fil/docs/snippets/kotlin/utils/token_reduction_example.md

import dev.kreuzberg.*
import java.nio.file.Paths
import java.util.Optional

fun main() {
    val tokenReduction = TokenReductionOptions.builder()
        .withMode("moderate")
        .withPreserveImportantWords(true)
        .build()

    val config = ExtractionConfig.builder()
        .withTokenReduction(Optional.of(tokenReduction))
        .build()

    val result = Kreuzberg.extractFileSync(Paths.get("verbose_document.pdf"), null, config)
    println("Reduced content length: ${result.content().length}")
}