Found yourself in need of a script that loops through all branches in a project and renames the package name? No?

It's a bit of a strange situation to be in I grant you, but if you have a project with a lot of branches for some strange reason, and you'd like to change the name / structure of your package. Then this is the script for you!

This scripts takes the paramaters:
Nested packages work just as well.

NOTE: it also pushes on all branches, please comment it out on the first try run.

#!/bin/bash

# Script made to loop though the src/main/java and the src/test/java directories and rename
# the packages based on the input. 
# Example: sh renameJavaPackages.sh to.dev.bacongubbe dev.bacongubbe.example /Users/bacongubbe/Projects/peskyproject

if [ "$1" == "--help" ] || [ "$1" == "-h" ]; then
    echo "Usage: $0 <old_package> <new_package> <repository_path>"
    echo "   - old_package: The old package name, ex: 'to.dev.bacongubbe'."
    echo "   - new_package: The new package name to use in package structure. Ex: dev.bacongubbe.example"
    echo "   - repository_path: Path to the repository where the changes should be made"
    exit 0
fi

if [ $# -ne 3 ]; then
    echo "Error: Incorrect number of arguments. Use --help for usage information."
    exit 1
fi

old_package="$1"
new_package="$2"
repository_path="$3"

cd "$repository_path" || exit

for branch in $(git for-each-ref --format='%(refname:short)' refs/heads/); do
    git checkout $branch
    echo "## Switched to $branch branch"

    for directory in "src/main/java" "src/test/java"; do
                echo "Package $old_package is being replaced with $new_package in $directory"

        find "$directory" -name "*.java" -type f -exec sed -i '' -e "s@$old_package@$new_package@g" {} +

        # Rename package directories
        old_dir="${old_package//.//}"
        new_dir="${new_package//.//}"
        mkdir -p "$directory/$new_dir"
        echo "dir $directory/$new_dir made"
        echo "copying contents from $directory/$old_dir to $directory/$new_dir"
        cp -r "$directory/$old_dir"/* "$directory/$new_dir"

        # Loop through the old and new structure and returns the first directory that is not matching. 
        # Ex: if you're renaming se.company.coolapp to se.company.coolerapp, it will return ../company/coolapp, since the se/company/ is the 
        # same in both.
        IFS="/" read -ra old_dir_parts <<< "$old_dir"
        IFS="/" read -ra new_dir_parts <<< "$new_dir"
        first_diff_dir=""
        for ((i=0; i<${#old_dir_parts[@]} && i<${#new_dir_parts[@]}; i++)); do
            if [ "${old_dir_parts[i]}" != "${new_dir_parts[i]}" ]; then
                first_diff_dir="${first_diff_dir}/${old_dir_parts[i]}"
                break
            fi
            first_diff_dir="${first_diff_dir}/${old_dir_parts[i]}"
        done
       echo "deleting: $directory${first_diff_dir}"

     # makes sure that there is a directory that needs to be deleted. 
       if [ -n "$first_diff_dir" ]; then
         rm -rf "$directory$first_diff_dir"
       fi
    done
    git add .
    git commit -m "Chore: Rename packages in $branch from $old_package to $new_package"

    # Push the changes to the remote repository
    git push origin $branch
done

Hope it helps!

From Java to Kotlin, first impression on for loops

Felix — Fri, 26 May 2023 05:49:13 +0000

What? This was supposed to be like Java? I can't even write a for loop?!

Or, at least, that was my first impression of Kotlin.
I tried to write for (int i = 0...) but then I quickly adapted to for (var i = 0...) and then for (var i = 0 : Int....) but realised that I was going nowhere.

After scratching my noggin and saying a few ~~curses~~ self-encouraging words. I opted in for the regular enhanced for loop instead. But alas for (n : ints) also resulted in a compilation error.

So lets dive in to what I should have done.

Quick disclaimer is that the IntRange object, as well as the other ranges in the kotlin.ranges package are easy, and often used in, for instance, loops. They can be easily created like the following examples:

val intRange = (1 .. 5)
val theEntireEnglishAlphabet = ('a' .. 'z')

// The ".." operator makes it an ascending range. 
// Another option, is to use the until or downTo keywords

val theSameAsIntRange = (1 until 5)
val descendingIntRange = (5 downTo 1)

So, back to what this article is all about, how do I use for loops?

Here are some examples!

// the regular, wonderful, enhanced for-loop

val someArr = intArrayOf(5, 3, 6, 32, 20, 1)

fun enhancedLoop() {
  for (num in someArr) {
    println(num) // -> console: 5, 3, 6, 32, 20, 1
  } 
}

Yes! A loop! 🙌 But how do I get the indexes? 🤔
There are 2 ways of easily achieving this. One using the .indicies property(?) that you'll find on all of the Arrays and Collections that returns a... you guessed it! IntRange.


// using the .indicies 

val someArr = intArrayOf(5, 3, 6, 32, 20, 1)

fun loopWithIndex() {
  for (i in someArr.indicies) {
    println(i) // -> 0, 1, 2, 3, 4, 5
  }
}

// and if you're looking to loop until a certain value.
// you can just create a new range

// top level: private const val MAX_LOOPS = 5 

fun loopWithInlineRange() {
  for (i in 0 .. MAX_LOOPS) { // you can also use the "until" keyword for increased readability 
    println(i) // -> 0, 1, 2, 3, 4, 5
  }
}

// and if you'd like to do it descending, most likely just 
// a "backwards loop" 

val someOtherArr = intArrayOf(1, 2, 10, 4, 5)

fun loopDescendingWithRange(){
  for (i in someOtherArr.size -1 downTo 0){
    println(someOtherArr[i]) // -> 5, 4, 10 2, 1
  }
}

"Nice! But that -1 feels very caveCode for being such a new language. Can't I just deconstruct the value and the index and keep it enhanced and safe from IndexOutOfBoundsExceptions?" you ask yourself.

Yes! Is the answer. By using the method withIndex() that you'll find on your arrays!


val someArr = intArrayOf(5, 3, 6, 32, 20, 1)

for ((i, num) in someArr.withIndex()) {
  println("$num has index $i") // -> "5 has index 0", "3 has index 1" .... 
}

// Important to note is that you're getting the index first, then the value

// If you want it reversed, you can just reverse the arr

for ((i, num) in someArr.reversed().withIndex()) {
  println("$num has index $i") // -> "1 has index 5", "20 has index 4" .... 
}

I hope you found this little read helpful. I missed my good 'ol for (int i = 0; i < whatHaveYou.length(); i++) in the beginning, but rather quickly grew fond of the usage of ranges instead.

GitHub Actions workflow for a Kotlin Spring app to GCP

Felix — Thu, 25 May 2023 16:57:22 +0000

If you, like I, are struggling to find a simple way to get a Kotlin spring boot application built and deployed to Google Cloud Platform (GCP), this might be a solution you can try!

I found it the easiest to opt in for using Google Artifact Registry (GAR), Google Cloud Run, and obviously in that case, Docker.

The app itself was built in Kotlin with Gradle, but should be practically the same if you're running a maven architecture and using Java.

Using the following Dockerfile:

FROM openjdk
COPY build/libs/**-0.0.1-SNAPSHOT.jar app.jar
EXPOSE 8080

ENTRYPOINT ["java","-jar","/app.jar"]

This was the steps I needed to take in order to get it up and spinning in the cloud.

First and foremost, you need to set up a Service Account in GCP IAM, with the principals Service Account User, Artifact Registry Service Agent, Cloud Run Service Agent and roles/artifactregistry.createOnPushRepoAdmin. Or at least trial and error lead me to those.

You'll also have to generate a key, there's a comment in the workflow about it! 😉

After that, using the following GitHub Actions workflow was the one that finally worked for me:

name: Build and deploy JAR to GCP

on:
  push:
    branches:
      - main
    paths-ignore:
      - 'README.md'

env:
  IMAGE_URL: {region}/{project-id}/{repo}/{app}:${{ github.sha }}

# replace region with your region, ex: europe-north1-docker.pkg.dev
# also replace all project-id etc to match your situation. 


jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - id: 'auth'
        name: Authenticate GCP
        uses: google-github-actions/auth@v1
        with:
          credentials_json: '${{ secrets.GCP_SA_KEY }}'
          # Generated from IAM > ServiceAccounts > Actions > Keys > ADD KEYS > JSON. 

      - name: Set up Cloud SDK
        uses: google-github-actions/setup-gcloud@v1
      - name: Use gcloud CLI
        run: gcloud info # Not requred, but can give you hints in the workflow if something goes wrong

      - name: Configure docker
        run: gcloud auth configure-docker {region} #Replace with your region

      - name: Set up Java version
        uses: actions/setup-java@v1
        with:
          java-version: '17'
      - name: Run tests
        run: ./gradlew test 
# you can replace with `mvn clean test` for maven
      - name: Build jar
        if: success() # only runs if all tests pass
        run: ./gradlew bootJar # mvn package work just fine for maven

      - name: Docker Build
        run: docker build -t $IMAGE_URL .
      - name: Docker Push
        run: docker push $IMAGE_URL

      - name: Update Cloud Run service
        uses: google-github-actions/deploy-cloudrun@v1
        with:
          service: {app} # same as the app in the $IMAGE_URL
          image: ${{ env.IMAGE_URL }}

Of course you can also test and build the application before configuring docker and GCP, and the order is not super important. The only really important order is that your - uses: actions/checkout@v3 is before

- id: 'auth'
  name: Authenticate GCP
  uses: google-github-actions/auth@v1

Else, it will not work.

Hope this helps you / anyone along their way in the CI/CD ~~nightmare~~ happyfuntimes.