Allow access from XSSFReader to sheet comments and headers/footers (related to bug #50076)

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@1037753 13f79535-47bb-0310-9956-ffa450edef68

src/documentation/content/xdocs/status.xml

@@ -34,6 +34,7 @@
 
     <changes>
         <release version="3.8-beta1" date="2010-??-??">
+           <action dev="poi-developers" type="add">50076 - Allow access from XSSFReader to sheet comments and headers/footers</action>
            <action dev="poi-developers" type="add">50076 - Refactor XSSFEventBasedExcelExtractor to make it easier for you to have control over outputting the cell contents</action>
            <action dev="poi-developers" type="fix">50258 - avoid corruption of XSSFWorkbook after applying XSSFRichTextRun#applyFont</action>
            <action dev="poi-developers" type="fix">50154 - Allow white spaces and unicode in OPC relationship targets </action>

src/ooxml/java/org/apache/poi/POIXMLTextExtractor.java

@@ -20,6 +20,7 @@ package org.apache.poi;
 import org.apache.poi.POIXMLProperties.CoreProperties;
 import org.apache.poi.POIXMLProperties.CustomProperties;
 import org.apache.poi.POIXMLProperties.ExtendedProperties;
+import org.apache.poi.openxml4j.opc.OPCPackage;
 
 public abstract class POIXMLTextExtractor extends POITextExtractor {
 	/** The POIXMLDocument that's open */
@@ -56,10 +57,16 @@ public abstract class POIXMLTextExtractor extends POITextExtractor {
 	/**
 	 * Returns opened document
 	 */
-	public final POIXMLDocument getDocument(){
+	public final POIXMLDocument getDocument() {
 		return _document;
 	}
 
+	/**
+	 * Returns the opened OPCPackage that contains the document
+	 */
+	public OPCPackage getPackage() {
+	   return _document.getPackage();
+	}
 
 	/**
 	 * Returns an OOXML properties text extractor for the

src/ooxml/java/org/apache/poi/xssf/eventusermodel/XSSFReader.java

@@ -30,8 +30,10 @@ import org.apache.poi.openxml4j.opc.OPCPackage;
 import org.apache.poi.openxml4j.opc.PackagePart;
 import org.apache.poi.openxml4j.opc.PackagePartName;
 import org.apache.poi.openxml4j.opc.PackageRelationship;
+import org.apache.poi.openxml4j.opc.PackageRelationshipCollection;
 import org.apache.poi.openxml4j.opc.PackageRelationshipTypes;
 import org.apache.poi.openxml4j.opc.PackagingURIHelper;
+import org.apache.poi.xssf.model.CommentsTable;
 import org.apache.poi.xssf.model.SharedStringsTable;
 import org.apache.poi.xssf.model.StylesTable;
 import org.apache.poi.xssf.usermodel.XSSFRelation;
@@ -155,7 +157,7 @@ public class XSSFReader {
          * Current CTSheet bean
          */
         private CTSheet ctSheet;
-
+        
         /**
          * Iterator over CTSheet objects, returns sheets in <tt>logical</tt> order.
          * We can't rely on the Ooxml4J's relationship iterator because it returns objects in physical order,
@@ -228,7 +230,40 @@ public class XSSFReader {
         public String getSheetName() {
             return ctSheet.getName();
         }
+        
+        /**
+         * Returns the comments associated with this sheet,
+         *  or null if there aren't any
+         */
+        public CommentsTable getSheetComments() {
+           PackagePart sheetPkg = getSheetPart();
+           
+           // Do we have a comments relationship? (Only ever one if so)
+           try {
+              PackageRelationshipCollection commentsList = 
+                   sheetPkg.getRelationshipsByType(XSSFRelation.SHEET_COMMENTS.getRelation());
+              if(commentsList.size() > 0) {
+                 PackageRelationship comments = commentsList.getRelationship(0);
+                 PackagePartName commentsName = PackagingURIHelper.createPartName(comments.getTargetURI());
+                 PackagePart commentsPart = sheetPkg.getPackage().getPart(commentsName);
+                 return new CommentsTable(commentsPart, comments);
+              }
+           } catch (InvalidFormatException e) {
+              return null;
+           } catch (IOException e) {
+              return null;
+           }
+           return null;
+        }
+        
+        public PackagePart getSheetPart() {
+           String sheetId = ctSheet.getId();
+           return sheetMap.get(sheetId);
+        }
 
+        /**
+         * We're read only, so remove isn't supported
+         */
         public void remove() {
             throw new IllegalStateException("Not supported");
         }

src/ooxml/java/org/apache/poi/xssf/eventusermodel/XSSFSheetXMLHandler.java

@@ -61,6 +61,8 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
    private boolean vIsOpen;
    // Set when F start element is seen
    private boolean fIsOpen;
+   // Set when a header/footer element is seen
+   private boolean hfIsOpen;
 
    // Set when cell start element is seen;
    // used when cell close element is seen.
@@ -76,26 +78,39 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
    // Gathers characters as they are seen.
    private StringBuffer value = new StringBuffer();
    private StringBuffer formula = new StringBuffer();
+   private StringBuffer headerFooter = new StringBuffer();
 
    /**
     * Accepts objects needed while parsing.
     *
     * @param styles  Table of styles
     * @param strings Table of shared strings
-    * @param cols    Minimum number of columns to show
-    * @param target  Sink for output
     */
    public XSSFSheetXMLHandler(
            StylesTable styles,
            ReadOnlySharedStringsTable strings,
            SheetContentsHandler sheetContentsHandler,
+           DataFormatter dataFormatter,
            boolean formulasNotResults) {
        this.stylesTable = styles;
        this.sharedStringsTable = strings;
        this.output = sheetContentsHandler;
        this.formulasNotResults = formulasNotResults;
        this.nextDataType = xssfDataType.NUMBER;
-       this.formatter = new DataFormatter();
+       this.formatter = dataFormatter;
+   }
+   /**
+    * Accepts objects needed while parsing.
+    *
+    * @param styles  Table of styles
+    * @param strings Table of shared strings
+    */
+   public XSSFSheetXMLHandler(
+           StylesTable styles,
+           ReadOnlySharedStringsTable strings,
+           SheetContentsHandler sheetContentsHandler,
+           boolean formulasNotResults) {
+       this(styles, strings, sheetContentsHandler, new DataFormatter(), formulasNotResults);
    }
 
    public void startElement(String uri, String localName, String name,
@@ -122,6 +137,13 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
              fIsOpen = true;
           }
        }
+       else if("oddHeader".equals(name) || "evenHeader".equals(name) ||
+             "firstHeader".equals(name) || "firstFooter".equals(name) ||
+             "oddFooter".equals(name) || "evenFooter".equals(name)) {
+          hfIsOpen = true;
+          // Clear contents cache
+          headerFooter.setLength(0);
+       }
        else if("row".equals(name)) {
            int rowNum = Integer.parseInt(attributes.getValue("r")) - 1;
            output.startRow(rowNum);
@@ -222,6 +244,16 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
        } else if ("row".equals(name)) {
           output.endRow();
        }
+       else if("oddHeader".equals(name) || "evenHeader".equals(name) ||
+             "firstHeader".equals(name)) {
+          hfIsOpen = false;
+          output.headerFooter(headerFooter.toString(), true, name);
+       }
+       else if("oddFooter".equals(name) || "evenFooter".equals(name) ||
+             "firstFooter".equals(name)) {
+          hfIsOpen = false;
+          output.headerFooter(headerFooter.toString(), false, name);
+       }
    }
 
    /**
@@ -236,6 +268,9 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
        if (fIsOpen) {
           formula.append(ch, start, length);
        }
+       if (hfIsOpen) {
+          headerFooter.append(ch, start, length);
+       }
    }
 
    /**
@@ -249,5 +284,7 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
       public void endRow();
       /** A cell, with the given formatted value, was encountered */
       public void cell(String cellReference, String formattedValue);
+      /** A header or footer has been encountered */
+      public void headerFooter(String text, boolean isHeader, String tagName);
    }
 }

src/ooxml/java/org/apache/poi/xssf/extractor/XSSFEventBasedExcelExtractor.java

@@ -18,14 +18,20 @@ package org.apache.poi.xssf.extractor;
 
 import java.io.IOException;
 import java.io.InputStream;
+import java.util.Locale;
 
 import javax.xml.parsers.ParserConfigurationException;
 import javax.xml.parsers.SAXParser;
 import javax.xml.parsers.SAXParserFactory;
 
+import org.apache.poi.POIXMLProperties;
 import org.apache.poi.POIXMLTextExtractor;
+import org.apache.poi.POIXMLProperties.CoreProperties;
+import org.apache.poi.POIXMLProperties.CustomProperties;
+import org.apache.poi.POIXMLProperties.ExtendedProperties;
 import org.apache.poi.openxml4j.exceptions.OpenXML4JException;
 import org.apache.poi.openxml4j.opc.OPCPackage;
+import org.apache.poi.ss.usermodel.DataFormatter;
 import org.apache.poi.xssf.eventusermodel.ReadOnlySharedStringsTable;
 import org.apache.poi.xssf.eventusermodel.XSSFReader;
 import org.apache.poi.xssf.eventusermodel.XSSFSheetXMLHandler;
@@ -43,6 +49,9 @@ import org.xml.sax.XMLReader;
  */
 public class XSSFEventBasedExcelExtractor extends POIXMLTextExtractor {
    private OPCPackage container;
+   private POIXMLProperties properties;
+   
+   private Locale locale;
 	private boolean includeSheetNames = true;
 	private boolean formulasNotResults = false;
 
@@ -52,6 +61,8 @@ public class XSSFEventBasedExcelExtractor extends POIXMLTextExtractor {
 	public XSSFEventBasedExcelExtractor(OPCPackage container) throws XmlException, OpenXML4JException, IOException {
 		super(null);
 		this.container = container;
+		
+		properties = new POIXMLProperties(container);
 	}
 
 	public static void main(String[] args) throws Exception {
@@ -79,22 +90,64 @@ public class XSSFEventBasedExcelExtractor extends POIXMLTextExtractor {
 		this.formulasNotResults = formulasNotResults;
 	}
 	
+	public void setLocale(Locale locale) {
+	   this.locale = locale;
+	}
+	
+	/**
+	 * Returns the opened OPCPackage container.
+	 */
+	@Override
+	public OPCPackage getPackage() {
+	   return container;
+	}
+	
+   /**
+    * Returns the core document properties
+    */
+   @Override
+   public CoreProperties getCoreProperties() {
+       return properties.getCoreProperties();
+   }
+   /**
+    * Returns the extended document properties
+    */
+   @Override
+   public ExtendedProperties getExtendedProperties() {
+      return properties.getExtendedProperties();
+   }
+   /**
+    * Returns the custom document properties
+    */
+   @Override
+   public CustomProperties getCustomProperties() {
+      return properties.getCustomProperties();
+   }
+   
    /**
     * Processes the given sheet
     */
    public void processSheet(
-           SheetTextExtractor sheetExtractor,
+           SheetContentsHandler sheetContentsExtractor,
            StylesTable styles,
            ReadOnlySharedStringsTable strings,
            InputStream sheetInputStream)
            throws IOException, SAXException {
 
+       DataFormatter formatter;
+       if(locale == null) {
+          formatter = new DataFormatter();
+       } else  {
+          formatter = new DataFormatter(locale);
+       }
+      
        InputSource sheetSource = new InputSource(sheetInputStream);
        SAXParserFactory saxFactory = SAXParserFactory.newInstance();
        try {
           SAXParser saxParser = saxFactory.newSAXParser();
           XMLReader sheetParser = saxParser.getXMLReader();
-          ContentHandler handler = new XSSFSheetXMLHandler(styles, strings, sheetExtractor, formulasNotResults);
+          ContentHandler handler = new XSSFSheetXMLHandler(
+                styles, strings, sheetContentsExtractor, formatter, formulasNotResults);
           sheetParser.setContentHandler(handler);
           sheetParser.parse(sheetSource);
        } catch(ParserConfigurationException e) {
@@ -162,5 +215,9 @@ public class XSSFEventBasedExcelExtractor extends POIXMLTextExtractor {
          }
          output.append(formattedValue);
       }
+      
+      public void headerFooter(String text, boolean isHeader, String tagName) {
+         // We don't include headers in the output yet, so ignore
+      }
    }
 }

src/ooxml/testcases/org/apache/poi/xssf/eventusermodel/TestXSSFReader.java

@@ -25,6 +25,7 @@ import junit.framework.TestCase;
 import org.apache.poi.openxml4j.opc.OPCPackage;
 import org.apache.poi.util.IOUtils;
 import org.apache.poi.xssf.XSSFTestDataSamples;
+import org.apache.poi.xssf.model.CommentsTable;
 import org.apache.poi.xssf.usermodel.XSSFRichTextString;
 import org.apache.poi.POIDataSamples;
 
@@ -117,6 +118,28 @@ public final class TestXSSFReader extends TestCase {
 		assertEquals(4, count);
 	}
 	
+	public void testComments() throws Exception {
+      OPCPackage pkg =  XSSFTestDataSamples.openSamplePackage("comments.xlsx");
+      XSSFReader r = new XSSFReader(pkg);
+      XSSFReader.SheetIterator it = (XSSFReader.SheetIterator)r.getSheetsData();
+      
+      int count = 0;
+      while(it.hasNext()) {
+         count++;
+         InputStream inp = it.next();
+         inp.close();
+
+         if(count == 1) {
+            assertNotNull(it.getSheetComments());
+            CommentsTable ct = it.getSheetComments();
+            assertEquals(1, ct.getNumberOfAuthors());
+            assertEquals(3, ct.getNumberOfComments());
+         } else {
+            assertNull(it.getSheetComments());
+         }
+      }
+      assertEquals(3, count);
+	}
    
    /**
     * Iterating over a workbook with chart sheets in it, using the