diff --git a/pom.xml b/pom.xml
index a5c01d4..d79948a 100644
--- a/pom.xml
+++ b/pom.xml
@@ -101,6 +101,12 @@
spring-ai-starter-vector-store-pgvector
1.0.0-M7
-->
+
+
+ org.springframework.ai
+ spring-ai-tika-document-reader
+ 1.0.0
+
com.baomidou
mybatis-plus-spring-boot3-starter
diff --git a/src/main/java/com/wok/supportbot/controller/AiController.java b/src/main/java/com/wok/supportbot/controller/AiController.java
index ef6c15b..1bd6f7e 100644
--- a/src/main/java/com/wok/supportbot/controller/AiController.java
+++ b/src/main/java/com/wok/supportbot/controller/AiController.java
@@ -18,7 +18,8 @@ import reactor.core.publisher.Flux;
import java.io.IOException;
-
+@RestController
+@RequestMapping("/ai")
public class AiController {
@Resource
diff --git a/src/main/java/com/wok/supportbot/controller/DocumentController.java b/src/main/java/com/wok/supportbot/controller/DocumentController.java
new file mode 100644
index 0000000..bcc7d34
--- /dev/null
+++ b/src/main/java/com/wok/supportbot/controller/DocumentController.java
@@ -0,0 +1,232 @@
+package com.wok.supportbot.controller;
+
+import com.wok.supportbot.document.extract.JsonDocumentLoader;
+import com.wok.supportbot.document.extract.MarkdownDocumentLoader;
+import com.wok.supportbot.document.extract.SimpleStringDocumentReader;
+import com.wok.supportbot.document.extract.TikaDocumentReader;
+import com.wok.supportbot.document.transform.MyKeywordEnricher;
+import com.wok.supportbot.document.transform.MyTokenTextSplitter;
+import org.springframework.ai.document.Document;
+import org.springframework.ai.vectorstore.VectorStore;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.http.ResponseEntity;
+import org.springframework.web.bind.annotation.*;
+import org.springframework.web.multipart.MultipartFile;
+
+import java.util.List;
+import java.util.Map;
+
+@RestController
+@RequestMapping("/document")
+public class DocumentController {
+
+ @Autowired
+ private TikaDocumentReader tikaDocumentReader;
+
+ @Autowired
+ private SimpleStringDocumentReader simpleStringDocumentReader;
+
+ @Autowired
+ private MarkdownDocumentLoader markdownDocumentLoader;
+
+ @Autowired
+ private JsonDocumentLoader jsonDocumentLoader;
+
+ @Autowired
+ private MyTokenTextSplitter myTokenTextSplitter;
+
+ @Autowired
+ private MyKeywordEnricher myKeywordEnricher;
+
+ @Autowired
+ private VectorStore pgVectorVectorStore;
+
+ /**
+ * 上传普通文件（支持多种格式），用 Tika 解析
+ */
+ @PostMapping("/upload/file")
+ public ResponseEntity> uploadFile(@RequestParam("file") MultipartFile file) {
+ try {
+ List documents = tikaDocumentReader.read(file);
+
+ // 拆分文档
+ List splitDocuments = myTokenTextSplitter.splitDocuments(documents);
+
+ // 添加元数据
+ List enrichedDocuments = myKeywordEnricher.enrichDocuments(splitDocuments);
+
+ // 转成向量并存入数据库
+ pgVectorVectorStore.add(enrichedDocuments);
+
+ return ResponseEntity.ok(Map.of(
+ "success", true,
+ "message", "文件上传并向量化成功",
+ "documentCount", enrichedDocuments.size()
+ ));
+ } catch (Exception e) {
+ return ResponseEntity.status(500).body(Map.of(
+ "success", false,
+ "message", "上传失败：" + e.getMessage()
+ ));
+ }
+ }
+
+ /**
+ * 上传字符串内容
+ */
+ @PostMapping("/upload/string")
+ public ResponseEntity> uploadString(@RequestBody String content) {
+ try {
+ List documents = simpleStringDocumentReader.read(content);
+
+ // 拆分文档
+ List splitDocuments = myTokenTextSplitter.splitDocuments(documents);
+
+ // 添加元数据
+ List enrichedDocuments = myKeywordEnricher.enrichDocuments(splitDocuments);
+
+ // 转成向量并存入数据库
+ pgVectorVectorStore.add(enrichedDocuments);
+
+ return ResponseEntity.ok(Map.of(
+ "success", true,
+ "message", "文本内容上传并向量化成功",
+ "documentCount", enrichedDocuments.size()
+ ));
+ } catch (Exception e) {
+ return ResponseEntity.status(500).body(Map.of(
+ "success", false,
+ "message", "上传失败：" + e.getMessage()
+ ));
+ }
+ }
+
+ /**
+ * 上传 Markdown 文件
+ */
+ @PostMapping("/upload/markdown")
+ public ResponseEntity> uploadMarkdown(@RequestParam("file") MultipartFile file) {
+ try {
+ List documents = markdownDocumentLoader.loadMarkdownFromFile(file);
+
+ // 拆分文档
+ List splitDocuments = myTokenTextSplitter.splitDocuments(documents);
+
+ // 添加元数据
+ List enrichedDocuments = myKeywordEnricher.enrichDocuments(splitDocuments);
+
+ // 转成向量并存入数据库
+ pgVectorVectorStore.add(enrichedDocuments);
+
+ return ResponseEntity.ok(Map.of(
+ "success", true,
+ "message", "Markdown文件上传并向量化成功",
+ "documentCount", enrichedDocuments.size()
+ ));
+ } catch (Exception e) {
+ return ResponseEntity.status(500).body(Map.of(
+ "success", false,
+ "message", "上传失败：" + e.getMessage()
+ ));
+ }
+ }
+
+ /**
+ * 上传 JSON 文件（基本方式）
+ * 把 JSON 根节点当成一个整体文档
+ */
+ @PostMapping("/upload/json/basic")
+ public ResponseEntity> uploadJsonBasic(@RequestParam("file") MultipartFile file) {
+ try {
+ List documents = jsonDocumentLoader.loadBasicJson(file);
+
+ // 拆分文档
+ List splitDocuments = myTokenTextSplitter.splitDocuments(documents);
+
+ // 添加元数据
+ List enrichedDocuments = myKeywordEnricher.enrichDocuments(splitDocuments);
+
+ // 转成向量并存入数据库
+ pgVectorVectorStore.add(enrichedDocuments);
+
+ return ResponseEntity.ok(Map.of(
+ "success", true,
+ "message", "JSON文件（基本方式）上传并向量化成功",
+ "documentCount", enrichedDocuments.size()
+ ));
+ } catch (Exception e) {
+ return ResponseEntity.status(500).body(Map.of(
+ "success", false,
+ "message", "上传失败：" + e.getMessage()
+ ));
+ }
+ }
+
+ /**
+ * 上传 JSON 文件（按字段提取）
+ * 用于提取指定字段文本
+ */
+ @PostMapping("/upload/json/fields")
+ public ResponseEntity> uploadJsonWithFields(
+ @RequestParam("file") MultipartFile file,
+ @RequestParam("fields") List fields) {
+ try {
+ List documents = jsonDocumentLoader.loadJsonByFields(file, fields.toArray(new String[0]));
+
+ // 拆分文档
+ List splitDocuments = myTokenTextSplitter.splitDocuments(documents);
+
+ // 添加元数据
+ List enrichedDocuments = myKeywordEnricher.enrichDocuments(splitDocuments);
+
+ // 转成向量并存入数据库
+ pgVectorVectorStore.add(enrichedDocuments);
+
+ return ResponseEntity.ok(Map.of(
+ "success", true,
+ "message", "JSON文件（按字段）上传并向量化成功",
+ "documentCount", enrichedDocuments.size(),
+ "extractedFields", fields
+ ));
+ } catch (Exception e) {
+ return ResponseEntity.status(500).body(Map.of(
+ "success", false,
+ "message", "上传失败：" + e.getMessage()
+ ));
+ }
+ }
+
+ /**
+ * 上传 JSON 文件（按指针拆分）
+ * 用于拆分数组元素，常用来分段成多文档
+ */
+ @PostMapping("/upload/json/pointer")
+ public ResponseEntity> uploadJsonWithPointer(
+ @RequestParam("file") MultipartFile file,
+ @RequestParam("pointer") String pointer) {
+ try {
+ List documents = jsonDocumentLoader.loadJsonByPointer(file, pointer);
+
+ // 拆分文档
+ List splitDocuments = myTokenTextSplitter.splitDocuments(documents);
+
+ // 添加元数据
+ List enrichedDocuments = myKeywordEnricher.enrichDocuments(splitDocuments);
+
+ // 转成向量并存入数据库
+ pgVectorVectorStore.add(enrichedDocuments);
+
+ return ResponseEntity.ok(Map.of(
+ "success", true,
+ "message", "JSON文件（按指针）上传并向量化成功",
+ "documentCount", enrichedDocuments.size(),
+ "pointer", pointer
+ ));
+ } catch (Exception e) {
+ return ResponseEntity.status(500).body(Map.of(
+ "success", false,
+ "message", "上传失败：" + e.getMessage()
+ ));
+ }
+ }
+}
diff --git a/src/main/java/com/wok/supportbot/document/extract/JsonDocumentLoader.java b/src/main/java/com/wok/supportbot/document/extract/JsonDocumentLoader.java
new file mode 100644
index 0000000..3fcdcd7
--- /dev/null
+++ b/src/main/java/com/wok/supportbot/document/extract/JsonDocumentLoader.java
@@ -0,0 +1,59 @@
+package com.wok.supportbot.document.extract;
+
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.ai.document.Document;
+import org.springframework.ai.reader.JsonReader;
+import org.springframework.core.io.FileSystemResource;
+import org.springframework.core.io.Resource;
+import org.springframework.stereotype.Component;
+import org.springframework.web.multipart.MultipartFile;
+
+import java.io.File;
+import java.io.IOException;
+import java.util.List;
+
+@Component
+@Slf4j
+public class JsonDocumentLoader {
+
+ /**
+ * 基本读取方式
+ */
+ public List loadBasicJson(MultipartFile file) {
+ Resource resource = toResource(file);
+ JsonReader reader = new JsonReader(resource);
+ return reader.get();
+ }
+
+ /**
+ * 指定字段读取方式（例如 description、features 字段）
+ */
+ public List loadJsonByFields(MultipartFile file, String... fields) {
+ Resource resource = toResource(file);
+ JsonReader reader = new JsonReader(resource, fields);
+ return reader.get();
+ }
+
+ /**
+ * 使用 JSON Pointer 提取数组路径内容（如 /items）
+ */
+ public List loadJsonByPointer(MultipartFile file, String pointer) {
+ Resource resource = toResource(file);
+ JsonReader reader = new JsonReader(resource);
+ return reader.get(pointer);
+ }
+
+ /**
+ * 将 MultipartFile 转换为 Resource
+ */
+ private Resource toResource(MultipartFile file) {
+ try {
+ File temp = File.createTempFile("upload-", file.getOriginalFilename());
+ file.transferTo(temp);
+ return new FileSystemResource(temp);
+ } catch (IOException e) {
+ log.error("JSON 文件转换失败", e);
+ throw new RuntimeException("JSON 文件转换失败", e);
+ }
+ }
+}
diff --git a/src/main/java/com/wok/supportbot/document/extract/MarkdownDocumentLoader.java b/src/main/java/com/wok/supportbot/document/extract/MarkdownDocumentLoader.java
index 494943d..e981088 100644
--- a/src/main/java/com/wok/supportbot/document/extract/MarkdownDocumentLoader.java
+++ b/src/main/java/com/wok/supportbot/document/extract/MarkdownDocumentLoader.java
@@ -2,52 +2,46 @@ package com.wok.supportbot.document.extract;
import lombok.extern.slf4j.Slf4j;
import org.springframework.ai.document.Document;
-
import org.springframework.ai.reader.markdown.MarkdownDocumentReader;
import org.springframework.ai.reader.markdown.config.MarkdownDocumentReaderConfig;
+import org.springframework.core.io.FileSystemResource;
import org.springframework.core.io.Resource;
-import org.springframework.core.io.support.ResourcePatternResolver;
import org.springframework.stereotype.Component;
+import org.springframework.web.multipart.MultipartFile;
+import java.io.File;
import java.io.IOException;
-import java.util.ArrayList;
import java.util.List;
/**
- * 文档加载器
+ * Markdown 文件上传解析器
*/
@Component
@Slf4j
public class MarkdownDocumentLoader {
- private final ResourcePatternResolver resourcePatternResolver;
+ public List loadMarkdownFromFile(MultipartFile file) {
+ try {
+ // 将 MultipartFile 保存为临时文件
+ File temp = File.createTempFile("upload-", file.getOriginalFilename());
+ file.transferTo(temp);
+ Resource resource = new FileSystemResource(temp);
- public MarkdownDocumentLoader(ResourcePatternResolver resourcePatternResolver) {
- this.resourcePatternResolver = resourcePatternResolver;
- }
+ // 配置文档解析
+ MarkdownDocumentReaderConfig config = MarkdownDocumentReaderConfig.builder()
+ .withHorizontalRuleCreateDocument(true)
+ .withIncludeCodeBlock(false)
+ .withIncludeBlockquote(false)
+ .withAdditionalMetadata("filename", file.getOriginalFilename())
+ .build();
+
+ // 读取文档内容
+ MarkdownDocumentReader reader = new MarkdownDocumentReader(resource, config);
+ return reader.get();
- /**
- * 加载多篇 Markdown 文档
- * @return
- */
- public List loadMarkdowns() {
- List allDocuments = new ArrayList<>();
- try {
- Resource[] resources = resourcePatternResolver.getResources("classpath:document/*.md");
- for (Resource resource : resources) {
- String filename = resource.getFilename();
- MarkdownDocumentReaderConfig config = MarkdownDocumentReaderConfig.builder()
- .withHorizontalRuleCreateDocument(true)
- .withIncludeCodeBlock(false)
- .withIncludeBlockquote(false)
- .withAdditionalMetadata("filename", filename)
- .build();
- MarkdownDocumentReader markdownDocumentReader = new MarkdownDocumentReader(resource, config);
- allDocuments.addAll(markdownDocumentReader.get());
- }
} catch (IOException e) {
- log.error("Markdown 文档加载失败", e);
+ log.error("Markdown 文件解析失败", e);
+ throw new RuntimeException("Markdown 文件解析失败", e);
}
- return allDocuments;
}
}
diff --git a/src/main/java/com/wok/supportbot/document/extract/MyJsonReader.java b/src/main/java/com/wok/supportbot/document/extract/MyJsonReader.java
deleted file mode 100644
index c7be227..0000000
--- a/src/main/java/com/wok/supportbot/document/extract/MyJsonReader.java
+++ /dev/null
@@ -1,37 +0,0 @@
-package com.wok.supportbot.document.extract;
-
-import org.springframework.ai.document.Document;
-import org.springframework.ai.reader.JsonReader;
-import org.springframework.beans.factory.annotation.Value;
-import org.springframework.core.io.Resource;
-import org.springframework.stereotype.Component;
-
-import java.util.List;
-
-// 从 classpath 下的 JSON 文件中读取文档
- @Component
- class MyJsonReader {
- private final Resource resource;
-
- MyJsonReader(@Value("classpath:products.json") Resource resource) {
- this.resource = resource;
- }
-
- // 基本用法
- List loadBasicJsonDocuments() {
- JsonReader jsonReader = new JsonReader(this.resource);
- return jsonReader.get();
- }
-
- // 指定使用哪些 JSON 字段作为文档内容
- List loadJsonWithSpecificFields() {
- JsonReader jsonReader = new JsonReader(this.resource, "description", "features");
- return jsonReader.get();
- }
-
- // 使用 JSON 指针精确提取文档内容
- List loadJsonWithPointer() {
- JsonReader jsonReader = new JsonReader(this.resource);
- return jsonReader.get("/items"); // 提取 items 数组内的内容
- }
- }
\ No newline at end of file
diff --git a/src/main/java/com/wok/supportbot/document/extract/SimpleStringDocumentReader.java b/src/main/java/com/wok/supportbot/document/extract/SimpleStringDocumentReader.java
new file mode 100644
index 0000000..b9d29cc
--- /dev/null
+++ b/src/main/java/com/wok/supportbot/document/extract/SimpleStringDocumentReader.java
@@ -0,0 +1,20 @@
+package com.wok.supportbot.document.extract;
+
+import org.springframework.ai.document.Document;
+import org.springframework.stereotype.Component;
+
+import java.util.Collections;
+import java.util.List;
+import java.util.UUID;
+
+@Component
+public class SimpleStringDocumentReader {
+
+ public List read(String content) {
+ Document doc = Document.builder()
+ .id(UUID.randomUUID().toString())
+ .text(content)
+ .build();
+ return Collections.singletonList(doc);
+ }
+}
diff --git a/src/main/java/com/wok/supportbot/document/extract/TikaDocumentReader.java b/src/main/java/com/wok/supportbot/document/extract/TikaDocumentReader.java
new file mode 100644
index 0000000..7ee7d2c
--- /dev/null
+++ b/src/main/java/com/wok/supportbot/document/extract/TikaDocumentReader.java
@@ -0,0 +1,44 @@
+package com.wok.supportbot.document.extract;
+
+import lombok.extern.slf4j.Slf4j;
+import org.apache.tika.Tika;
+import org.apache.tika.exception.TikaException;
+import org.springframework.ai.document.Document;
+import org.springframework.core.io.Resource;
+import org.springframework.core.io.FileSystemResource;
+import org.springframework.stereotype.Component;
+import org.springframework.web.multipart.MultipartFile;
+
+import java.io.File;
+import java.io.IOException;
+import java.util.Collections;
+import java.util.List;
+import java.util.UUID;
+
+@Component
+@Slf4j
+public class TikaDocumentReader {
+
+ public List read(MultipartFile file) {
+ try {
+ // MultipartFile 转 Resource
+ File tempFile = File.createTempFile("upload-", file.getOriginalFilename());
+ file.transferTo(tempFile);
+ Resource resource = new FileSystemResource(tempFile);
+
+ Tika tika = new Tika();
+ String text = tika.parseToString(resource.getInputStream());
+
+ Document doc = Document.builder()
+ .id(UUID.randomUUID().toString())
+ .text(text)
+ .build();
+
+ return Collections.singletonList(doc);
+
+ } catch (IOException | TikaException e) {
+ log.error("Tika 文件解析失败", e);
+ throw new RuntimeException("Tika 文件解析失败", e);
+ }
+ }
+}
diff --git a/src/main/java/com/wok/supportbot/document/transform/MyTokenTextSplitter.java b/src/main/java/com/wok/supportbot/document/transform/MyTokenTextSplitter.java
index 0584b96..3e13b15 100644
--- a/src/main/java/com/wok/supportbot/document/transform/MyTokenTextSplitter.java
+++ b/src/main/java/com/wok/supportbot/document/transform/MyTokenTextSplitter.java
@@ -10,7 +10,7 @@ import java.util.List;
* 自定义基于 Token 的切词器
*/
@Component
-class MyTokenTextSplitter {
+public class MyTokenTextSplitter {
/**
* 使用默认设置创建分割器。