With help from Yegor, fix bug #45537 - Include headers and footers (of slides and notes) in the extracted text from HSLF

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@683020 13f79535-47bb-0310-9956-ffa450edef68

src/documentation/content/xdocs/changes.xml

@@ -37,6 +37,7 @@
 
 		<!-- Don't forget to update status.xml too! -->
         <release version="3.1.1-alpha1" date="2008-??-??">
+           <action dev="POI-DEVELOPERS" type="add">45537 - Include headers and footers (of slides and notes) in the extracted text from HSLF</action>
            <action dev="POI-DEVELOPERS" type="fix">45472 - Fixed incorrect default row height in OpenOffice 2.3</action>
            <action dev="POI-DEVELOPERS" type="fix">44692 - HSSFPicture.resize() stretched image when there was a text next to it</action>
            <action dev="POI-DEVELOPERS" type="add">45543 - Optionally extract comment text with PowerPointExtractor, and initial hslf model support for comments</action>

src/documentation/content/xdocs/status.xml

@@ -34,6 +34,7 @@
 	<!-- Don't forget to update changes.xml too! -->
     <changes>
         <release version="3.1.1-alpha1" date="2008-??-??">
+           <action dev="POI-DEVELOPERS" type="add">45537 - Include headers and footers (of slides and notes) in the extracted text from HSLF</action>
            <action dev="POI-DEVELOPERS" type="fix">45472 - Fixed incorrect default row height in OpenOffice 2.3</action>
            <action dev="POI-DEVELOPERS" type="fix">44692 - HSSFPicture.resize() stretched image when there was a text next to it</action>
            <action dev="POI-DEVELOPERS" type="add">45543 - Optionally extract comment text with PowerPointExtractor, and initial hslf model support for comments</action>

src/scratchpad/src/org/apache/poi/hslf/extractor/PowerPointExtractor.java

@@ -176,11 +176,13 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 		for(int i=0; i<_slides.length; i++) {
 			Slide slide = _slides[i];
 			
+			// Slide header, if set
 			HeadersFooters hf = slide.getHeadersFooters();
-			if(hf != null && hf.getHeaderText() != null) {
+			if(hf != null && hf.isHeaderVisible() && hf.getHeaderText() != null) {
 				ret.append(hf.getHeaderText() + "\n");
 			}
 			
+			// Slide text
 			TextRun[] runs = slide.getTextRuns();
 			for(int j=0; j<runs.length; j++) {
 				TextRun run = runs[j];
@@ -193,10 +195,12 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 				}
 			}
 			
-			if(hf != null && hf.getFooterText() != null) {
+			// Slide footer, if set
+			if(hf != null && hf.isFooterVisible() && hf.getFooterText() != null) {
 				ret.append(hf.getFooterText() + "\n");
 			}
 			
+			// Comments, if requested and present
 			if(getCommentText) {
 				Comment[] comments = slide.getComments();
 				for(int j=0; j<comments.length; j++) {
@@ -219,6 +223,8 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 		//  master sheets in. Grab Slide list, then work from there,
 		//  but ensure no duplicates
 		HashSet seenNotes = new HashSet();
+		HeadersFooters hf = _show.getNotesHeadersFooters();
+		
 		for(int i=0; i<_slides.length; i++) {
 			Notes notes = _slides[i].getNotesSheet();
 			if(notes == null) { continue; }
@@ -226,6 +232,12 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 			if(seenNotes.contains(id)) { continue; }
 			seenNotes.add(id);
 			
+			// Repeat the Notes header, if set
+			if(hf != null && hf.isHeaderVisible() && hf.getHeaderText() != null) {
+				ret.append(hf.getHeaderText() + "\n");
+			}
+
+			// Notes text
 			TextRun[] runs = notes.getTextRuns();
 			if(runs != null && runs.length > 0) {
 				for(int j=0; j<runs.length; j++) {
@@ -237,6 +249,11 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 					}
 				}
 			}
+			
+			// Repeat the notes footer, if set
+			if(hf != null && hf.isFooterVisible() && hf.getFooterText() != null) {
+				ret.append(hf.getFooterText() + "\n");
+			}
 		}
 	}
 

src/scratchpad/src/org/apache/poi/hslf/model/HeadersFooters.java

@@ -23,6 +23,8 @@ import org.apache.poi.hslf.usermodel.SlideShow;
 /**
  * Header / Footer settings.
  * 
+ * You can get these on slides, or across all notes
+ *
  * @author Yegor Kozlov
  */
 public class HeadersFooters {

src/scratchpad/testcases/org/apache/poi/hslf/extractor/TextExtractor.java

@@ -253,32 +253,42 @@ public class TextExtractor extends TestCase {
     /**
      * From bug #45537
      */
-    public void DISABLEDtestHeaderFooter() throws Exception {
+    public void testHeaderFooter() throws Exception {
 		String filename, text;
 		
-		// With a header
+		// With a header on the notes
 		filename = dirname + "/45537_Header.ppt";
 		HSLFSlideShow hslf = new HSLFSlideShow(new FileInputStream(filename));
 		SlideShow ss = new SlideShow(hslf);
-		assertNotNull(ss.getSlides()[0].getHeadersFooters());
+		assertNotNull(ss.getNotesHeadersFooters());
-		assertEquals("testdoc test phrase", ss.getSlides()[0].getHeadersFooters().getHeaderText());
+		assertEquals("testdoc test phrase", ss.getNotesHeadersFooters().getHeaderText());
 		
 		ppe = new PowerPointExtractor(hslf);
 
+		text = ppe.getText();
+		assertFalse("Unable to find expected word in text\n" + text, text.contains("testdoc"));
+        assertFalse("Unable to find expected word in text\n" + text, text.contains("test phrase"));
+        
+        ppe.setNotesByDefault(true);
 		text = ppe.getText();
 		assertTrue("Unable to find expected word in text\n" + text, text.contains("testdoc"));
         assertTrue("Unable to find expected word in text\n" + text, text.contains("test phrase"));
 
         
-		// And with a footer
+		// And with a footer, also on notes
 		filename = dirname + "/45537_Footer.ppt";
 		hslf = new HSLFSlideShow(new FileInputStream(filename));
 		ss = new SlideShow(hslf);
-		assertNotNull(ss.getSlides()[0].getHeadersFooters());
+		assertNotNull(ss.getNotesHeadersFooters());
-		assertEquals("testdoc test phrase", ss.getSlides()[0].getHeadersFooters().getFooterText());
+		assertEquals("testdoc test phrase", ss.getNotesHeadersFooters().getFooterText());
 		
 		ppe = new PowerPointExtractor(filename);
 
+		text = ppe.getText();
+		assertFalse("Unable to find expected word in text\n" + text, text.contains("testdoc"));
+        assertFalse("Unable to find expected word in text\n" + text, text.contains("test phrase"));
+
+        ppe.setNotesByDefault(true);
 		text = ppe.getText();
 		assertTrue("Unable to find expected word in text\n" + text, text.contains("testdoc"));
         assertTrue("Unable to find expected word in text\n" + text, text.contains("test phrase"));