fil/packages/java/dev/kreuzberg/ChunkingConfig.java

// This file is auto-generated by alef — DO NOT EDIT.
// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
// To regenerate: alef generate
// To verify freshness: alef verify --exit-code
// Issues & docs: https://github.com/kreuzberg-dev/alef
package dev.kreuzberg;

import com.fasterxml.jackson.annotation.JsonProperty;
import com.fasterxml.jackson.annotation.JsonInclude;
import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
import org.jspecify.annotations.Nullable;

/**
 * Chunking configuration.
 *
 * Configures text chunking for document content, including chunk size,
 * overlap, trimming behavior, and optional embeddings.
 *
 * Use {@code ..Default.default()} when constructing to allow for future field additions:
 *
 * let config = ChunkingConfig {
 *     max_characters: 500,
 *     ..Default.default()
 * };
 *
 */
@JsonInclude(JsonInclude.Include.NON_ABSENT)
@JsonDeserialize(builder = ChunkingConfig.Builder.class)
public record ChunkingConfig(
    /**
     * Maximum size per chunk (in units determined by {@code sizing}).
     *
     * When {@code sizing} is {@code Characters} (default), this is the max character count.
     * When using token-based sizing, this is the max token count.
     *
     * Default: 1000
     */
    @Nullable @JsonProperty("max_chars") Long maxCharacters,
    /**
     * Overlap between chunks (in units determined by {@code sizing}).
     *
     * Default: 200
     */
    @Nullable @JsonProperty("max_overlap") Long overlap,
    /**
     * Whether to trim whitespace from chunk boundaries.
     *
     * Default: true
     */
    @Nullable @JsonProperty("trim") Boolean trim,
    /**
     * Type of chunker to use (Text or Markdown).
     *
     * Default: Text
     */
    @Nullable @JsonProperty("chunker_type") ChunkerType chunkerType,
    /**
     * Optional embedding configuration for chunk embeddings.
     */
    @Nullable @JsonProperty("embedding") EmbeddingConfig embedding,
    /**
     * Use a preset configuration (overrides individual settings if provided).
     */
    @Nullable @JsonProperty("preset") String preset,
    /**
     * How to measure chunk size.
     *
     * Default: {@code Characters} (Unicode character count).
     * Enable {@code chunking-tiktoken} or {@code chunking-tokenizers} features for token-based sizing.
     */
    @Nullable @JsonProperty("sizing") ChunkSizing sizing,
    /**
     * When {@code true} and {@code chunker_type} is {@code Markdown}, prepend the heading hierarchy
     * path (e.g. {@code "# Title > ## Section\n\n"}) to each chunk's content string.
     *
     * This is useful for RAG pipelines where each chunk needs self-contained
     * context about its position in the document structure.
     *
     * Default: {@code false}
     */
    @Nullable @JsonProperty("prepend_heading_context") Boolean prependHeadingContext,
    /**
     * Optional cosine similarity threshold for semantic topic boundary detection.
     *
     * Only used when {@code chunker_type} is {@code Semantic} and an {@code EmbeddingConfig} is
     * provided. You almost never need to set this. When omitted, defaults to
     * {@code 0.75} which works well for most documents. Lower values detect more
     * topic boundaries (more, smaller chunks); higher values detect fewer.
     * Range: {@code 0.0..=1.0}.
     */
    @Nullable @JsonProperty("topic_threshold") Float topicThreshold
) {
    public static Builder builder() {
        return new Builder();
    }
    public ChunkingConfig{
        if (maxCharacters == null) maxCharacters = 1000L;
        if (overlap == null) overlap = 200L;
    }

    // CPD-OFF
    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
    public static final class Builder {

        @JsonProperty("max_chars")
private Long maxCharacters = null;
        @JsonProperty("max_overlap")
private Long overlap = null;
private Boolean trim = null;
        @JsonProperty("chunker_type")
        @Nullable private ChunkerType chunkerType = ChunkerType.Text;
private EmbeddingConfig embedding = null;
private String preset = null;
        @Nullable private ChunkSizing sizing = new ChunkSizing.Characters();
        @JsonProperty("prepend_heading_context")
private Boolean prependHeadingContext = null;
        @JsonProperty("topic_threshold")
private Float topicThreshold = null;

        /** Sets the maxCharacters field. */
        @JsonProperty("max_chars")
        public Builder withMaxCharacters(final @Nullable Long value) {
            this.maxCharacters = value;
            return this;
        }

        /** Sets the overlap field. */
        @JsonProperty("max_overlap")
        public Builder withOverlap(final @Nullable Long value) {
            this.overlap = value;
            return this;
        }

        /** Sets the trim field. */
        @JsonProperty("trim")
        public Builder withTrim(final @Nullable Boolean value) {
            this.trim = value;
            return this;
        }

        /** Sets the chunkerType field. */
        @JsonProperty("chunker_type")
        public Builder withChunkerType(final @Nullable ChunkerType value) {
            this.chunkerType = value;
            return this;
        }

        /** Sets the embedding field. */
        @JsonProperty("embedding")
        public Builder withEmbedding(final @Nullable EmbeddingConfig value) {
            this.embedding = value;
            return this;
        }

        /** Sets the preset field. */
        @JsonProperty("preset")
        public Builder withPreset(final @Nullable String value) {
            this.preset = value;
            return this;
        }

        /** Sets the sizing field. */
        @JsonProperty("sizing")
        public Builder withSizing(final @Nullable ChunkSizing value) {
            this.sizing = value;
            return this;
        }

        /** Sets the prependHeadingContext field. */
        @JsonProperty("prepend_heading_context")
        public Builder withPrependHeadingContext(final @Nullable Boolean value) {
            this.prependHeadingContext = value;
            return this;
        }

        /** Sets the topicThreshold field. */
        @JsonProperty("topic_threshold")
        public Builder withTopicThreshold(final @Nullable Float value) {
            this.topicThreshold = value;
            return this;
        }

        /** Builds the ChunkingConfig instance. */
        public ChunkingConfig build() {
            return new ChunkingConfig(
                maxCharacters,
                overlap,
                trim,
                chunkerType,
                embedding,
                preset,
                sizing,
                prependHeadingContext,
                topicThreshold
            );
        }
    }
    // CPD-ON
    public static ChunkingConfig defaultInstance() {
        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
    }
}