xieyinlu/ruoyi-ai
1
0
Fork 0
mirror of https://gitcode.com/ageerle/ruoyi-ai.git synced 2026-03-18 23:23:43 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat: Weaviate改为langchain4j方式调用

Browse Source
This commit is contained in:
ageer
2025-05-07 22:53:21 +08:00
parent 1a645c6e10
commit 81c0bb5738
9 changed files with 44 additions and 317 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
ruoyi-modules-api/ruoyi-knowledge-api/src/main/java/org/ruoyi/service/VectorStoreService.java
View File

@@ -2,9 +2,13 @@ package org.ruoyi.service;
import java.util.List;
/**
* @author ageer
* 向量库管理
*/
public interface VectorStoreService {
void storeEmbeddings(List<String> chunkList, String kid);
void storeEmbeddings(List<String> chunkList, String kid,String docId,List<String> fids);
void removeByDocId(String kid,String docId);

13
ruoyi-modules-api/ruoyi-knowledge-api/src/main/java/org/ruoyi/service/VectorizationService.java
View File

@@ -1,13 +0,0 @@
package org.ruoyi.service;
import java.util.List;
/**
* 文本向量化
*/
public interface VectorizationService {
List<List<Double>> batchVectorization(List<String> chunkList, String kid);
List<Double> singleVectorization(String chunk, String kid);
}

81
ruoyi-modules-api/ruoyi-knowledge-api/src/main/java/org/ruoyi/service/impl/WeaviateVectorStoreImpl.java
View File

@@ -1,76 +1,64 @@
package org.ruoyi.service.impl;
import cn.hutool.core.util.RandomUtil;
import dev.langchain4j.data.embedding.Embedding;
import dev.langchain4j.data.segment.TextSegment;
import dev.langchain4j.model.embedding.EmbeddingModel;
import dev.langchain4j.model.openai.OpenAiEmbeddingModel;
import dev.langchain4j.model.output.Response;
import dev.langchain4j.store.embedding.EmbeddingMatch;
import dev.langchain4j.store.embedding.EmbeddingSearchRequest;
import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.store.embedding.filter.Filter;
import dev.langchain4j.store.embedding.filter.comparison.IsEqualTo;
import dev.langchain4j.store.embedding.weaviate.WeaviateEmbeddingStore;
import jakarta.annotation.PostConstruct;
import jakarta.annotation.Resource;
import lombok.RequiredArgsConstructor;
import lombok.extern.slf4j.Slf4j;
import org.ruoyi.common.core.service.ConfigService;
import org.ruoyi.service.VectorStoreService;
import org.ruoyi.service.IKnowledgeInfoService;
import org.springframework.context.annotation.Lazy;
import org.springframework.stereotype.Service;
import org.testcontainers.weaviate.WeaviateContainer;
import static dev.langchain4j.model.openai.OpenAiEmbeddingModelName.TEXT_EMBEDDING_3_SMALL;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
/**
* @author ageer
* Weaviate 向量库管理
*/
@Service
@Slf4j
@RequiredArgsConstructor
public class WeaviateVectorStoreImpl implements VectorStoreService {
private volatile String protocol;
private volatile String host;
private volatile String className;
@Lazy
@Resource
private IKnowledgeInfoService knowledgeInfoService;
@Lazy
@Resource
private ConfigService configService;
private EmbeddingStore<TextSegment> embeddingStore;
@PostConstruct
public void loadConfig() {
this.protocol = configService.getConfigValue("weaviate", "protocol");
this.host = configService.getConfigValue("weaviate", "host");
this.className = configService.getConfigValue("weaviate", "classname");
}
private EmbeddingStore<TextSegment> embeddingStore;
private final ConfigService configService;
@Override
public List<String> getQueryVector(String query, String kid) {
EmbeddingModel embeddingModel = OpenAiEmbeddingModel.builder()
.apiKey(System.getenv("OPENAI_API_KEY"))
.baseUrl(System.getenv("OPENAI_BASE_URL"))
.modelName("text-embedding-3-small")
.apiKey("sk-xxx")
.baseUrl("https://api.pandarobot.chat/v1/")
.modelName(TEXT_EMBEDDING_3_SMALL)
.build();
Filter simpleFilter = new IsEqualTo("kid", kid);
// Filter simpleFilter = new IsEqualTo("kid", kid);
Embedding queryEmbedding = embeddingModel.embed("What is your favourite sport?").content();
// createSchema(kid);
Embedding queryEmbedding = embeddingModel.embed("聊天补全模型").content();
EmbeddingSearchRequest embeddingSearchRequest = EmbeddingSearchRequest.builder()
.queryEmbedding(queryEmbedding)
.maxResults(3)
.maxResults(2)
// 添加过滤条件
.filter(simpleFilter)
// .filter(simpleFilter)
.build();
List<EmbeddingMatch<TextSegment>> matches = embeddingStore.search(embeddingSearchRequest).matches();
List<String> results = new ArrayList<>();
matches.forEach(embeddingMatch -> {
@@ -82,10 +70,11 @@ public class WeaviateVectorStoreImpl implements VectorStoreService {
@Override
public void createSchema(String kid) {
WeaviateContainer weaviate = new WeaviateContainer(protocol);
weaviate.start();
String protocol = configService.getConfigValue("weaviate", "protocol");
String host = configService.getConfigValue("weaviate", "host");
String className = configService.getConfigValue("weaviate", "classname");
this.embeddingStore = WeaviateEmbeddingStore.builder()
.scheme("http")
.scheme(protocol)
.host(host)
.objectClass(className+kid)
.scheme(protocol)
@@ -95,25 +84,23 @@ public class WeaviateVectorStoreImpl implements VectorStoreService {
}
@Override
public void storeEmbeddings(List<String> chunkList,String kid) {
public void storeEmbeddings(List<String> chunkList,String kid,String docId,List<String> fids) {
EmbeddingModel embeddingModel = OpenAiEmbeddingModel.builder()
.apiKey(System.getenv("OPENAI_API_KEY"))
.baseUrl(System.getenv("OPENAI_BASE_URL"))
.modelName("text-embedding-3-small")
.apiKey("sk-xxxx")
.baseUrl("https://api.pandarobot.chat/v1/")
.modelName(TEXT_EMBEDDING_3_SMALL)
.build();
// 生成文档id
String docId = RandomUtil.randomString(10);
chunkList.forEach(chunk -> {
// 生成知识块id
String fid = RandomUtil.randomString(10);
Map<String, Object> dataSchema = new HashMap<>();
dataSchema.put("kid", kid);
dataSchema.put("docId", docId);
dataSchema.put("fid", fid);
dataSchema.put("fid", fids.get(0));
Response<Embedding> response = embeddingModel.embed(chunk);
Embedding embedding = response.content();
TextSegment segment = TextSegment.from(chunk);
segment.metadata().putAll(dataSchema);
Embedding content = embeddingModel.embed(segment).content();
embeddingStore.add(content);
embeddingStore.add(embedding,segment);
});
}