]>
Commit | Line | Data |
---|---|---|
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 | #if defined(__GNUG__) && !defined(NO_GCC_PRAGMA) | |
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 | #if wxUSE_FONTMAP | |
22 | ||
23 | #include "wx/encconv.h" | |
24 | ||
25 | #include <stdlib.h> | |
26 | ||
27 | // conversion tables, generated by scripts in $(WXWIN)/misc/unictabl: | |
28 | #ifdef __BORLANDC__ | |
29 | #include "../common/unictabl.inc" | |
30 | #else | |
31 | #include "unictabl.inc" | |
32 | #endif | |
33 | ||
34 | #if wxUSE_WCHAR_T | |
35 | typedef wchar_t tchar; | |
36 | #else | |
37 | typedef char tchar; | |
38 | #endif | |
39 | ||
40 | #ifdef __WXWINCE__ | |
41 | #undef LINKAGEMODE | |
42 | #define LINKAGEMODE __cdecl | |
43 | #endif | |
44 | ||
45 | static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc) | |
46 | { | |
47 | for (int i = 0; encodings_list[i].table != NULL; i++) | |
48 | { | |
49 | if (encodings_list[i].encoding == enc) | |
50 | return encodings_list[i].table; | |
51 | } | |
52 | return NULL; | |
53 | } | |
54 | ||
55 | typedef struct { | |
56 | wxUint16 u; | |
57 | wxUint8 c; | |
58 | } CharsetItem; | |
59 | ||
60 | extern "C" int LINKAGEMODE CompareCharsetItems(const void *i1, const void *i2) | |
61 | { | |
62 | return ( ((CharsetItem*)i1) -> u - ((CharsetItem*)i2) -> u ); | |
63 | } | |
64 | ||
65 | ||
66 | static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl) | |
67 | { | |
68 | CharsetItem *rev = new CharsetItem[128]; | |
69 | ||
70 | for (int i = 0; i < 128; i++) | |
71 | rev[i].c = 128 + i, rev[i].u = tbl[i]; | |
72 | ||
73 | qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems); | |
74 | ||
75 | return rev; | |
76 | } | |
77 | ||
78 | ||
79 | ||
80 | wxEncodingConverter::wxEncodingConverter() | |
81 | { | |
82 | m_Table = NULL; | |
83 | m_UnicodeInput = m_UnicodeOutput = FALSE; | |
84 | m_JustCopy = FALSE; | |
85 | } | |
86 | ||
87 | ||
88 | ||
89 | bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method) | |
90 | { | |
91 | unsigned i; | |
92 | wxUint16 *in_tbl, *out_tbl = NULL; | |
93 | ||
94 | if (m_Table) {delete[] m_Table; m_Table = NULL;} | |
95 | ||
96 | #if !wxUSE_WCHAR_T | |
97 | if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return FALSE; | |
98 | #endif | |
99 | ||
100 | if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;} | |
101 | ||
102 | m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE); | |
103 | m_JustCopy = FALSE; | |
104 | ||
105 | if (input_enc == wxFONTENCODING_UNICODE) | |
106 | { | |
107 | if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE; | |
108 | ||
109 | m_Table = new tchar[65536]; | |
110 | for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII | |
111 | for (i = 128; i < 65536; i++) m_Table[i] = (tchar)'?'; | |
112 | // FIXME - this should be character that means `unicode to charset' impossible, not '?' | |
113 | ||
114 | if (method == wxCONVERT_SUBSTITUTE) | |
115 | { | |
116 | for (i = 0; i < encoding_unicode_fallback_count; i++) | |
117 | m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s; | |
118 | } | |
119 | ||
120 | for (i = 0; i < 128; i++) | |
121 | m_Table[out_tbl[i]] = (tchar)(128 + i); | |
122 | ||
123 | m_UnicodeInput = TRUE; | |
124 | } | |
125 | else // input !Unicode | |
126 | { | |
127 | if ((in_tbl = GetEncTable(input_enc)) == NULL) return FALSE; | |
128 | if (output_enc != wxFONTENCODING_UNICODE) | |
129 | if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE; | |
130 | ||
131 | m_UnicodeInput = FALSE; | |
132 | ||
133 | m_Table = new tchar[256]; | |
134 | for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII | |
135 | ||
136 | if (output_enc == wxFONTENCODING_UNICODE) | |
137 | { | |
138 | for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i]; | |
139 | return TRUE; | |
140 | } | |
141 | // FIXME: write a substitute for bsearch | |
142 | #ifndef __WXWINCE__ | |
143 | else // output !Unicode | |
144 | { | |
145 | CharsetItem *rev = BuildReverseTable(out_tbl); | |
146 | CharsetItem *item; | |
147 | CharsetItem key; | |
148 | ||
149 | for (i = 0; i < 128; i++) | |
150 | { | |
151 | key.u = in_tbl[i]; | |
152 | item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems); | |
153 | if (item == NULL && method == wxCONVERT_SUBSTITUTE) | |
154 | item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback, | |
155 | encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems); | |
156 | if (item) | |
157 | m_Table[128 + i] = (tchar)item -> c; | |
158 | else | |
159 | #if wxUSE_WCHAR_T | |
160 | m_Table[128 + i] = (wchar_t)(128 + i); | |
161 | #else | |
162 | m_Table[128 + i] = (char)(128 + i); | |
163 | #endif | |
164 | } | |
165 | ||
166 | delete[] rev; | |
167 | } | |
168 | #endif // !__WXWINCE__ | |
169 | } | |
170 | ||
171 | return TRUE; | |
172 | } | |
173 | ||
174 | ||
175 | ||
176 | void wxEncodingConverter::Convert(const char* input, char* output) const | |
177 | { | |
178 | wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!")); | |
179 | wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!")); | |
180 | ||
181 | const char *i; | |
182 | char *o; | |
183 | ||
184 | if (m_JustCopy) | |
185 | { | |
186 | strcpy(output, input); | |
187 | return; | |
188 | } | |
189 | ||
190 | wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); | |
191 | ||
192 | for (i = input, o = output; *i != 0;) | |
193 | *(o++) = (char)(m_Table[(wxUint8)*(i++)]); | |
194 | *o = 0; | |
195 | } | |
196 | ||
197 | ||
198 | #if wxUSE_WCHAR_T | |
199 | ||
200 | void wxEncodingConverter::Convert(const char* input, wchar_t* output) const | |
201 | { | |
202 | wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!")); | |
203 | wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!")); | |
204 | ||
205 | const char *i; | |
206 | wchar_t *o; | |
207 | ||
208 | if (m_JustCopy) | |
209 | { | |
210 | for (i = input, o = output; *i != 0;) | |
211 | *(o++) = (wchar_t)(*(i++)); | |
212 | *o = 0; | |
213 | return; | |
214 | } | |
215 | ||
216 | wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); | |
217 | ||
218 | for (i = input, o = output; *i != 0;) | |
219 | *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]); | |
220 | *o = 0; | |
221 | } | |
222 | ||
223 | ||
224 | ||
225 | void wxEncodingConverter::Convert(const wchar_t* input, char* output) const | |
226 | { | |
227 | wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!")); | |
228 | wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!")); | |
229 | ||
230 | const wchar_t *i; | |
231 | char *o; | |
232 | ||
233 | if (m_JustCopy) | |
234 | { | |
235 | for (i = input, o = output; *i != 0;) | |
236 | *(o++) = (char)(*(i++)); | |
237 | *o = 0; | |
238 | return; | |
239 | } | |
240 | ||
241 | wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); | |
242 | ||
243 | for (i = input, o = output; *i != 0;) | |
244 | *(o++) = (char)(m_Table[(wxUint16)*(i++)]); | |
245 | *o = 0; | |
246 | } | |
247 | ||
248 | ||
249 | ||
250 | void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) const | |
251 | { | |
252 | wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!")); | |
253 | wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!")); | |
254 | ||
255 | const wchar_t *i; | |
256 | wchar_t *o; | |
257 | ||
258 | if (m_JustCopy) | |
259 | { | |
260 | // wcscpy() is not guaranteed to exist | |
261 | for (i = input, o = output; *i != 0;) | |
262 | *(o++) = (*(i++)); | |
263 | *o = 0; | |
264 | return; | |
265 | } | |
266 | ||
267 | wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); | |
268 | ||
269 | for (i = input, o = output; *i != 0;) | |
270 | *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]); | |
271 | *o = 0; | |
272 | } | |
273 | ||
274 | #endif // wxUSE_WCHAR_T | |
275 | ||
276 | ||
277 | wxString wxEncodingConverter::Convert(const wxString& input) const | |
278 | { | |
279 | if (m_JustCopy) return input; | |
280 | ||
281 | wxString s; | |
282 | const wxChar *i; | |
283 | ||
284 | wxCHECK_MSG(m_Table != NULL, s, | |
285 | wxT("You must call wxEncodingConverter::Init() before actually converting!")); | |
286 | ||
287 | if (m_UnicodeInput) | |
288 | { | |
289 | for (i = input.c_str(); *i != 0; i++) | |
290 | s << (wxChar)(m_Table[(wxUint16)*i]); | |
291 | } | |
292 | else | |
293 | { | |
294 | for (i = input.c_str(); *i != 0; i++) | |
295 | s << (wxChar)(m_Table[(wxUint8)*i]); | |
296 | } | |
297 | ||
298 | return s; | |
299 | } | |
300 | ||
301 | ||
302 | ||
303 | ||
304 | ||
305 | ||
306 | ||
307 | // Following tables describe classes of encoding equivalence. | |
308 | // | |
309 | ||
310 | #define STOP wxFONTENCODING_SYSTEM | |
311 | ||
312 | #define NUM_OF_PLATFORMS 4 /*must conform to enum wxPLATFORM_XXXX !!!*/ | |
313 | #define ENC_PER_PLATFORM 5 | |
314 | // max no. of encodings for one language used on one platform | |
315 | // Anybody thinks 5 is not enough? ;-) | |
316 | ||
317 | static wxFontEncoding | |
318 | EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = { | |
319 | ||
320 | // *** Please put more common encodings as first! *** | |
321 | ||
322 | // Western European | |
323 | { | |
324 | /* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP}, | |
325 | /* windows */ {wxFONTENCODING_CP1252, STOP}, | |
326 | /* os2 */ {STOP}, | |
327 | /* mac */ {STOP} | |
328 | }, | |
329 | ||
330 | // Central European | |
331 | { | |
332 | /* unix */ {wxFONTENCODING_ISO8859_2, STOP}, | |
333 | /* windows */ {wxFONTENCODING_CP1250, STOP}, | |
334 | /* os2 */ {STOP}, | |
335 | /* mac */ {STOP} | |
336 | }, | |
337 | ||
338 | // Baltic | |
339 | { | |
340 | /* unix */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP}, | |
341 | /* windows */ {wxFONTENCODING_CP1257, STOP}, | |
342 | /* os2 */ {STOP}, | |
343 | /* mac */ {STOP} | |
344 | }, | |
345 | ||
346 | // Hebrew | |
347 | { | |
348 | /* unix */ {wxFONTENCODING_ISO8859_8, STOP}, | |
349 | /* windows */ {wxFONTENCODING_CP1255, STOP}, | |
350 | /* os2 */ {STOP}, | |
351 | /* mac */ {STOP} | |
352 | }, | |
353 | ||
354 | // Greek | |
355 | { | |
356 | /* unix */ {wxFONTENCODING_ISO8859_7, STOP}, | |
357 | /* windows */ {wxFONTENCODING_CP1253, STOP}, | |
358 | /* os2 */ {STOP}, | |
359 | /* mac */ {STOP} | |
360 | }, | |
361 | ||
362 | // Arabic | |
363 | { | |
364 | /* unix */ {wxFONTENCODING_ISO8859_6, STOP}, | |
365 | /* windows */ {wxFONTENCODING_CP1256, STOP}, | |
366 | /* os2 */ {STOP}, | |
367 | /* mac */ {STOP} | |
368 | }, | |
369 | ||
370 | // Turkish | |
371 | { | |
372 | /* unix */ {wxFONTENCODING_ISO8859_9, STOP}, | |
373 | /* windows */ {wxFONTENCODING_CP1254, STOP}, | |
374 | /* os2 */ {STOP}, | |
375 | /* mac */ {STOP} | |
376 | }, | |
377 | ||
378 | // Cyrillic | |
379 | { | |
380 | /* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP}, | |
381 | /* windows */ {wxFONTENCODING_CP1251, STOP}, | |
382 | /* os2 */ {STOP}, | |
383 | /* mac */ {STOP} | |
384 | }, | |
385 | ||
386 | {{STOP},{STOP},{STOP},{STOP}} /* Terminator */ | |
387 | /* no, _not_ Arnold! */ | |
388 | }; | |
389 | ||
390 | ||
391 | static bool FindEncoding(const wxFontEncodingArray& arr, wxFontEncoding f) | |
392 | { | |
393 | for (wxFontEncodingArray::const_iterator it = arr.begin(), en = arr.end(); | |
394 | it != en; ++it) | |
395 | if (*it == f) | |
396 | return true; | |
397 | return false; | |
398 | } | |
399 | ||
400 | wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform) | |
401 | { | |
402 | if (platform == wxPLATFORM_CURRENT) | |
403 | { | |
404 | #if defined(__WXMSW__) | |
405 | platform = wxPLATFORM_WINDOWS; | |
406 | #elif defined(__WXGTK__) || defined(__WXMOTIF__) | |
407 | platform = wxPLATFORM_UNIX; | |
408 | #elif defined(__WXOS2__) | |
409 | platform = wxPLATFORM_OS2; | |
410 | #elif defined(__WXMAC__) | |
411 | platform = wxPLATFORM_MAC; | |
412 | #endif | |
413 | } | |
414 | ||
415 | int i, clas, e ; | |
416 | wxFontEncoding *f; | |
417 | wxFontEncodingArray arr; | |
418 | ||
419 | clas = 0; | |
420 | while (EquivalentEncodings[clas][0][0] != STOP) | |
421 | { | |
422 | for (i = 0; i < NUM_OF_PLATFORMS; i++) | |
423 | for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++) | |
424 | if (EquivalentEncodings[clas][i][e] == enc) | |
425 | { | |
426 | for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) | |
427 | if (*f == enc) arr.push_back(enc); | |
428 | for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) | |
429 | if (!FindEncoding(arr, *f)) arr.push_back(*f); | |
430 | i = NUM_OF_PLATFORMS/*hack*/; break; | |
431 | } | |
432 | clas++; | |
433 | } | |
434 | ||
435 | return arr; | |
436 | } | |
437 | ||
438 | ||
439 | ||
440 | wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc) | |
441 | { | |
442 | int i, clas, e, j ; | |
443 | wxFontEncoding *f; | |
444 | wxFontEncodingArray arr; | |
445 | ||
446 | arr = GetPlatformEquivalents(enc); // we want them to be first items in array | |
447 | ||
448 | clas = 0; | |
449 | while (EquivalentEncodings[clas][0][0] != STOP) | |
450 | { | |
451 | for (i = 0; i < NUM_OF_PLATFORMS; i++) | |
452 | for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++) | |
453 | if (EquivalentEncodings[clas][i][e] == enc) | |
454 | { | |
455 | for (j = 0; j < NUM_OF_PLATFORMS; j++) | |
456 | for (f = EquivalentEncodings[clas][j]; *f != STOP; f++) | |
457 | if (!FindEncoding(arr, *f)) arr.push_back(*f); | |
458 | i = NUM_OF_PLATFORMS/*hack*/; break; | |
459 | } | |
460 | clas++; | |
461 | } | |
462 | ||
463 | return arr; | |
464 | } | |
465 | ||
466 | #endif // wxUSE_FONTMAP |