认证-上报库-word自定义导出: 解决表格不识别问题

jero-boot/jero-boot-module-certification/src/main/java/com/jero/modules/cert/report/service/impl/ParamsReportDetailEOServiceImpl.java

@@ -30,6 +30,7 @@ import com.jero.modules.cert.template.enums.ParamsIsMustEnum;
 import com.jero.modules.ocr.util.LineHumpUtil;
 import com.jero.modules.oss.entity.OSSFile;
 import com.jero.modules.oss.service.IOSSFileService;
+import com.jero.modules.project.util.Docx4jUtils;
 import com.jero.modules.project.util.ExcelUtil;
 import com.jero.modules.project.util.WordUtil;
 import com.jero.modules.split.common.ReadExcel;
@@ -48,6 +49,7 @@ import org.docx4j.Docx4J;
 import org.docx4j.openpackaging.exceptions.Docx4JException;
 import org.docx4j.openpackaging.packages.SpreadsheetMLPackage;
 import org.docx4j.openpackaging.packages.WordprocessingMLPackage;
+import org.docx4j.openpackaging.parts.WordprocessingML.MainDocumentPart;
 import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.beans.factory.annotation.Value;
 import org.springframework.mock.web.MockMultipartFile;
@@ -785,6 +787,8 @@ public class ParamsReportDetailEOServiceImpl extends ServiceImpl<ParamsReportDet
 			if (CosBootUtil.doesObjectExist(templateUrl)) {
 				InputStream wordTemplate = CosBootUtil.download(templateUrl);
 				WordprocessingMLPackage wordMLPackage = WordprocessingMLPackage.load(wordTemplate);
+				MainDocumentPart mainDocumentPart = wordMLPackage.getMainDocumentPart();
+				Docx4jUtils.cleanDocumentPartDouble(mainDocumentPart);
 
 				//下载关联文件内容
 				List<Map<String, Object>> fileListAll = (List<Map<String, Object>>) allParamsInfoList.get(2).get("fileListAll");
@@ -838,7 +842,7 @@ public class ParamsReportDetailEOServiceImpl extends ServiceImpl<ParamsReportDet
 
 				// 替换占位符内容
 				WordUtil.replacePictureBatch(wordMLPackage, imgListAll);
-				WordUtil.replaceVariable(wordMLPackage, params);
+				WordUtil.replaceVariableWithOutClear(wordMLPackage, params);
 				Docx4J.save(wordMLPackage, wordOS);
 
 				wordOS.flush();

jero-boot/jero-boot-modules/src/main/java/com/jero/modules/project/util/Docx4jUtils.java

@@ -100,6 +100,70 @@ public final class Docx4jUtils {
         return true;
     }
 
+    /**
+     * cleanDocumentPart
+     *
+     * @param documentPart
+     */
+    public static boolean cleanDocumentPartDouble(MainDocumentPart documentPart) throws Exception {
+        if (documentPart == null) {
+            return false;
+        }
+        Document document = documentPart.getContents();
+        String wmlTemplate =
+                XmlUtils.marshaltoString(document, true, false, Context.jc);
+        wmlTemplate = cleanXmlString(wmlTemplate);
+        document = (Document) XmlUtils.unwrap(DocxVariableClearUtils.doCleanDocumentPart(wmlTemplate, Context.jc));
+//        document = (Document) DocumentHelper.parseText(wmlTemplate);
+        documentPart.setContents(document);
+        return true;
+    }
+
+    public static String cleanXmlString(String string) {
+
+        StringBuilder appender = new StringBuilder();
+
+        //判断每一行数据中不包含'$'的数据先添加进新文件
+        if (!string.contains("$")) {
+            appender.append(string);
+            return appender.toString();
+        }
+        //如果一行数据中包含'$'变量符将替换为'#$'
+        string = string.replaceAll("\\$", "#\\$");
+        //然后以'#'切割成每一行（数组）,这样一来'$'都将在每一行的开头
+        String[] ss = string.split("#");
+
+        //遍历每一行（数组ss）
+        for (int i = 0; i < ss.length; i++) {
+            // 同一行的数据写到同一行,文件追加自动换行了(最后的完整数据)
+            StringBuilder sb = new StringBuilder();
+            //暂存数据
+            String s1 = ss[i];
+            //将不是以'$'开头的行数据放进StringBuilder
+            if (!s1.startsWith("$")) {
+                if (i > 0) {
+                    sb.append("#");
+                }
+                sb.append(s1);
+                appender.append(sb.toString());
+                continue;
+            }
+            //被分离的数据一般都是'${'这样被分开
+            //匹配以'$'开头的变量找到'}' 得到索引位置
+            int i1 = s1.indexOf("}");
+            //先切割得到这个完整体
+            String substr = s1.substring(0, i1 + 1);
+            //把变量追加到StringBuilder
+            sb.append(substr.replaceAll("\n","").replaceAll("<[^>]+>", ""));
+            //再将标签数据追加到StringBuilder包裹变量
+            sb.append(s1.substring(i1 + 1));
+
+            appender.append(sb.toString());
+        }
+
+        return appender.toString();
+    }
+
     /**
      * 清扫 docx4j 模板变量字符,通常以${variable}形式
      * <p>

jero-boot/jero-boot-modules/src/main/java/com/jero/modules/project/util/WordUtil.java

@@ -59,6 +59,11 @@ public class WordUtil {
         documentPart.variableReplace((HashMap<String, String>) map);
     }
 
+    public static void replaceVariableWithOutClear(WordprocessingMLPackage wordMLPackage, Map<String, String> map) throws Exception {
+        MainDocumentPart documentPart = wordMLPackage.getMainDocumentPart();
+        documentPart.variableReplace((HashMap<String, String>) map);
+    }
+
     public static void replaceTable(String templatePath, int tableNum, int rowNum, List<Map<String, Object>> dataList, String outPath) throws Exception {
         WordprocessingMLPackage wordMLPackage = WordprocessingMLPackage.load(new File(templatePath));
         MainDocumentPart mainDocumentPart = wordMLPackage.getMainDocumentPart();
@@ -203,7 +208,6 @@ public class WordUtil {
     public static void replacePictureBatch(WordprocessingMLPackage wordMLPackage, List<Map<String, Object>> picList) throws Exception {
         try {
             MainDocumentPart mainDocumentPart = wordMLPackage.getMainDocumentPart();
-            Docx4jUtils.cleanDocumentPart(mainDocumentPart);
             Document wmlDoc = (Document)mainDocumentPart.getJaxbElement();
             Body body = wmlDoc.getBody();
             List<Text> textList = getAllPlaceholderElementFromObject(body); // 获取文档中所有占位符
@@ -241,7 +245,6 @@ public class WordUtil {
      */
     public static Map<String, String> filterParams(WordprocessingMLPackage wordMLPackage, Map<String, String> params) throws Exception {
         MainDocumentPart mainDocumentPart = wordMLPackage.getMainDocumentPart();
-        Docx4jUtils.cleanDocumentPart(mainDocumentPart);
         Document wmlDoc = (Document)mainDocumentPart.getJaxbElement();
         Body body = wmlDoc.getBody();
         List<Text> textList = getAllPlaceholderElementFromObject(body); // 获取文档中所有占位符