windows上调通了将pdf转成图片，再使用ocr识别。

electrmangnetic/pom.xml

@@ -85,10 +85,11 @@
             </exclusions>
             <version>5.1.2</version>
         </dependency>
+        <!--引入druid数据源-->
         <dependency>
             <groupId>com.alibaba</groupId>
-            <artifactId>druid</artifactId>
+            <artifactId>druid-spring-boot-starter</artifactId>
-            <version>1.1.10</version>
+            <version>1.2.6</version>
         </dependency>
 
         <dependency>

electrmangnetic/src/main/java/com/electromagnetic/industry/software/manage/config/DruidDataSourceConfig.java

@@ -0,0 +1,23 @@
+package com.electromagnetic.industry.software.manage.config;
+
+import com.alibaba.druid.pool.DruidDataSource;
+import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
+import org.springframework.boot.context.properties.ConfigurationProperties;
+import org.springframework.context.annotation.Bean;
+import org.springframework.context.annotation.Configuration;
+
+import javax.sql.DataSource;
+
+@Configuration
+public class DruidDataSourceConfig {
+
+    /**
+     * 添加 DruidDataSource 组件到容器中，并绑定属性
+     */
+    @Bean
+    @ConfigurationProperties(prefix = "spring.datasource")
+    @ConditionalOnProperty(name = "spring.datasource.type", havingValue = "com.alibaba.druid.pool.DruidDataSource")
+    public DataSource druid(){
+        return new DruidDataSource();
+    }
+}

electrmangnetic/src/main/java/com/electromagnetic/industry/software/manage/service/serviceimpl/ChatService.java

@@ -146,11 +146,10 @@ public class ChatService {
                     }
                     pageFiles = OfficeFileUtil.parsePdfByPage(pdfPath);
                 }
-                for (PageFile pageFile : pageFiles) {
+                StringBuilder stringBuilder = new StringBuilder();
-                    Map<String, Object> metaData = Map.of("page_number", pageFile.getPageNumber(), "file_name", file.getOriginalFilename());
+                pageFiles.forEach(page -> stringBuilder.append(page.getContent()));
-                    Document document = new Document(pageFile.getContent(), metaData);
+                Document document = new Document(stringBuilder.toString(), Map.of("file_name", file.getOriginalFilename()));
-                    documents.add(document);
+                documents.add(document);
-                }
                 FileUtil.del(pdfPath);
                 FileUtil.del(srcPath);
             }

electromagnetic-common/pom.xml

@@ -115,13 +115,11 @@
             <artifactId>rapidocr-onnx-platform</artifactId>
             <version>0.0.7</version>
         </dependency>
-
         <dependency>
             <groupId>io.github.mymonstercat</groupId>
-            <artifactId>rapidocr-ncnn-platform</artifactId>
+            <artifactId>rapidocr-onnx-linux-x86_64</artifactId>
-            <version>0.0.7</version>
+            <version>1.2.2</version>
         </dependency>
-
         <dependency>
             <groupId>org.apache.pdfbox</groupId>
             <artifactId>jbig2-imageio</artifactId>

electromagnetic-common/src/main/java/com/electromagnetic/industry/software/common/util/OfficeFileUtil.java

@@ -15,11 +15,8 @@ import io.github.mymonstercat.Model;
 import io.github.mymonstercat.ocr.InferenceEngine;
 import lombok.extern.slf4j.Slf4j;
 import org.apache.pdfbox.Loader;
-import org.apache.pdfbox.cos.COSName;
 import org.apache.pdfbox.pdmodel.PDDocument;
-import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.rendering.PDFRenderer;
-import org.apache.pdfbox.pdmodel.PDResources;
-import org.apache.pdfbox.pdmodel.graphics.image.PDImage;
 import org.apache.pdfbox.text.PDFTextStripper;
 import org.apache.poi.hslf.usermodel.HSLFSlideShow;
 import org.apache.poi.hwpf.HWPFDocument;
@@ -288,73 +285,28 @@ public class OfficeFileUtil {
         return pageFiles;
     }
 
-    public static List<PageFile> parsePdfByPage(String filePath) throws IOException {
+    public static List<PageFile> parsePdfByPage(String pdfPath) throws IOException {
-        StringBuilder stringBuilder = new StringBuilder();
         List<PageFile> pageFiles = new ArrayList<>();
-        String fileName = new File(filePath).getName();
+        String fileName = new File(pdfPath).getName();
-        try (PDDocument document = Loader.loadPDF(new File(filePath))) {
+        try (PDDocument document = Loader.loadPDF(new File(pdfPath))) {
-            // 创建 PDFTextStripper 对象
+            PDFRenderer renderer = new PDFRenderer(document);
-            PDFTextStripper textStripper = new PDFTextStripper();
             // 遍历每一页
-            for (int pageIndex = 0; pageIndex < document.getNumberOfPages(); pageIndex++) {
+            for (int page = 0; page < document.getNumberOfPages(); page++) {
-                PDPage page = document.getPage(pageIndex);
+                // 渲染页面为图像（设置缩放比例，300 dpi）
-                int pageNumber = pageIndex + 1;
+                BufferedImage image = renderer.renderImageWithDPI(page, 300);
-                // 提取文本
+                // 生成输出路径
-                String text = extractTextFromPage(textStripper, document, pageIndex);
+                String imagePath = UserThreadLocal.getUser().getPrjTmpDir() + File.separator + IdUtil.fastSimpleUUID() + ".png";
-                // 提取图片
+                // 保存图像
-                String imageText = extractImagesFromPage(page, pageNumber);
+                ImageIO.write(image, "png", new File(imagePath));
-                stringBuilder.append(text).append("\n").append(imageText);
+                String content = getTextFromImage(imagePath);
-                PageFile pageFile = new PageFile(pageIndex, stringBuilder.toString(), fileName);
+                pageFiles.add(new PageFile(page, content, fileName));
-                pageFiles.add(pageFile);
+                FileUtil.del(imagePath);
             }
-        } catch (IOException e) {
-            log.error(e.getMessage(), e);
         }
         return pageFiles;
     }
 
-
+    private static String getTextFromImage(String imagePath) {
-    // 提取文本
-    private static String extractTextFromPage(PDFTextStripper textStripper, PDDocument document, int pageIndex) throws IOException {
-        textStripper.setStartPage(pageIndex + 1);  // PDFBox 页码从 1 开始
-        textStripper.setEndPage(pageIndex + 1);
-        return textStripper.getText(document);
-    }
-
-    // 提取图片
-    private static String extractImagesFromPage(PDPage page, int pageNumber) throws IOException {
-        PDResources resources = page.getResources();
-        if (resources == null) {
-            return "";
-        }
-        // 遍历所有 XObject 资源
-        Iterable<COSName> xObjectNames = resources.getXObjectNames();
-        StringBuilder stringBuilder = new StringBuilder();
-        for (COSName name : xObjectNames) {
-            // 获取 XObject 对象
-            org.apache.pdfbox.pdmodel.graphics.PDXObject xObject = resources.getXObject(name);
-            // 检查是否为图片对象
-            if (xObject instanceof PDImage) {
-                PDImage image = (PDImage) xObject;
-                String filePath = StrFormatter.format("D:/{}.png", IdUtil.fastSimpleUUID());
-                saveImage(image, filePath);
-                String textFromImage = getTextFromImage(filePath);
-                stringBuilder.append(textFromImage).append("\r\n");
-                FileUtil.del(filePath);
-            }
-        }
-        return stringBuilder.toString();
-    }
-
-    // 保存图片
-    private static void saveImage(PDImage image, String filename) throws IOException {
-        BufferedImage bufferedImage = image.getImage();
-        File outputFile = new File(filename);
-        outputFile.getParentFile().mkdirs();  // 确保目录存在
-        ImageIO.write(bufferedImage, "png", outputFile);
-    }
-
-    private static String getTextFromImage(String imagePath) throws IOException {
         InferenceEngine engine = InferenceEngine.getInstance(Model.ONNX_PPOCR_V3);
         OcrResult ocrResult = engine.runOcr(imagePath);
         return ocrResult.getStrRes().trim();