Merge pull request #73 from airslate-oss/ASP-28637-fix-nested-tables

[ASP-28637] fixed XWPFWordExtractor extracting nested table text

build.gradle

@@ -141,7 +141,7 @@ subprojects {
     apply plugin: 'de.thetaphi.forbiddenapis'
     apply plugin: 'com.github.spotbugs'
 
-    version = '5.2.3-AIRSLATE-26'
+    version = '5.2.3-AIRSLATE-26.2'
     ext {
         bouncyCastleVersion = '1.70'
         commonsCodecVersion = '1.15'

poi-ooxml/src/main/java/org/apache/poi/xwpf/usermodel/XWPFTableCell.java

@@ -541,13 +541,19 @@ private void appendBodyElementText(StringBuilder text, IBodyElement e, boolean i
         } else if (e instanceof XWPFTable) {
             XWPFTable eTable = (XWPFTable) e;
             for (XWPFTableRow row : eTable.getRows()) {
-                for (XWPFTableCell cell : row.getTableCells()) {
+                List<XWPFTableCell> tableCells = row.getTableCells();
+                for (int i = 0; i < tableCells.size(); i++) {
+                    XWPFTableCell cell = tableCells.get(i);
                     List<IBodyElement> localBodyElements = cell.getBodyElements();
-                    for (int i = 0; i < localBodyElements.size(); i++) {
-                        boolean localIsLast = (i == localBodyElements.size() - 1);
-                        appendBodyElementText(text, localBodyElements.get(i), localIsLast);
+                    for (int j = 0; j < localBodyElements.size(); j++) {
+                        boolean localIsLast = (j == localBodyElements.size() - 1);
+                        appendBodyElementText(text, localBodyElements.get(j), localIsLast);
+                    }
+                    if (i < tableCells.size() - 1) {
+                        text.append("\t");
                     }
                 }
+                text.append('\n');
             }
 
             if (!isLast) {

poi-ooxml/src/test/java/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java

@@ -27,10 +27,14 @@ Licensed to the Apache Software Foundation (ASF) under one or more
 import static org.junit.jupiter.api.Assertions.assertTrue;
 
 import java.io.IOException;
+import java.io.InputStream;
 import java.util.Locale;
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
+import org.apache.poi.extractor.POITextExtractor;
+import org.apache.poi.hwpf.extractor.Word6Extractor;
+import org.apache.poi.poifs.filesystem.POIFSFileSystem;
 import org.apache.poi.util.StringUtil;
 import org.apache.poi.xwpf.XWPFTestDataSamples;
 import org.apache.poi.xwpf.usermodel.XWPFDocument;
@@ -478,4 +482,41 @@ void bug55966() throws IOException  {
             assertEquals(expected, actual);
         }
     }
+
+    @Test
+    void testExtractorAddBreaksForNestedTable() throws Exception {
+        try (XWPFDocument doc = XWPFTestDataSamples.openSampleDocument("NestedTables.docx")) {
+            String expected = "Table 1\n" +
+                    "{{table(a)}}\t\n" +
+                    "{{h}}Name\t\n" +
+                    "{{name}}\t{{table(b)}}\t\n" +
+                    "{{h}}Product\tFillableField\n" +
+                    "{{Product}}\t{{t:t;r:y;l:\"text_field_1\";}} \n" +
+                    "{{endtable}}\t\n" +
+                    "\n" +
+                    "\n" +
+                    "{{endtable}}\t\n" +
+                    "\n" +
+                    "\n" +
+                    "\n" +
+                    "Table 2\n" +
+                    "{{table(t1)}}\t\n" +
+                    "{{table(t2)}}\n" +
+                    "{{name}}\n" +
+                    "{{endtable}}\n" +
+                    "\n" +
+                    "\t{{name}}\n" +
+                    "\t\n" +
+                    "{{endtable}}\t\n" +
+                    "\n" +
+                    "\n";
+
+            XWPFWordExtractor extractedDoc = new XWPFWordExtractor(doc);
+
+            String actual = extractedDoc.getText();
+
+            extractedDoc.close();
+            assertEquals(expected, actual);
+        }
+    }
 }