LangChain4j PGVector

<dependency>
    <groupId>dev.langchain4j</groupId>
    <artifactId>langchain4j-pgvector</artifactId>
    <version>1.11.0</version>
</dependency>

implementation 'dev.langchain4j:langchain4j-pgvector:1.11.0'

import dev.langchain4j.store.embedding.pgvector.PgVectorEmbeddingStore;
import dev.langchain4j.store.embedding.pgvector.PgVectorEmbeddingStore.SearchMode;
import dev.langchain4j.store.embedding.pgvector.MetadataStorageConfig;
import dev.langchain4j.store.embedding.pgvector.MetadataStorageMode;
import dev.langchain4j.store.embedding.pgvector.DefaultMetadataStorageConfig;
import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.store.embedding.EmbeddingSearchRequest;
import dev.langchain4j.store.embedding.EmbeddingSearchResult;
import dev.langchain4j.data.embedding.Embedding;
import dev.langchain4j.data.segment.TextSegment;

import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.data.segment.TextSegment;
import dev.langchain4j.store.embedding.pgvector.PgVectorEmbeddingStore;

// Create embedding store with minimal configuration
EmbeddingStore<TextSegment> embeddingStore = PgVectorEmbeddingStore.builder()
    .host("localhost")
    .port(5432)
    .database("postgres")
    .user("my_user")
    .password("my_password")
    .table("my_embeddings")
    .dimension(384)  // Must match your embedding model's dimension
    .build();

// Add embedding
String id = embeddingStore.add(embedding);

// Add embedding with text segment
String id2 = embeddingStore.add(embedding, textSegment);

// Search for similar embeddings
EmbeddingSearchRequest request = EmbeddingSearchRequest.builder()
    .queryEmbedding(queryEmbedding)
    .maxResults(5)
    .minScore(0.7)
    .build();

EmbeddingSearchResult<TextSegment> results = embeddingStore.search(request);

import javax.sql.DataSource;
import com.zaxxer.hikari.HikariConfig;
import com.zaxxer.hikari.HikariDataSource;

// Configure connection pool
HikariConfig config = new HikariConfig();
config.setJdbcUrl("jdbc:postgresql://localhost:5432/postgres");
config.setUsername("my_user");
config.setPassword("my_password");
config.setMaximumPoolSize(10);

DataSource dataSource = new HikariDataSource(config);

// Create embedding store with DataSource
EmbeddingStore<TextSegment> embeddingStore = PgVectorEmbeddingStore.datasourceBuilder()
    .datasource(dataSource)
    .table("my_embeddings")
    .dimension(384)
    .build();

/**
 * Creates a builder for PgVectorEmbeddingStore with individual connection parameters
 * @return PgVectorEmbeddingStoreBuilder instance
 */
public static PgVectorEmbeddingStoreBuilder builder();

/**
 * Creates a builder for PgVectorEmbeddingStore with DataSource
 * @return DatasourceBuilder instance
 */
public static DatasourceBuilder datasourceBuilder();

/**
 * Adds an embedding to the store with auto-generated ID
 * @param embedding The embedding to be added
 * @return The auto-generated ID
 */
String add(Embedding embedding);

/**
 * Adds an embedding with text segment to the store
 * @param embedding The embedding to be added
 * @param textSegment The original content that was embedded
 * @return The auto-generated ID
 */
String add(Embedding embedding, TextSegment textSegment);

/**
 * Adds multiple embeddings to the store
 * @param embeddings List of embeddings to be added
 * @return List of auto-generated IDs
 */
List<String> addAll(List<Embedding> embeddings);

/**
 * Removes a single embedding by its ID
 * @param id The ID of the embedding to remove
 */
void remove(String id);

/**
 * Removes all embeddings from the store
 */
void removeAll();

/**
 * Removes embeddings by their IDs
 * @param ids Collection of embedding IDs to remove
 */
void removeAll(Collection<String> ids);

/**
 * Searches for the most similar embeddings
 * @param request Search request containing query embedding, filters, and parameters
 * @return Search results with matching embeddings
 */
EmbeddingSearchResult<TextSegment> search(EmbeddingSearchRequest request);

/**
 * Metadata storage mode enumeration
 */
enum MetadataStorageMode {
    /** For static metadata when you know the list of keys in advance */
    COLUMN_PER_KEY,
    /** For dynamic metadata stored as JSON */
    COMBINED_JSON,
    /** For dynamic metadata stored as binary JSON (optimized for queries) */
    COMBINED_JSONB
}

/**
 * Creates a default metadata storage configuration
 * @return Default configuration with COMBINED_JSON mode
 */
static MetadataStorageConfig defaultConfig();

/**
 * Search mode enumeration for PgVectorEmbeddingStore
 */
enum SearchMode {
    /** Standard vector similarity search using cosine distance */
    VECTOR,
    /** Combines vector search with full-text keyword search using Reciprocal Rank Fusion (RRF) */
    HYBRID
}

/**
 * PGVector EmbeddingStore Implementation
 * Only cosine similarity is used for vector distance
 * Only IVFFlat index type is supported
 * Implements EmbeddingStore<TextSegment> interface
 */
class PgVectorEmbeddingStore implements EmbeddingStore<TextSegment> {
    // Main implementation - see capabilities for methods
}