qgb18030codec.cpp

奇趣公司比较新的qt/emd版本

/****************************************************************************
**
** Copyright (C) 1992-2007 Trolltech ASA. All rights reserved.
**
** This file is part of the plugins of the Qt Toolkit.
**
** This file may be used under the terms of the GNU General Public
** License version 2.0 as published by the Free Software Foundation
** and appearing in the file LICENSE.GPL included in the packaging of
** this file.  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 [e-mail].
**
** In addition, as a special exception, Trolltech gives you certain
** additional rights. These rights are described in the Trolltech GPL
** Exception version 1.0, which can be found at
** http://www.trolltech.com/products/qt/gplexception/ and in the file
** GPL_EXCEPTION.txt in this package.
**
** In addition, as a special exception, Trolltech, as the sole copyright
** holder for Qt Designer, grants users of the Qt/Eclipse Integration
** plug-in the right for the Qt/Eclipse Integration to link to
** functionality provided by Qt Designer and its related libraries.
**
** Trolltech reserves all rights not expressly granted herein.
**
** This file is provided AS IS with NO WARRANTY OF ANY KIND, INCLUDING THE
** WARRANTY OF DESIGN, MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
**
****************************************************************************/

/*! \class QGb18030Codec qgb18030codec.h
  \reentrant
  \internal
*/

#include "qgb18030codec.h"

#ifndef QT_NO_TEXTCODEC
#define InRange(c, lower, upper)  (((c) >= (lower)) && ((c) <= (upper)))
#define IsLatin(c)        ((c) <= 0x7F)
#define IsByteInGb2312(c)        (InRange((c), 0xA1, 0xFE))
#define Is1stByte(c)        (InRange((c), 0x81, 0xFE))
#define Is2ndByteIn2Bytes(c)        (InRange((c), 0x40, 0xFE) && (c) != 0x7F)
#define Is2ndByteIn4Bytes(c)        (InRange((c), 0x30, 0x39))
#define Is2ndByte(c)        (Is2ndByteIn2Bytes(c) || Is2ndByteIn4Bytes(c))
#define Is3rdByte(c)        (InRange((c), 0x81, 0xFE))
#define Is4thByte(c)        (InRange((c), 0x30, 0x39))

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

/* User-defined areas:        UDA 1: 0xAAA1 - 0xAFFE (564/0)
                        UDA 2: 0xF8A1 - 0xFEFE (658/0)
                        UDA 3: 0xA140 - 0xA7A0 (672/0) */
#define IsUDA1(a, b)        (InRange((a), 0xAA, 0xAF) && InRange((b), 0xA1, 0xFE))
#define IsUDA2(a, b)        (InRange((a), 0xF8, 0xFE) && InRange((b), 0xA1, 0xFE))
#define IsUDA3(a, b)        (InRange((a), 0xA1, 0xA7) && InRange((b), 0x40, 0xA0) && ((b) != 0x7F))

typedef struct {
    quint8        tblBegin;
    quint8        tblEnd;
    quint16        tblOffset;
    quint16        algOffset;
} indexTbl_t;

static uint qt_Gb18030ToUnicode(const uchar *gbstr, int& len);
static int qt_UnicodeToGb18030(uint unicode, uchar *gbchar);
int qt_UnicodeToGbk(uint unicode, uchar *gbchar);

QGb18030Codec::QGb18030Codec()
{
}

/*! \reimp */
QByteArray QGb18030Codec::convertFromUnicode(const QChar *uc, int len, ConverterState *state) const
{
    char replacement = '?';
    int high = -1;
    if (state) {
        if (state->flags & ConvertInvalidToNull)
            replacement = 0;
        if (state->remainingChars)
            high = state->state_data[0];
    }
    int invalid = 0;

    int rlen = 4 *len + 1;
    QByteArray rstr;
    rstr.resize(rlen);
    uchar* cursor = (uchar*)rstr.data();

    //qDebug("QGb18030Codec::fromUnicode(const QString& uc, int& lenInOut = %d)", lenInOut);
    for (int i = 0; i < len; i++) {
        unsigned short ch = uc[i].unicode();
        int len;
        uchar buf[4];
        if (high >= 0) {
            if (ch >= 0xdc00 && ch < 0xe000) {
                // valid surrogate pair
                ++i;
                uint u = (high-0xd800)*0x400+(ch-0xdc00)+0x10000;
                len = qt_UnicodeToGb18030(u, buf);
                if (len >= 2) {
                    for (int j=0; j<len; j++)
                        *cursor++ = buf[j];
                } else {
                    *cursor++ = replacement;
                    ++invalid;
                }
                high = -1;
                continue;
            } else {
                *cursor++ = replacement;
                ++invalid;
                high = -1;
            }
        }

        if (ch < 0x80) {
            // ASCII
            *cursor++ = ch;
        } else if ((ch >= 0xd800 && ch < 0xdc00)) {
            // surrogates area. check for correct encoding
            // we need at least one more character, first the high surrogate, then the low one
            high = ch;
        } else if ((len = qt_UnicodeToGb18030(ch, buf)) >= 2) {
            for (int j=0; j<len; j++)
                *cursor++ = buf[j];
        } else {
            // Error
            *cursor++ = replacement;
            ++invalid;
        }
    }
    rstr.resize(cursor - (uchar*)rstr.constData());

    if (state) {
        state->invalidChars += invalid;
        state->state_data[0] = high;
        if (high)
            state->remainingChars = 1;
    }
    return rstr;
}

QString QGb18030Codec::convertToUnicode(const char* chars, int len, ConverterState *state) const
{
    uchar buf[4];
    int nbuf = 0;
    QChar replacement = QChar::ReplacementCharacter;
    if (state) {
        if (state->flags & ConvertInvalidToNull)
            replacement = QChar::Null;
        nbuf = state->remainingChars;
        buf[0] = (state->state_data[0] >> 24) & 0xff;
        buf[1] = (state->state_data[0] >> 16) & 0xff;
        buf[2] = (state->state_data[0] >>  8) & 0xff;
        buf[3] = (state->state_data[0] >>  0) & 0xff;
    }
    int invalid = 0;

    QString result;
    //qDebug("QGb18030Decoder::toUnicode(const char* chars, int len = %d)", len);
    for (int i = 0; i < len; i++) {
        uchar ch = chars[i];
        switch (nbuf) {
        case 0:
            if (ch < 0x80) {
                // ASCII
                result += QLatin1Char(ch);
            } else if (Is1stByte(ch)) {
                // GB18030?
                buf[0] = ch;
                nbuf = 1;
            } else {
                // Invalid
                result += replacement;
                ++invalid;
            }
            break;
        case 1:
            // GB18030 2 bytes
            if (Is2ndByteIn2Bytes(ch)) {
                buf[1] = ch;
                int clen = 2;
                uint u = qt_Gb18030ToUnicode(buf, clen);
                if (clen == 2) {
                    result += QValidChar(u);
                } else {
                    result += replacement;
                    ++invalid;
                }
                nbuf = 0;
            } else if (Is2ndByteIn4Bytes(ch)) {
                buf[1] = ch;
                nbuf = 2;
            } else {
                // Error
                result += replacement;
                ++invalid;
                nbuf = 0;
            }
            break;
        case 2:
            // GB18030 3 bytes
            if (Is3rdByte(ch)) {
                buf[2] = ch;
                nbuf = 3;
            } else {
                result += replacement;
                ++invalid;
                nbuf = 0;
            }
            break;
        case 3:
            // GB18030 4 bytes
            if (Is4thByte(ch)) {
                buf[3] = ch;