fix for extraction paragraphs and sections from headers/footers with XWPFWordExtractor, see Bugzilla 47727

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@809662 13f79535-47bb-0310-9956-ffa450edef68

src/documentation/content/xdocs/status.xml

@@ -33,7 +33,8 @@
 
     <changes>
         <release version="3.5-beta7" date="2009-??-??">
-           <action dev="POI-DEVELOPERS" type="fix">47773 - Support for extraction of header / footer images in HWPF</action>
+           <action dev="POI-DEVELOPERS" type="fix">47773 - Fix for extraction paragraphs and sections from headers/footers with XWPFWordExtractor</action>
+           <action dev="POI-DEVELOPERS" type="fix">47727 - Support for extraction of header / footer images in HWPF</action>
            <action dev="POI-DEVELOPERS" type="fix">moved all test data to a top-level directory</action>
            <action dev="POI-DEVELOPERS" type="add">47721 - Added implementation for INDIRECT()</action>
            <action dev="POI-DEVELOPERS" type="add">45583 - Avoid exception when reading ClipboardData packet in OLE property sets</action>

src/ooxml/java/org/apache/poi/xwpf/extractor/XWPFWordExtractor.java

@@ -21,6 +21,7 @@ import java.util.Iterator;
 
 import org.apache.poi.POIXMLDocument;
 import org.apache.poi.POIXMLTextExtractor;
+import org.apache.poi.POIXMLException;
 import org.apache.poi.openxml4j.exceptions.OpenXML4JException;
 import org.apache.poi.openxml4j.opc.OPCPackage;
 import org.apache.poi.xwpf.model.XWPFCommentsDecorator;
@@ -31,6 +32,7 @@ import org.apache.poi.xwpf.usermodel.XWPFDocument;
 import org.apache.poi.xwpf.usermodel.XWPFParagraph;
 import org.apache.poi.xwpf.usermodel.XWPFTable;
 import org.apache.xmlbeans.XmlException;
+import org.openxmlformats.schemas.wordprocessingml.x2006.main.CTSectPr;
 
 /**
  * Helper class to extract text from an OOXML Word file
@@ -74,43 +76,75 @@ public class XWPFWordExtractor extends POIXMLTextExtractor {
 		XWPFHeaderFooterPolicy hfPolicy = document.getHeaderFooterPolicy();
 
 		// Start out with all headers
-		// TODO - put them in where they're needed
+                extractHeaders(text, hfPolicy);
-		if(hfPolicy.getFirstPageHeader() != null) {
-			text.append( hfPolicy.getFirstPageHeader().getText() );
-		}
-		if(hfPolicy.getEvenPageHeader() != null) {
-			text.append( hfPolicy.getEvenPageHeader().getText() );
-		}
-		if(hfPolicy.getDefaultHeader() != null) {
-			text.append( hfPolicy.getDefaultHeader().getText() );
-		}
 		
 		// First up, all our paragraph based text
 		Iterator<XWPFParagraph> i = document.getParagraphsIterator();
 		while(i.hasNext()) {
-			XWPFParagraphDecorator decorator = new XWPFCommentsDecorator(
+                        XWPFParagraph paragraph = i.next();
-					new XWPFHyperlinkDecorator(i.next(), null, fetchHyperlinks));
+
-			text.append(decorator.getText()+"\n");
+
-		}
+                        try {
+                                CTSectPr ctSectPr = null;
+                                if (paragraph.getCTP().getPPr()!=null) {
+                                        ctSectPr = paragraph.getCTP().getPPr().getSectPr();
+                                }
+
+                                XWPFHeaderFooterPolicy headerFooterPolicy = null;
+
+                                if (ctSectPr!=null) {
+                                        headerFooterPolicy = new XWPFHeaderFooterPolicy(document, ctSectPr);
+
+                                        extractHeaders(text, headerFooterPolicy);
+                                }
+
+                                XWPFParagraphDecorator decorator = new XWPFCommentsDecorator(
+                                                new XWPFHyperlinkDecorator(paragraph, null, fetchHyperlinks));
+                                text.append(decorator.getText()).append('\n');
+
+                                if (ctSectPr!=null) {
+                                        extractFooters(text, headerFooterPolicy);
+                                }
+                        } catch (IOException e) {
+                                throw new POIXMLException(e);
+                        } catch (XmlException e) {
+                                throw new POIXMLException(e);
+                        }
+                }
 
 		// Then our table based text
 		Iterator<XWPFTable> j = document.getTablesIterator();
 		while(j.hasNext()) {
-			text.append(j.next().getText()+"\n");
+                        text.append(j.next().getText()).append('\n');
 		}
 		
 		// Finish up with all the footers
-		// TODO - put them in where they're needed
+                extractFooters(text, hfPolicy);
-		if(hfPolicy.getFirstPageFooter() != null) {
-			text.append( hfPolicy.getFirstPageFooter().getText() );
-		}
-		if(hfPolicy.getEvenPageFooter() != null) {
-			text.append( hfPolicy.getEvenPageFooter().getText() );
-		}
-		if(hfPolicy.getDefaultFooter() != null) {
-			text.append( hfPolicy.getDefaultFooter().getText() );
-		}
 		
 		return text.toString();
 	}
+
+        private void extractFooters(StringBuffer text, XWPFHeaderFooterPolicy hfPolicy) {
+                if(hfPolicy.getFirstPageFooter() != null) {
+                        text.append( hfPolicy.getFirstPageFooter().getText() );
+                }
+                if(hfPolicy.getEvenPageFooter() != null) {
+                        text.append( hfPolicy.getEvenPageFooter().getText() );
+                }
+                if(hfPolicy.getDefaultFooter() != null) {
+                        text.append( hfPolicy.getDefaultFooter().getText() );
+                }
+        }
+
+        private void extractHeaders(StringBuffer text, XWPFHeaderFooterPolicy hfPolicy) {
+                if(hfPolicy.getFirstPageHeader() != null) {
+                        text.append( hfPolicy.getFirstPageHeader().getText() );
+                }
+                if(hfPolicy.getEvenPageHeader() != null) {
+                        text.append( hfPolicy.getEvenPageHeader().getText() );
+                }
+                if(hfPolicy.getDefaultHeader() != null) {
+                        text.append( hfPolicy.getDefaultHeader().getText() );
+                }
+        }
 }

src/ooxml/java/org/apache/poi/xwpf/model/XWPFHeaderFooterPolicy.java

@@ -83,19 +83,26 @@ public class XWPFHeaderFooterPolicy {
 	private XWPFHeader defaultHeader;
 	private XWPFFooter defaultFooter;
 	
+        /**
+         * Figures out the policy for the given document,
+         *  and creates any header and footer objects
+         *  as required.
+         */
+        public XWPFHeaderFooterPolicy(XWPFDocument doc) throws IOException, XmlException {
+                this(doc, doc.getDocument().getBody().getSectPr());
+        }
 
 	/**
 	 * Figures out the policy for the given document,
 	 *  and creates any header and footer objects
 	 *  as required.
 	 */
-	public XWPFHeaderFooterPolicy(XWPFDocument doc) throws IOException, XmlException {
+	public XWPFHeaderFooterPolicy(XWPFDocument doc, CTSectPr sectPr) throws IOException, XmlException {
 		// Grab what headers and footers have been defined
 		// For now, we don't care about different ranges, as it
 		//  doesn't seem that .docx properly supports that
 		//  feature of the file format yet
 		this.doc = doc;
-		CTSectPr sectPr = doc.getDocument().getBody().getSectPr();
 		for(int i=0; i<sectPr.sizeOfHeaderReferenceArray(); i++) {
 			// Get the header
 			CTHdrFtrRef ref = sectPr.getHeaderReferenceArray(i);

src/ooxml/testcases/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java

@@ -198,4 +198,13 @@ public class TestXWPFWordExtractor extends TestCase {
         assertTrue(extractor.getText().contains("extremely well"));
     }
 
+    public void testParagraphHeader() {
+        XWPFDocument doc = XWPFTestDataSamples.openSampleDocument("Headers.docx");
+        XWPFWordExtractor extractor = new XWPFWordExtractor(doc);
+
+        assertTrue(extractor.getText().contains("Section 1"));
+        assertTrue(extractor.getText().contains("Section 2"));
+        assertTrue(extractor.getText().contains("Section 3"));
+    }
+
 }