public void TestSSTRecordBug()
{
// create an SSTRecord and Write a certain pattern of strings
// to it ... then Serialize it and verify the content
SSTRecord record = new SSTRecord();
// the record will start with two integers, then this string
// ... that will eat up 16 of the 8224 bytes that the record
// can hold
record.AddString(new UnicodeString("Hello"));
// now we have an additional 8208 bytes, which is1 an exact
// multiple of 16 bytes
long testvalue = 1000000000000L;
for (int k = 0; k < 2000; k++)
{
record.AddString(new UnicodeString((testvalue++).ToString()));
}
byte[] content = new byte[record.RecordSize];
record.Serialize(0, content);
Assert.AreEqual(8224, LittleEndian.GetShort(content, 2));
Assert.AreEqual(ContinueRecord.sid, LittleEndian.GetShort(content, 8228));
Assert.AreEqual(8224, LittleEndian.GetShort(content, 8228 + 2));
Assert.AreEqual((byte)13, content[4 + 8228]);
Assert.AreEqual(ContinueRecord.sid, LittleEndian.GetShort(content, 2 * 8228));
Assert.AreEqual(8224, LittleEndian.GetShort(content, 8228 * 2 + 2));
Assert.AreEqual((byte)13, content[4 + 8228 * 2]);
Assert.AreEqual(ContinueRecord.sid, LittleEndian.GetShort(content, 3 * 8228));
Assert.AreEqual((byte)13, content[4 + 8228 * 3]);
}
/// <summary>
/// Creates a stub Workbook from the supplied records,
/// suitable for use with the {@link HSSFFormulaParser}
/// </summary>
/// <param name="externs">The ExternSheetRecords in your file</param>
/// <param name="bounds">The BoundSheetRecords in your file</param>
/// <param name="sst">TThe SSTRecord in your file.</param>
/// <returns>A stub Workbook suitable for use with HSSFFormulaParser</returns>
public static InternalWorkbook CreateStubWorkbook(ExternSheetRecord[] externs,
BoundSheetRecord[] bounds, SSTRecord sst)
{
List <Record> wbRecords = new List <Record>();
// Core Workbook records go first
if (bounds != null)
{
for (int i = 0; i < bounds.Length; i++)
{
wbRecords.Add(bounds[i]);
}
}
if (sst != null)
{
wbRecords.Add(sst);
}
// Now we can have the ExternSheetRecords,
// preceded by a SupBookRecord
if (externs != null)
{
wbRecords.Add(SupBookRecord.CreateInternalReferences(
(short)externs.Length));
for (int i = 0; i < externs.Length; i++)
{
wbRecords.Add(externs[i]);
}
}
// Finally we need an EoF record
wbRecords.Add(EOFRecord.instance);
return(InternalWorkbook.CreateWorkbook(wbRecords));
}
/// <summary>
/// Process the record ourselves, but do not
/// pass it on to the child Listener.
/// </summary>
/// <param name="record">The record.</param>
public void ProcessRecordInternally(Record record)
{
if (record is BoundSheetRecord)
{
boundSheetRecords.Add(record);
}
else if (record is ExternSheetRecord)
{
externSheetRecords.Add(record);
}
else if (record is SSTRecord)
{
sstRecord = (SSTRecord)record;
}
}
public void TestSimpleAddString()
{
SSTRecord record = new SSTRecord();
UnicodeString s1 = new UnicodeString("Hello world");
// \u2122 is1 the encoding of the trademark symbol ...
UnicodeString s2 = new UnicodeString("Hello world\u2122");
Assert.AreEqual(0, record.AddString(s1));
Assert.AreEqual(s1, record.GetString(0));
Assert.AreEqual(1, record.CountStrings);
Assert.AreEqual(1, record.NumStrings);
Assert.AreEqual(1, record.NumUniqueStrings);
Assert.AreEqual(0, record.AddString(s1));
Assert.AreEqual(s1, record.GetString(0));
Assert.AreEqual(1, record.CountStrings);
Assert.AreEqual(2, record.NumStrings);
Assert.AreEqual(1, record.NumUniqueStrings);
Assert.AreEqual(1, record.AddString(s2));
Assert.AreEqual(s2, record.GetString(1));
Assert.AreEqual(2, record.CountStrings);
Assert.AreEqual(3, record.NumStrings);
Assert.AreEqual(2, record.NumUniqueStrings);
System.Collections.IEnumerator iter = record.GetStrings();
while (iter.MoveNext())
{
UnicodeString ucs = (UnicodeString)iter.Current;
if (ucs.Equals(s1))
{
Assert.AreEqual((byte)0, ucs.OptionFlags);
}
else if (ucs.Equals(s2))
{
Assert.AreEqual((byte)1, ucs.OptionFlags);
}
else
{
Assert.Fail("cannot match string: " + ucs.String);
}
}
}
public void TestSimpleConstructor()
{
SSTRecord record = new SSTRecord();
Assert.AreEqual(0, record.NumStrings);
Assert.AreEqual(0, record.NumUniqueStrings);
Assert.AreEqual(0, record.CountStrings);
byte[] output = record.Serialize();
byte[] expected =
{
(byte)record.Sid, (byte)(record.Sid >> 8),
(byte)8, (byte)0, (byte)0, (byte)0, (byte)0,
(byte)0, (byte)0, (byte)0, (byte)0, (byte)0
};
Assert.AreEqual(expected.Length, output.Length);
for (int k = 0; k < expected.Length; k++)
{
Assert.AreEqual(expected[k], output[k], k.ToString());
}
}
/**
* Writes out the shared string table
*
* @param outputFile the binary output file
* @exception IOException
*/
public void write(File outputFile)
{
// Thanks to Guenther for contributing the ExtSST implementation portion
// of this method
int charsLeft = 0;
string curString = null;
SSTRecord sst = new SSTRecord(totalOccurrences, stringList.Count);
ExtendedSSTRecord extsst = new ExtendedSSTRecord(stringList.Count);
int bucketSize = extsst.getNumberOfStringsPerBucket();
int stringIndex = 0;
// CML: this one is nasty....
IEnumerator iterator = stringList.GetEnumerator();
while (iterator.MoveNext() && charsLeft == 0)
{
curString = (string)iterator.Current;
// offset + header bytes
int relativePosition = sst.getOffset() + 4;
charsLeft = sst.add(curString);
if ((stringIndex % bucketSize) == 0)
extsst.addString(outputFile.getPos(), relativePosition);
stringIndex++;
}
outputFile.write(sst);
if (charsLeft != 0 || iterator.MoveNext())
{
// Add the remainder of the string to the continue record
SSTContinueRecord cont = createContinueRecord(curString,charsLeft,outputFile);
// Carry on looping through the array until all the strings are done
do
{
curString = (string)iterator.Current;
int relativePosition = cont.getOffset() + 4;
charsLeft = cont.add(curString);
if ((stringIndex % bucketSize) == 0)
extsst.addString(outputFile.getPos(), relativePosition);
stringIndex++;
if (charsLeft != 0)
{
outputFile.write(cont);
cont = createContinueRecord(curString, charsLeft, outputFile);
}
}
while (iterator.MoveNext());
outputFile.write(cont);
}
outputFile.write(extsst);
}
/// <summary>
/// Process an HSSF Record. Called when a record occurs in an HSSF file.
/// </summary>
/// <param name="record"></param>
public void ProcessRecord(Record record)
{
String thisText = null;
int thisRow = -1;
switch (record.Sid)
{
case BoundSheetRecord.sid:
BoundSheetRecord sr = (BoundSheetRecord)record;
sheetNames.Add(sr.Sheetname);
break;
case BOFRecord.sid:
BOFRecord bof = (BOFRecord)record;
if (bof.Type == BOFRecordType.Worksheet)
{
sheetNum++;
rowNum = -1;
if (includeSheetNames)
{
if (text.Length > 0)
{
text.Append("\n");
}
text.Append(sheetNames[sheetNum]);
}
}
break;
case SSTRecord.sid:
sstRecord = (SSTRecord)record;
break;
case FormulaRecord.sid:
FormulaRecord frec = (FormulaRecord)record;
thisRow = frec.Row;
if (formulasNotResults)
{
thisText = HSSFFormulaParser.ToFormulaString((HSSFWorkbook)null, frec.ParsedExpression);
}
else
{
if (frec.HasCachedResultString)
{
// Formula result is a string
// This is stored in the next record
outputNextStringValue = true;
nextRow = frec.Row;
}
else
{
thisText = FormatNumberDateCell(frec, frec.Value);
}
}
break;
case StringRecord.sid:
if (outputNextStringValue)
{
// String for formula
StringRecord srec = (StringRecord)record;
thisText = srec.String;
thisRow = nextRow;
outputNextStringValue = false;
}
break;
case LabelRecord.sid:
LabelRecord lrec = (LabelRecord)record;
thisRow = lrec.Row;
thisText = lrec.Value;
break;
case LabelSSTRecord.sid:
LabelSSTRecord lsrec = (LabelSSTRecord)record;
thisRow = lsrec.Row;
if (sstRecord == null)
{
throw new Exception("No SST record found");
}
thisText = sstRecord.GetString(lsrec.SSTIndex).ToString();
break;
case NoteRecord.sid:
NoteRecord nrec = (NoteRecord)record;
thisRow = nrec.Row;
// TODO: Find object to match nrec.GetShapeId()
break;
case NumberRecord.sid:
NumberRecord numrec = (NumberRecord)record;
thisRow = numrec.Row;
thisText = FormatNumberDateCell(numrec, numrec.Value);
break;
default:
break;
}
if (thisText != null)
{
if (thisRow != rowNum)
{
rowNum = thisRow;
if (text.Length > 0)
{
text.Append("\n");
}
}
else
{
text.Append("\t");
}
text.Append(thisText);
}
}
/// <summary>
/// Main HSSFListener method, processes events, and outputs the
/// strings as the file is processed
/// </summary>
public void ProcessRecord(Record record)
{
var thisRow = -1;
string thisStr = null;
switch (record.Sid)
{
case BoundSheetRecord.sid:
_boundSheetRecords.Add(record);
break;
case BOFRecord.sid:
var br = (BOFRecord)record;
if (br.Type == BOFRecord.TYPE_WORKSHEET)
{
// Create sub workbook if required
if (_workbookBuildingListener != null && _stubWorkbook == null)
{
_stubWorkbook = _workbookBuildingListener.GetStubHSSFWorkbook();
}
// Output the worksheet name
// Works by ordering the BSRs by the location of
// their BOFRecords, and then knowing that we
// process BOFRecords in byte offset order
if (_orderedBsRs == null)
{
_orderedBsRs = BoundSheetRecord.OrderByBofPosition(_boundSheetRecords);
}
}
break;
case SSTRecord.sid:
_sstRecord = (SSTRecord)record;
break;
case BlankRecord.sid:
var brec = (BlankRecord)record;
thisRow = brec.Row;
thisStr = "";
break;
case BoolErrRecord.sid:
var berec = (BoolErrRecord)record;
thisRow = berec.Row;
thisStr = "";
break;
case FormulaRecord.sid:
var frec = (FormulaRecord)record;
thisRow = frec.Row;
if (OutputFormulaValues)
{
if (double.IsNaN(frec.Value))
{
// Formula result is a string
// This is stored in the next record
_outputNextStringRecord = true;
_nextRow = frec.Row;
}
else
{
thisStr = _formatListener.FormatNumberDateCell(frec);
}
}
else
{
thisStr = HSSFFormulaParser.ToFormulaString(_stubWorkbook, frec.ParsedExpression);
}
break;
case StringRecord.sid:
if (_outputNextStringRecord)
{
// String for formula
var srec = (StringRecord)record;
thisStr = srec.String;
thisRow = _nextRow;
_outputNextStringRecord = false;
}
break;
case LabelRecord.sid:
var lrec = (LabelRecord)record;
thisRow = lrec.Row;
thisStr = lrec.Value;
break;
case LabelSSTRecord.sid:
var lsrec = (LabelSSTRecord)record;
thisRow = lsrec.Row;
if (_sstRecord == null)
{
thisStr = "(No SST Record, can't identify string)";
}
else
{
thisStr = _sstRecord.GetString(lsrec.SSTIndex).ToString();
}
break;
case NoteRecord.sid:
var nrec = (NoteRecord)record;
thisRow = nrec.Row;
thisStr = "";
break;
case NumberRecord.sid:
var numrec = (NumberRecord)record;
thisRow = numrec.Row;
// Format
thisStr = _formatListener.FormatNumberDateCell(numrec);
break;
case RKRecord.sid:
var rkrec = (RKRecord)record;
thisRow = rkrec.Row;
thisStr = "";
break;
}
// Handle new row
if (thisRow != -1 && thisRow != _lastRowNumber)
{
_row = new List <string>();
}
// Handle missing column
if (record is MissingCellDummyRecord)
{
var mc = (MissingCellDummyRecord)record;
thisRow = mc.Row;
thisStr = "";
}
// If we got something to print out, do so
if (thisStr != null)
{
_row.Add(thisStr);
}
// Update column and row count
if (thisRow > -1)
{
_lastRowNumber = thisRow;
}
// Handle end of row
if (record is LastCellOfRowDummyRecord)
{
// We're onto a new row
Output.Add(_row);
}
}
public void TestHugeStrings()
{
SSTRecord record = new SSTRecord();
byte[][] bstrings =
{
new byte[9000], new byte[7433], new byte[9002],
new byte[16998]
};
UnicodeString[] strings = new UnicodeString[bstrings.Length];
int total_length = 0;
for (int k = 0; k < bstrings.Length; k++)
{
Arrays.Fill(bstrings[k], (byte)(Convert.ToInt32('a') + k));
strings[k] = new UnicodeString(ConvertByteArrayToString(bstrings[k]));
record.AddString(strings[k]);
total_length += 3 + bstrings[k].Length;
}
// add overhead of SST record
total_length += 8;
// add overhead of broken strings
total_length += 4;
// add overhead of six records
total_length += (6 * 4);
byte[] content = new byte[record.RecordSize];
record.Serialize(0, content);
Assert.AreEqual(total_length, content.Length);
//DeSerialize the record.
RecordInputStream recStream = new RecordInputStream(new MemoryStream(content));
recStream.NextRecord();
record = new SSTRecord(recStream);
Assert.AreEqual(strings.Length, record.NumStrings);
Assert.AreEqual(strings.Length, record.NumUniqueStrings);
Assert.AreEqual(strings.Length, record.CountStrings);
for (int k = 0; k < strings.Length; k++)
{
Assert.AreEqual(strings[k], record.GetString(k));
}
record = new SSTRecord();
bstrings[1] = new byte[bstrings[1].Length - 1];
for (int k = 0; k < bstrings.Length; k++)
{
if ((bstrings[k].Length % 2) == 1)
{
Arrays.Fill(bstrings[k], (byte)('a' + k));
strings[k] = new UnicodeString(ConvertByteArrayToString(bstrings[k]));
}
else
{
char[] data = new char[bstrings[k].Length / 2];
Arrays.Fill(data, (char)(Convert.ToInt32('\u2122') + k));
strings[k] = new UnicodeString(new String(data));
}
record.AddString(strings[k]);
}
content = new byte[record.RecordSize];
record.Serialize(0, content);
total_length--;
Assert.AreEqual(total_length, content.Length);
recStream = new RecordInputStream(new MemoryStream(content));
recStream.NextRecord();
record = new SSTRecord(recStream);
Assert.AreEqual(strings.Length, record.NumStrings);
Assert.AreEqual(strings.Length, record.NumUniqueStrings);
Assert.AreEqual(strings.Length, record.CountStrings);
for (int k = 0; k < strings.Length; k++)
{
Assert.AreEqual(strings[k], record.GetString(k));
}
}
