sl@0: /*
sl@0: * Copyright (c) 1997-2009 Nokia Corporation and/or its subsidiary(-ies).
sl@0: * All rights reserved.
sl@0: * This component and the accompanying materials are made available
sl@0: * under the terms of "Eclipse Public License v1.0"
sl@0: * which accompanies this distribution, and is available
sl@0: * at the URL "http://www.eclipse.org/legal/epl-v10.html".
sl@0: *
sl@0: * Initial Contributors:
sl@0: * Nokia Corporation - initial contribution.
sl@0: *
sl@0: * Contributors:
sl@0: *
sl@0: * Description:       
sl@0: *
sl@0: */
sl@0: 
sl@0: 
sl@0: #include <e32std.h>
sl@0: #include <charconv.h>
sl@0: #include "SHIFTJIS_2.H"
sl@0: #include <ecom/implementationproxy.h>
sl@0: #include "charactersetconverter.h"
sl@0: #include "featmgr/featmgr.h"
sl@0: 
sl@0: /**
sl@0: Shift-JIS character converter wrapper
sl@0: 
sl@0: @internalTechnology 
sl@0: @released 9.1
sl@0: */
sl@0: class CShiftJisConverterImpl : public CCharacterSetConverterPluginInterface
sl@0: 	{
sl@0: 
sl@0: public:
sl@0: 	virtual const TDesC8& ReplacementForUnconvertibleUnicodeCharacters();
sl@0: 
sl@0: 	virtual TInt ConvertFromUnicode(
sl@0: 		CCnvCharacterSetConverter::TEndianness aDefaultEndiannessOfForeignCharacters, 
sl@0: 		const TDesC8& aReplacementForUnconvertibleUnicodeCharacters, 
sl@0: 		TDes8& aForeign, 
sl@0: 		const TDesC16& aUnicode, 
sl@0: 		CCnvCharacterSetConverter::TArrayOfAscendingIndices& aIndicesOfUnconvertibleCharacters);
sl@0: 
sl@0: 	virtual TInt ConvertToUnicode(
sl@0: 		CCnvCharacterSetConverter::TEndianness aDefaultEndiannessOfForeignCharacters, 
sl@0: 		TDes16& aUnicode, 
sl@0: 		const TDesC8& aForeign, 
sl@0: 		TInt& aState, 
sl@0: 		TInt& aNumberOfUnconvertibleCharacters, 
sl@0: 		TInt& aIndexOfFirstByteOfFirstUnconvertibleCharacter);
sl@0: 
sl@0: 	virtual TBool IsInThisCharacterSetL(
sl@0: 		TBool& aSetToTrue, 
sl@0: 		TInt& aConfidenceLevel, 
sl@0: 		const TDesC8& aSample);
sl@0: 
sl@0: 	static CShiftJisConverterImpl* NewL();
sl@0: 	virtual ~CShiftJisConverterImpl();
sl@0: 
sl@0: private:
sl@0: 	CShiftJisConverterImpl();
sl@0: 	void ConstructL();
sl@0: 
sl@0: 	};
sl@0: 
sl@0: /**
sl@0: Get the the Shift-JIS byte sequence which will replace any Unicode characters which can't be converted.
sl@0: 
sl@0: @return The Shift-JIS byte sequence which will replace any Unicode characters which can't be converted.
sl@0: @internalTechnology 
sl@0: */
sl@0: const TDesC8& CShiftJisConverterImpl::ReplacementForUnconvertibleUnicodeCharacters()
sl@0: 	{
sl@0: 	return CnvShiftJis::ReplacementForUnconvertibleUnicodeCharacters();
sl@0: 	}
sl@0: 
sl@0: TInt CShiftJisConverterImpl::ConvertFromUnicode(
sl@0: 		CCnvCharacterSetConverter::TEndianness aDefaultEndiannessOfForeignCharacters, 
sl@0: 		const TDesC8& aReplacementForUnconvertibleUnicodeCharacters, 
sl@0: 		TDes8& aForeign, 
sl@0: 		const TDesC16& aUnicode, 
sl@0: 		CCnvCharacterSetConverter::TArrayOfAscendingIndices& aIndicesOfUnconvertibleCharacters)
sl@0: 	{
sl@0: 	return CnvShiftJis::ConvertFromUnicode(aDefaultEndiannessOfForeignCharacters, aReplacementForUnconvertibleUnicodeCharacters, aForeign, aUnicode, aIndicesOfUnconvertibleCharacters);
sl@0: 	}
sl@0: 
sl@0: 
sl@0: /**
sl@0:  Converts Shift-JIS encoded input text to Unicode
sl@0:  
sl@0:  NOTE: For debugging the selected character set is returned in the state.
sl@0:  
sl@0:   @released  9.1
sl@0:   @param     aDefaultEndiannessOfForeignCharacters The default endian-ness to use when reading characters
sl@0:              in the foreign character set.
sl@0:   @param     aUnicode On return, contains the text converted into Unicode.
sl@0:   @param     aForeign The non-Unicode source text to be converted.
sl@0:   @param     aState Used to save state information across multiple calls
sl@0:              to <code>ConvertToUnicode()</code>.
sl@0:   @param     aNumberOfUnconvertibleCharacters On return, contains the number of bytes which were not
sl@0:              converted.
sl@0:   @param     aIndexOfFirstByteOfFirstUnconvertibleCharacter On return, contains the index of the first bytein the
sl@0:              input text that could not be converted. A negative
sl@0:              value indicates that all the characters were
sl@0:              converted.
sl@0:   @return 	 The number of unconverted bytes left at the end of the input descriptor 
sl@0:  		     (e.g. because the output descriptor is not long enough to hold all the text), 
sl@0:  		     or one of the error values defined in TError. 
sl@0:   @internalTechnology 
sl@0: */
sl@0: TInt CShiftJisConverterImpl::ConvertToUnicode(
sl@0: 		CCnvCharacterSetConverter::TEndianness aDefaultEndiannessOfForeignCharacters, 
sl@0: 		TDes16& aUnicode, 
sl@0: 		const TDesC8& aForeign, 
sl@0: 		TInt& /*aState*/, 
sl@0: 		TInt& aNumberOfUnconvertibleCharacters, 
sl@0: 		TInt& aIndexOfFirstByteOfFirstUnconvertibleCharacter)
sl@0: 	{
sl@0: 	return CnvShiftJis::ConvertToUnicode(aDefaultEndiannessOfForeignCharacters, aUnicode, aForeign, aNumberOfUnconvertibleCharacters, aIndexOfFirstByteOfFirstUnconvertibleCharacter);
sl@0: 	}
sl@0: 
sl@0: 
sl@0: /**
sl@0:  This API is used by CCnvCharacterSetConverter::AutoDetectCharacterSetL(). 
sl@0:  This method returns a value between 0 and 100, indicating how likely it 
sl@0:  is that this is the correct converter, for the text supplied.  
sl@0:  @internalTechnology 
sl@0:  */
sl@0: TBool CShiftJisConverterImpl::IsInThisCharacterSetL(
sl@0: 		TBool& aSetToTrue, 
sl@0: 		TInt& aConfidenceLevel, 
sl@0: 		const TDesC8& aSample)
sl@0: 	{
sl@0: 	aSetToTrue=ETrue;
sl@0: 	TInt sampleLength = aSample.Length();
sl@0: 	aConfidenceLevel = 0;
sl@0: 	TInt numberOfShiftJis=0;
sl@0: 	TInt occurrence=0;
sl@0: 	for (TInt i = 0; i < sampleLength; ++i)
sl@0: 		{
sl@0: 		// Check for JISX 0208:1997 Charset
sl@0: 		// First Byte in range 0x81-0x9f, 0xe0-0xef
sl@0: 		if (((aSample[i] >= 0x81) && (aSample[i] <= 0x9f)) ||
sl@0: 			((aSample[i] >= 0xe0) && (aSample[i] <= 0xef)))
sl@0: 			{
sl@0: 			// check that the second byte is in range as well 
sl@0: 			TInt increment1 = i+1;
sl@0: 			if(increment1 >= sampleLength)
sl@0: 				break;
sl@0: 			if (((aSample[increment1] >= 0x40) && (aSample[increment1] <= 0x7e)) ||
sl@0: 				((aSample[increment1] >= 0x80) && (aSample[increment1] <= 0xfc)))
sl@0: 				{
sl@0: 				// increase the confidence of this sample as ShiftJis
sl@0: 				aConfidenceLevel=(aConfidenceLevel >0)?aConfidenceLevel+5:60;
sl@0: 	
sl@0: 				TUint charShiftJis=(aSample[i]<<8)|(aSample[increment1]);
sl@0: 				if ((charShiftJis>=0x829f)&&(charShiftJis<=0x82f1)||
sl@0: 					(charShiftJis>=0x8340)&&(charShiftJis<=0x8396))//those are kanas range
sl@0: 					occurrence++;
sl@0: 				numberOfShiftJis++;
sl@0: 				i++;
sl@0: 				}
sl@0: 			}
sl@0: 		// Check That no other Japanese escape sequence occur... if they do, cancel this and return 0
sl@0: 		// eg EUC-JP's SS(Single shift) characters followed by the
sl@0: 		if(aSample[i]==0x8e)
sl@0: 			{
sl@0: 			TInt increment1 = i+1;
sl@0: 			if(increment1 >= sampleLength)
sl@0: 				break;
sl@0: 			if ((aSample[increment1] >= 0xa1) && (aSample[increment1] <= 0xdf))
sl@0: 				{
sl@0: 				// This could be EUC-JP format..
sl@0: 				aConfidenceLevel=0;
sl@0: 				i++;
sl@0: 				}
sl@0: 			}
sl@0: 		if(aSample[i]==0x8f)
sl@0: 			{
sl@0: 			TInt increment1 = i+1;
sl@0: 			TInt increment2 = i+2;
sl@0: 			if((increment1 >= sampleLength) || (increment2 >= sampleLength))
sl@0: 				break;
sl@0: 			if (((aSample[increment1] >= 0xa1) && (aSample[increment1] <= 0xfe)) && 
sl@0: 				((aSample[increment2] >= 0xa1) && (aSample[increment2] <= 0xfe)))
sl@0: 				{
sl@0: 				// 	This is definitely EUC-JP format. 
sl@0: 				aConfidenceLevel=0;
sl@0: 				break;
sl@0: 				}
sl@0: 			}
sl@0:         // Check the half width Katakana
sl@0:         if (aSample[i]>=0xa1 && aSample[i]<=0xdf)
sl@0:             {
sl@0:             // increase the confidence of this sample as ShiftJis
sl@0:             aConfidenceLevel=(aConfidenceLevel > 0) ? aConfidenceLevel+5 : 75;
sl@0:             occurrence++;
sl@0:             numberOfShiftJis++;
sl@0:             }
sl@0:         else if (aSample[i]>=0xf0)
sl@0:             {
sl@0:             aConfidenceLevel=0;
sl@0:             }
sl@0: 		} // for 
sl@0: 
sl@0: 	if(numberOfShiftJis)
sl@0: 		{
sl@0: 		aConfidenceLevel=(aConfidenceLevel >100)?100:((aConfidenceLevel <0)?0:aConfidenceLevel);
sl@0: 		aConfidenceLevel=aConfidenceLevel-Max(0,(30-occurrence*100/numberOfShiftJis));
sl@0: 		}
sl@0: 	aConfidenceLevel=(aConfidenceLevel < 0)?0:aConfidenceLevel;
sl@0: 	return ETrue;
sl@0: 	}
sl@0: 
sl@0: 
sl@0: CShiftJisConverterImpl* CShiftJisConverterImpl::NewL()
sl@0: 	{
sl@0: 	CShiftJisConverterImpl* self = new(ELeave) CShiftJisConverterImpl();
sl@0: 	CleanupStack::PushL(self);
sl@0: 	self->ConstructL();
sl@0: 	CleanupStack::Pop(self);
sl@0: 	return self;
sl@0: 	}
sl@0: 
sl@0: 
sl@0: CShiftJisConverterImpl::~CShiftJisConverterImpl()
sl@0: 	{
sl@0:     FeatureManager::UnInitializeLib();	
sl@0: 	}
sl@0: 
sl@0: CShiftJisConverterImpl::CShiftJisConverterImpl()
sl@0: 	{
sl@0: 	}
sl@0: 
sl@0: 
sl@0: void CShiftJisConverterImpl::ConstructL()
sl@0: 	{
sl@0:     FeatureManager::InitializeLibL();	
sl@0: 	}
sl@0: 
sl@0: const TImplementationProxy ImplementationTable[] = 
sl@0: 	{
sl@0: 		IMPLEMENTATION_PROXY_ENTRY(0x10000FBD,	CShiftJisConverterImpl::NewL)
sl@0: 	};
sl@0: 
sl@0: 
sl@0: EXPORT_C const TImplementationProxy* ImplementationGroupProxy(TInt& aTableCount)
sl@0: 	{
sl@0: 	aTableCount = sizeof(ImplementationTable) / sizeof(TImplementationProxy);
sl@0: 
sl@0: 	return ImplementationTable;
sl@0: 	}