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