commit 4cfd83f2a5ef12eb1d6acfebea3dda52aa926943
Author: maxcom <maxcom@jet.msk.su>
Date:   Fri Jul 24 15:15:29 2009 +0400

    docx: text extraction from deleted/inserted blocks

diff --git a/src/ooxml/java/org/apache/poi/xwpf/usermodel/XWPFParagraph.java b/src/ooxml/java/org/apache/poi/xwpf/usermodel/XWPFParagraph.java
index 81a4add..4906fcf 100644
--- a/src/ooxml/java/org/apache/poi/xwpf/usermodel/XWPFParagraph.java
+++ b/src/ooxml/java/org/apache/poi/xwpf/usermodel/XWPFParagraph.java
@@ -18,6 +18,7 @@ package org.apache.poi.xwpf.usermodel;
 
 import java.math.BigInteger;
 import java.util.ArrayList;
+import java.util.Arrays;
 
 import org.apache.xmlbeans.XmlCursor;
 import org.apache.xmlbeans.XmlObject;
@@ -56,23 +57,24 @@ public class XWPFParagraph {
             // TODO - replace this with some sort of XPath expression
             // to directly find all the CTRs, in the right order
             ArrayList<CTR> rs = new ArrayList<CTR>();
-            CTR[] tmp;
 
             // Get the main text runs
-            tmp = paragraph.getRArray();
-            for (int i = 0; i < tmp.length; i++) {
-                rs.add(tmp[i]);
-            }
+            rs.addAll(Arrays.asList(paragraph.getRArray()));
 
             // Not sure quite what these are, but they hold
             // more text runs
             CTSdtRun[] sdts = paragraph.getSdtArray();
             for (int i = 0; i < sdts.length; i++) {
                 CTSdtContentRun run = sdts[i].getSdtContent();
-                tmp = run.getRArray();
-                for (int j = 0; j < tmp.length; j++) {
-                    rs.add(tmp[j]);
-                }
+                rs.addAll(Arrays.asList(run.getRArray()));
+            }
+
+            for (CTRunTrackChange c : paragraph.getDelArray()) {
+                rs.addAll(Arrays.asList(c.getRArray()));
+            }
+
+            for (CTRunTrackChange c : paragraph.getInsArray()) {
+                rs.addAll(Arrays.asList(c.getRArray()));
             }
 
             // Get text of the paragraph
diff --git a/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java b/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
index 9a0e648..f35dfb0 100644
--- a/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
+++ b/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
@@ -191,6 +191,13 @@ public class TestXWPFWordExtractor extends TestCase {
         assertTrue(extractor.getText().contains("XXX"));
     }
 
+    public void testInsertedDeletedText() throws Exception {
+        XWPFDocument doc = open("delins.docx");
+        XWPFWordExtractor extractor = new XWPFWordExtractor(doc);
+
+        assertTrue(extractor.getText().contains("pendant worn"));
+        assertTrue(extractor.getText().contains("extremely well"));
+    }
 
     //TODO use the same logic for opening test files as in HSSFTestDataSamples
     private XWPFDocument open(String sampleFileName) throws IOException {
diff --git a/src/scratchpad/testcases/org/apache/poi/hwpf/data/delins.docx b/src/scratchpad/testcases/org/apache/poi/hwpf/data/delins.docx
new file mode 100644
index 0000000..b530691
Binary files /dev/null and b/src/scratchpad/testcases/org/apache/poi/hwpf/data/delins.docx differ