Fix bug #45543 - Optionally extract comment text with PowerPointExtractor, and initial hslf model support for comments

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@682533 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">45543 - Optionally extract comment text with PowerPointExtractor, and initial hslf model support for comments</action>
            <action dev="POI-DEVELOPERS" type="fix">45538 - Include excel headers and footers in the output of ExcelExtractor</action>
            <action dev="POI-DEVELOPERS" type="fix">44894 - refactor duplicate logic from EventRecordFactory to RecordFactory</action>
            <action dev="POI-DEVELOPERS" type="add">Support for Headers / Footers in HSLF</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">45543 - Optionally extract comment text with PowerPointExtractor, and initial hslf model support for comments</action>
            <action dev="POI-DEVELOPERS" type="fix">45538 - Include excel headers and footers in the output of ExcelExtractor</action>
            <action dev="POI-DEVELOPERS" type="fix">44894 - refactor duplicate logic from EventRecordFactory to RecordFactory</action>
            <action dev="POI-DEVELOPERS" type="add">Support for Headers / Footers in HSLF</action>

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

@@ -27,6 +27,8 @@ import org.apache.poi.POIOLE2TextExtractor;
 import org.apache.poi.poifs.filesystem.POIFSFileSystem;
 import org.apache.poi.hslf.*;
 import org.apache.poi.hslf.model.*;
+import org.apache.poi.hslf.record.Comment2000;
+import org.apache.poi.hslf.record.Record;
 import org.apache.poi.hslf.usermodel.*;
 
 /**
@@ -44,6 +46,7 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 	
 	private boolean slidesByDefault = true;
 	private boolean notesByDefault = false;
+	private boolean commentsByDefault = false;
 
   /**
    * Basic extractor. Returns all the text, and optionally all the notes
@@ -57,16 +60,20 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 	}
 
 	boolean notes = false;
+	boolean comments = false;
 	String file;
 	if(args.length > 1) {
 		notes = true;
 		file = args[1];
+		if(args.length > 2) {
+			comments = true;
+		}
 	} else {
 		file = args[0];
 	}
 
 	PowerPointExtractor ppe = new PowerPointExtractor(file);
-	System.out.println(ppe.getText(true,notes));
+	System.out.println(ppe.getText(true,notes,comments));
 	ppe.close();
   }
 
@@ -127,6 +134,13 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 	public void setNotesByDefault(boolean notesByDefault) {
 		this.notesByDefault = notesByDefault;
 	}
+	/**
+	 * Should a call to getText() return comments text?
+	 * Default is no
+	 */
+	public void setCommentsByDefault(boolean commentsByDefault) {
+		this.commentsByDefault = commentsByDefault;
+	}
 
 	/**
 	 * Fetches all the slide text from the slideshow, 
@@ -135,7 +149,7 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 	 *  to change this
 	 */
 	public String getText() {
-		return getText(slidesByDefault,notesByDefault);
+		return getText(slidesByDefault,notesByDefault,commentsByDefault);
 	}
 
 	/**
@@ -153,6 +167,9 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
    * @param getNoteText fetch note text
    */
   public String getText(boolean getSlideText, boolean getNoteText) {
+	  return getText(getSlideText, getNoteText, commentsByDefault);
+  }
+  public String getText(boolean getSlideText, boolean getNoteText, boolean getCommentText) {
 	StringBuffer ret = new StringBuffer(); 
 
 	if(getSlideText) {
@@ -169,6 +186,18 @@ public class PowerPointExtractor extends POIOLE2TextExtractor
 					}
 				}
 			}
+			
+			if(getCommentText) {
+				Comment[] comments = slide.getComments();
+				for(int j=0; j<comments.length; j++) {
+					ret.append(
+							comments[j].getAuthor() + 
+							" - " +
+							comments[j].getText() + 
+							"\n"
+					);
+				}
+			}
 		}
 		if(getNoteText) {
 			ret.append("\n");

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

@@ -0,0 +1,54 @@
+package org.apache.poi.hslf.model;
+
+import org.apache.poi.hslf.record.Comment2000;
+
+public class Comment {
+	private Comment2000 comment2000;
+	
+	public Comment(Comment2000 comment2000) {
+		this.comment2000 = comment2000;
+	}
+	
+	protected Comment2000 getComment2000() {
+		return comment2000;
+	}
+	
+	/**
+	 * Get the Author of this comment
+	 */
+	public String getAuthor() {
+		return comment2000.getAuthor();
+	}
+	/**
+	 * Set the Author of this comment
+	 */
+	public void setAuthor(String author) {
+		comment2000.setAuthor(author);
+	}
+
+	/**
+	 * Get the Author's Initials of this comment
+	 */
+	public String getAuthorInitials() {
+		return comment2000.getAuthorInitials();
+	}
+	/**
+	 * Set the Author's Initials of this comment
+	 */
+	public void setAuthorInitials(String initials) {
+		comment2000.setAuthorInitials(initials);
+	}
+
+	/**
+	 * Get the text of this comment
+	 */
+	public String getText() {
+		return comment2000.getText();
+	}
+	/**
+	 * Set the text of this comment
+	 */
+	public void setText(String text) {
+		comment2000.setText(text);
+	}
+}

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

@@ -362,6 +362,59 @@ public class Slide extends Sheet
         return super.getColorScheme();
     }
     
+    /**
+     * Get the comment(s) for this slide.
+     * Note - for now, only works on PPT 2000 and 
+     *  PPT 2003 files. Doesn't work for PPT 97
+     *  ones, as they do their comments oddly.
+     */
+    public Comment[] getComments() {
+    	// If there are any, they're in
+    	//  ProgTags -> ProgBinaryTag -> BinaryTagData
+    	RecordContainer progTags = (RecordContainer)
+    			getSheetContainer().findFirstOfType(
+    						RecordTypes.ProgTags.typeID
+    	);
+    	if(progTags != null) {
+    		RecordContainer progBinaryTag = (RecordContainer)
+    			progTags.findFirstOfType(
+    					RecordTypes.ProgBinaryTag.typeID
+    		);
+    		if(progBinaryTag != null) {
+    			RecordContainer binaryTags = (RecordContainer)
+    				progBinaryTag.findFirstOfType(
+    						RecordTypes.BinaryTagData.typeID
+    			);
+    			if(binaryTags != null) {
+    				// This is where they'll be
+    				int count = 0;
+    				for(int i=0; i<binaryTags.getChildRecords().length; i++) {
+    					if(binaryTags.getChildRecords()[i] instanceof Comment2000) {
+    						count++;
+    					}
+    				}
+    				
+    				// Now build
+    				Comment[] comments = new Comment[count];
+    				count = 0;
+    				for(int i=0; i<binaryTags.getChildRecords().length; i++) {
+    					if(binaryTags.getChildRecords()[i] instanceof Comment2000) {
+    						comments[i] = new Comment(
+    								(Comment2000)binaryTags.getChildRecords()[i]
+    						);
+    						count++;
+    					}
+    				}
+    				
+    				return comments;
+    			}
+    		}
+    	}
+    	
+    	// None found
+    	return new Comment[0];
+    }
+
     public void draw(Graphics2D graphics){
         MasterSheet master = getMasterSheet();
         if(getFollowMasterBackground()) master.getBackground().draw(graphics);

src/scratchpad/src/org/apache/poi/hslf/record/RecordContainer.java

@@ -123,6 +123,20 @@ public abstract class RecordContainer extends Record
 	}
 	
 	
+	/**
+	 * Finds the first child record of the given type,
+	 *  or null if none of the child records are of the
+	 *  given type. Does not descend.
+	 */
+	public Record findFirstOfType(long type) {
+		for(int i=0; i<_children.length; i++) {
+			if(_children[i].getRecordType() == type) {
+				return _children[i];
+			}
+		}
+		return null;
+	}
+	
 	/* ===============================================================
 	 *                   External Move Methods
 	 * ===============================================================

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

@@ -216,4 +216,36 @@ public class TextExtractor extends TestCase {
 				ppe.getText(true, false)
 		);
     }
+    
+    /**
+     * From bug #45543
+     */
+    public void testWithComments() throws Exception {
+		String filename;
+		
+		// New file
+		filename = dirname + "/WithComments.ppt";
+		ppe = new PowerPointExtractor(filename);
+
+		String text = ppe.getText();
+		assertFalse("Comments not in by default", text.contains("This is a test comment"));
+		
+		ppe.setCommentsByDefault(true);
+		
+		text = ppe.getText();
+		assertTrue("Unable to find expected word in text\n" + text, text.contains("This is a test comment"));
+
+		
+		// And another file
+		filename = dirname + "/45543.ppt";
+		ppe = new PowerPointExtractor(filename);
+
+		text = ppe.getText();
+		assertFalse("Comments not in by default", text.contains("testdoc"));
+		
+		ppe.setCommentsByDefault(true);
+		
+		text = ppe.getText();
+		assertTrue("Unable to find expected word in text\n" + text, text.contains("testdoc"));
+    }
 }