public static String GetDocumentSource(ref HTMLDocument doc, Encoding enc)
{
if (doc == null)
{
return(null);
}
bool IsUnicodeDetermined = false;
Encoding theEncoding = enc;
if (theEncoding == null)
{
theEncoding = Encoding.GetEncoding(0);
//Windows default
}
if (theEncoding != Encoding.GetEncoding(0))
{
//Don't try to detect unicode if we were
//passed an encoding other than the default
IsUnicodeDetermined = true;
}
// use the routine from htmlwrapper
MemoryStream memstream = new MemoryStream();
ComStream cstream = new ComStream(memstream);
IPersistStreamInit pStreamInit = (IPersistStreamInit)doc;
pStreamInit.Save(cstream, false);
StringBuilder Result = new StringBuilder();
//goto start of stream
memstream.Seek(0, SeekOrigin.Begin);
int iSize = 2048;
byte[] bytedata = new byte[2048];
int iBOMLength = 0;
while (true)
{
iSize = memstream.Read(bytedata, 0, bytedata.Length);
if (iSize > 0)
{
if (!IsUnicodeDetermined)
{
//look for byte order mark
bool IsUTF16LE = false;
bool IsUTF16BE = false;
bool IsUTF8 = false;
bool IsBOMPresent = false;
if ((bytedata[0] == 0xFF) & (bytedata[1] == 0xFE))//UTF16LE
{
IsUTF16LE = true;
IsBOMPresent = true;
}
if ((bytedata[0] == 0xFE) & (bytedata[1] == 0xFF))// UTF16BE
{
IsUTF16BE = true;
IsBOMPresent = true;
}
if ((bytedata[0] == 0xEF) & (bytedata[1] == 0xBB) & (bytedata[2] == 0xBF)) //UTF8
{
IsUTF8 = true;
IsBOMPresent = true;
}
//look for alternate zeroes
if (!IsUTF16LE & !IsUTF16BE & !IsUTF8)
{
if ((bytedata[1] == 0) & (bytedata[3] == 0) & (bytedata[5] == 0) & (bytedata[7] == 0))
{
IsUTF16LE = true; //best guess
}
}
if (IsUTF16LE)
{
theEncoding = Encoding.Unicode;
}
else if (IsUTF16BE)
{
theEncoding = Encoding.BigEndianUnicode;
}
else if (IsUTF8)
{
theEncoding = Encoding.UTF8;
}
if (IsBOMPresent)
{
//strip out the BOM
iBOMLength = theEncoding.GetPreamble().Length;
}
//don't repeat the test
IsUnicodeDetermined = true;
}
Result.Append(theEncoding.GetString(bytedata, iBOMLength, iSize));
}
else
{
break;
}
}
memstream.Close();
return(Result.ToString());
}
public static String GetDocumentSource(ref HTMLDocument doc, Encoding enc)
{
if (doc == null) return null;
bool IsUnicodeDetermined = false;
Encoding theEncoding = enc;
if (theEncoding == null)
{
theEncoding = Encoding.GetEncoding(0);
//Windows default
}
if (theEncoding != Encoding.GetEncoding(0))
{
//Don't try to detect unicode if we were
//passed an encoding other than the default
IsUnicodeDetermined = true;
}
// use the routine from htmlwrapper
MemoryStream memstream = new MemoryStream();
ComStream cstream = new ComStream(memstream);
IPersistStreamInit pStreamInit = (IPersistStreamInit)doc;
pStreamInit.Save(cstream, false);
StringBuilder Result = new StringBuilder();
//goto start of stream
memstream.Seek(0, SeekOrigin.Begin);
int iSize = 2048;
byte[] bytedata = new byte[2048];
int iBOMLength = 0;
while (true)
{
iSize = memstream.Read(bytedata, 0, bytedata.Length);
if (iSize > 0)
{
if (!IsUnicodeDetermined)
{
//look for byte order mark
bool IsUTF16LE = false;
bool IsUTF16BE = false;
bool IsUTF8 = false;
bool IsBOMPresent = false;
if ((bytedata[0] == 0xFF) & (bytedata[1] == 0xFE))//UTF16LE
{
IsUTF16LE = true;
IsBOMPresent = true;
}
if ((bytedata[0] == 0xFE) & (bytedata[1] == 0xFF))// UTF16BE
{
IsUTF16BE = true;
IsBOMPresent = true;
}
if ((bytedata[0] == 0xEF) & (bytedata[1] == 0xBB) & (bytedata[2] == 0xBF)) //UTF8
{
IsUTF8 = true;
IsBOMPresent = true;
}
//look for alternate zeroes
if (!IsUTF16LE & !IsUTF16BE & !IsUTF8)
{
if ((bytedata[1] == 0) & (bytedata[3] == 0) & (bytedata[5] == 0) & (bytedata[7] == 0))
{
IsUTF16LE = true; //best guess
}
}
if (IsUTF16LE)
{
theEncoding = Encoding.Unicode;
}
else if (IsUTF16BE)
{
theEncoding = Encoding.BigEndianUnicode;
}
else if (IsUTF8)
{
theEncoding = Encoding.UTF8;
}
if (IsBOMPresent)
{
//strip out the BOM
iBOMLength = theEncoding.GetPreamble().Length;
}
//don't repeat the test
IsUnicodeDetermined = true;
}
Result.Append(theEncoding.GetString(bytedata, iBOMLength, iSize));
}
else
{
break;
}
}
memstream.Close();
return Result.ToString();
}
