Fix bug #49446 - Don't consider 17.16.23 field codes as properly part of the paragraph's text

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@958965 13f79535-47bb-0310-9956-ffa450edef68

2 files changed, 20 insertions, 1 deletions

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 6aabadef95..e0741e782a 100644
--- a/src/ooxml/java/org/apache/poi/xwpf/usermodel/XWPFParagraph.java
+++ b/src/ooxml/java/org/apache/poi/xwpf/usermodel/XWPFParagraph.java
@@ -133,7 +133,13 @@ public class XWPFParagraph implements IBodyElement{
           while (c.toNextSelection()) {
               XmlObject o = c.getObject();
               if (o instanceof CTText) {
-                  text.append(((CTText) o).getStringValue());
+                  String tagName = o.getDomNode().getNodeName();
+                  // Field Codes (w:instrText, defined in spec sec. 17.16.23)
+                  //  come up as instances of CTText, but we don't want them
+                  //  in the normal text output
+                  if (!"w:instrText".equals(tagName)) {
+                     text.append(((CTText) o).getStringValue());
+                  }
               }
               if (o instanceof CTPTab) {
                   text.append("\t");
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 f98d5e69ca..2277ecec9d 100644
--- a/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
+++ b/src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
@@ -237,4 +237,17 @@ public class TestXWPFWordExtractor extends TestCase {
        // Now check the first paragraph in total
        assertTrue(extractor.getText().contains("a\tb\n"));
     }
+    
+    /**
+     * The output should not contain field codes, e.g. those specified in the
+     * w:instrText tag (spec sec. 17.16.23)
+     */
+    public void testNoFieldCodes() {
+        XWPFDocument doc = XWPFTestDataSamples.openSampleDocument("FieldCodes.docx");
+        XWPFWordExtractor extractor = new XWPFWordExtractor(doc);
+        String text = extractor.getText();
+        assertTrue(text.length() > 0);
+        assertFalse(text.contains("AUTHOR"));
+        assertFalse(text.contains("CREATEDATE"));
+    }
 }