[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

#if wxUSE_FONTMAP

#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;
}

#endif // wxUSE_FONTMAP
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
1e6feb95 VZ	21	#if wxUSE_FONTMAP
1e6feb95 VZ	22
c958260b VS	23	#include "wx/encconv.h"
	24
	25	#include <stdlib.h>
	26
94fc5183 VS	27	// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl:
	28	#ifdef __BORLANDC__
	29	#include "../common/unictabl.inc"
	30	#else
03424b1b	31	#include "unictabl.inc"
94fc5183	32	#endif
c958260b	33
f6bcfd97 BP	34	#if wxUSE_WCHAR_T
	35	typedef wchar_t tchar;
	36	#else
	37	typedef char tchar;
	38	#endif
c958260b	39
eda22ec3	40	static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc)
c958260b VS	41	{
	42	for (int i = 0; encodings_list[i].table != NULL; i++)
	43	{
03424b1b	44	if (encodings_list[i].encoding == enc)
c958260b VS	45	return encodings_list[i].table;
	46	}
	47	return NULL;
	48	}
	49
	50	typedef struct {
	51	wxUint16 u;
	52	wxUint8 c;
	53	} CharsetItem;
	54
	55
	56
eda22ec3	57	static int LINKAGEMODE CompareCharsetItems(const void i1, const void i2)
c958260b VS	58	{
	59	return ( ((CharsetItem)i1) -> u - ((CharsetItem)i2) -> u );
	60	}
	61
	62
eda22ec3	63	static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl)
c958260b VS	64	{
c958260b VS	65	CharsetItem *rev = new CharsetItem[128];
03424b1b	66
c958260b VS	67	for (int i = 0; i < 128; i++)
	68	rev[i].c = 128 + i, rev[i].u = tbl[i];
	69
	70	qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems);
03424b1b	71
c958260b VS	72	return rev;
	73	}
	74
	75
	76
	77	wxEncodingConverter::wxEncodingConverter()
	78	{
	79	m_Table = NULL;
47e55c2f	80	m_UnicodeInput = m_UnicodeOutput = FALSE;
c958260b VS	81	m_JustCopy = FALSE;
	82	}
	83
	84
	85
	86	bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
	87	{
	88	unsigned i;
	89	wxUint16 in_tbl = NULL, out_tbl = NULL;
	90
	91	if (m_Table) {delete[] m_Table; m_Table = NULL;}
	92
f6bcfd97	93	#if !wxUSE_WCHAR_T
c958260b VS	94	if (input_enc == wxFONTENCODING_UNICODE \|\| output_enc == wxFONTENCODING_UNICODE) return FALSE;
	95	#endif
	96
	97	if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;}
03424b1b	98
47e55c2f	99	m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE);
c958260b	100	m_JustCopy = FALSE;
03424b1b	101
c958260b VS	102	if (input_enc == wxFONTENCODING_UNICODE)
	103	{
	104	if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
	105
f6bcfd97 BP	106	m_Table = new tchar[65536];
f6bcfd97 BP	107	for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
03424b1b	108	for (i = 128; i < 65536; i++) m_Table[i] = (tchar)'?';
c958260b VS	109	// FIXME - this should be character that means `unicode to charset' impossible, not '?'
	110
	111	if (method == wxCONVERT_SUBSTITUTE)
	112	{
	113	for (i = 0; i < encoding_unicode_fallback_count; i++)
f6bcfd97	114	m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s;
c958260b VS	115	}
	116
	117	for (i = 0; i < 128; i++)
f6bcfd97	118	m_Table[out_tbl[i]] = (tchar)(128 + i);
c958260b VS	119
	120	m_UnicodeInput = TRUE;
	121	return TRUE;
	122	}
03424b1b	123
c958260b VS	124	else
	125	{
	126	if ((in_tbl = GetEncTable(input_enc)) == NULL) return FALSE;
	127	if (output_enc != wxFONTENCODING_UNICODE)
	128	if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
	129
	130	m_UnicodeInput = FALSE;
03424b1b	131
f6bcfd97 BP	132	m_Table = new tchar[256];
f6bcfd97 BP	133	for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
03424b1b	134
c958260b VS	135	if (output_enc == wxFONTENCODING_UNICODE)
c958260b VS	136	{
f6bcfd97	137	for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i];
c958260b VS	138	return TRUE;
c958260b VS	139	}
03424b1b	140	else
c958260b VS	141	{
c958260b VS	142	CharsetItem *rev = BuildReverseTable(out_tbl);
33ac7e6f KB	143	CharsetItem *item;
33ac7e6f KB	144	CharsetItem key;
03424b1b VZ	145
03424b1b VZ	146	for (i = 0; i < 128; i++)
c958260b VS	147	{
	148	key.u = in_tbl[i];
	149	item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems);
	150	if (item == NULL && method == wxCONVERT_SUBSTITUTE)
03424b1b	151	item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback,
c958260b VS	152	encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems);
c958260b VS	153	if (item)
f6bcfd97	154	m_Table[128 + i] = (tchar)item -> c;
c958260b	155	else
33ac7e6f KB	156	#if wxUSE_WCHAR_T
	157	m_Table[128 + i] = (wchar_t)(128 + i);
	158	#else
	159	m_Table[128 + i] = (char)(128 + i);
	160	#endif
c958260b	161	}
03424b1b	162
c958260b VS	163	delete[] rev;
	164	return TRUE;
	165	}
	166	}
	167	}
	168
	169
47e55c2f	170
f6bcfd97	171	void wxEncodingConverter::Convert(const char* input, char* output)
c958260b	172	{
f6bcfd97 BP	173	wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
	174	wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
	175
	176	const char *i;
	177	char *o;
	178
c958260b VS	179	if (m_JustCopy)
c958260b VS	180	{
f6bcfd97	181	strcpy(output, input);
c958260b VS	182	return;
c958260b VS	183	}
03424b1b	184
c958260b	185	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	186
f6bcfd97 BP	187	for (i = input, o = output; *i != 0;)
f6bcfd97 BP	188	(o++) = (char)(m_Table[(wxUint8)(i++)]);
c958260b VS	189	*o = 0;
	190	}
	191
	192
f6bcfd97	193	#if wxUSE_WCHAR_T
47e55c2f	194
f6bcfd97	195	void wxEncodingConverter::Convert(const char* input, wchar_t* output)
47e55c2f	196	{
f6bcfd97	197	wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
47e55c2f VS	198	wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
	199
	200	const char *i;
f6bcfd97	201	wchar_t *o;
47e55c2f VS	202
	203	if (m_JustCopy)
	204	{
	205	for (i = input, o = output; *i != 0;)
f6bcfd97	206	(o++) = (wchar_t)((i++));
47e55c2f VS	207	*o = 0;
	208	return;
	209	}
03424b1b	210
47e55c2f	211	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	212
47e55c2f	213	for (i = input, o = output; *i != 0;)
f6bcfd97	214	(o++) = (wchar_t)(m_Table[(wxUint8)(i++)]);
47e55c2f VS	215	*o = 0;
	216	}
	217
	218
	219
f6bcfd97	220	void wxEncodingConverter::Convert(const wchar_t* input, char* output)
47e55c2f VS	221	{
47e55c2f VS	222	wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
f6bcfd97	223	wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
47e55c2f	224
f6bcfd97	225	const wchar_t *i;
47e55c2f VS	226	char *o;
	227
	228	if (m_JustCopy)
	229	{
	230	for (i = input, o = output; *i != 0;)
	231	(o++) = (char)((i++));
	232	*o = 0;
	233	return;
	234	}
03424b1b	235
47e55c2f	236	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	237
f6bcfd97 BP	238	for (i = input, o = output; *i != 0;)
f6bcfd97 BP	239	(o++) = (char)(m_Table[(wxUint16)(i++)]);
47e55c2f VS	240	*o = 0;
	241	}
	242
	243
	244
f6bcfd97	245	void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output)
47e55c2f	246	{
f6bcfd97 BP	247	wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
f6bcfd97 BP	248	wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
47e55c2f	249
f6bcfd97 BP	250	const wchar_t *i;
f6bcfd97 BP	251	wchar_t *o;
47e55c2f VS	252
	253	if (m_JustCopy)
	254	{
f6bcfd97 BP	255	// wcscpy() is not guaranteed to exist
	256	for (i = input, o = output; *i != 0;)
	257	(o++) = ((i++));
	258	*o = 0;
47e55c2f VS	259	return;
47e55c2f VS	260	}
03424b1b	261
47e55c2f	262	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	263
47e55c2f	264	for (i = input, o = output; *i != 0;)
f6bcfd97	265	(o++) = (wchar_t)(m_Table[(wxUint8)(i++)]);
47e55c2f VS	266	*o = 0;
	267	}
	268
f6bcfd97	269	#endif // wxUSE_WCHAR_T
47e55c2f	270
c958260b VS	271
	272	wxString wxEncodingConverter::Convert(const wxString& input)
	273	{
	274	if (m_JustCopy) return input;
03424b1b	275
c958260b VS	276	wxString s;
c958260b VS	277	const wxChar *i;
03424b1b	278
c958260b	279	wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
03424b1b	280
c958260b VS	281	if (m_UnicodeInput)
	282	for (i = input.c_str(); *i != 0; i++)
	283	s << (wxChar)(m_Table[(wxUint16)*i]);
	284	else
	285	for (i = input.c_str(); *i != 0; i++)
	286	s << (wxChar)(m_Table[(wxUint8)*i]);
	287	return s;
	288	}
	289
	290
	291
	292
47e55c2f VS	293
	294
	295
c958260b	296	// Following tables describe classes of encoding equivalence.
03424b1b	297	//
c958260b VS	298
	299	#define STOP wxFONTENCODING_SYSTEM
	300
	301	#define NUM_OF_PLATFORMS 4 /must conform to enum wxPLATFORM_XXXX !!!/
617eb021	302	#define ENC_PER_PLATFORM 5
c958260b	303	// max no. of encodings for one language used on one platform
617eb021	304	// Anybody thinks 5 is not enough? ;-)
c958260b	305
03424b1b	306	static wxFontEncoding
c958260b VS	307	EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = {
c958260b VS	308
47e55c2f VS	309	// * Please put more common encodings as first! *
47e55c2f VS	310
03424b1b	311	// Western European
c958260b VS	312	{
	313	/* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP},
	314	/* windows */ {wxFONTENCODING_CP1252, STOP},
	315	/* os2 */ {STOP},
03424b1b	316	/* mac */ {STOP}
c958260b VS	317	},
c958260b VS	318
47e55c2f	319	// Central European
c958260b VS	320	{
	321	/* unix */ {wxFONTENCODING_ISO8859_2, STOP},
	322	/* windows */ {wxFONTENCODING_CP1250, STOP},
	323	/* os2 */ {STOP},
03424b1b	324	/* mac */ {STOP}
c958260b	325	},
03424b1b	326
47e55c2f VS	327	// Baltic
47e55c2f VS	328	{
03424b1b	329	/* unix */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP},
47e55c2f VS	330	/* windows */ {wxFONTENCODING_CP1257, STOP},
47e55c2f VS	331	/* os2 */ {STOP},
03424b1b	332	/* mac */ {STOP}
47e55c2f VS	333	},
	334
	335	// Hebrew
	336	{
	337	/* unix */ {wxFONTENCODING_ISO8859_8, STOP},
	338	/* windows */ {wxFONTENCODING_CP1255, STOP},
	339	/* os2 */ {STOP},
03424b1b	340	/* mac */ {STOP}
47e55c2f VS	341	},
	342
	343	// Greek
	344	{
	345	/* unix */ {wxFONTENCODING_ISO8859_7, STOP},
	346	/* windows */ {wxFONTENCODING_CP1253, STOP},
	347	/* os2 */ {STOP},
03424b1b	348	/* mac */ {STOP}
47e55c2f VS	349	},
	350
	351	// Arabic
	352	{
	353	/* unix */ {wxFONTENCODING_ISO8859_6, STOP},
	354	/* windows */ {wxFONTENCODING_CP1256, STOP},
	355	/* os2 */ {STOP},
03424b1b	356	/* mac */ {STOP}
47e55c2f VS	357	},
	358
	359	// Turkish
	360	{
	361	/* unix */ {wxFONTENCODING_ISO8859_9, STOP},
	362	/* windows */ {wxFONTENCODING_CP1254, STOP},
	363	/* os2 */ {STOP},
03424b1b	364	/* mac */ {STOP}
47e55c2f VS	365	},
	366
	367	// Cyrillic
3b61656e VS	368	{
3b61656e VS	369	/* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP},
47e55c2f VS	370	/* windows */ {wxFONTENCODING_CP1251, STOP},
47e55c2f VS	371	/* os2 */ {STOP},
03424b1b	372	/* mac */ {STOP}
47e55c2f	373	},
c958260b VS	374
	375	{{STOP},{STOP},{STOP},{STOP}} /* Terminator */
	376	/* no, _not_ Arnold! */
	377	};
	378
	379
	380
	381
	382	wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform)
	383	{
	384	if (platform == wxPLATFORM_CURRENT)
	385	{
	386	#if defined(__WXMSW__)
	387	platform = wxPLATFORM_WINDOWS;
	388	#elif defined(__WXGTK__) \|\| defined(__WXMOTIF__)
	389	platform = wxPLATFORM_UNIX;
	390	#elif defined(__WXOS2__)
	391	platform = wxPLATFORM_OS2;
	392	#elif defined(__WXMAC__)
	393	platform = wxPLATFORM_MAC;
	394	#endif
	395	}
03424b1b	396
c958260b VS	397	int i, clas, e ;
	398	wxFontEncoding *f;
	399	wxFontEncodingArray arr;
	400
	401	clas = 0;
	402	while (EquivalentEncodings[clas][0][0] != STOP)
	403	{
	404	for (i = 0; i < NUM_OF_PLATFORMS; i++)
	405	for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
	406	if (EquivalentEncodings[clas][i][e] == enc)
	407	{
47e55c2f VS	408	for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
	409	if (*f == enc) arr.Add(enc);
	410	for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
c958260b	411	if (arr.Index(f) == wxNOT_FOUND) arr.Add(f);
03424b1b	412	i = NUM_OF_PLATFORMS/hack/; break;
c958260b VS	413	}
	414	clas++;
	415	}
03424b1b	416
c958260b VS	417	return arr;
	418	}
	419
	420
	421
	422	wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
	423	{
	424	int i, clas, e, j ;
	425	wxFontEncoding *f;
	426	wxFontEncodingArray arr;
03424b1b	427
47e55c2f	428	arr = GetPlatformEquivalents(enc); // we want them to be first items in array
c958260b VS	429
	430	clas = 0;
	431	while (EquivalentEncodings[clas][0][0] != STOP)
	432	{
	433	for (i = 0; i < NUM_OF_PLATFORMS; i++)
	434	for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
	435	if (EquivalentEncodings[clas][i][e] == enc)
	436	{
	437	for (j = 0; j < NUM_OF_PLATFORMS; j++)
03424b1b VZ	438	for (f = EquivalentEncodings[clas][j]; *f != STOP; f++)
	439	if (arr.Index(f) == wxNOT_FOUND) arr.Add(f);
	440	i = NUM_OF_PLATFORMS/hack/; break;
c958260b VS	441	}
	442	clas++;
	443	}
03424b1b	444
c958260b VS	445	return arr;
c958260b VS	446	}
1e6feb95 VZ	447
1e6feb95 VZ	448	#endif // wxUSE_FONTMAP