[wxWidgets.git] / src / common / encconv.cpp

/////////////////////////////////////////////////////////////////////////////
// Name:        encconv.cpp
// Purpose:     wxEncodingConverter class for converting between different
//              font encodings
// Author:      Vaclav Slavik
// Copyright:   (c) 1999 Vaclav Slavik
// Licence:     wxWindows Licence
/////////////////////////////////////////////////////////////////////////////

#ifdef __GNUG__
#pragma implementation "encconv.h"
#endif

// For compilers that support precompilation, includes "wx.h".
#include "wx/wxprec.h"

#ifdef __BORLANDC__
  #pragma hdrstop
#endif

#include "wx/encconv.h"

#include <stdlib.h>

// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl:
#ifdef __BORLANDC__
#include "../common/unictabl.inc"
#else
#include "unictabl.inc"
#endif

#if wxUSE_WCHAR_T
typedef wchar_t tchar;
#else
typedef char tchar;
#endif

static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc)
{
    for (int i = 0; encodings_list[i].table != NULL; i++)
    {
        if (encodings_list[i].encoding == enc)
            return encodings_list[i].table;
    }
    return NULL;
}

typedef struct {
    wxUint16 u;
    wxUint8  c;
} CharsetItem;


static int LINKAGEMODE CompareCharsetItems(const void *i1, const void *i2)
{
    return ( ((CharsetItem*)i1) -> u - ((CharsetItem*)i2) -> u );
}


static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl)
{
    CharsetItem *rev = new CharsetItem[128];

    for (int i = 0; i < 128; i++)
        rev[i].c = 128 + i, rev[i].u = tbl[i];

    qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems);

    return rev;
}


wxEncodingConverter::wxEncodingConverter()
{
    m_Table = NULL;
    m_UnicodeInput = m_UnicodeOutput = FALSE;
    m_JustCopy = FALSE;
}


bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
{
    unsigned i;
    wxUint16 *in_tbl = NULL, *out_tbl = NULL;

    if (m_Table) {delete[] m_Table; m_Table = NULL;}

#if !wxUSE_WCHAR_T
    if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return FALSE;
#endif

    if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;}

    m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE);
    m_JustCopy = FALSE;

    if (input_enc == wxFONTENCODING_UNICODE)
    {
        if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;

        m_Table = new tchar[65536];
        for (i = 0; i < 128; i++)  m_Table[i] = (tchar)i; // 7bit ASCII
        for (i = 128; i < 65536; i++)  m_Table[i] = (tchar)'?';
                // FIXME - this should be character that means `unicode to charset' impossible, not '?'

        if (method == wxCONVERT_SUBSTITUTE)
        {
            for (i = 0; i < encoding_unicode_fallback_count; i++)
                m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s;
        }

        for (i = 0; i < 128; i++)
            m_Table[out_tbl[i]] = (tchar)(128 + i);

        m_UnicodeInput = TRUE;
        return TRUE;
    }

    else
    {
        if ((in_tbl = GetEncTable(input_enc)) == NULL) return FALSE;
        if (output_enc != wxFONTENCODING_UNICODE)
            if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;

        m_UnicodeInput = FALSE;

        m_Table = new tchar[256];
        for (i = 0; i < 128; i++)  m_Table[i] = (tchar)i; // 7bit ASCII

        if (output_enc == wxFONTENCODING_UNICODE)
        {
            for (i = 0; i < 128; i++)  m_Table[128 + i] = (tchar)in_tbl[i];
            return TRUE;
        }
        else
        {
            CharsetItem *rev = BuildReverseTable(out_tbl);
            CharsetItem *item;
            CharsetItem key;

            for (i = 0; i < 128; i++)
            {
                key.u = in_tbl[i];
                item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems);
                if (item == NULL && method == wxCONVERT_SUBSTITUTE)
                    item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback,
                                encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems);
                if (item)
                    m_Table[128 + i] = (tchar)item -> c;
                else
#if wxUSE_WCHAR_T
                    m_Table[128 + i] = (wchar_t)(128 + i);
#else
                    m_Table[128 + i] = (char)(128 + i);
#endif					
            }

            delete[] rev;
            return TRUE;
        }
    }
}


void wxEncodingConverter::Convert(const char* input, char* output)
{
    wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
    wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));

    const char *i;
    char *o;

    if (m_JustCopy)
    {
        strcpy(output, input);
        return;
    }

    wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));

    for (i = input, o = output; *i != 0;)
        *(o++) = (char)(m_Table[(wxUint8)*(i++)]);
    *o = 0;
}


#if wxUSE_WCHAR_T

void wxEncodingConverter::Convert(const char* input, wchar_t* output)
{
    wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
    wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));

    const char *i;
    wchar_t *o;

    if (m_JustCopy)
    {
        for (i = input, o = output; *i != 0;)
            *(o++) = (wchar_t)(*(i++));
        *o = 0;
        return;
    }

    wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));

    for (i = input, o = output; *i != 0;)
        *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]);
    *o = 0;
}


void wxEncodingConverter::Convert(const wchar_t* input, char* output)
{
    wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
    wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));

    const wchar_t *i;
    char *o;

    if (m_JustCopy)
    {
        for (i = input, o = output; *i != 0;)
            *(o++) = (char)(*(i++));
        *o = 0;
        return;
    }

    wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));

    for (i = input, o = output; *i != 0;)
        *(o++) = (char)(m_Table[(wxUint16)*(i++)]);
    *o = 0;
}


void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output)
{
    wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
    wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));

    const wchar_t *i;
    wchar_t *o;

    if (m_JustCopy)
    {
        // wcscpy() is not guaranteed to exist
        for (i = input, o = output; *i != 0;)
            *(o++) = (*(i++));
        *o = 0;
        return;
    }

    wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));

    for (i = input, o = output; *i != 0;)
        *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]);
    *o = 0;
}

#endif // wxUSE_WCHAR_T


wxString wxEncodingConverter::Convert(const wxString& input)
{
    if (m_JustCopy) return input;

    wxString s;
    const wxChar *i;

    wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));

    if (m_UnicodeInput)
        for (i = input.c_str(); *i != 0; i++)
            s << (wxChar)(m_Table[(wxUint16)*i]);
    else
        for (i = input.c_str(); *i != 0; i++)
            s << (wxChar)(m_Table[(wxUint8)*i]);
    return s;
}


// Following tables describe classes of encoding equivalence.
//

#define STOP wxFONTENCODING_SYSTEM

#define NUM_OF_PLATFORMS  4 /*must conform to enum wxPLATFORM_XXXX !!!*/
#define ENC_PER_PLATFORM  5
           // max no. of encodings for one language used on one platform
           // Anybody thinks 5 is not enough? ;-)

static wxFontEncoding
    EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = {

    // *** Please put more common encodings as first! ***

    // Western European
    {
        /* unix    */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP},
        /* windows */ {wxFONTENCODING_CP1252, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    // Central European
    {
        /* unix    */ {wxFONTENCODING_ISO8859_2, STOP},
        /* windows */ {wxFONTENCODING_CP1250, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    // Baltic
    {
        /* unix    */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP},
        /* windows */ {wxFONTENCODING_CP1257, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    // Hebrew
    {
        /* unix    */ {wxFONTENCODING_ISO8859_8, STOP},
        /* windows */ {wxFONTENCODING_CP1255, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    // Greek
    {
        /* unix    */ {wxFONTENCODING_ISO8859_7, STOP},
        /* windows */ {wxFONTENCODING_CP1253, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    // Arabic
    {
        /* unix    */ {wxFONTENCODING_ISO8859_6, STOP},
        /* windows */ {wxFONTENCODING_CP1256, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    // Turkish
    {
        /* unix    */ {wxFONTENCODING_ISO8859_9, STOP},
        /* windows */ {wxFONTENCODING_CP1254, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    // Cyrillic
    {
        /* unix    */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP},
        /* windows */ {wxFONTENCODING_CP1251, STOP},
        /* os2     */ {STOP},
        /* mac     */ {STOP}
    },

    {{STOP},{STOP},{STOP},{STOP}} /* Terminator */
    /* no, _not_ Arnold! */
};


wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform)
{
    if (platform == wxPLATFORM_CURRENT)
    {
#if defined(__WXMSW__)
        platform = wxPLATFORM_WINDOWS;
#elif defined(__WXGTK__) || defined(__WXMOTIF__)
        platform = wxPLATFORM_UNIX;
#elif defined(__WXOS2__)
        platform = wxPLATFORM_OS2;
#elif defined(__WXMAC__)
        platform = wxPLATFORM_MAC;
#endif
    }

    int i, clas, e ;
    wxFontEncoding *f;
    wxFontEncodingArray arr;

    clas = 0;
    while (EquivalentEncodings[clas][0][0] != STOP)
    {
        for (i = 0; i < NUM_OF_PLATFORMS; i++)
            for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
                if (EquivalentEncodings[clas][i][e] == enc)
                {
                    for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
                        if (*f == enc) arr.Add(enc);
                    for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
                        if (arr.Index(*f) == wxNOT_FOUND) arr.Add(*f);
                    i = NUM_OF_PLATFORMS/*hack*/; break;
                }
        clas++;
    }

    return arr;
}


wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
{
    int i, clas, e, j ;
    wxFontEncoding *f;
    wxFontEncodingArray arr;

    arr = GetPlatformEquivalents(enc); // we want them to be first items in array

    clas = 0;
    while (EquivalentEncodings[clas][0][0] != STOP)
    {
        for (i = 0; i < NUM_OF_PLATFORMS; i++)
            for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
                if (EquivalentEncodings[clas][i][e] == enc)
                {
                    for (j = 0; j < NUM_OF_PLATFORMS; j++)
                        for (f = EquivalentEncodings[clas][j]; *f != STOP; f++)
                            if (arr.Index(*f) == wxNOT_FOUND) arr.Add(*f);
                    i = NUM_OF_PLATFORMS/*hack*/; break;
                }
        clas++;
    }

    return arr;
}
Commit	Line	Data
c958260b VS	1	/////////////////////////////////////////////////////////////////////////////
	2	// Name: encconv.cpp
	3	// Purpose: wxEncodingConverter class for converting between different
	4	// font encodings
	5	// Author: Vaclav Slavik
	6	// Copyright: (c) 1999 Vaclav Slavik
	7	// Licence: wxWindows Licence
	8	/////////////////////////////////////////////////////////////////////////////
	9
	10	#ifdef __GNUG__
	11	#pragma implementation "encconv.h"
	12	#endif
	13
	14	// For compilers that support precompilation, includes "wx.h".
	15	#include "wx/wxprec.h"
	16
	17	#ifdef __BORLANDC__
	18	#pragma hdrstop
	19	#endif
	20
	21	#include "wx/encconv.h"
	22
	23	#include <stdlib.h>
	24
94fc5183 VS	25	// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl:
	26	#ifdef __BORLANDC__
	27	#include "../common/unictabl.inc"
	28	#else
03424b1b	29	#include "unictabl.inc"
94fc5183	30	#endif
c958260b	31
f6bcfd97 BP	32	#if wxUSE_WCHAR_T
	33	typedef wchar_t tchar;
	34	#else
	35	typedef char tchar;
	36	#endif
c958260b	37
eda22ec3	38	static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc)
c958260b VS	39	{
	40	for (int i = 0; encodings_list[i].table != NULL; i++)
	41	{
03424b1b	42	if (encodings_list[i].encoding == enc)
c958260b VS	43	return encodings_list[i].table;
	44	}
	45	return NULL;
	46	}
	47
	48	typedef struct {
	49	wxUint16 u;
	50	wxUint8 c;
	51	} CharsetItem;
	52
	53
	54
eda22ec3	55	static int LINKAGEMODE CompareCharsetItems(const void i1, const void i2)
c958260b VS	56	{
	57	return ( ((CharsetItem)i1) -> u - ((CharsetItem)i2) -> u );
	58	}
	59
	60
eda22ec3	61	static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl)
c958260b VS	62	{
c958260b VS	63	CharsetItem *rev = new CharsetItem[128];
03424b1b	64
c958260b VS	65	for (int i = 0; i < 128; i++)
	66	rev[i].c = 128 + i, rev[i].u = tbl[i];
	67
	68	qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems);
03424b1b	69
c958260b VS	70	return rev;
	71	}
	72
	73
	74
	75	wxEncodingConverter::wxEncodingConverter()
	76	{
	77	m_Table = NULL;
47e55c2f	78	m_UnicodeInput = m_UnicodeOutput = FALSE;
c958260b VS	79	m_JustCopy = FALSE;
	80	}
	81
	82
	83
	84	bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
	85	{
	86	unsigned i;
	87	wxUint16 in_tbl = NULL, out_tbl = NULL;
	88
	89	if (m_Table) {delete[] m_Table; m_Table = NULL;}
	90
f6bcfd97	91	#if !wxUSE_WCHAR_T
c958260b VS	92	if (input_enc == wxFONTENCODING_UNICODE \|\| output_enc == wxFONTENCODING_UNICODE) return FALSE;
	93	#endif
	94
	95	if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;}
03424b1b	96
47e55c2f	97	m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE);
c958260b	98	m_JustCopy = FALSE;
03424b1b	99
c958260b VS	100	if (input_enc == wxFONTENCODING_UNICODE)
	101	{
	102	if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
	103
f6bcfd97 BP	104	m_Table = new tchar[65536];
f6bcfd97 BP	105	for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
03424b1b	106	for (i = 128; i < 65536; i++) m_Table[i] = (tchar)'?';
c958260b VS	107	// FIXME - this should be character that means `unicode to charset' impossible, not '?'
	108
	109	if (method == wxCONVERT_SUBSTITUTE)
	110	{
	111	for (i = 0; i < encoding_unicode_fallback_count; i++)
f6bcfd97	112	m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s;
c958260b VS	113	}
	114
	115	for (i = 0; i < 128; i++)
f6bcfd97	116	m_Table[out_tbl[i]] = (tchar)(128 + i);
c958260b VS	117
	118	m_UnicodeInput = TRUE;
	119	return TRUE;
	120	}
03424b1b	121
c958260b VS	122	else
	123	{
	124	if ((in_tbl = GetEncTable(input_enc)) == NULL) return FALSE;
	125	if (output_enc != wxFONTENCODING_UNICODE)
	126	if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
	127
	128	m_UnicodeInput = FALSE;
03424b1b	129
f6bcfd97 BP	130	m_Table = new tchar[256];
f6bcfd97 BP	131	for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
03424b1b	132
c958260b VS	133	if (output_enc == wxFONTENCODING_UNICODE)
c958260b VS	134	{
f6bcfd97	135	for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i];
c958260b VS	136	return TRUE;
c958260b VS	137	}
03424b1b	138	else
c958260b VS	139	{
c958260b VS	140	CharsetItem *rev = BuildReverseTable(out_tbl);
33ac7e6f KB	141	CharsetItem *item;
33ac7e6f KB	142	CharsetItem key;
03424b1b VZ	143
03424b1b VZ	144	for (i = 0; i < 128; i++)
c958260b VS	145	{
	146	key.u = in_tbl[i];
	147	item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems);
	148	if (item == NULL && method == wxCONVERT_SUBSTITUTE)
03424b1b	149	item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback,
c958260b VS	150	encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems);
c958260b VS	151	if (item)
f6bcfd97	152	m_Table[128 + i] = (tchar)item -> c;
c958260b	153	else
33ac7e6f KB	154	#if wxUSE_WCHAR_T
	155	m_Table[128 + i] = (wchar_t)(128 + i);
	156	#else
	157	m_Table[128 + i] = (char)(128 + i);
	158	#endif
c958260b	159	}
03424b1b	160
c958260b VS	161	delete[] rev;
	162	return TRUE;
	163	}
	164	}
	165	}
	166
	167
47e55c2f	168
f6bcfd97	169	void wxEncodingConverter::Convert(const char* input, char* output)
c958260b	170	{
f6bcfd97 BP	171	wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
	172	wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
	173
	174	const char *i;
	175	char *o;
	176
c958260b VS	177	if (m_JustCopy)
c958260b VS	178	{
f6bcfd97	179	strcpy(output, input);
c958260b VS	180	return;
c958260b VS	181	}
03424b1b	182
c958260b	183	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	184
f6bcfd97 BP	185	for (i = input, o = output; *i != 0;)
f6bcfd97 BP	186	(o++) = (char)(m_Table[(wxUint8)(i++)]);
c958260b VS	187	*o = 0;
	188	}
	189
	190
f6bcfd97	191	#if wxUSE_WCHAR_T
47e55c2f	192
f6bcfd97	193	void wxEncodingConverter::Convert(const char* input, wchar_t* output)
47e55c2f	194	{
f6bcfd97	195	wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
47e55c2f VS	196	wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
	197
	198	const char *i;
f6bcfd97	199	wchar_t *o;
47e55c2f VS	200
	201	if (m_JustCopy)
	202	{
	203	for (i = input, o = output; *i != 0;)
f6bcfd97	204	(o++) = (wchar_t)((i++));
47e55c2f VS	205	*o = 0;
	206	return;
	207	}
03424b1b	208
47e55c2f	209	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	210
47e55c2f	211	for (i = input, o = output; *i != 0;)
f6bcfd97	212	(o++) = (wchar_t)(m_Table[(wxUint8)(i++)]);
47e55c2f VS	213	*o = 0;
	214	}
	215
	216
	217
f6bcfd97	218	void wxEncodingConverter::Convert(const wchar_t* input, char* output)
47e55c2f VS	219	{
47e55c2f VS	220	wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
f6bcfd97	221	wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
47e55c2f	222
f6bcfd97	223	const wchar_t *i;
47e55c2f VS	224	char *o;
	225
	226	if (m_JustCopy)
	227	{
	228	for (i = input, o = output; *i != 0;)
	229	(o++) = (char)((i++));
	230	*o = 0;
	231	return;
	232	}
03424b1b	233
47e55c2f	234	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	235
f6bcfd97 BP	236	for (i = input, o = output; *i != 0;)
f6bcfd97 BP	237	(o++) = (char)(m_Table[(wxUint16)(i++)]);
47e55c2f VS	238	*o = 0;
	239	}
	240
	241
	242
f6bcfd97	243	void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output)
47e55c2f	244	{
f6bcfd97 BP	245	wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
f6bcfd97 BP	246	wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
47e55c2f	247
f6bcfd97 BP	248	const wchar_t *i;
f6bcfd97 BP	249	wchar_t *o;
47e55c2f VS	250
	251	if (m_JustCopy)
	252	{
f6bcfd97 BP	253	// wcscpy() is not guaranteed to exist
	254	for (i = input, o = output; *i != 0;)
	255	(o++) = ((i++));
	256	*o = 0;
47e55c2f VS	257	return;
47e55c2f VS	258	}
03424b1b	259
47e55c2f	260	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	261
47e55c2f	262	for (i = input, o = output; *i != 0;)
f6bcfd97	263	(o++) = (wchar_t)(m_Table[(wxUint8)(i++)]);
47e55c2f VS	264	*o = 0;
	265	}
	266
f6bcfd97	267	#endif // wxUSE_WCHAR_T
47e55c2f	268
c958260b VS	269
	270	wxString wxEncodingConverter::Convert(const wxString& input)
	271	{
	272	if (m_JustCopy) return input;
03424b1b	273
c958260b VS	274	wxString s;
c958260b VS	275	const wxChar *i;
03424b1b	276
c958260b	277	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	278
c958260b VS	279	if (m_UnicodeInput)
	280	for (i = input.c_str(); *i != 0; i++)
	281	s << (wxChar)(m_Table[(wxUint16)*i]);
	282	else
	283	for (i = input.c_str(); *i != 0; i++)
	284	s << (wxChar)(m_Table[(wxUint8)*i]);
	285	return s;
	286	}
	287
	288
	289
	290
47e55c2f VS	291
	292
	293
c958260b	294	// Following tables describe classes of encoding equivalence.
03424b1b	295	//
c958260b VS	296
	297	#define STOP wxFONTENCODING_SYSTEM
	298
	299	#define NUM_OF_PLATFORMS 4 /must conform to enum wxPLATFORM_XXXX !!!/
617eb021	300	#define ENC_PER_PLATFORM 5
c958260b	301	// max no. of encodings for one language used on one platform
617eb021	302	// Anybody thinks 5 is not enough? ;-)
c958260b	303
03424b1b	304	static wxFontEncoding
c958260b VS	305	EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = {
c958260b VS	306
47e55c2f VS	307	// * Please put more common encodings as first! *
47e55c2f VS	308
03424b1b	309	// Western European
c958260b VS	310	{
	311	/* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP},
	312	/* windows */ {wxFONTENCODING_CP1252, STOP},
	313	/* os2 */ {STOP},
03424b1b	314	/* mac */ {STOP}
c958260b VS	315	},
c958260b VS	316
47e55c2f	317	// Central European
c958260b VS	318	{
	319	/* unix */ {wxFONTENCODING_ISO8859_2, STOP},
	320	/* windows */ {wxFONTENCODING_CP1250, STOP},
	321	/* os2 */ {STOP},
03424b1b	322	/* mac */ {STOP}
c958260b	323	},
03424b1b	324
47e55c2f VS	325	// Baltic
47e55c2f VS	326	{
03424b1b	327	/* unix */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP},
47e55c2f VS	328	/* windows */ {wxFONTENCODING_CP1257, STOP},
47e55c2f VS	329	/* os2 */ {STOP},
03424b1b	330	/* mac */ {STOP}
47e55c2f VS	331	},
	332
	333	// Hebrew
	334	{
	335	/* unix */ {wxFONTENCODING_ISO8859_8, STOP},
	336	/* windows */ {wxFONTENCODING_CP1255, STOP},
	337	/* os2 */ {STOP},
03424b1b	338	/* mac */ {STOP}
47e55c2f VS	339	},
	340
	341	// Greek
	342	{
	343	/* unix */ {wxFONTENCODING_ISO8859_7, STOP},
	344	/* windows */ {wxFONTENCODING_CP1253, STOP},
	345	/* os2 */ {STOP},
03424b1b	346	/* mac */ {STOP}
47e55c2f VS	347	},
	348
	349	// Arabic
	350	{
	351	/* unix */ {wxFONTENCODING_ISO8859_6, STOP},
	352	/* windows */ {wxFONTENCODING_CP1256, STOP},
	353	/* os2 */ {STOP},
03424b1b	354	/* mac */ {STOP}
47e55c2f VS	355	},
	356
	357	// Turkish
	358	{
	359	/* unix */ {wxFONTENCODING_ISO8859_9, STOP},
	360	/* windows */ {wxFONTENCODING_CP1254, STOP},
	361	/* os2 */ {STOP},
03424b1b	362	/* mac */ {STOP}
47e55c2f VS	363	},
	364
	365	// Cyrillic
3b61656e VS	366	{
3b61656e VS	367	/* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP},
47e55c2f VS	368	/* windows */ {wxFONTENCODING_CP1251, STOP},
47e55c2f VS	369	/* os2 */ {STOP},
03424b1b	370	/* mac */ {STOP}
47e55c2f	371	},
c958260b VS	372
	373	{{STOP},{STOP},{STOP},{STOP}} /* Terminator */
	374	/* no, _not_ Arnold! */
	375	};
	376
	377
	378
	379
	380	wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform)
	381	{
	382	if (platform == wxPLATFORM_CURRENT)
	383	{
	384	#if defined(__WXMSW__)
	385	platform = wxPLATFORM_WINDOWS;
	386	#elif defined(__WXGTK__) \|\| defined(__WXMOTIF__)
	387	platform = wxPLATFORM_UNIX;
	388	#elif defined(__WXOS2__)
	389	platform = wxPLATFORM_OS2;
	390	#elif defined(__WXMAC__)
	391	platform = wxPLATFORM_MAC;
	392	#endif
	393	}
03424b1b	394
c958260b VS	395	int i, clas, e ;
	396	wxFontEncoding *f;
	397	wxFontEncodingArray arr;
	398
	399	clas = 0;
	400	while (EquivalentEncodings[clas][0][0] != STOP)
	401	{
	402	for (i = 0; i < NUM_OF_PLATFORMS; i++)
	403	for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
	404	if (EquivalentEncodings[clas][i][e] == enc)
	405	{
47e55c2f VS	406	for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
	407	if (*f == enc) arr.Add(enc);
	408	for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
c958260b	409	if (arr.Index(f) == wxNOT_FOUND) arr.Add(f);
03424b1b	410	i = NUM_OF_PLATFORMS/hack/; break;
c958260b VS	411	}
	412	clas++;
	413	}
03424b1b	414
c958260b VS	415	return arr;
	416	}
	417
	418
	419
	420	wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
	421	{
	422	int i, clas, e, j ;
	423	wxFontEncoding *f;
	424	wxFontEncodingArray arr;
03424b1b	425
47e55c2f	426	arr = GetPlatformEquivalents(enc); // we want them to be first items in array
c958260b VS	427
	428	clas = 0;
	429	while (EquivalentEncodings[clas][0][0] != STOP)
	430	{
	431	for (i = 0; i < NUM_OF_PLATFORMS; i++)
	432	for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
	433	if (EquivalentEncodings[clas][i][e] == enc)
	434	{
	435	for (j = 0; j < NUM_OF_PLATFORMS; j++)
03424b1b VZ	436	for (f = EquivalentEncodings[clas][j]; *f != STOP; f++)
	437	if (arr.Index(f) == wxNOT_FOUND) arr.Add(f);
	438	i = NUM_OF_PLATFORMS/hack/; break;
c958260b VS	439	}
	440	clas++;
	441	}
03424b1b	442
c958260b VS	443	return arr;
c958260b VS	444	}