/* AbiWord
 * Copyright (C) 1998 AbiSource, Inc.
 * 
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; either version 2
 * of the License, or (at your option) any later version.
 * 
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 * 
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  
 * 02111-1307, USA.
 */


#include <stdio.h>
#include <stdlib.h>
#include "ut_types.h"
#include "ut_assert.h"
#include "ut_debugmsg.h"
#include "ut_string.h"
#include "ie_imp_UCS2.h"
#include "pd_Document.h"
#include "ut_growbuf.h"

/*****************************************************************/
/*****************************************************************/

#ifdef ENABLE_PLUGINS

// completely generic code to allow this to be a plugin

#include "xap_Module.h"

#define SUPPORTS_ABI_VERSION(a,b,c) (((a==0)&&(b==7)&&(c==15)) ? 1 : 0)

// we use a reference-counted sniffer
static IE_Imp_UCS2_Sniffer * m_sniffer = 0;
static UT_sint32 m_refs = 0;

ABI_FAR extern "C"
int abi_plugin_register (XAP_ModuleInfo * mi)
{

	if (!m_refs && !m_sniffer)
	{
		m_sniffer = new IE_Imp_UCS2_Sniffer ();
		m_refs++;
	}
	else if (m_refs && m_sniffer)
	{
		m_refs++;
	}
	else
	{
		UT_ASSERT(UT_SHOULD_NOT_HAPPEN);
	}

	mi->name = "UCS2 Importer";
	mi->desc = "Import UCS2 Documents";
	mi->version = "0.7.15";
	mi->author = "Andrew Dunbar";
	mi->usage = "No Usage";

	IE_Imp::registerImporter (m_sniffer);
	return 1;
}

ABI_FAR extern "C"
int abi_plugin_unregister (XAP_ModuleInfo * mi)
{
	mi->name = 0;
	mi->desc = 0;
	mi->version = 0;
	mi->author = 0;
	mi->usage = 0;

	UT_ASSERT (m_refs && m_sniffer);

	m_refs--;
	IE_Imp::unregisterImporter (m_sniffer);
	if (!m_refs)
	{
		delete m_sniffer;
		m_sniffer = 0;
	}

	return 1;
}

ABI_FAR extern "C"
int abi_plugin_supports_version (UT_uint32 major, UT_uint32 minor, 
								 UT_uint32 release)
{
	return SUPPORTS_ABI_VERSION(major, minor, release);
}

#endif

/*****************************************************************/
/*****************************************************************/

bool IE_Imp_UCS2_Sniffer::recognizeContents(const char * szBuf, UT_uint32 iNumbytes)
{
	bool bSuccess = false;
	
	if (iNumbytes >= 2)
	{
		const unsigned char *p = reinterpret_cast<const unsigned char *>(szBuf);

		if (p[0] == 0xfe && p[1] == 0xff)
		{
			// Big endian
			UT_DEBUGMSG(("BOM big endian\n"));
			bSuccess = true;
		}
		else if (p[0] == 0xff && p[1] == 0xfe)
		{
			// Little endian
			UT_DEBUGMSG(("BOM little endian\n"));
			bSuccess = true;
		}
	}

	return bSuccess;
}

bool IE_Imp_UCS2_Sniffer::recognizeSuffix(const char * szSuffix)
{
	return (UT_stricmp(szSuffix,".ucs2") == 0);
}

UT_Error IE_Imp_UCS2_Sniffer::constructImporter(PD_Document * pDocument,
												IE_Imp ** ppie)
{
	IE_Imp_UCS2 * p = new IE_Imp_UCS2(pDocument);
	*ppie = p;
	return UT_OK;
}

bool	IE_Imp_UCS2_Sniffer::getDlgLabels(const char ** pszDesc,
										  const char ** pszSuffixList,
										  IEFileType * ft)
{
	*pszDesc = "UCS2 (.ucs2)";
	*pszSuffixList = "*.ucs2";
	*ft = getFileType();
	return true;
}

/*****************************************************************/
/*****************************************************************/

/*
  Import a plain text file formatted in UCS2.
  We allow either LF or CR or CRLF line termination.
  Each line terminator is taken to be a paragraph break.
*/

/*****************************************************************/
/*****************************************************************/

#define X_CleanupIfError(error,exp)	do { if (((error)=(exp)) != UT_OK) goto Cleanup; } while (0)

UT_Error IE_Imp_UCS2::importFile(const char * szFilename)
{
	FILE *fp = fopen(szFilename, "r");
	if (!fp)
	{
		UT_DEBUGMSG(("Could not open file %s\n",szFilename));
		return UT_IE_FILENOTFOUND;
	}
	
	UT_Error error;

	X_CleanupIfError(error,_writeHeader(fp));
	X_CleanupIfError(error,_parseFile(fp));

	error = UT_OK;

Cleanup:
	fclose(fp);
	return error;
}

#undef X_CleanupIfError

/*****************************************************************/
/*****************************************************************/

IE_Imp_UCS2::~IE_Imp_UCS2()
{
}

IE_Imp_UCS2::IE_Imp_UCS2(PD_Document * pDocument)
	: IE_Imp(pDocument)
{
}

/*****************************************************************/
/*****************************************************************/

#define X_ReturnIfFail(exp,error)		do { bool b = (exp); if (!b) return (error); } while (0)
#define X_ReturnNoMemIfError(exp)	X_ReturnIfFail(exp,UT_IE_NOMEMORY)

UT_Error IE_Imp_UCS2::_writeHeader(FILE * /* fp */)
{
	X_ReturnNoMemIfError(m_pDocument->appendStrux(PTX_Section, NULL));

	return UT_OK;
}

UT_Error IE_Imp_UCS2::_parseFile(FILE * fp)
{
	UT_GrowBuf gbBlock(1024);
	bool bEatLF = false;
	bool bEmptyFile = true;
	unsigned char b1, b2;
	UT_UCSChar c;
	bool bBigEndian = false;	// TODO Without BOMs should this be platform-dependant?

	while (1)
	{
		if (fread(&b1, 1, sizeof(b1), fp) > 0)
		{
			if (fread(&b2, 1, sizeof(b2), fp) > 0)
			{
				// Handle BOM
				if (b1 == 0xfe && b2 == 0xff)
				{
					bBigEndian = true;
					continue;
				}
				else if (b1 == 0xff && b2 == 0xfe)
				{
					bBigEndian = false;
					continue;
				}
				else
				{
					if (bBigEndian)
					{
						c = b1;
						c <<= 8;
						c |= b2;
					}
					else
					{
						c = b2;
						c <<= 8;
						c |= b1;
					}
				}
			}
			else
			{
				// Handle half character missing
				c = (UT_UCSChar)0xfffd;
			}
		}
		else
		{
			// TODO
			// Handle no more bytes
			break;
		}

		switch (c)
		{
		case '\r':
		case '\n':
			if ((c == '\n') && bEatLF)
			{
				bEatLF = false;
				break;
			}

			if (c == '\r')
			{
				bEatLF = true;
			}
			
			// we interprete either CRLF, CR, or LF as a paragraph break.
			
			// start a paragraph and emit any text that we
			// have accumulated.
			X_ReturnNoMemIfError(m_pDocument->appendStrux(PTX_Block, NULL));
			bEmptyFile = false;
			if (gbBlock.getLength() > 0)
			{
				X_ReturnNoMemIfError(m_pDocument->appendSpan(gbBlock.getPointer(0), gbBlock.getLength()));
				gbBlock.truncate(0);
			}
			break;

		default:
			bEatLF = false;
			X_ReturnNoMemIfError(gbBlock.ins(gbBlock.getLength(),&c,1));
			break;
		}
	}

	if (gbBlock.getLength() > 0 || bEmptyFile)
	{
		// if we have text left over (without final CR/LF),
		// or if we read an empty file,
		// create a paragraph and emit the text now.
		X_ReturnNoMemIfError(m_pDocument->appendStrux(PTX_Block, NULL));
		if (gbBlock.getLength() > 0)
			X_ReturnNoMemIfError(m_pDocument->appendSpan(gbBlock.getPointer(0), gbBlock.getLength()));
	}

	return UT_OK;
}

#undef X_ReturnNoMemIfError
#undef X_ReturnIfFail

/*****************************************************************/
/*****************************************************************/

void IE_Imp_UCS2::pasteFromBuffer(PD_DocumentRange * pDocRange,
								  unsigned char * pData, UT_uint32 lenData)
{
}