#57796 - Support hyperlink extraction when rendering slides

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@1724338 13f79535-47bb-0310-9956-ffa450edef68

src/examples/src/org/apache/poi/hslf/examples/CreateHyperlink.java

@@ -46,7 +46,7 @@ public abstract class CreateHyperlink {
         String text = textBox1.getText();
         HSLFHyperlink link = new HSLFHyperlink();
         link.setAddress("http://www.apache.org");
-        link.setTitle(textBox1.getText());
+        link.setLabel(textBox1.getText());
         int linkId = ppt.addHyperlink(link);
 
         // apply link to the text

src/examples/src/org/apache/poi/hslf/examples/Hyperlinks.java

@@ -70,6 +70,6 @@ public final class Hyperlinks {
         //in ppt end index is inclusive
         String formatStr = "title: %1$s, address: %2$s" + (rawText == null ? "" : ", start: %3$s, end: %4$s, substring: %5$s");
         String substring = (rawText == null) ? "" : rawText.substring(link.getStartIndex(), link.getEndIndex()-1);
-        return String.format(formatStr, link.getTitle(), link.getAddress(), link.getStartIndex(), link.getEndIndex(), substring);
+        return String.format(formatStr, link.getLabel(), link.getAddress(), link.getStartIndex(), link.getEndIndex(), substring);
     }
 }

src/java/org/apache/poi/sl/draw/DrawTextParagraph.java

@@ -24,6 +24,7 @@ import java.awt.font.LineBreakMeasurer;
 import java.awt.font.TextAttribute;
 import java.awt.font.TextLayout;
 import java.awt.geom.Rectangle2D;
+import java.io.InvalidObjectException;
 import java.text.AttributedCharacterIterator;
 import java.text.AttributedCharacterIterator.Attribute;
 import java.text.AttributedString;
@@ -32,6 +33,7 @@ import java.util.List;
 import java.util.Map;
 
 import org.apache.poi.sl.usermodel.AutoNumberingScheme;
+import org.apache.poi.sl.usermodel.Hyperlink;
 import org.apache.poi.sl.usermodel.Insets2D;
 import org.apache.poi.sl.usermodel.PaintStyle;
 import org.apache.poi.sl.usermodel.PlaceableShape;
@@ -46,6 +48,10 @@ import org.apache.poi.util.StringUtil;
 import org.apache.poi.util.Units;
 
 public class DrawTextParagraph implements Drawable {
+    /** Keys for passing hyperlinks to the graphics context */
+    public static final XlinkAttribute HYPERLINK_HREF = new XlinkAttribute("href");
+    public static final XlinkAttribute HYPERLINK_LABEL = new XlinkAttribute("label");
+
     protected TextParagraph<?,?,?> paragraph;
     double x, y;
     protected List<DrawTextFragment> lines = new ArrayList<DrawTextFragment>();
@@ -58,6 +64,31 @@ public class DrawTextParagraph implements Drawable {
      */
     protected double maxLineHeight;
 
+    /**
+     * Defines an attribute used for storing the hyperlink associated with
+     * some renderable text.
+     */
+    private static class XlinkAttribute extends Attribute {
+
+        XlinkAttribute(String name) {
+            super(name);
+        }
+
+        /**
+         * Resolves instances being deserialized to the predefined constants.
+         */
+        protected Object readResolve() throws InvalidObjectException {
+            if (HYPERLINK_HREF.getName().equals(getName())) {
+                return HYPERLINK_HREF;
+            }
+            if (HYPERLINK_LABEL.getName().equals(getName())) {
+                return HYPERLINK_LABEL;
+            }
+            throw new InvalidObjectException("unknown attribute name");
+        }
+    }
+
+
     public DrawTextParagraph(TextParagraph<?,?,?> paragraph) {
         this.paragraph = paragraph;
     }
@@ -498,6 +529,12 @@ public class DrawTextParagraph implements Drawable {
             if(run.isSuperscript()) {
                 attList.add(new AttributedStringData(TextAttribute.SUPERSCRIPT, TextAttribute.SUPERSCRIPT_SUPER, beginIndex, endIndex));
             }
+            
+            Hyperlink hl = run.getHyperlink();
+            if (hl != null) {
+                attList.add(new AttributedStringData(HYPERLINK_HREF, hl.getAddress(), beginIndex, endIndex));
+                attList.add(new AttributedStringData(HYPERLINK_LABEL, hl.getLabel(), beginIndex, endIndex));
+            }
         }
 
         // ensure that the paragraph contains at least one character

src/java/org/apache/poi/sl/usermodel/TextRun.java

@@ -156,4 +156,11 @@ public interface TextRun {
      * @return the pitch and family id or -1 if not applicable
      */
     byte getPitchAndFamily();
+
+    /**
+     * Return the associated hyperlink
+     * 
+     * @return the associated hyperlink or null if no hyperlink was set
+     */
+    Hyperlink getHyperlink();
 }

src/ooxml/java/org/apache/poi/xslf/usermodel/XSLFHyperlink.java

@@ -18,15 +18,13 @@ package org.apache.poi.xslf.usermodel;
 
 import org.apache.poi.openxml4j.opc.PackageRelationship;
 import org.apache.poi.openxml4j.opc.TargetMode;
+import org.apache.poi.sl.usermodel.Hyperlink;
 import org.apache.poi.util.Internal;
 import org.openxmlformats.schemas.drawingml.x2006.main.CTHyperlink;
 
 import java.net.URI;
 
-/**
+public class XSLFHyperlink implements Hyperlink {
- * @author Yegor Kozlov
- */
-public class XSLFHyperlink {
     final XSLFTextRun _r;
     final CTHyperlink _link;
 
@@ -40,13 +38,37 @@ public class XSLFHyperlink {
         return _link;
     }
 
+    @Override
     public void setAddress(String address){
         XSLFSheet sheet = _r.getParentParagraph().getParentShape().getSheet();
         PackageRelationship rel =
                 sheet.getPackagePart().
                         addExternalRelationship(address, XSLFRelation.HYPERLINK.getRelation());
         _link.setId(rel.getId());
+    }
     
+    @Override
+    public String getAddress() {
+        return getTargetURI().toASCIIString();
+    }
+
+    @Override
+    public String getLabel() {
+        return _link.getTooltip();
+    }
+    
+    @Override
+    public void setLabel(String label) {
+        _link.setTooltip(label);
+    }
+
+    @Override
+    public int getType() {
+        // TODO: currently this just returns nonsense
+        if ("ppaction://hlinksldjump".equals(_link.getAction())) {
+            return LINK_DOCUMENT;
+        }
+        return LINK_URL;
     }
     
     public void setAddress(XSLFSlide slide){

src/ooxml/java/org/apache/poi/xslf/usermodel/XSLFTextRun.java

@@ -449,6 +449,7 @@ public class XSLFTextRun implements TextRun {
         return link;
     }
 
+    @Override
     public XSLFHyperlink getHyperlink(){
         if(!_r.getRPr().isSetHlinkClick()) return null;
 

src/scratchpad/src/org/apache/poi/hslf/model/PPGraphics2D.java

@@ -282,7 +282,7 @@ public final class PPGraphics2D extends Graphics2D implements Cloneable {
      */
     public void drawString(String s, float x, float y) {
         HSLFTextBox txt = new HSLFTextBox(_group);
-        txt.getTextParagraphs().get(0).supplySheet(_group.getSheet());
+        txt.setSheet(_group.getSheet());
         txt.setText(s);
 
         HSLFTextRun rt = txt.getTextParagraphs().get(0).getTextRuns().get(0);

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFHyperlink.java

@@ -29,13 +29,12 @@ import org.apache.poi.hslf.record.InteractiveInfo;
 import org.apache.poi.hslf.record.InteractiveInfoAtom;
 import org.apache.poi.hslf.record.Record;
 import org.apache.poi.hslf.record.TxInteractiveInfoAtom;
+import org.apache.poi.sl.usermodel.Hyperlink;
 
 /**
  * Represents a hyperlink in a PowerPoint document
- *
- * @author Yegor Kozlov
  */
-public final class HSLFHyperlink {
+public final class HSLFHyperlink implements Hyperlink {
     public static final byte LINK_NEXTSLIDE = InteractiveInfoAtom.LINK_NextSlide;
     public static final byte LINK_PREVIOUSSLIDE = InteractiveInfoAtom.LINK_PreviousSlide;
     public static final byte LINK_FIRSTSLIDE = InteractiveInfoAtom.LINK_FirstSlide;
@@ -47,7 +46,7 @@ public final class HSLFHyperlink {
     private int id=-1;
     private int type;
     private String address;
-    private String title;
+    private String label;
     private int startIndex, endIndex;
 
     /**
@@ -57,6 +56,7 @@ public final class HSLFHyperlink {
      * @return the hyperlink URL
      * @see InteractiveInfoAtom
      */
+    @Override
     public int getType() {
         return type;
     }
@@ -65,35 +65,31 @@ public final class HSLFHyperlink {
         type = val;
         switch(type){
             case LINK_NEXTSLIDE:
-                title = "NEXT";
+                label = "NEXT";
                 address = "1,-1,NEXT";
                 break;
             case LINK_PREVIOUSSLIDE:
-                title = "PREV";
+                label = "PREV";
                 address = "1,-1,PREV";
                 break;
             case LINK_FIRSTSLIDE:
-                title = "FIRST";
+                label = "FIRST";
                 address = "1,-1,FIRST";
                 break;
             case LINK_LASTSLIDE:
-                title = "LAST";
+                label = "LAST";
                 address = "1,-1,LAST";
                 break;
             case LINK_SLIDENUMBER:
                 break;
             default:
-                title = "";
+                label = "";
                 address = "";
                 break;
         }
     }
 
-    /**
+    @Override
-     * Gets the hyperlink URL
-     *
-     * @return the hyperlink URL
-     */
     public String getAddress() {
         return address;
     }
@@ -101,10 +97,11 @@ public final class HSLFHyperlink {
     public void setAddress(HSLFSlide slide) {
         String href = slide._getSheetNumber() + ","+slide.getSlideNumber()+",Slide " + slide.getSlideNumber();
         setAddress(href);;
-        setTitle("Slide " + slide.getSlideNumber());
+        setLabel("Slide " + slide.getSlideNumber());
         setType(HSLFHyperlink.LINK_SLIDENUMBER);
     }
 
+    @Override
     public void setAddress(String str) {
         address = str;
     }
@@ -117,17 +114,14 @@ public final class HSLFHyperlink {
         this.id = id;
     }
 
-    /**
+    @Override
-     * Gets the hyperlink user-friendly title (if different from URL)
+    public String getLabel() {
-     *
+        return label;
-     * @return the  hyperlink user-friendly title
-     */
-    public String getTitle() {
-        return title;
     }
 
-    public void setTitle(String str) {
+    @Override
-        title = str;
+    public void setLabel(String str) {
+        label = str;
     }
 
     /**
@@ -139,6 +133,15 @@ public final class HSLFHyperlink {
         return startIndex;
     }
 
+    /**
+     * Sets the beginning character position
+     *
+     * @param startIndex the beginning character position
+     */
+    public void setStartIndex(int startIndex) {
+        this.startIndex = startIndex;
+    }
+    
     /**
      * Gets the ending character position
      *
@@ -148,6 +151,15 @@ public final class HSLFHyperlink {
         return endIndex;
     }
 
+    /**
+     * Sets the ending character position
+     *
+     * @param endIndex the ending character position
+     */
+    public void setEndIndex(int endIndex) {
+        this.endIndex = endIndex;
+    }
+    
     /**
      * Find hyperlinks in a text shape
      *
@@ -222,9 +234,10 @@ public final class HSLFHyperlink {
             }
             
             HSLFHyperlink link = new HSLFHyperlink();
-            link.title = linkRecord.getLinkTitle();
+            link.setId(id);
-            link.address = linkRecord.getLinkURL();
+            link.setType(info.getAction());
-            link.type = info.getAction();
+            link.setLabel(linkRecord.getLinkTitle());
+            link.setAddress(linkRecord.getLinkURL());
             out.add(link);
 
             if (iter.hasNext()) {
@@ -234,8 +247,8 @@ public final class HSLFHyperlink {
                     continue;
                 }
                 TxInteractiveInfoAtom txinfo = (TxInteractiveInfoAtom)r;
-                link.startIndex = txinfo.getStartIndex();
+                link.setStartIndex(txinfo.getStartIndex());
-                link.endIndex = txinfo.getEndIndex();
+                link.setEndIndex(txinfo.getEndIndex());
             }
         }
     }

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFNotes.java

@@ -56,13 +56,6 @@ public final class HSLFNotes extends HSLFSheet implements Notes<HSLFShape,HSLFTe
         if (_paragraphs.isEmpty()) {
             logger.log(POILogger.WARN, "No text records found for notes sheet");
         }
-
-        // Set the sheet on each TextRun
-        for (List<HSLFTextParagraph> ltp : _paragraphs) {
-            for (HSLFTextParagraph tp : ltp) {
-                tp.supplySheet(this);
-            }
-        }
     }
 
     /**

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFSheet.java

@@ -133,9 +133,7 @@ public abstract class HSLFSheet implements HSLFShapeContainer, Sheet<HSLFShape,H
         List<List<HSLFTextParagraph>> trs = getTextParagraphs();
         if (trs == null) return;
         for (List<HSLFTextParagraph> ltp : trs) {
-            for (HSLFTextParagraph tp : ltp) {
+            HSLFTextParagraph.supplySheet(ltp, this);
-                tp.supplySheet(this);
-            }
         }
     }
 

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFSimpleShape.java

@@ -310,7 +310,7 @@ public abstract class HSLFSimpleShape extends HSLFShape implements SimpleShape<H
                 infoAtom.setHyperlinkType(InteractiveInfoAtom.LINK_SlideNumber);
                 break;
             default:
-                logger.log(POILogger.WARN, "Ignore unknown hyperlink type : "+link.getTitle());
+                logger.log(POILogger.WARN, "Ignore unknown hyperlink type : "+link.getLabel());
                 break;
         }
 

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFSlide.java

@@ -94,12 +94,6 @@ public final class HSLFSlide extends HSLFSheet implements Slide<HSLFShape,HSLFTe
 		for (List<HSLFTextParagraph> l : HSLFTextParagraph.findTextParagraphs(getPPDrawing(), this)) {
 		    if (!_paragraphs.contains(l)) _paragraphs.add(l);
 		}
-
-        for(List<HSLFTextParagraph> ltp : _paragraphs) {
-            for (HSLFTextParagraph tp : ltp) {
-                tp.supplySheet(this);
-            }
-        }
 	}
 
 	/**

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFSlideMaster.java

@@ -51,12 +51,6 @@ public final class HSLFSlideMaster extends HSLFMasterSheet {
         for (List<HSLFTextParagraph> l : HSLFTextParagraph.findTextParagraphs(getPPDrawing(), this)) {
             if (!_paragraphs.contains(l)) _paragraphs.add(l);
         }
-
-        for (List<HSLFTextParagraph> p : _paragraphs) {
-            for (HSLFTextParagraph htp : p) {
-                htp.supplySheet(this);
-            }
-        }
     }
 
     /**

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFSlideShow.java

@@ -33,6 +33,7 @@ import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
 
+import org.apache.poi.hslf.record.MainMaster;
 import org.apache.poi.ddf.EscherBSERecord;
 import org.apache.poi.ddf.EscherContainerRecord;
 import org.apache.poi.ddf.EscherOptRecord;
@@ -59,12 +60,14 @@ import org.apache.poi.hslf.record.ExVideoContainer;
 import org.apache.poi.hslf.record.FontCollection;
 import org.apache.poi.hslf.record.FontEntityAtom;
 import org.apache.poi.hslf.record.HeadersFootersContainer;
+import org.apache.poi.hslf.record.Notes;
 import org.apache.poi.hslf.record.PersistPtrHolder;
 import org.apache.poi.hslf.record.PositionDependentRecord;
 import org.apache.poi.hslf.record.PositionDependentRecordContainer;
 import org.apache.poi.hslf.record.Record;
 import org.apache.poi.hslf.record.RecordContainer;
 import org.apache.poi.hslf.record.RecordTypes;
+import org.apache.poi.hslf.record.Slide;
 import org.apache.poi.hslf.record.SlideListWithText;
 import org.apache.poi.hslf.record.SlideListWithText.SlideAtomsSet;
 import org.apache.poi.hslf.record.SlidePersistAtom;
@@ -120,12 +123,6 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
 	private POILogger logger = POILogFactory.getLogger(this.getClass());
 
 
-	/* ===============================================================
-	 *                       Setup Code
-	 * ===============================================================
-	 */
-
-
 	/**
 	 * Constructs a Powerpoint document from the underlying
 	 * HSLFSlideShow object. Finds the model stuff from this
@@ -338,46 +335,53 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
 		// We always use the latest versions of these records, and use the
 		// SlideAtom/NotesAtom to match them with the StyleAtomSet
 
-		SlideListWithText masterSLWT = _documentRecord.getMasterSlideListWithText();
+		findMasterSlides();
-		SlideListWithText slidesSLWT = _documentRecord.getSlideSlideListWithText();
+		
-		SlideListWithText notesSLWT = _documentRecord.getNotesSlideListWithText();
+		// Having sorted out the masters, that leaves the notes and slides
+        Map<Integer,Integer> slideIdToNotes = new HashMap<Integer,Integer>();
+
+        // Start by finding the notes records
+        findNotesSlides(slideIdToNotes);
+
+		// Now, do the same thing for our slides
+		findSlides(slideIdToNotes);
+	}
+
+	/**
+     * Find master slides
+     * These can be MainMaster records, but oddly they can also be
+     * Slides or Notes, and possibly even other odd stuff....
+     * About the only thing you can say is that the master details are in the first SLWT.
+	 */
+	private void findMasterSlides() {
+        SlideListWithText masterSLWT = _documentRecord.getMasterSlideListWithText();
+        if (masterSLWT == null) {
+            return;
+        }
         
-		// Find master slides
-		// These can be MainMaster records, but oddly they can also be
-		// Slides or Notes, and possibly even other odd stuff....
-		// About the only thing you can say is that the master details are in
-		// the first SLWT.
-		if (masterSLWT != null) {
         for (SlideAtomsSet sas : masterSLWT.getSlideAtomsSets()) {
             Record r = getCoreRecordForSAS(sas);
             int sheetNo = sas.getSlidePersistAtom().getSlideIdentifier();
-				if (r instanceof org.apache.poi.hslf.record.Slide) {
+            if (r instanceof Slide) {
-					HSLFTitleMaster master = new HSLFTitleMaster((org.apache.poi.hslf.record.Slide) r,
+                HSLFTitleMaster master = new HSLFTitleMaster((Slide)r, sheetNo);
-							sheetNo);
                 master.setSlideShow(this);
                 _titleMasters.add(master);
-				} else if (r instanceof org.apache.poi.hslf.record.MainMaster) {
+            } else if (r instanceof MainMaster) {
-					HSLFSlideMaster master = new HSLFSlideMaster((org.apache.poi.hslf.record.MainMaster) r,
+                HSLFSlideMaster master = new HSLFSlideMaster((MainMaster)r, sheetNo);
-							sheetNo);
                 master.setSlideShow(this);
                 _masters.add(master);
             }
         }
 	}
 	
-		// Having sorted out the masters, that leaves the notes and slides
+	private void findNotesSlides(Map<Integer,Integer> slideIdToNotes) {
+        SlideListWithText notesSLWT = _documentRecord.getNotesSlideListWithText();
 
-		// Start by finding the notes records to go with the entries in
-		// notesSLWT
-		org.apache.poi.hslf.record.Notes[] notesRecords;
-		Map<Integer,Integer> slideIdToNotes = new HashMap<Integer,Integer>();
         if (notesSLWT == null) {
-			// None
+            return;
-			notesRecords = new org.apache.poi.hslf.record.Notes[0];
+        }
-		} else {
+        
         // Match up the records and the SlideAtomSets
-			List<org.apache.poi.hslf.record.Notes> notesRecordsL = 
-			   new ArrayList<org.apache.poi.hslf.record.Notes>();
         int idx = -1;
         for (SlideAtomsSet notesSet : notesSLWT.getSlideAtomsSets()) {
             idx++;
@@ -387,76 +391,63 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
 
             String loggerLoc = "A Notes SlideAtomSet at "+idx+" said its record was at refID "+spa.getRefID();
 
-				// Ensure it really is a notes record
+            // we need to add null-records, otherwise the index references to other existing don't work anymore
-				if (r == null || r instanceof org.apache.poi.hslf.record.Notes) {
             if (r == null) {
                 logger.log(POILogger.WARN, loggerLoc+", but that record didn't exist - record ignored.");
+                continue;
             }
-				    // we need to add also null-records, otherwise the index references to other existing
+
-				    // don't work anymore
+            // Ensure it really is a notes record
-					org.apache.poi.hslf.record.Notes notesRecord = (org.apache.poi.hslf.record.Notes) r;
+            if (!(r instanceof Notes)) {
-					notesRecordsL.add(notesRecord);
+                logger.log(POILogger.ERROR, loggerLoc+", but that was actually a " + r);
+                continue;
+            }
+            
+            Notes notesRecord = (Notes) r;
 
             // Record the match between slide id and these notes
             int slideId = spa.getSlideIdentifier();
             slideIdToNotes.put(slideId, idx);
-				} else {
-					logger.log(POILogger.ERROR, loggerLoc+", but that was actually a " + r);
-				}
-			}
-			notesRecords = new org.apache.poi.hslf.record.Notes[notesRecordsL.size()];
-			notesRecords = notesRecordsL.toArray(notesRecords);
-		}
             
-		// Now, do the same thing for our slides
+            HSLFNotes hn = new HSLFNotes(notesRecord);
-		org.apache.poi.hslf.record.Slide[] slidesRecords;
-		SlideAtomsSet[] slidesSets = new SlideAtomsSet[0];
-		if (slidesSLWT == null) {
-			// None
-			slidesRecords = new org.apache.poi.hslf.record.Slide[0];
-		} else {
-			// Match up the records and the SlideAtomSets
-			slidesSets = slidesSLWT.getSlideAtomsSets();
-			slidesRecords = new org.apache.poi.hslf.record.Slide[slidesSets.length];
-			for (int i = 0; i < slidesSets.length; i++) {
-				// Get the right core record
-				Record r = getCoreRecordForSAS(slidesSets[i]);
-
-				// Ensure it really is a slide record
-				if (r instanceof org.apache.poi.hslf.record.Slide) {
-					slidesRecords[i] = (org.apache.poi.hslf.record.Slide) r;
-				} else {
-					logger.log(POILogger.ERROR, "A Slide SlideAtomSet at " + i
-							+ " said its record was at refID "
-							+ slidesSets[i].getSlidePersistAtom().getRefID()
-							+ ", but that was actually a " + r);
-				}
-			}
-		}
-
-		// Finally, generate model objects for everything
-		// Notes first
-		for (org.apache.poi.hslf.record.Notes n : notesRecords) {
-		    HSLFNotes hn = null;
-		    if (n != null) {
-    		    hn = new HSLFNotes(n);
             hn.setSlideShow(this);
-		    }
             _notes.add(hn);
         }
-		// Then slides
+	}
-		for (int i = 0; i < slidesRecords.length; i++) {
+
-			SlideAtomsSet sas = slidesSets[i];
+	private void findSlides(Map<Integer,Integer> slideIdToNotes) {
-			int slideIdentifier = sas.getSlidePersistAtom().getSlideIdentifier();
+        SlideListWithText slidesSLWT = _documentRecord.getSlideSlideListWithText();
+        if (slidesSLWT == null) {
+            return;
+        }
+
+        // Match up the records and the SlideAtomSets
+        int idx = -1;
+        for (SlideAtomsSet sas : slidesSLWT.getSlideAtomsSets()) {
+            idx++;
+            // Get the right core record
+            SlidePersistAtom spa = sas.getSlidePersistAtom();
+            Record r = getCoreRecordForSAS(sas);
+
+            // Ensure it really is a slide record
+            if (!(r instanceof Slide)) {
+                logger.log(POILogger.ERROR, "A Slide SlideAtomSet at " + idx
+                        + " said its record was at refID "
+                        + spa.getRefID()
+                        + ", but that was actually a " + r);
+                continue;
+            }
+            
+            Slide slide = (Slide)r;
 
             // Do we have a notes for this?
             HSLFNotes notes = null;
             // Slide.SlideAtom.notesId references the corresponding notes slide.
             // 0 if slide has no notes.
-			int noteId = slidesRecords[i].getSlideAtom().getNotesID();
+            int noteId = slide.getSlideAtom().getNotesID();
             if (noteId != 0) {
                 Integer notesPos = slideIdToNotes.get(noteId);
-				if (notesPos != null) {
+                if (notesPos != null && 0 <= notesPos && notesPos < _notes.size()) {
                     notes = _notes.get(notesPos);
                 } else {
                     logger.log(POILogger.ERROR, "Notes not found for noteId=" + noteId);
@@ -464,7 +455,8 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
             }
 
             // Now, build our slide
-			HSLFSlide hs = new HSLFSlide(slidesRecords[i], notes, sas, slideIdentifier, (i + 1));
+            int slideIdentifier = spa.getSlideIdentifier();
+            HSLFSlide hs = new HSLFSlide(slide, notes, sas, slideIdentifier, (idx + 1));
             hs.setSlideShow(this);
             _slides.add(hs);
         }
@@ -512,12 +504,6 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
         }
 	}
 
-	/*
-	 * ===============================================================
-	 *                         Accessor Code
-	 * ===============================================================
-	 */
-
 	/**
 	 * Returns an array of the most recent version of all the interesting
 	 * records
@@ -604,12 +590,6 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
 		return _documentRecord;
 	}
 
-	/*
-	 * ===============================================================
-	 * Re-ordering Code
-	 * ===============================================================
-	 */
-
 	/**
 	 * Re-orders a slide, to a new position.
 	 *
@@ -720,12 +700,6 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
 		return removedSlide;
 	}
 
-	/*
-	 * ===============================================================
-	 *  Addition Code
-	 * ===============================================================
-	 */
-
 	/**
 	 * Create a blank <code>Slide</code>.
 	 *
@@ -786,7 +760,7 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
 				+ " and identifier " + sp.getSlideIdentifier());
 
 		// Add the core records for this new Slide to the record tree
-		org.apache.poi.hslf.record.Slide slideRecord = slide.getSlideRecord();
+		Slide slideRecord = slide.getSlideRecord();
 		int psrId = addPersistentObject(slideRecord);
 		sp.setRefID(psrId);
 		slideRecord.setSheetId(psrId);
@@ -1049,7 +1023,7 @@ public final class HSLFSlideShow implements SlideShow<HSLFShape,HSLFTextParagrap
         } else {
             ctrl.setLinkURL(link.getAddress());
         }
-		ctrl.setLinkTitle(link.getTitle());
+		ctrl.setLinkTitle(link.getLabel());
 
 		int objectId = addToObjListAtom(ctrl);
 		link.setId(objectId);

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFTextParagraph.java

@@ -103,6 +103,8 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
 
     private boolean _dirty = false;
     
+    private final List<HSLFTextParagraph> parentList;
+
     /**
     * Constructs a Text Run from a Unicode text block.
     * Either a {@link TextCharsAtom} or a {@link TextBytesAtom} needs to be provided.
@@ -110,11 +112,13 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
     * @param tha the TextHeaderAtom that defines what's what
     * @param tba the TextBytesAtom containing the text or null if {@link TextCharsAtom} is provided
     * @param tca the TextCharsAtom containing the text or null if {@link TextBytesAtom} is provided
+    * @param parentList the list which contains this paragraph
      */
     /* package */ HSLFTextParagraph(
         TextHeaderAtom tha,
         TextBytesAtom tba,
-        TextCharsAtom tca
+        TextCharsAtom tca,
+        List<HSLFTextParagraph> parentList
     ) {
         if (tha == null) {
             throw new IllegalArgumentException("TextHeaderAtom must be set.");
@@ -122,6 +126,7 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
         _headerAtom = tha;
         _byteAtom = tba;
         _charAtom = tca;
+        this.parentList = parentList;
     }
 
     /* package */HSLFTextParagraph(HSLFTextParagraph other) {
@@ -133,6 +138,7 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
         _ruler = other._ruler;
         shapeId = other.shapeId;
         _paragraphStyle.copy(other._paragraphStyle);
+        parentList = other.parentList;
     }
 
     public void addTextRun(HSLFTextRun run) {
@@ -168,7 +174,23 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
      * Supply the Sheet we belong to, which might have an assigned SlideShow
      * Also passes it on to our child RichTextRuns
      */
-    public void supplySheet(HSLFSheet sheet) {
+    public static void supplySheet(List<HSLFTextParagraph> paragraphs, HSLFSheet sheet) {
+        if (paragraphs == null) {
+            return;
+        }
+        for (HSLFTextParagraph p : paragraphs) {
+            p.supplySheet(sheet);
+        }
+
+        assert(sheet.getSlideShow() != null);
+        applyHyperlinks(paragraphs);
+    }
+    
+    /**
+     * Supply the Sheet we belong to, which might have an assigned SlideShow
+     * Also passes it on to our child RichTextRuns
+     */
+    private void supplySheet(HSLFSheet sheet) {
         this._sheet = sheet;
 
         if (_runs == null) return;
@@ -942,7 +964,7 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
             if (addParagraph && !lastParaEmpty) {
                 TextPropCollection tpc = htp.getParagraphStyle();
                 HSLFTextParagraph prevHtp = htp;
-                htp = new HSLFTextParagraph(htp._headerAtom, htp._byteAtom, htp._charAtom);
+                htp = new HSLFTextParagraph(htp._headerAtom, htp._byteAtom, htp._charAtom, paragraphs);
                 htp.getParagraphStyle().copy(tpc);
                 htp.setParentShape(prevHtp.getParentShape());
                 htp.setShapeId(prevHtp.getShapeId());
@@ -1211,7 +1233,7 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
 
             // split, but keep delimiter
             for (String para : rawText.split("(?<=\r)")) {
-                HSLFTextParagraph tpara = new HSLFTextParagraph(header, tbytes, tchars);
+                HSLFTextParagraph tpara = new HSLFTextParagraph(header, tbytes, tchars, paragraphs);
                 paragraphs.add(tpara);
                 tpara._ruler = ruler;
                 tpara.getParagraphStyle().updateTextSize(para.length());
@@ -1235,6 +1257,22 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
         return paragraphCollection;
     }
 
+    protected static void applyHyperlinks(List<HSLFTextParagraph> paragraphs) {
+        List<HSLFHyperlink> links = HSLFHyperlink.find(paragraphs);
+        
+        for (HSLFHyperlink h : links) {
+            int csIdx = 0;
+            for (HSLFTextParagraph p : paragraphs) {
+                for (HSLFTextRun r : p) {
+                    if (h.getStartIndex() <= csIdx && csIdx < h.getEndIndex()) {
+                        r.setHyperlinkId(h.getId());
+                    }
+                    csIdx += r.getLength();
+                }
+            }
+        }
+    }
+    
     protected static void applyCharacterStyles(List<HSLFTextParagraph> paragraphs, List<TextPropCollection> charStyles) {
         int paraIdx = 0, runIdx = 0;
         HSLFTextRun trun;
@@ -1340,15 +1378,17 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
         TextPropCollection charStyle = sta.addCharacterTextPropCollection(1);
         wrapper.appendChildRecord(sta);
 
-        HSLFTextParagraph htp = new HSLFTextParagraph(tha, tba, null);
+        List<HSLFTextParagraph> paragraphs = new ArrayList<HSLFTextParagraph>(1);
+        HSLFTextParagraph htp = new HSLFTextParagraph(tha, tba, null, paragraphs);
         htp.setParagraphStyle(paraStyle);
+        paragraphs.add(htp);
 
         HSLFTextRun htr = new HSLFTextRun(htp);
         htr.setCharacterStyle(charStyle);
         htr.setText("");
         htp.addTextRun(htr);
 
-        return Arrays.asList(htp);
+        return paragraphs;
     }
 
     public EscherTextboxWrapper getTextboxWrapper() {
@@ -1397,4 +1437,23 @@ public final class HSLFTextParagraph implements TextParagraph<HSLFShape,HSLFText
     public boolean isDirty() {
         return _dirty;
     }
+    
+    /**
+     * Calculates the start index of the given text run
+     *
+     * @param textrun the text run to search for
+     * @return the start index with the paragraph collection or -1 if not found
+     */
+    /* package */ int getStartIdxOfTextRun(HSLFTextRun textrun) {
+        int idx = 0;
+        for (HSLFTextParagraph p : parentList) {
+            for (HSLFTextRun r : p) {
+                if (r == textrun) {
+                    return idx;
+                }
+                idx += r.getLength();
+            }
+        }
+        return -1;
+    }
 }

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFTextRun.java

@@ -27,6 +27,10 @@ import org.apache.poi.hslf.model.textproperties.CharFlagsTextProp;
 import org.apache.poi.hslf.model.textproperties.TextProp;
 import org.apache.poi.hslf.model.textproperties.TextPropCollection;
 import org.apache.poi.hslf.model.textproperties.TextPropCollection.TextPropType;
+import org.apache.poi.hslf.record.ExHyperlink;
+import org.apache.poi.hslf.record.InteractiveInfo;
+import org.apache.poi.hslf.record.InteractiveInfoAtom;
+import org.apache.poi.hslf.record.Record;
 import org.apache.poi.sl.draw.DrawPaint;
 import org.apache.poi.sl.usermodel.PaintStyle;
 import org.apache.poi.sl.usermodel.PaintStyle.SolidPaint;
@@ -47,6 +51,7 @@ public final class HSLFTextRun implements TextRun {
 	private HSLFTextParagraph parentParagraph;
 	private String _runText = "";
 	private String _fontFamily;
+	private int hyperlinkId = -1;
 	
 	/**
 	 * Our paragraph and character style.
@@ -390,4 +395,64 @@ public final class HSLFTextRun implements TextRun {
     public byte getPitchAndFamily() {
         return 0;
     }
+    
+    /**
+     * Sets the associated hyperlink id - currently this is only used while parsing and
+     * can't be used for update a ppt
+     *
+     * @param hyperlink the id or -1 to unset it
+     */
+    public void setHyperlinkId(int hyperlinkId) {
+        this.hyperlinkId = hyperlinkId;
+    }
+    
+    /**
+     * Returns the associated hyperlink id
+     *
+     * @return the hyperlink id
+     */
+    public int getHyperlinkId() {
+        return hyperlinkId;
+    }
+    
+
+    @Override
+    public HSLFHyperlink getHyperlink() {
+        if (hyperlinkId == -1) {
+            return null;
+        }
+        
+        ExHyperlink linkRecord = parentParagraph.getSheet().getSlideShow().getDocumentRecord().getExObjList().get(hyperlinkId);
+        if (linkRecord == null) {
+            return null;
+        }
+        
+        InteractiveInfoAtom info = null;
+        for (Record r : parentParagraph.getRecords()) {
+            if (r instanceof InteractiveInfo) {
+                InteractiveInfo ii = (InteractiveInfo)r;
+                InteractiveInfoAtom iia = ii.getInteractiveInfoAtom();
+                if (iia.getHyperlinkID() == hyperlinkId) {
+                    info = iia;
+                    break;
+                }
+            }
+        }
+        if (info == null) {
+            return null;
+        }
+        
+        // TODO: check previous/next sibling runs for same hyperlink id and return the whole string length
+        int startIdx = parentParagraph.getStartIdxOfTextRun(this);
+
+        HSLFHyperlink link = new HSLFHyperlink();
+        link.setId(hyperlinkId);
+        link.setType(info.getAction());
+        link.setLabel(linkRecord.getLinkTitle());
+        link.setAddress(linkRecord.getLinkURL());
+        link.setStartIndex(startIdx);
+        link.setEndIndex(startIdx+getLength());
+        
+        return link;
+    }
 }

src/scratchpad/src/org/apache/poi/hslf/usermodel/HSLFTextShape.java

@@ -594,13 +594,8 @@ implements TextShape<HSLFShape,HSLFTextParagraph> {
         // (We can't do it in the constructor because the sheet
         //  is not assigned then, it's only built once we have
         //  all the records)
-        List<HSLFTextParagraph> paras = getTextParagraphs();
+        List<HSLFTextParagraph> ltp = getTextParagraphs();
-        if (paras != null) {
+        HSLFTextParagraph.supplySheet(ltp, sheet);
-            for (HSLFTextParagraph htp : paras) {
-                // Supply the sheet to our child RichTextRuns
-                htp.supplySheet(_sheet);
-            }
-        }
     }
 
     /**

src/scratchpad/testcases/org/apache/poi/hslf/model/TestHyperlink.java

@@ -57,11 +57,11 @@ public final class TestHyperlink {
         assertNotNull(links);
         assertEquals(2, links.size());
 
-        assertEquals("http://jakarta.apache.org/poi/", links.get(0).getTitle());
+        assertEquals("http://jakarta.apache.org/poi/", links.get(0).getLabel());
         assertEquals("http://jakarta.apache.org/poi/", links.get(0).getAddress());
         assertEquals("http://jakarta.apache.org/poi/", rawText.substring(links.get(0).getStartIndex(), links.get(0).getEndIndex()-1));
 
-        assertEquals("http://slashdot.org/", links.get(1).getTitle());
+        assertEquals("http://slashdot.org/", links.get(1).getLabel());
         assertEquals("http://slashdot.org/", links.get(1).getAddress());
         assertEquals("http://slashdot.org/", rawText.substring(links.get(1).getStartIndex(), links.get(1).getEndIndex()-1));
 
@@ -77,7 +77,7 @@ public final class TestHyperlink {
         assertNotNull(links);
         assertEquals(1, links.size());
 
-        assertEquals("http://jakarta.apache.org/poi/hssf/", links.get(0).getTitle());
+        assertEquals("Open Jakarta POI HSSF module test  ", links.get(0).getLabel());
         assertEquals("http://jakarta.apache.org/poi/hssf/", links.get(0).getAddress());
         assertEquals("Jakarta HSSF", rawText.substring(links.get(0).getStartIndex(), links.get(0).getEndIndex()-1));
     }

src/scratchpad/testcases/org/apache/poi/hslf/usermodel/TestBugs.java

@@ -26,6 +26,10 @@ import static org.junit.Assert.assertTrue;
 import java.awt.Color;
 import java.io.File;
 import java.io.IOException;
+import java.io.OutputStream;
+import java.io.PrintStream;
+import java.text.AttributedCharacterIterator;
+import java.text.AttributedCharacterIterator.Attribute;
 import java.util.ArrayList;
 import java.util.Date;
 import java.util.HashMap;
@@ -47,7 +51,9 @@ import org.apache.poi.hslf.record.Record;
 import org.apache.poi.hslf.record.SlideListWithText;
 import org.apache.poi.hslf.record.SlideListWithText.SlideAtomsSet;
 import org.apache.poi.hslf.record.TextHeaderAtom;
+import org.apache.poi.hssf.usermodel.DummyGraphics2d;
 import org.apache.poi.sl.draw.DrawPaint;
+import org.apache.poi.sl.draw.DrawTextParagraph;
 import org.apache.poi.sl.usermodel.PaintStyle;
 import org.apache.poi.sl.usermodel.PaintStyle.SolidPaint;
 import org.apache.poi.sl.usermodel.PictureData.PictureType;
@@ -67,8 +73,6 @@ import org.junit.Test;
 /**
  * Testcases for bugs entered in bugzilla
  * the Test name contains the bugzilla bug id
- *
- * @author Yegor Kozlov
  */
 public final class TestBugs {
     /**
@@ -823,6 +827,59 @@ public final class TestBugs {
         ppt.close();
     }
 
+    @Test
+    public void bug57796() throws IOException {
+        HSLFSlideShow ppt = open("WithLinks.ppt");
+        HSLFSlide slide = ppt.getSlides().get(0);
+        HSLFTextShape shape = (HSLFTextShape)slide.getShapes().get(1);
+        List<HSLFHyperlink> hlList =  HSLFHyperlink.find(shape);
+        HSLFHyperlink hlShape = hlList.get(0);
+        HSLFTextRun r = shape.getTextParagraphs().get(1).getTextRuns().get(0);
+        HSLFHyperlink hlRun = r.getHyperlink();
+        assertEquals(hlRun.getId(), hlShape.getId());
+        assertEquals(hlRun.getAddress(), hlShape.getAddress());
+        assertEquals(hlRun.getLabel(), hlShape.getLabel());
+        assertEquals(hlRun.getType(), hlShape.getType());
+        assertEquals(hlRun.getStartIndex(), hlShape.getStartIndex());
+        assertEquals(hlRun.getEndIndex(), hlShape.getEndIndex());
+
+        OutputStream nullOutput = new OutputStream(){
+            public void write(int b) throws IOException {}
+        };
+        
+        final boolean found[] = { false }; 
+        DummyGraphics2d dgfx = new DummyGraphics2d(new PrintStream(nullOutput)){
+            public void drawString(AttributedCharacterIterator iterator, float x, float y) {
+                // For the test file, common sl draws textruns one by one and not mixed
+                // so we evaluate the whole iterator
+                Map<Attribute, Object> attributes = null;
+                StringBuffer sb = new StringBuffer();
+                
+                for (char c = iterator.first();
+                        c != AttributedCharacterIterator.DONE;
+                        c = iterator.next()) {
+                    sb.append(c);
+                    attributes = iterator.getAttributes();
+                }
+
+                if ("Jakarta HSSF".equals(sb.toString())) {
+                    // this is a test for a manually modified ppt, for real hyperlink label
+                    // one would need to access the screen tip record
+                    String href = (String)attributes.get(DrawTextParagraph.HYPERLINK_HREF);
+                    String label = (String)attributes.get(DrawTextParagraph.HYPERLINK_LABEL);
+                    assertEquals("http://jakarta.apache.org/poi/hssf/", href);
+                    assertEquals("Open Jakarta POI HSSF module test  ", label);
+                    found[0] = true;
+                }
+            }
+        };
+        
+        ppt.getSlides().get(1).draw(dgfx);
+        assertTrue(found[0]);
+        
+        ppt.close();
+    }
+    
     private static HSLFSlideShow open(String fileName) throws IOException {
         File sample = HSLFTestDataSamples.getSampleFile(fileName);
         return (HSLFSlideShow)SlideShowFactory.create(sample);