Skip to content

karmakrafts/Kompress

Repository files navigation

🗜️ Kompress

Lightweight zlib (de)compression API for Kotlin Multiplatform.

Features

  • Supports all Kotlin Multiplatform targets
  • Support for DEFLATE and DEFLATE RAW compression
  • Synchronous streaming API inspired by Java's Inflater/Deflater APIs
  • Integration with kotlinx.io
  • Customizable compression-level
  • Extra lightweight on JVM and native because it wraps available platform APIs

How to use it

First, add the official Maven Central repository to your settings.gradle.kts:

dependencyResolutionManagement {
    repositories {
        maven("https://central.sonatype.com/repository/maven-snapshots")
        mavenCentral()
    }
}

Then add a dependency on the library in your root buildscript:

kotlin {
    sourceSets {
        commonMain {
            dependencies {
                implementation("dev.karmakrafts.kompress:kompress-core:<version>")
            }
        }
    }
}

Or, if you are only using Kotlin/JVM, add it to your top-level dependencies block instead.

Inflater and Deflater interfaces

Bulk compression

If you just want to (de)compress one large blob of data, the Deflater.deflate and Inflater.inflate functions are what you're probably looking for:

fun main() {
    val myData = "Hello, World! This is an important message."
    val compressedData = Deflater.deflate(
        data = myData.encodeToByteArray(),
        raw = false, // Control if you want the gzip/pkzip header
        level = 9, // Control the compression level
        bufferSize = 1024 // Control the internal buffer size
    )
    val decompressedData = Inflater.inflate(
        data = compressedData,
        raw = false,
        bufferSize = 1024
    )
    println(myData == decompressedData.decodeToString())
}

Streaming compression

Streaming compression is what you want if your data exceeds a certain size,
that size usually being the limit of the underlying runtime's array size.
With Kompress, that limit is about 2.147GB because the index type of an array in Kotlin is a signed integer. Streaming allows you to split up the data into discrete chunks and compress those chunks sequentially until you processed all the data.

You can either use the Deflater and Inflater interfaces from the core module manually:

fun main() {
    val deflater = Deflater(
        raw = false,
        level = 9,
        // ...
    )
    val outputBuffer = ByteArray(1024)
    while(deflater.needsInput) {
        if(!hasMoreInput) deflater.finish() // Signal that we are done compressing
        deflater.input = getInputChunk()
        while(!deflater.finished) {
            deflater.deflate(outputBuffer) // Deflate data into the buffer
            copyChunkToSomeTarget(outputBuffer)
        }
    }
    deflater.close() // Always close Deflater/Inflater, it is recommended to use .use{}
}

Or you can use the recommended way of kotlinx.io wrappers:

fun main() {
    val buffer = Buffer()
    buffer.writeInt(42)
    buffer.writeFloat(4.20F)
    
    val compressedBuffer = Buffer()
    buffer.deflating().use(compressedBuffer::transferFrom)
    
    val decompressedBuffer = Buffer()
    compressedBuffer.inflating().use(decompressedBuffer::transferFrom)
}