SpringAI(GA)：PGvector向量数据库存储快速上手_org.springframework.ai.pgvector.hibernate.pgvector-优快云博客

[!TIP]
PGvector 是 PostgreSQL 的开源扩展，支持存储和搜索机器学习生成的扩展

实战代码可见：https://github.com/GTyingzi/spring-ai-tutorial 下的 vector/vector-pgvector 模块

Docker 启动 pgvector

推荐使用 ankane/pgvector 提供的官方支持镜像，它已经内置了 pgvector 扩展

docker pull ankane/pgvector:latest

启动容器

docker run -d \
  --name pgvector-db \
  -e POSTGRESPASSWORD=yingzipassword \
  -e POSTGRESUSER=yingziuser \
  -e POSTGRESDB=yingzidb \
  -p 5432:5432 \
  ankane/pgvector:latest

name：容器名称
POSTGRESPASSWORD：数据库密码
POSTGRESUSER：数据库用户
POSTGRESDB：初始化数据库名
p：映射主机端口到容器 PostgreSQL 端口

虽然镜像已包含 pgvector，但仍需在目标数据库中创建 pgvector 扩展：

docker exec -it pgvector-db psql -U yingziuser -d yingzidb

在 PostgreSQL 命令行中执行

CREATE EXTENSION IF NOT EXISTS vector;

校验示例

CREATE TABLE items (
  id bigserial PRIMARY KEY,
  embedding vector(3)  -- 3维向量，可根据需要调整维度
);

INSERT INTO items (embedding) VALUES ('[1,2,3]'), ('[4,5,6]');

-- 查询最相似的向量
SELECT * FROM items ORDER BY embedding <-> '[1,2,3]' LIMIT 5;

Spring AI 连接 PGvector

pom 依赖

<dependencies>
    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-web</artifactId>
    </dependency>

    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-autoconfigure-model-openai</artifactId>
    </dependency>

    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-vector-store</artifactId>
    </dependency>

    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-jdbc</artifactId>
    </dependency>

    <dependency>
        <groupId>org.postgresql</groupId>
        <artifactId>postgresql</artifactId>
        <scope>runtime</scope>
    </dependency>

    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-starter-vector-store-pgvector</artifactId>
    </dependency>

</dependencies>

application.yml

server:
  port: 8080

spring:
  application:
    name: vector-pgvector

  ai:
    openai:
      api-key: ${DASHSCOPEAPIKEY}
      base-url: https://dashscope.aliyuncs.com/compatible-mode
      embedding:
        options:
          model: text-embedding-v1

    vectorstore:
      pgvector:
        index-type: HNSW
        distance-type: COSINEDISTANCE
        dimensions: 1536
        max-document-batch-size: 10000 # Optional: Maximum number of documents per batch
---
spring:
  datasource:
    url: jdbc:postgresql://localhost:5432/yingzidb
    username: yingziuser
    password: yingzipassword

PgvectorConfig

package com.spring.ai.tutorial.vector.config;

import org.springframework.ai.embedding.EmbeddingModel;
import org.springframework.ai.vectorstore.pgvector.PgVectorStore;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.jdbc.core.JdbcTemplate;

/**
 * @author yingzi
 * @since 2025/9/3
 */
@Configuration
public class PgvectorConfig {

    @Value("${spring.ai.vectorstore.pgvector.index-type}")
    private PgVectorStore.PgIndexType indexType;
    @Value("${spring.ai.vectorstore.pgvector.distance-type}")
    private PgVectorStore.PgDistanceType distanceType;
    @Value("${spring.ai.vectorstore.pgvector.dimensions}")
    private int dimensions;
    @Value("${spring.ai.vectorstore.pgvector.max-document-batch-size}")
    private int maxDocumentBatchSize;

    @Bean("pgVectorStore")
    public PgVectorStore vectorStore(JdbcTemplate jdbcTemplate, EmbeddingModel embeddingModel) {
        return PgVectorStore.builder(jdbcTemplate, embeddingModel)
                .dimensions(dimensions)                    // Optional: defaults to model dimensions or 1536
                .distanceType(distanceType)       // Optional: defaults to COSINEDISTANCE
                .indexType(indexType)                     // Optional: defaults to HNSW
                .initializeSchema(true)              // Optional: defaults to false
                .schemaName("public")                // Optional: defaults to "public"
                .vectorTableName("vectorstore")     // Optional: defaults to "vectorstore"
                .maxDocumentBatchSize(maxDocumentBatchSize)         // Optional: defaults to 10000
                .build();
    }
}

PgvectorController

package com.spring.ai.tutorial.vector.controller;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.ai.document.Document;
import org.springframework.ai.vectorstore.SearchRequest;
import org.springframework.ai.vectorstore.pgvector.PgVectorStore;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.beans.factory.annotation.Qualifier;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RestController;

import java.util.HashMap;
import java.util.List;
import java.util.Map;

/**
 * @author yingzi
 * @since 2025/9/3
 */
@RestController
@RequestMapping("/vector/pgvector")
public class PgvectorController {

    private static final Logger logger = LoggerFactory.getLogger(PgvectorController.class);

    private final PgVectorStore pgVectorStore;

    @Autowired
    public PgvectorController(@Qualifier("pgVectorStore") PgVectorStore vectorStore) {
        this.pgVectorStore = vectorStore;
    }

    @GetMapping("/add")
    public void add() {
        logger.info("start add data");

        HashMap<String, Object> map = new HashMap<>();
        map.put("id", "12345");
        map.put("year", 2025);
        map.put("name", "yingzi");
        List<Document> documents = List.of(
                new Document("The World is Big and Salvation Lurks Around the Corner"),
                new Document("You walk forward facing the past and you turn back toward the future.", Map.of("year", 2024)),
                new Document("Spring AI rocks!! Spring AI rocks!! Spring AI rocks!! Spring AI rocks!! Spring AI rocks!!", map));
        pgVectorStore.add(documents);
    }

    @GetMapping("/search")
    public List<Document> search() {
        logger.info("start search data");
        return pgVectorStore.similaritySearch(SearchRequest
                .builder()
                .query("Spring")
                .topK(2)
                .build());
    }

}