1
/////////////////////////////////////////////////////////////////////////////
2
// Name: src/common/encconv.cpp
3
// Purpose: wxEncodingConverter class for converting between different
5
// Author: Vaclav Slavik
6
// Copyright: (c) 1999 Vaclav Slavik
7
// Licence: wxWindows licence
8
/////////////////////////////////////////////////////////////////////////////
10
// For compilers that support precompilation, includes "wx.h".
11
#include "wx/wxprec.h"
17
#include "wx/encconv.h"
21
// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl:
22
#if defined( __BORLANDC__ ) || defined(__DARWIN__)
23
#include "../common/unictabl.inc"
25
#include "unictabl.inc"
29
#include "wx/osx/core/cfstring.h"
30
#include <CoreFoundation/CFStringEncodingExt.h>
32
wxUint16 gMacEncodings[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1][128] ;
33
bool gMacEncodingsInited[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1] ;
37
#include "wx/msw/wince/missing.h" // for bsearch()
40
static const wxUint16* GetEncTable(wxFontEncoding enc)
43
if( enc >= wxFONTENCODING_MACMIN && enc <= wxFONTENCODING_MACMAX )
45
int i = enc-wxFONTENCODING_MACMIN ;
46
if ( gMacEncodingsInited[i] == false )
49
CFStringEncoding cfencoding = wxMacGetSystemEncFromFontEnc( enc ) ;
50
if( !CFStringIsEncodingAvailable( cfencoding ) )
53
memset( gMacEncodings[i] , 0 , 128 * 2 );
54
char s[2] = { 0 , 0 };
55
CFRange firstchar = CFRangeMake( 0, 1 );
56
for( unsigned char c = 255 ; c >= 128 ; --c )
59
wxCFStringRef cfref( CFStringCreateWithCStringNoCopy( NULL, s, cfencoding , kCFAllocatorNull ) );
60
CFStringGetCharacters( cfref, firstchar, (UniChar*) &gMacEncodings[i][c-128] );
62
gMacEncodingsInited[i]=true;
64
return gMacEncodings[i] ;
68
for (int i = 0; encodings_list[i].table != NULL; i++)
70
if (encodings_list[i].encoding == enc)
71
return encodings_list[i].table;
83
static int wxCMPFUNC_CONV
84
CompareCharsetItems(const void *i1, const void *i2)
86
return ( ((CharsetItem*)i1) -> u - ((CharsetItem*)i2) -> u );
90
static CharsetItem* BuildReverseTable(const wxUint16 *tbl)
92
CharsetItem *rev = new CharsetItem[128];
94
for (int i = 0; i < 128; i++)
95
rev[i].c = wxUint8(128 + i), rev[i].u = tbl[i];
97
qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems);
104
wxEncodingConverter::wxEncodingConverter()
107
m_UnicodeInput = m_UnicodeOutput = false;
113
bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
116
const wxUint16 *in_tbl;
117
const wxUint16 *out_tbl = NULL;
121
if (input_enc == output_enc) {m_JustCopy = true; return true;}
123
m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE);
126
if (input_enc == wxFONTENCODING_UNICODE)
128
if ((out_tbl = GetEncTable(output_enc)) == NULL) return false;
130
m_Table = new wchar_t[65536];
131
for (i = 0; i < 128; i++) m_Table[i] = (wchar_t)i; // 7bit ASCII
132
for (i = 128; i < 65536; i++) m_Table[i] = (wchar_t)0;
134
if (method == wxCONVERT_SUBSTITUTE)
136
for (i = 0; i < encoding_unicode_fallback_count; i++)
137
m_Table[encoding_unicode_fallback[i].c] = (wchar_t) encoding_unicode_fallback[i].s;
140
for (i = 0; i < 128; i++)
141
m_Table[out_tbl[i]] = (wchar_t)(128 + i);
143
m_UnicodeInput = true;
145
else // input !Unicode
147
if ((in_tbl = GetEncTable(input_enc)) == NULL) return false;
148
if (output_enc != wxFONTENCODING_UNICODE)
149
if ((out_tbl = GetEncTable(output_enc)) == NULL) return false;
151
m_UnicodeInput = false;
153
m_Table = new wchar_t[256];
154
for (i = 0; i < 128; i++) m_Table[i] = (wchar_t)i; // 7bit ASCII
156
if (output_enc == wxFONTENCODING_UNICODE)
158
for (i = 0; i < 128; i++) m_Table[128 + i] = (wchar_t)in_tbl[i];
161
else // output !Unicode
163
CharsetItem *rev = BuildReverseTable(out_tbl);
167
for (i = 0; i < 128; i++)
170
item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems);
171
if (item == NULL && method == wxCONVERT_SUBSTITUTE)
172
item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback,
173
encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems);
175
m_Table[128 + i] = (wchar_t)item -> c;
177
m_Table[128 + i] = (wchar_t)(128 + i);
188
#define REPLACEMENT_CHAR (L'?')
190
inline wchar_t GetTableValue(const wchar_t *table, wchar_t value, bool& repl)
192
wchar_t r = table[value];
193
if (r == 0 && value != 0)
195
r = REPLACEMENT_CHAR;
202
bool wxEncodingConverter::Convert(const char* input, char* output) const
204
wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
205
wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
212
strcpy(output, input);
216
wxCHECK_MSG(m_Table != NULL, false,
217
wxT("You must call wxEncodingConverter::Init() before actually converting!"));
219
bool replaced = false;
221
for (i = input, o = output; *i != 0;)
222
*(o++) = (char)(GetTableValue(m_Table, (wxUint8)*(i++), replaced));
229
bool wxEncodingConverter::Convert(const char* input, wchar_t* output) const
231
wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
232
wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
239
for (i = input, o = output; *i != 0;)
240
*(o++) = (wchar_t)(*(i++));
245
wxCHECK_MSG(m_Table != NULL, false,
246
wxT("You must call wxEncodingConverter::Init() before actually converting!"));
248
bool replaced = false;
250
for (i = input, o = output; *i != 0;)
251
*(o++) = (wchar_t)(GetTableValue(m_Table, (wxUint8)*(i++), replaced));
259
bool wxEncodingConverter::Convert(const wchar_t* input, char* output) const
261
wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
262
wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
269
for (i = input, o = output; *i != 0;)
270
*(o++) = (char)(*(i++));
275
wxCHECK_MSG(m_Table != NULL, false,
276
wxT("You must call wxEncodingConverter::Init() before actually converting!"));
278
bool replaced = false;
280
for (i = input, o = output; *i != 0;)
281
*(o++) = (char)(GetTableValue(m_Table, (wxUint16)*(i++), replaced));
289
bool wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) const
291
wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
292
wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
299
// wcscpy() is not guaranteed to exist
300
for (i = input, o = output; *i != 0;)
306
wxCHECK_MSG(m_Table != NULL, false,
307
wxT("You must call wxEncodingConverter::Init() before actually converting!"));
309
bool replaced = false;
311
for (i = input, o = output; *i != 0;)
312
*(o++) = (wchar_t)(GetTableValue(m_Table, (wxUint8)*(i++), replaced));
319
wxString wxEncodingConverter::Convert(const wxString& input) const
321
if (m_JustCopy) return input;
326
wxCHECK_MSG(m_Table != NULL, s,
327
wxT("You must call wxEncodingConverter::Init() before actually converting!"));
331
for (i = input.c_str(); *i != 0; i++)
332
s << (wxChar)(m_Table[(wxUint16)*i]);
336
for (i = input.c_str(); *i != 0; i++)
337
s << (wxChar)(m_Table[(wxUint8)*i]);
349
// Following tables describe classes of encoding equivalence.
352
#define STOP wxFONTENCODING_SYSTEM
354
#define NUM_OF_PLATFORMS 4 /*must conform to enum wxPLATFORM_XXXX !!!*/
355
#define ENC_PER_PLATFORM 3
356
// max no. of encodings for one language used on one platform.
357
// Using maximum of everything at the current moment to not make the
358
// library larger than necessary. Make larger only if necessary - MR
360
static const wxFontEncoding
361
EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = {
363
// *** Please put more common encodings as first! ***
367
/* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP},
368
/* windows */ {wxFONTENCODING_CP1252, STOP},
370
/* mac */ {wxFONTENCODING_MACROMAN, STOP}
375
/* unix */ {wxFONTENCODING_ISO8859_2, STOP},
376
/* windows */ {wxFONTENCODING_CP1250, STOP},
378
/* mac */ {wxFONTENCODING_MACCENTRALEUR, STOP}
383
/* unix */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP},
384
/* windows */ {wxFONTENCODING_CP1257, STOP},
391
/* unix */ {wxFONTENCODING_ISO8859_8, STOP},
392
/* windows */ {wxFONTENCODING_CP1255, STOP},
394
/* mac */ {wxFONTENCODING_MACHEBREW, STOP}
399
/* unix */ {wxFONTENCODING_ISO8859_7, STOP},
400
/* windows */ {wxFONTENCODING_CP1253, STOP},
402
/* mac */ {wxFONTENCODING_MACGREEK, STOP}
407
/* unix */ {wxFONTENCODING_ISO8859_6, STOP},
408
/* windows */ {wxFONTENCODING_CP1256, STOP},
410
/* mac */ {wxFONTENCODING_MACARABIC, STOP}
415
/* unix */ {wxFONTENCODING_ISO8859_9, STOP},
416
/* windows */ {wxFONTENCODING_CP1254, STOP},
418
/* mac */ {wxFONTENCODING_MACTURKISH, STOP}
423
/* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_KOI8_U, wxFONTENCODING_ISO8859_5, STOP},
424
/* windows */ {wxFONTENCODING_CP1251, STOP},
426
/* mac */ {wxFONTENCODING_MACCYRILLIC, STOP}
429
{{STOP},{STOP},{STOP},{STOP}} /* Terminator */
430
/* no, _not_ Arnold! */
434
static bool FindEncoding(const wxFontEncodingArray& arr, wxFontEncoding f)
436
for (wxFontEncodingArray::const_iterator it = arr.begin(), en = arr.end();
443
wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform)
445
if (platform == wxPLATFORM_CURRENT)
447
#if defined(__WINDOWS__)
448
platform = wxPLATFORM_WINDOWS;
449
#elif defined(__WXGTK__) || defined(__WXMOTIF__)
450
platform = wxPLATFORM_UNIX;
451
#elif defined(__WXPM__)
452
platform = wxPLATFORM_OS2;
453
#elif defined(__WXMAC__)
454
platform = wxPLATFORM_MAC;
459
const wxFontEncoding *f;
460
wxFontEncodingArray arr;
463
while (EquivalentEncodings[clas][0][0] != STOP)
465
for (i = 0; i < NUM_OF_PLATFORMS; i++)
466
for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
467
if (EquivalentEncodings[clas][i][e] == enc)
469
for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
470
if (*f == enc) arr.push_back(enc);
471
for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
472
if (!FindEncoding(arr, *f)) arr.push_back(*f);
473
i = NUM_OF_PLATFORMS/*hack*/; break;
483
wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
486
const wxFontEncoding *f;
487
wxFontEncodingArray arr;
489
arr = GetPlatformEquivalents(enc); // we want them to be first items in array
492
while (EquivalentEncodings[clas][0][0] != STOP)
494
for (i = 0; i < NUM_OF_PLATFORMS; i++)
495
for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
496
if (EquivalentEncodings[clas][i][e] == enc)
498
for (j = 0; j < NUM_OF_PLATFORMS; j++)
499
for (f = EquivalentEncodings[clas][j]; *f != STOP; f++)
500
if (!FindEncoding(arr, *f)) arr.push_back(*f);
501
i = NUM_OF_PLATFORMS/*hack*/; break;