Bug 55966: Include content control text in word extraction also if it is part of a paragraph

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@1875802 13f79535-47bb-0310-9956-ffa450edef68

2 files changed, 32 insertions, 14 deletions

diff --git a/src/ooxml/java/org/apache/poi/xwpf/extractor/XWPFWordExtractor.java b/src/ooxml/java/org/apache/poi/xwpf/extractor/XWPFWordExtractor.java
index cacac24323..4b61c09d19 100644
--- a/src/ooxml/java/org/apache/poi/xwpf/extractor/XWPFWordExtractor.java
+++ b/src/ooxml/java/org/apache/poi/xwpf/extractor/XWPFWordExtractor.java
@@ -90,7 +90,7 @@ public class XWPFWordExtractor extends POIXMLTextExtractor {
 
     /**
      * Should we concatenate phonetic runs in extraction.  Default is <code>true</code>
-     * @param concatenatePhoneticRuns
+     * @param concatenatePhoneticRuns If phonetic runs should be concatenated
      */
     public void setConcatenatePhoneticRuns(boolean concatenatePhoneticRuns) {
         this.concatenatePhoneticRuns = concatenatePhoneticRuns;
@@ -138,9 +138,10 @@ public class XWPFWordExtractor extends POIXMLTextExtractor {
             extractHeaders(text, headerFooterPolicy);
         }
 
-
-        for (IRunElement run : paragraph.getRuns()) {
-            if (! concatenatePhoneticRuns && run instanceof XWPFRun) {
+        for (IRunElement run : paragraph.getIRuns()) {
+            if (run instanceof XWPFSDT) {
+                text.append(((XWPFSDT) run).getContent().getText());
+            } else if (! concatenatePhoneticRuns && run instanceof XWPFRun) {
                 text.append(((XWPFRun)run).text());
             } else {
                 text.append(run);
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 ecab432c10..8a14b25ba0 100644
--- a/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
+++ b/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
@@ -17,6 +17,16 @@
 
 package org.apache.poi.xwpf.extractor;
 
+import org.apache.poi.util.StringUtil;
+import org.apache.poi.xwpf.XWPFTestDataSamples;
+import org.apache.poi.xwpf.usermodel.XWPFDocument;
+import org.junit.Test;
+
+import java.io.IOException;
+import java.util.Locale;
+import java.util.regex.Matcher;
+import java.util.regex.Pattern;
+
 import static org.apache.poi.POITestCase.assertContains;
 import static org.apache.poi.POITestCase.assertEndsWith;
 import static org.apache.poi.POITestCase.assertNotContained;
@@ -25,16 +35,6 @@ import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertTrue;
 
-import java.io.IOException;
-import java.util.Locale;
-import java.util.regex.Matcher;
-import java.util.regex.Pattern;
-
-import org.apache.poi.util.StringUtil;
-import org.apache.poi.xwpf.XWPFTestDataSamples;
-import org.apache.poi.xwpf.usermodel.XWPFDocument;
-import org.junit.Test;
-
 /**
  * Tests for HXFWordExtractor
  */
@@ -460,4 +460,21 @@ public class TestXWPFWordExtractor {
             assertContains(txt, "footer 1");
         }
     }
+
+    @Test
+    public void bug55966() throws IOException  {
+        try (XWPFDocument doc = XWPFTestDataSamples.openSampleDocument("55966.docx")) {
+            String expected = "Content control within a paragraph is here text content from within a paragraph second control with a new\n" +
+                    "line\n" +
+                    "\n" +
+                    "Content control that is the entire paragraph\n";
+
+            XWPFWordExtractor extractedDoc = new XWPFWordExtractor(doc);
+
+            String actual = extractedDoc.getText();
+
+            extractedDoc.close();
+            assertEquals(expected, actual);
+        }
+    }
 }