http://nagoya.apache.org/bugzilla/show_bug.cgi?id=11010 - great patch by

sergei.... if he'd learn to do diff -u patches instead of the nasty kind I'd be
loving his work alot right now ;-)
PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/jakarta/poi/trunk@352831 13f79535-47bb-0310-9956-ffa450edef68

src/java/org/apache/poi/hssf/dev/BiffViewer.java

@@ -624,6 +624,9 @@ public class BiffViewer {
             case StringRecord.sid:
                 retval = new StringRecord( rectype, size, data );
                 break;
+            case NameRecord.sid:
+                retval = new NameRecord( rectype, size, data );
+                break;
             default:
                 retval = new UnknownRecord( rectype, size, data );
         }

src/java/org/apache/poi/hssf/eventmodel/HSSFEventFactory.java

@@ -185,6 +185,24 @@ public class HSSFEventFactory
 			{
 
 				sid = LittleEndian.getShort(sidbytes);
+                
+                //
+                // for some reasons we have to make the workbook to be at least 4096 bytes
+                // but if we have such workbook we fill the end of it with zeros (many zeros)
+                //
+                // it is not good:
+                // if the length( all zero records ) % 4 = 1
+                // e.g.: any zero record would be readed as  4 bytes at once ( 2 - id and 2 - size ).
+                // And the last 1 byte will be readed WRONG ( the id must be 2 bytes )
+                //
+                // So we should better to check if the sid is zero and not to read more data
+                // The zero sid shows us that rest of the stream data is a fake to make workbook 
+                // certain size
+                //
+                if ( sid == 0 )
+                    break;
+
+
 				if ((rec != null) && (sid != ContinueRecord.sid))
 				{
 					userCode = req.processRecord(rec);

src/java/org/apache/poi/hssf/record/FormatRecord.java

@@ -73,7 +73,8 @@ public class FormatRecord
     public final static short sid = 0x41e;
     private short             field_1_index_code;
     private byte              field_2_formatstring_len;
-    private short             field_3_zero;   // undocumented 2 bytes of 0
+    private short             field_3_unicode_len;      // unicode string length
+    private boolean          field_3_unicode_flag;     // it is not undocumented - it is unicode flag
     private String            field_4_formatstring;
 
     public FormatRecord()
@@ -118,10 +119,19 @@ public class FormatRecord
     protected void fillFields(byte [] data, short size, int offset)
     {
         field_1_index_code       = LittleEndian.getShort(data, 0 + offset);
-        field_2_formatstring_len = data[ 2 + offset ];
+        // field_2_formatstring_len = data[ 2 + offset ];
-        field_3_zero             = LittleEndian.getShort(data, 3 + offset);
+        field_3_unicode_len      = LittleEndian.getShort( data, 2 + offset );
-        field_4_formatstring     = new String(data, 5 + offset,
+        field_3_unicode_flag     = ( data[ 4 + offset ] & (byte)0x01 ) != 0;
-                                              LittleEndian.ubyteToInt(field_2_formatstring_len));
+                                              
+                                              
+      if ( field_3_unicode_flag  ) {
+          // unicode
+          field_4_formatstring = StringUtil.getFromUnicodeHigh( data, 5 + offset, field_3_unicode_len );
+      }
+      else {
+          // not unicode
+          field_4_formatstring = new String(data, 5 + offset, field_3_unicode_len );
+      }
     }
 
     /**
@@ -203,11 +213,15 @@ public class FormatRecord
         buffer.append("[FORMAT]\n");
         buffer.append("    .indexcode       = ")
             .append(Integer.toHexString(getIndexCode())).append("\n");
+        /*
         buffer.append("    .formatstringlen = ")
             .append(Integer.toHexString(getFormatStringLength()))
             .append("\n");
-        buffer.append("    .zero            = ")
+        */
-            .append(Integer.toHexString(field_3_zero)).append("\n");
+        buffer.append("    .unicode length  = ")
+            .append(Integer.toHexString(field_3_unicode_len)).append("\n");
+        buffer.append("    .isUnicode       = ")
+            .append( field_3_unicode_flag ).append("\n");
         buffer.append("    .formatstring    = ").append(getFormatString())
             .append("\n");
         buffer.append("[/FORMAT]\n");
@@ -217,20 +231,29 @@ public class FormatRecord
     public int serialize(int offset, byte [] data)
     {
         LittleEndian.putShort(data, 0 + offset, sid);
-        LittleEndian.putShort(data, 2 + offset,
+        LittleEndian.putShort(data, 2 + offset, (short)( 2 + 2 + 1 + ( (field_3_unicode_flag) 
-                              ( short ) (5 + getFormatStringLength()));
+                                                                  ? 2 * field_3_unicode_len 
-
+                                                                  : field_3_unicode_len ) ) );
-        // 9 - 4(len/sid) + format string length
+                                                  // index + len + flag + format string length
         LittleEndian.putShort(data, 4 + offset, getIndexCode());
-        data[ 6 + offset ] = getFormatStringLength();
+        LittleEndian.putShort(data, 6 + offset, field_3_unicode_len);
-        LittleEndian.putShort(data, 7 + offset, ( short ) 0);
+        data[ 8 + offset ] = (byte)( (field_3_unicode_flag) ? 0x01 : 0x00 );
+
+      if ( field_3_unicode_flag ) {
+          // unicode
+          StringUtil.putUncompressedUnicode( getFormatString(), data, 9 + offset );
+      }
+      else {
+          // not unicode
           StringUtil.putCompressedUnicode( getFormatString(), data, 9 + offset );
+      }
+      
         return getRecordSize();
     }
 
     public int getRecordSize()
     {
-        return 9 + getFormatStringLength();
+        return 9 + ( ( field_3_unicode_flag ) ? 2 * field_3_unicode_len : field_3_unicode_len );
     }
 
     public short getSid()

src/java/org/apache/poi/hssf/record/NameRecord.java

@@ -55,6 +55,7 @@
 
 package org.apache.poi.hssf.record;
 
+import org.apache.poi.util.HexDump;
 import org.apache.poi.util.LittleEndian;
 import org.apache.poi.util.StringUtil;
 import java.util.Stack;
@@ -70,6 +71,7 @@ import org.apache.poi.hssf.util.SheetReferences;
  * Description:  Defines a named range within a workbook. <P>
  * REFERENCE:  <P>
  * @author Libin Roman (Vista Portal LDT. Developer)
+ * @author  Sergei Kozello (sergeikozello at mail.ru)
  * @version 1.0-pre
  */
 
@@ -88,9 +90,10 @@ public class NameRecord extends Record {
     private byte              field_9_length_help_topic_text;
     private byte              field_10_length_status_bar_text;
     private byte              field_11_compressed_unicode_flag;   // not documented
+    private byte              field_12_builtIn_name;
     private String            field_12_name_text;
     private Stack             field_13_name_definition;
-    private byte[]            field_13_raw_name_definition = null; // raw data
+    private byte[]            field_13_raw_name_definition;       // raw data
     private String            field_14_custom_menu_text;
     private String            field_15_description_text;
     private String            field_16_help_topic_text;
@@ -378,7 +381,7 @@ public class NameRecord extends Record {
      */
     public int serialize(int offset, byte[] data) {
         LittleEndian.putShort(data, 0 + offset, sid);
-        LittleEndian.putShort(data, 2 + offset, (short)( 15 + getTextsLength()));
+        // size defined below
         LittleEndian.putShort(data, 4 + offset, getOptionFlag());
         data[6 + offset] = getKeyboardShortcut();
         data[7 + offset] = getNameTextLength();
@@ -391,6 +394,18 @@ public class NameRecord extends Record {
         data [17 + offset] =  getStatusBarLength();
         data [18 + offset] =  getCompressedUnicodeFlag();
         
+      if ( ( field_1_option_flag & (short)0x20 ) != 0 ) {
+          LittleEndian.putShort(data, 2 + offset, (short)( 16 + field_13_raw_name_definition.length ));
+          
+            data [19 + offset] =  field_12_builtIn_name;
+            System.arraycopy( field_13_raw_name_definition, 0, data, 20 + offset, field_13_raw_name_definition.length );
+            
+            return 20 + field_13_raw_name_definition.length;
+      }
+      else {
+          LittleEndian.putShort(data, 2 + offset, (short)( 15 + getTextsLength()));
+          
+          
             StringUtil.putCompressedUnicode(getNameText(), data , 19 + offset);
     
             int start_of_name_definition    = 19  + field_3_length_name_text;
@@ -413,9 +428,9 @@ public class NameRecord extends Record {
             int start_of_status_bar_text       = start_of_help_topic_text + field_10_length_status_bar_text;
             StringUtil.putCompressedUnicode(getStatusBarText(), data , start_of_status_bar_text + offset);
             
-
           return getRecordSize();
       }
+    }
 
     private void serializePtgs(byte [] data, int offset) {
         int pos = offset;
@@ -579,6 +594,29 @@ public class NameRecord extends Record {
         field_9_length_help_topic_text  = data [12 + offset];
         field_10_length_status_bar_text = data [13 + offset];
 
+        
+        if ( ( field_1_option_flag & (short)0x20 ) != 0 ) {
+            // DEBUG
+            // System.out.println( "Built-in name" );
+            
+            field_11_compressed_unicode_flag = data[ 14 + offset ];
+            field_12_builtIn_name = data[ 15 + offset ];
+
+            if ( (field_12_builtIn_name & (short)0x07) != 0 ) {
+                field_12_name_text = "Print_Titles";
+                
+                // DEBUG
+                // System.out.println( field_12_name_text );
+                
+                field_13_raw_name_definition = new byte[ field_4_length_name_definition ];
+                System.arraycopy( data, 16 + offset, field_13_raw_name_definition, 0, field_13_raw_name_definition.length );
+                
+                // DEBUG
+                // System.out.println( HexDump.toHex( field_13_raw_name_definition ) );
+            }
+        }
+        else {
+    
             field_11_compressed_unicode_flag= data [14 + offset];
             field_12_name_text = new String(data, 15 + offset,
             LittleEndian.ubyteToInt(field_3_length_name_text));
@@ -602,7 +640,7 @@ public class NameRecord extends Record {
             int start_of_status_bar_text       = start_of_help_topic_text + field_10_length_status_bar_text;
             field_17_status_bar_text        = new String(data, start_of_status_bar_text +  offset,
             LittleEndian.ubyteToInt(field_10_length_status_bar_text));
-
+        }
     }
 
     private Stack getParsedExpressionTokens(byte [] data, short size,
@@ -635,5 +673,101 @@ public class NameRecord extends Record {
     public short getSid() {
         return this.sid;
     }
+    /*
+      20 00 
+      00 
+      01 
+      1A 00 // sz = 0x1A = 26
+      00 00 
+      01 00 
+      00 
+      00 
+      00 
+      00 
+      00 // unicode flag
+      07 // name
+      
+      29 17 00 3B 00 00 00 00 FF FF 00 00 02 00 3B 00 //{ 26
+      00 07 00 07 00 00 00 FF 00 10                   //  }
+      
+      
+      
+      20 00 
+      00 
+      01 
+      0B 00 // sz = 0xB = 11
+      00 00 
+      01 00 
+      00 
+      00 
+      00 
+      00 
+      00 // unicode flag
+      07 // name
+      
+      3B 00 00 07 00 07 00 00 00 FF 00   // { 11 }
+  */
+    /*
+      18, 00, 
+      1B, 00, 
+      
+      20, 00, 
+      00, 
+      01, 
+      0B, 00, 
+      00, 
+      00, 
+      00, 
+      00, 
+      00, 
+      07, 
+      3B 00 00 07 00 07 00 00 00 FF 00 ]     
+     */
+
+    /**
+     * @see Object#toString()
+     */
+    public String toString() {
+        StringBuffer buffer = new StringBuffer();
+
+        buffer.append("[NAME]\n");
+        buffer.append("    .option flags         = ").append( HexDump.toHex( field_1_option_flag ) )
+            .append("\n");
+        buffer.append("    .keyboard shortcut    = ").append( HexDump.toHex( field_2_keyboard_shortcut ) )
+            .append("\n");
+        buffer.append("    .length of the name   = ").append( field_3_length_name_text )
+            .append("\n");
+        buffer.append("    .size of the formula data = ").append( field_4_length_name_definition )
+            .append("\n");
+        buffer.append("    .unused                 = ").append( field_5_index_to_sheet )
+            .append("\n");
+        buffer.append("    .( 0 = Global name, otherwise index to sheet (one-based) ) = ").append( field_6_equals_to_index_to_sheet )
+            .append("\n");
+        buffer.append("    .Length of menu text (character count)        = ").append( field_7_length_custom_menu )
+            .append("\n");
+        buffer.append("    .Length of description text (character count) = ").append( field_8_length_description_text )
+            .append("\n");
+        buffer.append("    .Length of help topic text (character count)  = ").append( field_9_length_help_topic_text )
+            .append("\n");
+        buffer.append("    .Length of status bar text (character count)  = ").append( field_10_length_status_bar_text )
+            .append("\n");
+        buffer.append("    .Name (Unicode flag)  = ").append( field_11_compressed_unicode_flag )
+            .append("\n");
+        buffer.append("    .Name (Unicode text)  = ").append( field_12_name_text )
+            .append("\n");
+        buffer.append("    .Formula data (RPN token array without size field)      = ").append( HexDump.toHex( field_13_raw_name_definition ) )
+            .append("\n");
+        buffer.append("    .Menu text (Unicode string without length field)        = ").append( field_14_custom_menu_text )
+            .append("\n");
+        buffer.append("    .Description text (Unicode string without length field) = ").append( field_15_description_text )
+            .append("\n");
+        buffer.append("    .Help topic text (Unicode string without length field)  = ").append( field_16_help_topic_text )
+            .append("\n");
+        buffer.append("    .Status bar text (Unicode string without length field)  = ").append( field_17_status_bar_text )
+            .append("\n");
+        buffer.append("[/NAME]\n");
+        
+        return buffer.toString();
+    }
 
 }

src/java/org/apache/poi/hssf/usermodel/HSSFWorkbook.java

@@ -202,6 +202,9 @@ public class HSSFWorkbook
         // none currently
     }
 
+    public final static byte ENCODING_COMPRESSED_UNICODE = 0;
+    public final static byte ENCODING_UTF_16             = 1;
+    
     /**
      * set the sheet name.
      * @param sheet number (0 based)
@@ -209,12 +212,28 @@ public class HSSFWorkbook
      */
 
     public void setSheetName(int sheet, String name)
+    {
+        workbook.setSheetName( sheet, name, ENCODING_COMPRESSED_UNICODE );
+    }
+
+    public void setSheetName( int sheet, String name, short encoding )
     {
         if (sheet > (sheets.size() - 1))
         {
             throw new RuntimeException("Sheet out of bounds");
         }
-        workbook.setSheetName(sheet, name);
+        
+        switch ( encoding ) {
+        case ENCODING_COMPRESSED_UNICODE:
+        case ENCODING_UTF_16:
+            break;
+            
+        default:
+            // TODO java.io.UnsupportedEncodingException
+            throw new RuntimeException( "Unsupported encoding" );
+        }
+        
+        workbook.setSheetName( sheet, name, encoding );
     }
 
     /**

src/java/org/apache/poi/util/StringUtil.java

@@ -63,7 +63,13 @@ import java.text.FieldPosition;
 /**
  *  Title: String Utility Description: Collection of string handling utilities
  * 
+ * Now it is quite confusing: the method pairs, in which
+ * one of them write data and other read written data are:
+ * putUncompressedUnicodeHigh and getFromUnicode
+ * putUncompressedUnicode     and getFromUnicodeHigh
+ *
  *@author     Andrew C. Oliver
+ *@author     Sergei Kozello (sergeikozello at mail.ru)
  *@created    May 10, 2002
  *@version    1.0
  */
@@ -79,6 +85,8 @@ public class StringUtil {
      *  given a byte array of 16-bit unicode characters, compress to 8-bit and
      *  return a string
      *
+     * { 0x16, 0x00 } -> 0x16
+     * 
      *@param  string                              the byte array to be converted
      *@param  offset                              the initial offset into the
      *      byte array. it is assumed that string[ offset ] and string[ offset +
@@ -103,23 +111,38 @@ public class StringUtil {
         if ((len < 0) || (((string.length - offset) / 2) < len)) {
             throw new IllegalArgumentException("Illegal length");
         }
-        byte[] bstring = new byte[len];
-        int index = offset;
-        // start with high bits.
         
-        for (int k = 0; k < len; k++) {
+        char[] chars = new char[ len ];
-            bstring[k] = string[index];
+        for ( int i = 0; i < chars.length; i++ ) {
-            index += 2;
+            chars[i] = (char)( string[ offset + ( 2*i ) ] + 
-        }
+                             ( string[ offset + ( 2*i+1 ) ] << 8 ) );
-        return new String(bstring);
         }
 
+        return new String( chars );
+    }
     
     
     /**
      *  given a byte array of 16-bit unicode characters, compress to 8-bit and
      *  return a string
      * 
+     * { 0x16, 0x00 } -> 0x16
+     *
+     *@param  string  the byte array to be converted
+     *@return         the converted string
+     */
+
+    public static String getFromUnicodeHigh( final byte[] string ) {
+        return getFromUnicodeHigh( string, 0, string.length / 2 );
+    }
+
+
+    /**
+     *  given a byte array of 16-bit unicode characters, compress to 8-bit and
+     *  return a string
+     * 
+     * { 0x00, 0x16 } -> 0x16
+     *
      *@param  string                              the byte array to be converted
      *@param  offset                              the initial offset into the
      *      byte array. it is assumed that string[ offset ] and string[ offset +
@@ -144,15 +167,15 @@ public class StringUtil {
         if ((len < 0) || (((string.length - offset) / 2) < len)) {
             throw new IllegalArgumentException("Illegal length");
         }
-        byte[] bstring = new byte[len];
-        int index = offset + 1;
-        // start with low bits.
 
-        for (int k = 0; k < len; k++) {
+        
-            bstring[k] = string[index];
+        char[] chars = new char[ len ];
-            index += 2;
+        for ( int i = 0; i < chars.length; i++ ) {
+            chars[i] = (char)( ( string[ offset + ( 2*i ) ] << 8 ) +
+                              string[ offset + ( 2*i+1 ) ] );
         }
-        return new String(bstring);
+        
+        return new String( chars );
     }
 
 
@@ -160,6 +183,8 @@ public class StringUtil {
      *  given a byte array of 16-bit unicode characters, compress to 8-bit and
      *  return a string
      * 
+     * { 0x00, 0x16 } -> 0x16
+     *
      *@param  string  the byte array to be converted
      *@return         the converted string
      */

src/testcases/org/apache/poi/hssf/usermodel/TestCellStyle.java

@@ -129,7 +129,6 @@ public class TestCellStyle
         out.close();
         SanityChecker sanityChecker = new SanityChecker();
         sanityChecker.checkHSSFWorkbook(wb);
-        assertEquals("FILE LENGTH == 87040", file.length(), 87040);  // remove me
         assertEquals("LAST ROW == 99", 99, s.getLastRowNum());
         assertEquals("FIRST ROW == 0", 0, s.getFirstRowNum());
 
@@ -169,7 +168,6 @@ public class TestCellStyle
         SanityChecker sanityChecker = new SanityChecker();
         sanityChecker.checkHSSFWorkbook(wb);
 
-        assertEquals("FILE LENGTH ", 5632, file.length());      // remove me
         assertEquals("LAST ROW ", 0, s.getLastRowNum());
         assertEquals("FIRST ROW ", 0, s.getFirstRowNum());
 
@@ -232,7 +230,6 @@ public class TestCellStyle
         out.close();
         SanityChecker sanityChecker = new SanityChecker();
         sanityChecker.checkHSSFWorkbook(wb);
-        assertEquals("FILE LENGTH == 87040", file.length(), 87040);     // remove me
         assertEquals("LAST ROW == 99", 99, s.getLastRowNum());
         assertEquals("FIRST ROW == 0", 0, s.getFirstRowNum());
 

src/testcases/org/apache/poi/hssf/usermodel/TestWorkbook.java

@@ -142,8 +142,6 @@ public class TestWorkbook
         wb.write(out);
         out.close();
         sanityChecker.checkHSSFWorkbook(wb);
-        assertEquals("FILE LENGTH == 87040", 87040,
-                     file.length());                            // remove me
         assertEquals("LAST ROW == 99", 99, s.getLastRowNum());
         assertEquals("FIRST ROW == 0", 0, s.getFirstRowNum());
 
@@ -201,8 +199,6 @@ public class TestWorkbook
         out.close();
 
         sanityChecker.checkHSSFWorkbook(wb);
-        assertEquals("FILE LENGTH == 45568", 45568,
-                     file.length());   // remove
         assertEquals("LAST ROW == 74", 74, s.getLastRowNum());
         assertEquals("FIRST ROW == 25", 25, s.getFirstRowNum());
     }

src/testcases/org/apache/poi/util/TestStringUtil.java

@@ -64,6 +64,7 @@ import java.text.NumberFormat;
  *
  * @author  Marc Johnson (mjohnson at apache dot org
  * @author  Glen Stampoultzis (glens at apache.org)
+ * @author  Sergei Kozello (sergeikozello at mail.ru)
  */
 
 public class TestStringUtil
@@ -99,6 +100,48 @@ public class TestStringUtil
                      StringUtil.getFromUnicode(test_data));
     }
 
+    /**
+     * test simple form of getFromUnicode with symbols with code below and more 127
+     */
+
+    public void testGetFromUnicodeSymbolsWithCodesMoreThan127()
+    {
+        byte[] test_data = new byte[] {   0x04, 0x22,
+                                          0x04, 0x35,
+                                          0x04, 0x41,
+                                          0x04, 0x42,
+                                          0x00, 0x20,
+                                          0x00, 0x74,
+                                          0x00, 0x65,
+                                          0x00, 0x73,
+                                          0x00, 0x74,
+                                          };
+
+        assertEquals("\u0422\u0435\u0441\u0442 test",
+                     StringUtil.getFromUnicode(test_data));
+    }
+
+    /**
+     * test getFromUnicodeHigh for symbols with code below and more 127
+     */
+
+    public void testGetFromUnicodeHighSymbolsWithCodesMoreThan127()
+    {
+        byte[] test_data = new byte[] {   0x22, 0x04,
+                                          0x35, 0x04,
+                                          0x41, 0x04,
+                                          0x42, 0x04,
+                                          0x20, 0x00,
+                                          0x74, 0x00,
+                                          0x65, 0x00,
+                                          0x73, 0x00,
+                                          0x74, 0x00,
+                                          };
+
+        assertEquals("\u0422\u0435\u0441\u0442 test",
+                     StringUtil.getFromUnicodeHigh( test_data ) );
+    }
+
     /**
      * Test more complex form of getFromUnicode
      */