/*

* Copyright (c) 2005 Nokia Corporation and/or its subsidiary(-ies).

* All rights reserved.

* This component and the accompanying materials are made available

* under the terms of the License "Eclipse Public License v1.0"

* which accompanies this distribution, and is available

* at the URL "http://www.eclipse.org/legal/epl-v10.html".

*

* Initial Contributors:

* Nokia Corporation - initial contribution.

*

* Contributors:

*

* Description:  A misc. char-encoding related utilities.

*

*/

#include <charconv.h>

// -----------------------------------------------------------------------------

// CXmlEncoding::NewL

//

// Two-phased constructor.

// -----------------------------------------------------------------------------

//

EXPORT_C CXmlEncoding* CXmlEncoding::NewL()

    {

    CXmlEncoding* self = new (ELeave) CXmlEncoding();

    CleanupStack::PushL(self);

    self->ConstructL();

    CleanupStack::Pop();

    return self;

    }

// -----------------------------------------------------------------------------

// CXmlEncoding::CXmlEncoding

//

// C++ default constructor can NOT contain any code, that

// might leave.

// -----------------------------------------------------------------------------

//

CXmlEncoding::CXmlEncoding():

        iCharEncodings(5)

    {

    }

// -----------------------------------------------------------------------------

// CXmlEncoding::ConstructL

//

// Symbian 2nd phase constructor can leave.

// -----------------------------------------------------------------------------

//

void CXmlEncoding::ConstructL()

    {

    User::LeaveIfError(iRfs.Connect());

    iConverter = CCnvCharacterSetConverter::NewL();

    }

// -----------------------------------------------------------------------------

// CXmlEncoding::~CXmlEncoding

//

// Deconstructor.

// -----------------------------------------------------------------------------

//

CXmlEncoding::~CXmlEncoding()

    {

    iRfs.Close();

    delete iConverter;

    // Delete the cached encoding-map.

    for (TInt i = 0; i < iCharEncodings.Count(); i++)

        {

        delete iCharEncodings[i].charEncoding;

        }

    iCharEncodings.Close();

    }

// -----------------------------------------------------------------------------

// CXmlEncoding::ResolveCharEncodingL

// 

// Resolves the given char-encoding into its uid.

// -----------------------------------------------------------------------------

//

TBool CXmlEncoding::ResolveCharEncodingL(const TDesC8& aCharEncoding, TUint& aUid) const

    {

    TBool  found = EFalse;

    // First look it up in the cached encodings.

    for (TInt i = 0; i < iCharEncodings.Count(); i++)

        {

        if (iCharEncodings[i].charEncoding->CompareF(aCharEncoding) == 0)

            {

            aUid = iCharEncodings[i].uid;

            found = ETrue;

            break;

            }

        }

    // Otherwise use the CCnvCharacterSetConverter

    if (!found)

        {

        TUint  tid;

        // Look it up.

        tid = iConverter->ConvertStandardNameOfCharacterSetToIdentifierL(aCharEncoding, 

                const_cast<CXmlEncoding*>(this)->iRfs);

        // If found add it to the cached encodings.

        if (tid != 0)

            {

            SupportedEncodings  encoding;

            TInt                err;

            encoding.charEncoding = aCharEncoding.AllocL();

            encoding.uid = tid;

            err = const_cast<CXmlEncoding*>(this)->iCharEncodings.Append(encoding);

            if (err != KErrNone)

                {

                delete encoding.charEncoding;

                User::Leave(err);

                }

            aUid = tid;

            found = ETrue;

            }

        }

    return found;

    }

// -----------------------------------------------------------------------------

// XmlEncoding::DetermineCharEncoding

//

// Determine the char-encoding.

// -----------------------------------------------------------------------------

//

EXPORT_C TBool CXmlEncoding::DetermineCharEncodingL(const TDesC8& aBuffer, 

        const TDesC& aCharSet, TUint& aEncoding) const

    {

    TBool  foundEncoding = EFalse;

    // Try to determine the encoding via the BOM (byte order mask).

    foundEncoding = DetermineCharEncodingFromBom(aBuffer, aEncoding);

    // Try to determine the encoding via the xml-prolog.

    if (!foundEncoding)

        {

        foundEncoding = DetermineCharEncodingFromXmlProlog(aBuffer, aEncoding);

        }

    // Try to determine the encoding via the char-set provided by 

    // the orignal source.

    if (!foundEncoding)

        {

        HBufC8*  str = NULL;

        // Convert it to 8bit first.

        str = HBufC8::NewL(aCharSet.Length());

        CleanupStack::PushL(str);

        str->Des().Append(aCharSet);

        foundEncoding = ResolveCharEncodingL(*str, aEncoding);

        CleanupStack::PopAndDestroy(str);

        }

    return foundEncoding;

    }

// -----------------------------------------------------------------------------

// XmlEncoding::DetermineCharEncodingFromBom

//

// Determine the char-encoding from the BOM.

// -----------------------------------------------------------------------------

//

TBool CXmlEncoding::DetermineCharEncodingFromBom(const TDesC8& aBuffer, 

        TUint& aEncoding) const

    {

    _LIT8(KUcs2Big, "UTF-16BE");

    _LIT8(KUcs2Little, "UTF-16LE");

    _LIT8(KUtf8, "UTF-8");

    TBool  foundEncoding = EFalse;

    if (aBuffer.Length() < 3)

        {    

        return EFalse;

        }

    // Extract the first three bytes.

    TUint8 c1 = aBuffer[0];

    TUint8 c2 = aBuffer[1];

    TUint8 c3 = aBuffer[2];

    // Check for the BOM.

    if ((c1 == 0xFE) && (c2 == 0xFF))

        {

        TRAP_IGNORE(foundEncoding = ResolveCharEncodingL(KUcs2Big, aEncoding));

        }

    else if ((c1 == 0xFF) && (c2 == 0xFE))

        {

        TRAP_IGNORE(foundEncoding = ResolveCharEncodingL(KUcs2Little, aEncoding));

        } 

    else if ((c1 == 0xEF) && (c2 == 0xBB) && (c3 == 0xBF))

        {

        TRAP_IGNORE(foundEncoding = ResolveCharEncodingL(KUtf8, aEncoding));

        }

    return foundEncoding;

    }

// -----------------------------------------------------------------------------

// XmlEncoding::DetermineCharEncodingFromXmlProlog

//

// Determine the char-encoding from the char-encoding in the xml-prolog.

// -----------------------------------------------------------------------------

//

TBool CXmlEncoding::DetermineCharEncodingFromXmlProlog(const TDesC8& aBuffer, 

        TUint& aEncoding) const

    {

    TBool  foundEncoding = EFalse;

    TInt   begin;

    TInt   end;

    TInt   valueBegin;

    TInt   valueEnd;

    // Try to determine the encoding via the xml-prolog.

    if (FindEncoding(aBuffer, begin, end, valueBegin, valueEnd))

        {

        TPtrC8  str(aBuffer.Ptr() + valueBegin, valueEnd - valueBegin + 1);

        TRAP_IGNORE(foundEncoding = ResolveCharEncodingL(str, aEncoding));

        }

    return foundEncoding;

    }

// -----------------------------------------------------------------------------

// XmlEncoding::StripCharEncoding

//

// Erase the char-encoding attribute (if any) from the xml-prolog.

// -----------------------------------------------------------------------------

//

void CXmlEncoding::StripCharEncoding(HBufC8& aUtf8Buffer) const

    {

    TInt  begin;

    TInt  end;

    TInt  valueBegin;

    TInt  valueEnd;

    if (FindEncoding(aUtf8Buffer, begin, end, valueBegin, valueEnd))

        {

        aUtf8Buffer.Des().Delete(begin, end - begin + 1);

        }

    }

// -----------------------------------------------------------------------------

// XmlEncoding::StripCharEncoding

//

// Erase the char-encoding attribute (if any) from the xml-prolog.

// -----------------------------------------------------------------------------

//

void CXmlEncoding::StripCharEncoding(HBufC16& aUcs2Buffer) const

    {

    TInt  begin;

    TInt  end;

    TInt  valueBegin;

    TInt  valueEnd;

    if (FindEncoding(aUcs2Buffer, begin, end, valueBegin, valueEnd))

        {

        aUcs2Buffer.Des().Delete(begin, end - begin + 1);

        }

    }

// -----------------------------------------------------------------------------

// XmlEncoding::ConvertToUtf8L

//

// Converts the given native buffer into utf8.

// -----------------------------------------------------------------------------

//

EXPORT_C HBufC8* CXmlEncoding::ConvertToUtf8L(TUint aEncoding, const TDesC8& aBuffer) const

	{

	HBufC16*  usc2Buffer = NULL;

	HBufC8*   utf8Buffer = NULL;

    // First convert it to ucs2.

    usc2Buffer = ConvertToUcs2L(aEncoding, aBuffer);

    CleanupStack::PushL(usc2Buffer);

    // Then convert it to utf8.

    utf8Buffer = ConvertToUtf8L(*usc2Buffer);

    CleanupStack::PopAndDestroy(usc2Buffer);

    usc2Buffer = NULL;

    // Erase the char-encoding attribute (if any) from the xml-prolog.  If forces

    // Libxml2 to use the default encoding, utf8.

    StripCharEncoding(*utf8Buffer);

    return utf8Buffer;

	}

// -----------------------------------------------------------------------------

// XmlEncoding::ConvertToUcs2L

//

// Converts the given native buffer into ucs2.

// -----------------------------------------------------------------------------

//

EXPORT_C HBufC16* CXmlEncoding::ConvertToUcs2L(TUint aEncoding, const TDesC8& aBuffer) const

	{

	TBuf<100>  temp16Buffer;

	HBufC*     unicode = NULL;

	TPtrC8	   source8Ptr(aBuffer);

    // Init the converter and ensure the encoding is supported.

	if (iConverter->PrepareToConvertToOrFromL(aEncoding, const_cast<CXmlEncoding*>(this)->iRfs) != 

            CCnvCharacterSetConverter::EAvailable)

		{

		User::Leave(KErrNotSupported);

		}

	for(;;) // conversion loop

        {

        TInt  returnValue;

		TInt  state = CCnvCharacterSetConverter::KStateDefault;

		returnValue = iConverter->ConvertToUnicode(temp16Buffer, source8Ptr, state);

        if (returnValue == CCnvCharacterSetConverter::EErrorIllFormedInput)

			{

            User::Leave(KErrCorrupt);

			}

        else

			{

			if (returnValue < 0) // future-proof against "TError" expanding

				{

				User::Leave(KErrGeneral);

				}

			}

		if (!unicode)

			{

			unicode = temp16Buffer.AllocLC();

			}

		else

			{

			HBufC* tmp = unicode->ReAllocL(unicode->Length() + temp16Buffer.Length());

			CleanupStack::Pop(unicode);

			unicode = tmp;

			CleanupStack::PushL(unicode);

			unicode->Des().Append(temp16Buffer);

			}

        if (returnValue == 0)	// All is converted without Errors

			{

            break;

			}

		// There is "returnValue" bytes not converted yet

        source8Ptr.Set(source8Ptr.Right(returnValue));

        }

    // Erase the char-encoding attribute (if any) from the xml-prolog -- as the

    // encoding is no longer valid.

    StripCharEncoding(*unicode);

	CleanupStack::Pop(unicode);

	return unicode;

	}

// -----------------------------------------------------------------------------

// XmlEncoding::ConvertToUtf8L

//

// Converts the given ucs2 buffer into utf8.

// -----------------------------------------------------------------------------

//

HBufC8* CXmlEncoding::ConvertToUtf8L(HBufC16& aUsc2Buffer) const

	{

	TBuf8<100>	temp8Buffer;

	HBufC8*		target = NULL;

	TPtrC		source16Ptr(aUsc2Buffer);

    // Init the converter and ensure the encoding is supported.

	if (iConverter->PrepareToConvertToOrFromL(KCharacterSetIdentifierUtf8, 

	        const_cast<CXmlEncoding*>(this)->iRfs) != CCnvCharacterSetConverter::EAvailable)

		{

		User::Leave(KErrNotSupported);

		}

	for(;;) // conversion loop

        {

		TInt  returnValue;

        TInt  state = CCnvCharacterSetConverter::KStateDefault;

		returnValue = iConverter->ConvertFromUnicode(temp8Buffer, source16Ptr, state);

        if (returnValue == CCnvCharacterSetConverter::EErrorIllFormedInput)

			{

            User::Leave(KErrCorrupt);

			}

        else

			{

			if (returnValue < 0) // future-proof against "TError" expanding

				{

				User::Leave(KErrGeneral);

				}

			}

		if (!target)

			{

			target = temp8Buffer.AllocLC();

			}

		else

			{

			HBufC8* tmp = target->ReAllocL(target->Length() + temp8Buffer.Length());

			CleanupStack::Pop(target);

			target = tmp;

			CleanupStack::PushL(target);

			target->Des().Append(temp8Buffer);

			}

        if (returnValue == 0)	// All is converted without Errors

			{

            break;

			}

		// There is "returnValue" bytes not converted yet

        source16Ptr.Set(source16Ptr.Right(returnValue));

        }

	CleanupStack::Pop(target);

	return target;

	}

// -----------------------------------------------------------------------------

// XmlEncoding::FindEncoding

//

// Finds the location of the encoding attribute in the xml-prolog.

// Refer to: http://www.w3.org/TR/2000/REC-xml-20001006#sec-prolog-dtd.

// -----------------------------------------------------------------------------

//

TBool CXmlEncoding::FindEncoding(const TDesC8& aBuffer, TInt& aBegin, TInt& aEnd, 

        TInt& aValueBegin, TInt& aValueEnd) const

    {

    _LIT8(KPiStart, "<?xml");

    _LIT8(KPiEnd, "?>");

    _LIT8(KEncoding, "encoding");

    TInt    piStartLoc = KErrNotFound;

    TInt    piEndLoc = KErrNotFound;

    aBegin = KErrNotFound;

    aEnd = KErrNotFound;

    aValueBegin = KErrNotFound;

    aValueEnd = KErrNotFound;

    // Determine if the xml-prolog contains a char-encoding attribute.

    piStartLoc = aBuffer.FindF(KPiStart);

    piEndLoc = aBuffer.FindF(KPiEnd);

    aBegin = aBuffer.FindF(KEncoding);

    if ((piStartLoc != KErrNotFound) && (piEndLoc != KErrNotFound) && 

            (aBegin != KErrNotFound) && (piStartLoc < piEndLoc) && (aBegin < piEndLoc))

        { 

        TInt   i;

        TBool  foundEquals = EFalse;

        TInt   quoteCount = 0;

        // If so, find the end of the attribute.

        i = aBegin + KEncoding().Length();

        while (i < piEndLoc)

            {

            TInt8  c;

            c = aBuffer[i];

            // The '=' char must be the first non-whitespace after "encoding".

            if (!foundEquals)

                {

                if (c == '=')

                    {

                    foundEquals = ETrue;

                    }

                else if ((c != 0x20) && (c != 0x09) && (c != 0x0D) && (c != 0x0A))

                    {

                    // Give up.

                    return EFalse;

                    }

                }

            // Otherwise look for the two quotes

            else if ((c == '\"') || (c == '\''))

                {

                quoteCount++;

                if (quoteCount == 1)

                    {

                    aValueBegin = i + 1;

                    }

                else if (quoteCount == 2)

                    {

                    aEnd = i;

                    aValueEnd = i - 1;

                    break;

                    }

                }

            i++;

            }

        }