PieceDescriptorCollection GetPieceDescriptors(BinaryReader reader, int offset, uint length)
{
PieceDescriptorCollection result = new PieceDescriptorCollection(offset, length);
result.Read(reader);
return(result);
}
bool LoadText(OleStorage storage, out string text)
{
text = string.Empty;
if (storage == null)
{
return(false);
}
BinaryReader documentReader = GetDocumentStreamReader(storage);
if (documentReader == null)
{
return(false);
}
int pdcOffset;
uint pdcLength;
string tableName;
GetDataFromFib(documentReader, out tableName, out pdcOffset, out pdcLength);
BinaryReader tableReader = GetTableStreamReader(storage, tableName);
if (tableReader == null)
{
return(false);
}
PieceDescriptorCollection pieces = GetPieceDescriptors(tableReader, pdcOffset, pdcLength);
if (pieces == null)
{
return(false);
}
//EDIT: Chris Wilson 04/01/16
//We have had some performance issues with legacy word docs, and I suspect that
//it's the use of overloaded operator concatenation in the below loop
//rather than using a stringbuilder
//change is to address this speed issue as we are encountering documents with word counts
//in the several 100ks
//ADDED: as per comment above
StringBuilder builder = new StringBuilder();
int count = pieces.Count;
for (int i = 0; i < count; i++)
{
uint pieceStart;
uint pieceEnd;
bool isUnicode = pieces.GetPieceFileBounds(i, out pieceStart, out pieceEnd);
documentReader.BaseStream.Seek(pieceStart, SeekOrigin.Begin);
//REMOVED: Chris Wilson as per comment above
//text += ReadString(documentReader, pieceEnd - pieceStart, isUnicode);
//ADDED: Chris Wilson as per comment above
builder.Append(ReadString(documentReader, pieceEnd - pieceStart, isUnicode));
}
//ADDED: Chris Wilson as per comment above
text = builder.ToString();
return(true);
}