Hanserwei/snails-ai-backend
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(document): 实现多格式文档上传与解析功能

Browse Source 
- 移除 AiChatController 中的 PDF 读取相关逻辑与依赖- 新增 DocumentController 支持文件上传接口
- 新增 DocumentIngestionService 接口及实现，负责文档处理流程
- 抽象 DocumentParser 接口统一各类文档解析器行为
- 重构所有具体文档读取器（PDF、HTML、JSON 等）实现新的解析接口- 引入 MultipartFileResource 工具类以适配 Spring AI 读取器
- 添加 DocumentUploadResponse 响应模型类
- 各文档读取器增加对文件扩展名和 MIME 类型的支持判断
This commit is contained in:
Hanserwei
2025-10-31 21:31:44 +08:00
parent 5ee2a0f11c
commit a9fce282ed
14 changed files with 269 additions and 102 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
snails-chat/src/main/java/com/hanserwei/chat/controller/AiChatController.java
View File

@@ -2,20 +2,15 @@ package com.hanserwei.chat.controller;
import com.hanserwei.chat.model.dto.ChatMessageDTO;
import com.hanserwei.chat.model.vo.AIResponse;
import com.hanserwei.chat.reader.MyPdfReader;
import com.hanserwei.chat.utils.ConversationContext;
import jakarta.annotation.Resource;
import lombok.extern.slf4j.Slf4j;
import org.springframework.ai.chat.client.ChatClient;
import org.springframework.ai.chat.memory.ChatMemory;
import org.springframework.ai.document.Document;
import org.springframework.ai.vectorstore.VectorStore;
import org.springframework.http.MediaType;
import org.springframework.web.bind.annotation.*;
import reactor.core.publisher.Flux;
import java.util.List;
@Slf4j
@RestController
@RequestMapping("/ai")
@@ -23,10 +18,6 @@ public class AiChatController {
@Resource
private ChatClient dashScopeChatClient;
@Resource
private MyPdfReader myPdfReader;
@Resource
private VectorStore vectorStore;
@PostMapping(path = "/chat", produces = MediaType.TEXT_EVENT_STREAM_VALUE)
public Flux<AIResponse> chatWithAi(@RequestBody ChatMessageDTO chatMessageDTO) {
@@ -44,11 +35,4 @@ public class AiChatController {
.contextWrite(ctx -> ConversationContext.withConversationId(chatMessageDTO.getConversionId()))
.doFinally(signalType -> ConversationContext.clear());
}
@GetMapping("/readpdf")
public String readPdf() {
List<Document> docsFromPdf = myPdfReader.getDocsFromPdf();
vectorStore.add(docsFromPdf);
return "ok!";
}
}

29
snails-chat/src/main/java/com/hanserwei/chat/controller/DocumentController.java Normal file
View File

@@ -0,0 +1,29 @@
package com.hanserwei.chat.controller;
import com.hanserwei.chat.model.vo.DocumentUploadResponse;
import com.hanserwei.chat.service.DocumentIngestionService;
import lombok.RequiredArgsConstructor;
import lombok.extern.slf4j.Slf4j;
import org.springframework.http.MediaType;
import org.springframework.http.ResponseEntity;
import org.springframework.web.bind.annotation.PostMapping;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;
import org.springframework.web.multipart.MultipartFile;
@Slf4j
@RestController
@RequestMapping("/documents")
@RequiredArgsConstructor
public class DocumentController {
private final DocumentIngestionService documentIngestionService;
@PostMapping(value = "/upload", consumes = MediaType.MULTIPART_FORM_DATA_VALUE)
public ResponseEntity<DocumentUploadResponse> upload(@RequestParam("file") MultipartFile file) {
int documentCount = documentIngestionService.ingest(file);
log.info("文件 {} 上传成功。", file.getOriginalFilename());
return ResponseEntity.ok(new DocumentUploadResponse(file.getOriginalFilename(), documentCount, "ok"));
}
}

4
snails-chat/src/main/java/com/hanserwei/chat/model/vo/DocumentUploadResponse.java Normal file
View File

@@ -0,0 +1,4 @@
package com.hanserwei.chat.model.vo;
public record DocumentUploadResponse(String filename, int documentCount, String message) {
}

59
snails-chat/src/main/java/com/hanserwei/chat/reader/DocumentParser.java Normal file
View File

@@ -0,0 +1,59 @@
package com.hanserwei.chat.reader;
import org.springframework.ai.document.Document;
import org.springframework.lang.Nullable;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
/**
* 将上传文件转换为Spring AI {@link Document}实例的策略接口。
*/
public interface DocumentParser {
/**
* 判断此解析器是否可以处理提供的文件。
*
* @param filename 客户端提供的原始文件名（可能为{@code null}）
* @param contentType 从上传中派生的MIME类型（可能为{@code null}）
* @return 如果此解析器应该处理该文件则返回{@code true}
*/
boolean supports(@Nullable String filename, @Nullable String contentType);
/**
* 将文件解析为Spring AI文档。
*
* @param file 多部分上传载荷
* @return 解析后的文档列表
*/
List<Document> parse(MultipartFile file);
default boolean hasExtension(@Nullable String filename, String... extensions) {
if (filename == null) {
return false;
}
int index = filename.lastIndexOf('.');
if (index < 0 || index == filename.length() - 1) {
return false;
}
String actualExtension = filename.substring(index + 1);
for (String extension : extensions) {
if (actualExtension.equalsIgnoreCase(extension)) {
return true;
}
}
return false;
}
default boolean matchesContentType(@Nullable String contentType, String... supportedContentTypes) {
if (contentType == null) {
return false;
}
for (String supportedContentType : supportedContentTypes) {
if (contentType.equalsIgnoreCase(supportedContentType)) {
return true;
}
}
return false;
}
}

36
snails-chat/src/main/java/com/hanserwei/chat/reader/MultipartFileResource.java Normal file
View File

@@ -0,0 +1,36 @@
package com.hanserwei.chat.reader;
import org.springframework.core.io.ByteArrayResource;
import org.springframework.web.multipart.MultipartFile;
import java.io.IOException;
import java.io.UncheckedIOException;
import java.util.Objects;
/**
* 简单的 {@link ByteArrayResource}，用于保留多部分文件的文件名。
*/
final class MultipartFileResource extends ByteArrayResource {
private final String filename;
private MultipartFileResource(byte[] byteArray, String filename) {
super(byteArray);
this.filename = filename;
}
static MultipartFileResource of(MultipartFile file) {
try {
String originalFilename = Objects.requireNonNullElse(file.getOriginalFilename(), "upload");
return new MultipartFileResource(file.getBytes(), originalFilename);
}
catch (IOException ex) {
throw new UncheckedIOException("读取多部分文件内容失败", ex);
}
}
@Override
public String getFilename() {
return filename;
}
}

21
snails-chat/src/main/java/com/hanserwei/chat/reader/MyHtmlReader.java
View File

@@ -3,32 +3,35 @@ package com.hanserwei.chat.reader;
import org.springframework.ai.document.Document;
import org.springframework.ai.reader.jsoup.JsoupDocumentReader;
import org.springframework.ai.reader.jsoup.config.JsoupDocumentReaderConfig;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.core.io.Resource;
import org.springframework.stereotype.Component;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Component
public class MyHtmlReader {
public class MyHtmlReader implements DocumentParser {
@Value("classpath:/document/my-page.html")
private Resource resource;
public List<Document> loadHtml() {
@Override
public List<Document> parse(MultipartFile file) {
// JsoupDocumentReader 阅读器配置类
JsoupDocumentReaderConfig config = JsoupDocumentReaderConfig.builder()
.selector("article p") // 提取 <article> 标签内的 p 段落
.charset("UTF-8") // 使用 UTF-8 编码
.includeLinkUrls(true) // 在元数据中包含链接 URL（绝对链接）
.metadataTags(List.of("author", "date")) // 提取 author 和 date 元标签
.additionalMetadata("source", "my-page.html") // 添加自定义元数据
.additionalMetadata("source", file.getOriginalFilename()) // 添加自定义元数据
.build();
// 新建 JsoupDocumentReader 阅读器
JsoupDocumentReader reader = new JsoupDocumentReader(resource, config);
JsoupDocumentReader reader = new JsoupDocumentReader(MultipartFileResource.of(file), config);
// 读取并转换为 Document 文档集合
return reader.get();
}
@Override
public boolean supports(String filename, String contentType) {
return hasExtension(filename, "html", "htm") ||
matchesContentType(contentType, "text/html", "application/xhtml+xml");
}
}

18
snails-chat/src/main/java/com/hanserwei/chat/reader/MyJsonReader.java
View File

@@ -2,26 +2,28 @@ package com.hanserwei.chat.reader;
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 org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Component
public class MyJsonReader {
@Value("classpath:/document/tv.json")
private Resource resource;
public class MyJsonReader implements DocumentParser {
/**
* 读取 Json 文件
* @return
*/
public List<Document> loadJson() {
@Override
public List<Document> parse(MultipartFile file) {
// 创建 JsonReader 阅读器实例，配置需要读取的字段
JsonReader jsonReader = new JsonReader(resource, "description", "content", "title");
JsonReader jsonReader = new JsonReader(MultipartFileResource.of(file), "description", "content", "title");
// 执行读取操作，并转换为 Document 对象集合
return jsonReader.get();
}
@Override
public boolean supports(String filename, String contentType) {
return hasExtension(filename, "json") || matchesContentType(contentType, "application/json");
}
}

21
snails-chat/src/main/java/com/hanserwei/chat/reader/MyMarkdownReader.java
View File

@@ -3,31 +3,34 @@ package com.hanserwei.chat.reader;
import org.springframework.ai.document.Document;
import org.springframework.ai.reader.markdown.MarkdownDocumentReader;
import org.springframework.ai.reader.markdown.config.MarkdownDocumentReaderConfig;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.core.io.Resource;
import org.springframework.stereotype.Component;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Component
public class MyMarkdownReader {
public class MyMarkdownReader implements DocumentParser {
@Value("classpath:/document/code.md")
private Resource resource;
public List<Document> loadMarkdown() {
@Override
public List<Document> parse(MultipartFile file) {
// MarkdownDocumentReader 阅读器配置类
MarkdownDocumentReaderConfig config = MarkdownDocumentReaderConfig.builder()
.withHorizontalRuleCreateDocument(true) // 遇到水平线 ---，则创建新文档
.withIncludeCodeBlock(false) // 排除代码块（代码块生成单独文档）
.withIncludeBlockquote(false) // 排除块引用（块引用生成单独文档）
.withAdditionalMetadata("filename", "code.md") // 添加自定义元数据，如文件名称
.withAdditionalMetadata("filename", file.getOriginalFilename()) // 添加自定义元数据，如文件名称
.build();
// 新建 MarkdownDocumentReader 阅读器
MarkdownDocumentReader reader = new MarkdownDocumentReader(resource, config);
MarkdownDocumentReader reader = new MarkdownDocumentReader(MultipartFileResource.of(file), config);
// 读取并转换为 Document 文档集合
return reader.get();
}
@Override
public boolean supports(String filename, String contentType) {
return hasExtension(filename, "md", "markdown") ||
matchesContentType(contentType, "text/markdown", "text/x-markdown");
}
}

22
snails-chat/src/main/java/com/hanserwei/chat/reader/MyPdfReader.java
View File

@@ -5,24 +5,30 @@ import org.springframework.ai.reader.ExtractedTextFormatter;
import org.springframework.ai.reader.pdf.PagePdfDocumentReader;
import org.springframework.ai.reader.pdf.config.PdfDocumentReaderConfig;
import org.springframework.stereotype.Component;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Component
public class MyPdfReader {
public class MyPdfReader implements DocumentParser {
public List<Document> getDocsFromPdf() {
// 新建 PagePdfDocumentReader 阅读器
PagePdfDocumentReader pdfReader = new PagePdfDocumentReader("classpath:/document/profile.pdf", // PDF 文件路径
@Override
public List<Document> parse(MultipartFile file) {
PagePdfDocumentReader pdfReader = new PagePdfDocumentReader(
MultipartFileResource.of(file),
PdfDocumentReaderConfig.builder()
.withPageTopMargin(0) // 设置页面顶边距为0
.withPageTopMargin(0)
.withPageExtractedTextFormatter(ExtractedTextFormatter.builder()
.withNumberOfTopTextLinesToDelete(0) // 设置删除顶部文本行数为0
.withNumberOfTopTextLinesToDelete(0)
.build())
.withPagesPerDocument(1) // 设置每个文档包含1页
.withPagesPerDocument(1)
.build());
// 读取并转换为 Document 文档集合
return pdfReader.read();
}
@Override
public boolean supports(String filename, String contentType) {
return hasExtension(filename, "pdf") || matchesContentType(contentType, "application/pdf");
}
}

38
snails-chat/src/main/java/com/hanserwei/chat/reader/MyTextReader.java
View File

@@ -2,47 +2,27 @@ package com.hanserwei.chat.reader;
import org.springframework.ai.document.Document;
import org.springframework.ai.reader.TextReader;
import org.springframework.ai.transformer.splitter.TokenTextSplitter;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.core.io.Resource;
import org.springframework.stereotype.Component;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Component
public class MyTextReader {
@Value("classpath:/document/manual.txt")
private Resource resource;
public class MyTextReader implements DocumentParser {
/**
* 读取 Txt 文档
* @return 读取的文档集合
*/
public List<Document> loadText() {
// 创建 TextReader 对象，用于读取指定资源 (resource) 的文本内容
TextReader textReader = new TextReader(resource);
// 添加自定义元数据，如文件名称
textReader.getCustomMetadata()
.put("filename", "manual.txt");
// 读取并转换为 Document 文档集合
@Override
public List<Document> parse(MultipartFile file) {
TextReader textReader = new TextReader(MultipartFileResource.of(file));
textReader.getCustomMetadata().put("filename", file.getOriginalFilename());
return textReader.read();
}
/**
* 读取 Txt 文档并分块拆分
* @return 文档分块集合
*/
public List<Document> loadTextAndSplit() {
// 创建 TextReader 对象，用于读取指定资源 (resource) 的文本内容
TextReader textReader = new TextReader(resource);
// 将资源内容解析为 Document 对象集合
List<Document> documents = textReader.get();
// 使用 TokenTextSplitter 对文档列表进行分块处理
// 返回拆分后的文档分块集合
return new TokenTextSplitter().apply(documents);
@Override
public boolean supports(String filename, String contentType) {
return hasExtension(filename, "txt") || matchesContentType(contentType, "text/plain");
}
}

21
snails-chat/src/main/java/com/hanserwei/chat/reader/MyTikaPptReader.java
View File

@@ -3,21 +3,18 @@ package com.hanserwei.chat.reader;
import org.springframework.ai.document.Document;
import org.springframework.ai.reader.tika.TikaDocumentReader;
import org.springframework.ai.transformer.splitter.TokenTextSplitter;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.core.io.Resource;
import org.springframework.stereotype.Component;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Component
public class MyTikaPptReader {
public class MyTikaPptReader implements DocumentParser {
@Value("classpath:/document/XX牌云感变频空调说明书.pptx")
private Resource resource;
public List<Document> loadPpt() {
@Override
public List<Document> parse(MultipartFile file) {
// 新建 TikaDocumentReader 阅读器
TikaDocumentReader tikaDocumentReader = new TikaDocumentReader(resource);
TikaDocumentReader tikaDocumentReader = new TikaDocumentReader(MultipartFileResource.of(file));
// 读取并转换为 Document 文档集合
List<Document> documents = tikaDocumentReader.get();
@@ -26,4 +23,12 @@ public class MyTikaPptReader {
TokenTextSplitter splitter = new TokenTextSplitter(1000, 400, 10, 5000, true);
return splitter.apply(documents);
}
@Override
public boolean supports(String filename, String contentType) {
return hasExtension(filename, "ppt", "pptx") ||
matchesContentType(contentType,
"application/vnd.ms-powerpoint",
"application/vnd.openxmlformats-officedocument.presentationml.presentation");
}
}

21
snails-chat/src/main/java/com/hanserwei/chat/reader/MyTikaWordReader.java
View File

@@ -3,21 +3,18 @@ package com.hanserwei.chat.reader;
import org.springframework.ai.document.Document;
import org.springframework.ai.reader.tika.TikaDocumentReader;
import org.springframework.ai.transformer.splitter.TokenTextSplitter;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.core.io.Resource;
import org.springframework.stereotype.Component;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Component
public class MyTikaWordReader {
public class MyTikaWordReader implements DocumentParser {
@Value("classpath:/document/55f79946a0964b89bc7ab9b55e4a49ff.docx")
private Resource resource;
public List<Document> loadWord() {
@Override
public List<Document> parse(MultipartFile file) {
// 新建 TikaDocumentReader 阅读器
TikaDocumentReader tikaDocumentReader = new TikaDocumentReader(resource);
TikaDocumentReader tikaDocumentReader = new TikaDocumentReader(MultipartFileResource.of(file));
// 读取并转换为 Document 文档集合
List<Document> documents = tikaDocumentReader.get();
@@ -25,4 +22,12 @@ public class MyTikaWordReader {
TokenTextSplitter splitter = new TokenTextSplitter(); // 不设置任何构造参数，表示使用默认设置
return splitter.apply(documents);
}
@Override
public boolean supports(String filename, String contentType) {
return hasExtension(filename, "doc", "docx") ||
matchesContentType(contentType,
"application/msword",
"application/vnd.openxmlformats-officedocument.wordprocessingml.document");
}
}

8
snails-chat/src/main/java/com/hanserwei/chat/service/DocumentIngestionService.java Normal file
View File

@@ -0,0 +1,8 @@
package com.hanserwei.chat.service;
import org.springframework.web.multipart.MultipartFile;
public interface DocumentIngestionService {
int ingest(MultipartFile file);
}

43
snails-chat/src/main/java/com/hanserwei/chat/service/impl/DocumentIngestionServiceImpl.java Normal file
View File

@@ -0,0 +1,43 @@
package com.hanserwei.chat.service.impl;
import com.hanserwei.chat.reader.DocumentParser;
import com.hanserwei.chat.service.DocumentIngestionService;
import lombok.RequiredArgsConstructor;
import lombok.extern.slf4j.Slf4j;
import org.springframework.ai.document.Document;
import org.springframework.ai.vectorstore.VectorStore;
import org.springframework.stereotype.Service;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
@Slf4j
@Service
@RequiredArgsConstructor
public class DocumentIngestionServiceImpl implements DocumentIngestionService {
private final List<DocumentParser> documentParsers;
private final VectorStore vectorStore;
@Override
public int ingest(MultipartFile file) {
if (file == null || file.isEmpty()) {
throw new IllegalArgumentException("上传文件不能为空");
}
DocumentParser parser = documentParsers.stream()
.filter(candidate -> candidate.supports(file.getOriginalFilename(), file.getContentType()))
.findFirst()
.orElseThrow(() -> new IllegalArgumentException("不支持的文件类型：" + file.getOriginalFilename()));
List<Document> documents = parser.parse(file);
if (documents.isEmpty()) {
log.warn("文件 {} 解析后未生成任何文档，跳过入库。", file.getOriginalFilename());
return 0;
}
vectorStore.add(documents);
log.info("文件 {} 入库成功，共写入 {} 条向量。", file.getOriginalFilename(), documents.size());
return documents.size();
}
}