/****************************************************************************
**
** Implementation of QJisCodec class
**
** Created : 990225
**
** Copyright (C) 2000-2008 Trolltech ASA.  All rights reserved.
**
** This file is part of the tools module of the Qt GUI Toolkit.
**
** This file may be used under the terms of the GNU General
** Public License versions 2.0 or 3.0 as published by the Free
** Software Foundation and appearing in the files LICENSE.GPL2
** and LICENSE.GPL3 included in the packaging of this file.
** Alternatively you may (at your option) use any later version
** of the GNU General Public License if such license has been
** publicly approved by Trolltech ASA (or its successors, if any)
** and the KDE Free Qt Foundation.
**
** Please review the following information to ensure GNU General
** Public Licensing requirements will be met:
** http://trolltech.com/products/qt/licenses/licensing/opensource/.
** If you are unsure which license is appropriate for your use, please
** review the following information:
** http://trolltech.com/products/qt/licenses/licensing/licensingoverview
** or contact the sales department at sales@trolltech.com.
**
** This file may be used under the terms of the Q Public License as
** defined by Trolltech ASA and appearing in the file LICENSE.QPL
** included in the packaging of this file.  Licensees holding valid Qt
** Commercial licenses may use this file in accordance with the Qt
** Commercial License Agreement provided with the Software.
**
** This file is provided "AS IS" with NO WARRANTY OF ANY KIND,
** INCLUDING THE WARRANTIES OF DESIGN, MERCHANTABILITY AND FITNESS FOR
** A PARTICULAR PURPOSE. Trolltech reserves all rights not granted
** herein.
**
**********************************************************************/

// Most of the code here was originally written by Serika Kurusugawa,
// a.k.a. Junji Takagi, and is included in Qt with the author's permission
// and the grateful thanks of the Trolltech team.

/*! \class QJisCodec
    \reentrant
    \ingroup i18n

  \brief The QJisCodec class provides conversion to and from JIS character sets.

  More precisely, the QJisCodec class subclasses QTextCodec to
  provide support for JIS X 0201 Latin, JIS X 0201 Kana, JIS X 0208
  and JIS X 0212.

  The environment variable UNICODEMAP_JP can be used to fine-tune
  QJisCodec, QSjisCodec and QEucJpCodec. The mapping names are as for
  the Japanese XML working group's \link
  http://www.y-adagio.com/public/standards/tr_xml_jpf/toc.htm XML
  Japanese Profile\endlink, because it names and explains all the
  widely used mappings. Here are brief descriptions, written by
  Serika Kurusugawa:

  \list

  \i "unicode-0.9" or "unicode-0201" for Unicode style. This assumes
  JISX0201 for 0x00-0x7f. (0.9 is a table version of jisx02xx mapping
  used for Uniocde spec version 1.1.)

  \i "unicode-ascii" This assumes US-ASCII for 0x00-0x7f; some
  chars (JISX0208 0x2140 and JISX0212 0x2237) are different from
  Unicode 1.1 to avoid conflict.

  \i "open-19970715-0201" ("open-0201" for convenience) or
  "jisx0221-1995" for JISX0221-JISX0201 style. JIS X 0221 is JIS
  version of Unicode, but a few chars (0x5c, 0x7e, 0x2140, 0x216f,
  0x2131) are different from Unicode 1.1. This is used when 0x5c is
  treated as YEN SIGN.

  \i "open-19970715-ascii" ("open-ascii" for convenience) for
  JISX0221-ASCII style. This is used when 0x5c is treated as REVERSE
  SOLIDUS.

  \i "open-19970715-ms" ("open-ms" for convenience) or "cp932" for
  Microsoft Windows style. Windows Code Page 932. Some chars (0x2140,
  0x2141, 0x2142, 0x215d, 0x2171, 0x2172) are different from Unicode
  1.1.

  \i "jdk1.1.7" for Sun's JDK style. Same as Unicode 1.1, except that
  JIS 0x2140 is mapped to UFF3C. Either ASCII or JISX0201 can be used
  for 0x00-0x7f.

  \endlist

  In addition, the extensions "nec-vdc", "ibm-vdc" and "udc" are
  supported.

  For example, if you want to use Unicode style conversion but with
  NEC's extension, set \c UNICODEMAP_JP to
  <nobr>\c {unicode-0.9, nec-vdc}.</nobr> (You will probably
  need to quote that in a shell command.)

  Most of the code here was written by Serika Kurusugawa,
  a.k.a. Junji Takagi, and is included in Qt with the author's
  permission and the grateful thanks of the Trolltech team. Here is
  the copyright statement for that code:

  \legalese

  Copyright (C) 1999 Serika Kurusugawa. All rights reserved.

  Redistribution and use in source and binary forms, with or without
  modification, are permitted provided that the following conditions
  are met:
  \list 1
  \i Redistributions of source code must retain the above copyright
     notice, this list of conditions and the following disclaimer.
  \i Redistributions in binary form must reproduce the above copyright
     notice, this list of conditions and the following disclaimer in the
     documentation and/or other materials provided with the distribution.
  \endlist

  THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS "AS IS".
  ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  SUCH DAMAGE.
*/

#include "qjiscodec.h"

#ifndef QT_NO_BIG_CODECS

static const uchar Esc = 0x1b;
static const uchar So = 0x0e;	// Shift Out
static const uchar Si = 0x0f;	// Shift In

static const uchar ReverseSolidus = 0x5c;
static const uchar YenSign = 0x5c;
static const uchar Tilde = 0x7e;
static const uchar Overline = 0x7e;

#define	IsKana(c)	(((c) >= 0xa1) && ((c) <= 0xdf))
#define	IsJisChar(c)	(((c) >= 0x21) && ((c) <= 0x7e))

#define	QValidChar(u)	((u) ? QChar((ushort)(u)) : QChar::replacement)

enum Iso2022State{ Ascii, MinState = Ascii,
		   JISX0201_Latin, JISX0201_Kana,
		   JISX0208_1978, JISX0208_1983,
		   JISX0212, MaxState = JISX0212,
		   UnknownState };

static const char Esc_CHARS[] = "()*+-./";

static const char Esc_Ascii[]		= {Esc, '(', 'B', 0 };
static const char Esc_JISX0201_Latin[]	= {Esc, '(', 'J', 0 };
static const char Esc_JISX0201_Kana[]	= {Esc, '(', 'I', 0 };
static const char Esc_JISX0208_1978[]	= {Esc, '$', '@', 0 };
static const char Esc_JISX0208_1983[]	= {Esc, '$', 'B', 0 };
static const char Esc_JISX0212[]	= {Esc, '$', '(', 'D', 0 };
static const char * const Esc_SEQ[] = { Esc_Ascii,
					Esc_JISX0201_Latin,
					Esc_JISX0201_Kana,
					Esc_JISX0208_1978,
					Esc_JISX0208_1983,
					Esc_JISX0212 };

/*! \internal */
QJisCodec::QJisCodec() : conv(QJpUnicodeConv::newConverter(QJpUnicodeConv::Default))
{
}


/*! \internal */
QJisCodec::~QJisCodec()
{
    delete (QJpUnicodeConv*)conv;
    conv = 0;
}


/*! \internal */
int QJisCodec::mibEnum() const
{
    /*
    Name: JIS_Encoding
    MIBenum: 16
    Source: JIS X 0202-1991.  Uses ISO 2022 escape sequences to
	    shift code sets as documented in JIS X 0202-1991.
    Alias: csJISEncoding
    */
    return 16;
}

/*! \internal */
QCString QJisCodec::fromUnicode(const QString& uc, int& lenInOut) const
{
    int l = QMIN((int)uc.length(),lenInOut);
    QCString result;
    Iso2022State state = Ascii;
    Iso2022State prev = Ascii;
    for (int i=0; i<l; i++) {
	QChar ch = uc[i];
	uint j;
	if ( ch.row() == 0x00 && ch.cell() < 0x80 ) {
	    // Ascii
	    if (state != JISX0201_Latin ||
		ch.cell() == ReverseSolidus || ch.cell() == Tilde) {
		state = Ascii;
	    }
	    j = ch.cell();
	} else if ((j = conv->unicodeToJisx0201(ch.row(), ch.cell())) != 0) {
	    if (j < 0x80) {
		// JIS X 0201 Latin
		if (state != Ascii ||
		    ch.cell() == YenSign || ch.cell() == Overline) {
		    state = JISX0201_Latin;
		}
	    } else {
		// JIS X 0201 Kana
		state = JISX0201_Kana;
		j &= 0x7f;
	    }
	} else if ((j = conv->unicodeToJisx0208(ch.row(), ch.cell())) != 0) {
	    // JIS X 0208
	    state = JISX0208_1983;
	} else if ((j = conv->unicodeToJisx0212(ch.row(), ch.cell())) != 0) {
	    // JIS X 0212
	    state = JISX0212;
	} else {
	    // Invalid
	    state = UnknownState;
	    j = '?';
	}
	if (state != prev) {
	    if (state == UnknownState) {
		result += Esc_Ascii;
	    } else {
		result += Esc_SEQ[state - MinState];
	    }
	    prev = state;
	}
	if (j < 0x0100) {
	    result += j & 0xff;
	} else {
	    result += (j >> 8) & 0xff;
	    result += j & 0xff;
	}
    }
    if (prev != Ascii) {
	result += Esc_Ascii;
    }
    lenInOut = result.length();
    return result;
}

/*! \internal */
QString QJisCodec::toUnicode(const char* chars, int len) const
{
    QString result;
    Iso2022State state = Ascii, prev = Ascii;
    for (int i=0; i<len; i++) {
	uchar ch = chars[i];
	if ( ch == Esc ) {
	    // Escape sequence
	    state = UnknownState;
	    if ( i < len-1 ) {
		uchar c2 = chars[++i];
		if (c2 == '$') {
		    if ( i < len-1 ) {
			uchar c3 = chars[++i];
			if (strchr(Esc_CHARS, c3)) {
			    if ( i < len-1 ) {
				uchar c4 = chars[++i];
				if (c4 == '(') {
				    switch (c4) {
				      case 'D':
					state = JISX0212;	// Esc $ ( D
					break;
				    }
				}
			    }
			} else {
			    switch (c3) {
			      case '@':
				state = JISX0208_1978;	// Esc $ @
				break;
			      case 'B':
				state = JISX0208_1983;	// Esc $ B
				break;
			    }
			}
		    }
		} else {
		    if (strchr(Esc_CHARS, c2)) {
			if ( i < len-1 ) {
			    uchar c3 = chars[++i];
			    if (c2 == '(') {
				switch (c3) {
				  case 'B':
				    state = Ascii;	// Esc ( B
				    break;
				  case 'I':
				    state = JISX0201_Kana;	// Esc ( I
				    break;
				  case 'J':
				    state = JISX0201_Latin;	// Esc ( J
				    break;
				}
			    }
			}
		    }
		}
	    }
	} else if (ch == So) {
	    // Shift out
	    prev = state;
	    state = JISX0201_Kana;
	} else if (ch == Si) {
	    // Shift in
	    if (prev == Ascii || prev == JISX0201_Latin) {
		state = prev;
	    } else {
		state = Ascii;
	    }
	} else {
	    uint u;
	    switch (state) {
	      case Ascii:
		if (ch < 0x80) {
		    result += QChar(ch);
		    break;
		}
		/* fall throught */
	      case JISX0201_Latin:
		u = conv->jisx0201ToUnicode(ch);
		result += QValidChar(u);
		break;
	      case JISX0201_Kana:
		u = conv->jisx0201ToUnicode(ch | 0x80);
		result += QValidChar(u);
		break;
	      case JISX0208_1978:
	      case JISX0208_1983:
		if ( i < len-1 ) {
		    uchar c2 = chars[++i];
		    u = conv->jisx0208ToUnicode(ch & 0x7f, c2 & 0x7f);
		    result += QValidChar(u);
		}
		break;
	      case JISX0212:
		if ( i < len-1 ) {
		    uchar c2 = chars[++i];
		    u = conv->jisx0212ToUnicode(ch & 0x7f, c2 & 0x7f);
		    result += QValidChar(u);
		}
		break;
	      default:
		result += QChar::replacement;
		break;
	    }
	}
    }
    return result;
}

/*! \internal */
const char* QJisCodec::name() const
{
    return "JIS7";
}

/*!
    Returns the codec's mime name.
*/
const char* QJisCodec::mimeName() const
{
    return "ISO-2022-JP";
}

/*! \internal */
int QJisCodec::heuristicNameMatch(const char* hint) const
{
    if ( qstrnicmp( hint, "ISO-2022-JP", 11 ) == 0 )
	return 10000;
    if ( simpleHeuristicNameMatch( "ISO-2022-JP-2", hint ) > 0 )
	return 10;

    int score = 0;
    bool ja = FALSE;
    if (qstrnicmp(hint, "ja_JP", 5) == 0 || qstrnicmp(hint, "japan", 5) == 0) {
	score += 3;
	ja = TRUE;
    } else if (qstrnicmp(hint, "ja", 2) == 0) {
	score += 2;
	ja = TRUE;
    }
    const char *p;
    if (ja) {
	p = strchr(hint, '.');
	if (p == 0) {
	    return score - 2;
	}
	p++;
    } else {
	p = hint;
    }
    if (p) {
	if ((qstricmp(p, "JIS") == 0) ||
	    (qstricmp(p, "JIS7") == 0) ||
	    (simpleHeuristicNameMatch("ISO-2022-JP", p) > 0)) {
	    return score + 4;
	}
    }
    return QTextCodec::heuristicNameMatch(hint);
}

/*! \internal */
int QJisCodec::heuristicContentMatch(const char* chars, int len) const
{
    int score = 0;
    Iso2022State state = Ascii, prev = Ascii;
    for (int i=0; i<len; i++) {
	uchar ch = chars[i];
	// No nulls allowed.
	if ( !ch )
	    return -1;
	if ( ch == Esc ) {
	    // Escape sequence
	    state = UnknownState;
	    if ( i < len-1 ) {
		uchar c2 = chars[++i];
		if (c2 == '$') {
		    if ( i < len-1 ) {
			uchar c3 = chars[++i];
			if (strchr(Esc_CHARS, c3)) {
			    if ( i < len-1 ) {
				uchar c4 = chars[++i];
				if (c4 == '(') {
				    switch (c4) {
				      case 'D':
					state = JISX0212;	// Esc $ ( D
					score++;
					break;
				    }
				}
			    }
			    score++;
			} else {
			    switch (c3) {
			      case '@':
				state = JISX0208_1978;	// Esc $ @
				score++;
				break;
			      case 'B':
				state = JISX0208_1983;	// Esc $ B
				score++;
				break;
			    }
			}
		    }
		    score++;
		} else {
		    if (strchr(Esc_CHARS, c2)) {
			if ( i < len-1 ) {
			    uchar c3 = chars[++i];
			    if (c2 == '(') {
				switch (c3) {
				  case 'B':
				    state = Ascii;	// Esc ( B
				    score++;
				    break;
				  case 'I':
				    state = JISX0201_Kana;	// Esc ( I
				    score++;
				    break;
				  case 'J':
				    state = JISX0201_Latin;	// Esc ( J
				    score++;
				    break;
				}
			    }
			}
			score++;
		    }
		}
	    }
	    if ( state == UnknownState ) {
		return -1;
	    }
	    score++;
	} else if (ch == So) {
	    // Shift out
	    prev = state;
	    state = JISX0201_Kana;
	    score++;
	} else if (ch == Si) {
	    // Shift in
	    if (prev == Ascii || prev == JISX0201_Latin) {
		state = prev;
	    } else {
		state = Ascii;
	    }
	    score++;
	} else {
	    switch (state) {
	      case Ascii:
	      case JISX0201_Latin:
		if ( ch < 32 && ch != '\t' && ch != '\n' && ch != '\r' ) {
		    // Suspicious
		    if ( score )
		      score--;
		} else {
		    // Inconclusive
		}
		break;
	      case JISX0201_Kana:
		if ( !IsKana(ch | 0x80) ) {
		    return -1;
		}
		score++;
		break;
	      case JISX0208_1978:
	      case JISX0208_1983:
	      case JISX0212:
		if ( !IsJisChar(ch) ) {
		    // Invalid
		    return -1;
		}
		if ( i < len-1 ) {
		    uchar c2 = chars[++i];
		    if ( !IsJisChar(c2) ) {
			// Invalid
			return -1;
		    }
		    score++;
		}
		score++;
		break;
	      default:
		return -1;
	    }
	}
    }
    return score;
}

class QJisDecoder : public QTextDecoder {
    uchar buf[4];
    int nbuf;
    Iso2022State state, prev;
    bool esc;
    const QJpUnicodeConv * const conv;
public:
    QJisDecoder(const QJpUnicodeConv *c) : nbuf(0), state(Ascii), prev(Ascii), esc(FALSE), conv(c)
    {
    }

    QString toUnicode(const char* chars, int len)
    {
	QString result;
	for (int i=0; i<len; i++) {
	    uchar ch = chars[i];
	    if (esc) {
		// Escape sequence
		state = UnknownState;
		switch (nbuf) {
		  case 0:
		    if (ch == '$' || strchr(Esc_CHARS, ch)) {
			buf[nbuf++] = ch;
		    } else {
			nbuf = 0;
			esc = FALSE;
		    }
		    break;
		  case 1:
		    if (buf[0] == '$') {
			if (strchr(Esc_CHARS, ch)) {
			    buf[nbuf++] = ch;
			} else {
			    switch (ch) {
			      case '@':
				state = JISX0208_1978;	// Esc $ @
				break;
			      case 'B':
				state = JISX0208_1983;	// Esc $ B
				break;
			    }
			    nbuf = 0;
			    esc = FALSE;
			}
		    } else {
			if (buf[0] == '(') {
			    switch (ch) {
			      case 'B':
				state = Ascii;	// Esc ( B
				break;
			      case 'I':
				state = JISX0201_Kana;	// Esc ( I
				break;
			      case 'J':
				state = JISX0201_Latin;	// Esc ( J
				break;
			    }
			}
			nbuf = 0;
			esc = FALSE;
		    }
		    break;
		  case 2:
		    if (buf[1] == '(') {
			switch (ch) {
			  case 'D':
			    state = JISX0212;	// Esc $ ( D
			    break;
			}
		    }
		    nbuf = 0;
		    esc = FALSE;
		    break;
		}
	    } else {
		if (ch == Esc) {
		    // Escape sequence
		    nbuf = 0;
		    esc = TRUE;
		} else if (ch == So) {
		    // Shift out
		    prev = state;
		    state = JISX0201_Kana;
		    nbuf = 0;
		} else if (ch == Si) {
		    // Shift in
		    if (prev == Ascii || prev == JISX0201_Latin) {
			state = prev;
		    } else {
			state = Ascii;
		    }
		    nbuf = 0;
		} else {
		    uint u;
		    switch (nbuf) {
		      case 0:
			switch (state) {
			  case Ascii:
			    if (ch < 0x80) {
				result += QChar(ch);
				break;
			    }
			    /* fall throught */
			  case JISX0201_Latin:
			    u = conv->jisx0201ToUnicode(ch);
			    result += QValidChar(u);
			    break;
			  case JISX0201_Kana:
			    u = conv->jisx0201ToUnicode(ch | 0x80);
			    result += QValidChar(u);
			    break;
			  case JISX0208_1978:
			  case JISX0208_1983:
			  case JISX0212:
			    buf[nbuf++] = ch;
			    break;
			  default:
			    result += QChar::replacement;
			    break;
			}
			break;
		      case 1:
			switch (state) {
			  case JISX0208_1978:
			  case JISX0208_1983:
			    u = conv->jisx0208ToUnicode(buf[0] & 0x7f, ch & 0x7f);
			    result += QValidChar(u);
			    break;
			  case JISX0212:
			    u = conv->jisx0212ToUnicode(buf[0] & 0x7f, ch & 0x7f);
			    result += QValidChar(u);
			    break;
			  default:
			    result += QChar::replacement;
			    break;
			}
			nbuf = 0;
			break;
		    }
		}
	    }
	}
	return result;
    }
};

/*! \internal */
QTextDecoder* QJisCodec::makeDecoder() const
{
    return new QJisDecoder(conv);
}

#endif