支持ppt格式

src/main/java/com/zzsn/knowbase/service/impl/KnowledgeServiceImpl.java

@@ -27,7 +27,15 @@ import org.apache.pdfbox.pdmodel.PDDocument;
 import org.apache.pdfbox.pdmodel.PDPage;
 import org.apache.pdfbox.text.PDFTextStripper;
 import org.apache.pdfbox.text.TextPosition;
+import org.apache.poi.hslf.usermodel.HSLFShape;
+import org.apache.poi.hslf.usermodel.HSLFSlide;
+import org.apache.poi.hslf.usermodel.HSLFSlideShow;
+import org.apache.poi.hslf.usermodel.HSLFTextShape;
 import org.apache.poi.ss.usermodel.Workbook;
+import org.apache.poi.xslf.usermodel.XMLSlideShow;
+import org.apache.poi.xslf.usermodel.XSLFShape;
+import org.apache.poi.xslf.usermodel.XSLFSlide;
+import org.apache.poi.xslf.usermodel.XSLFTextShape;
 import org.apache.poi.xssf.usermodel.XSSFWorkbook;
 import org.elasticsearch.action.search.SearchRequest;
 import org.elasticsearch.action.search.SearchResponse;
@@ -175,19 +183,69 @@ class KnowledgeServiceImpl implements IKnowledgeService {
                         pdfTextStripper.setStartPage(0);
                         pdfTextStripper.setEndPage(document.getNumberOfPages());
                         String text = pdfTextStripper.getText(document);
-                        knowledge.setContentAll(text);
-                        log.info("allBuilder:{}", allBuilder);
+                        //knowledge.setContentAll(text);
+                        //log.info("allBuilder:{}", allBuilder.toString().substring(1000));
                         contentStringList = Arrays.asList(allBuilder.toString().split("😀"));
                     } catch (Exception e) {
                         log.error("parsing pdf error :{}", e.getMessage());
                     }
+                } else if(".pptx".equals(knowFile.getFileType())){
+                    knowledge.setImportData(2);
+                    filePath = path;
+                    log.info("pptx dealing filePath{}", filePath);
+                    // 加载PPT文件
+                    XMLSlideShow ppt = new XMLSlideShow(new FileInputStream(filePath));
+                    // 遍历幻灯片
+                    for (XSLFSlide slide : ppt.getSlides()) {
+                        StringBuffer sb = new StringBuffer();
+                        // 遍历形状
+                        List<XSLFShape> shapes = slide.getShapes();
+                        for (XSLFShape shape : shapes) {
+                            // 检查形状是否包含文本
+                            if (shape instanceof XSLFTextShape) {
+                                XSLFTextShape textShape = (XSLFTextShape) shape;
+                                String text = textShape.getText();
+                                if (text != null) {
+                                    sb.append(text);
+                                    log.info("text :{}", text);
+                                }
+                            }
+                        }
+                        contentStringList.add(sb.toString());
+                    }
+                    ppt.close(); // 关闭PPT
+                } else if(".ppt".equals(knowFile.getFileType())){
+                    knowledge.setImportData(2);
+                    filePath = path;
+                    log.info("pptx dealing filePath{}", filePath);
+                    // 加载PPT文件
+                    HSLFSlideShow ppt = new HSLFSlideShow(new FileInputStream(filePath));
+                    // 遍历幻灯片
+                    for (HSLFSlide slide : ppt.getSlides()) {
+                        StringBuffer sb = new StringBuffer();
+                        // 遍历形状
+                        List<HSLFShape> shapes = slide.getShapes();
+                        for (HSLFShape shape : shapes) {
+                            // 检查形状是否包含文本
+                            if (shape instanceof HSLFTextShape) {
+                                HSLFTextShape textShape = (HSLFTextShape) shape;
+                                String text = textShape.getText();
+                                if (text != null) {
+                                    sb.append(text);
+                                    log.info("text :{}", text);
+                                }
+                            }
+                        }
+                        contentStringList.add(sb.toString());
+                    }
+                    ppt.close(); // 关闭PPT
                 } else {
                     filePath = path;
                     File file = new File(filePath);
 
                     log.info("word dealing filePath{}", filePath);
                     html = DocUtil.convertDocStream2Html(new FileInputStream(file));
-                    log.info("html :{}", html);
+                    //log.info("html :{}", html);
                     String htmlWithTable = html.replace("</p>", "######</p>");
                     htmlWithTable = htmlWithTable.replace("</title>", "######</title>");
                     htmlWithTable = htmlWithTable.replace("</h1>", "######</h1>");
@@ -264,7 +322,9 @@ class KnowledgeServiceImpl implements IKnowledgeService {
             knowledge.setKnowledgeProjectId(knowledgeVO.getKbKnowledgeId());
         }
         deleteForPython(knowledge.getId(), knowledge.getKnowledgeProjectId());
-        log.info("send message:{}", knowledgeMessage);
+        //log.info("send message:{}", knowledgeMessage);
+        String msg = JSON.toJSONString(knowledgeMessage);
+        log.info(msg);
         produceInfo.sendKnowledgeContents(knowledgeMessage);
     }