1 | /**********************************************************************
|
---|
2 | *
|
---|
3 | * gsdlunicode.cpp --
|
---|
4 | * Copyright (C) 1999 The New Zealand Digital Library Project
|
---|
5 | *
|
---|
6 | * A component of the Greenstone digital library software
|
---|
7 | * from the New Zealand Digital Library Project at the
|
---|
8 | * University of Waikato, New Zealand.
|
---|
9 | *
|
---|
10 | * This program is free software; you can redistribute it and/or modify
|
---|
11 | * it under the terms of the GNU General Public License as published by
|
---|
12 | * the Free Software Foundation; either version 2 of the License, or
|
---|
13 | * (at your option) any later version.
|
---|
14 | *
|
---|
15 | * This program is distributed in the hope that it will be useful,
|
---|
16 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
17 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
18 | * GNU General Public License for more details.
|
---|
19 | *
|
---|
20 | * You should have received a copy of the GNU General Public License
|
---|
21 | * along with this program; if not, write to the Free Software
|
---|
22 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
23 | *
|
---|
24 | *********************************************************************/
|
---|
25 |
|
---|
26 | #include "gsdlunicode.h"
|
---|
27 |
|
---|
28 |
|
---|
29 | // unitool is currently in mg, if mg is not being used it should
|
---|
30 | // be moved into GSDLHOME/lib
|
---|
31 | // A copy of mgpp's unitool has now been moved into common-src/src/lib/
|
---|
32 | #include "unitool.h"
|
---|
33 |
|
---|
34 | #include "fileutil.h"
|
---|
35 |
|
---|
36 | #include <stdio.h>
|
---|
37 |
|
---|
38 | #if defined(GSDL_USE_OBJECTSPACE)
|
---|
39 | # include <ospace\std\iostream>
|
---|
40 | # include <ospace\std\fstream>
|
---|
41 | #elif defined(GSDL_USE_IOS_H)
|
---|
42 | # include <iostream.h>
|
---|
43 | # include <fstream.h>
|
---|
44 | #else
|
---|
45 | # include <iostream>
|
---|
46 | # include <fstream>
|
---|
47 | #endif
|
---|
48 |
|
---|
49 |
|
---|
50 | // converts a unicode encode text_t string to a utf-8
|
---|
51 | // encoded text_t string
|
---|
52 | text_t to_utf8 (text_t::const_iterator here, text_t::const_iterator end) {
|
---|
53 | text_t out;
|
---|
54 |
|
---|
55 | unsigned char thischar[MAXUTF8CHARLEN];
|
---|
56 | int i, charlen;
|
---|
57 |
|
---|
58 | while (here != end) {
|
---|
59 | charlen = output_utf8_char (*here, thischar, &thischar[MAXUTF8CHARLEN-1]);
|
---|
60 | for (i=0; i<charlen; ++i) out.push_back(thischar[i]);
|
---|
61 | ++here;
|
---|
62 | }
|
---|
63 |
|
---|
64 | return out;
|
---|
65 | }
|
---|
66 |
|
---|
67 | // converts a utf-8 encoded text_t string to a unicode
|
---|
68 | // encoded text_t string
|
---|
69 | text_t to_uni (const text_t &in) {
|
---|
70 | text_t out;
|
---|
71 | unsigned char *in_cstr = (unsigned char *)in.getcstr();
|
---|
72 | unsigned char *here = in_cstr;
|
---|
73 | unsigned char *end = in_cstr;
|
---|
74 |
|
---|
75 | unsigned short unichar;
|
---|
76 | int charlen = 0;
|
---|
77 |
|
---|
78 | // get the last valid character in the string
|
---|
79 | while (*end != '\0') ++end;
|
---|
80 | --end;
|
---|
81 |
|
---|
82 | while ((charlen = parse_utf8_char (here, end, &unichar)) > 0) {
|
---|
83 | out.push_back(unichar);
|
---|
84 | here += charlen;
|
---|
85 | }
|
---|
86 |
|
---|
87 | delete []in_cstr;
|
---|
88 |
|
---|
89 | return out;
|
---|
90 | }
|
---|
91 |
|
---|
92 |
|
---|
93 | // this works for all unicode values < 65536...
|
---|
94 | void utf16outconvertclass::convert (char *out, size_t maxlen, size_t &len, status_t &status) {
|
---|
95 | // we should already have text_t* input set...
|
---|
96 | if (input == NULL || out == NULL)
|
---|
97 | {
|
---|
98 | status = finished;
|
---|
99 | return;
|
---|
100 | }
|
---|
101 | unsigned char *output = (unsigned char *)out;
|
---|
102 | text_t::iterator textend = input->end();
|
---|
103 | len = 0;
|
---|
104 | if (maxlen % 2) --maxlen; // we need an even number of output bytes...
|
---|
105 | while ((len < maxlen) && (texthere != textend)) {
|
---|
106 | unsigned short int uni_char=(unsigned short int) *texthere;
|
---|
107 | // big endian utf-16...
|
---|
108 | if (uni_char < 256) {
|
---|
109 | out[len]=0;
|
---|
110 | out[len+1]=uni_char;
|
---|
111 | } else {
|
---|
112 | out[len]=uni_char >> 8;
|
---|
113 | out[len+1]=uni_char & 255;
|
---|
114 | }
|
---|
115 | len+=2;
|
---|
116 | ++texthere;
|
---|
117 | }
|
---|
118 | if (texthere==textend)
|
---|
119 | status=finished;
|
---|
120 | else
|
---|
121 | status=unfinished;
|
---|
122 | }
|
---|
123 |
|
---|
124 |
|
---|
125 | utf8inconvertclass::utf8inconvertclass () {
|
---|
126 | utf8buflen = 0;
|
---|
127 | }
|
---|
128 |
|
---|
129 | void utf8inconvertclass::reset () {
|
---|
130 | start = NULL;
|
---|
131 | len = 0;
|
---|
132 | utf8buflen=0;
|
---|
133 | }
|
---|
134 |
|
---|
135 | void utf8inconvertclass::convert (text_t &output, status_t &status) {
|
---|
136 | output.clear();
|
---|
137 | output.reserve (len/3);
|
---|
138 |
|
---|
139 | if (start == NULL || len == 0) {
|
---|
140 | if (utf8buflen == 0) status = finished;
|
---|
141 | else status = stopped;
|
---|
142 | return;
|
---|
143 | }
|
---|
144 |
|
---|
145 | // don't want any funny sign conversions happening
|
---|
146 | unsigned char *here = (unsigned char *)start;
|
---|
147 | unsigned char *end = here+len-1;
|
---|
148 | unsigned short c;
|
---|
149 | size_t realcharlen;
|
---|
150 |
|
---|
151 | size_t charlen = getutf8charlen ();
|
---|
152 | while (len > 0) {
|
---|
153 | if (charlen == 0) {
|
---|
154 | // start parsing a new character
|
---|
155 | utf8buflen = 0;
|
---|
156 |
|
---|
157 | // fast common case
|
---|
158 | while (len > 3) {
|
---|
159 | realcharlen = parse_utf8_char (here, end, &c);
|
---|
160 | output.push_back (c);
|
---|
161 | here += realcharlen;
|
---|
162 | len -= realcharlen;
|
---|
163 | }
|
---|
164 |
|
---|
165 | utf8buf[utf8buflen++] = *here;
|
---|
166 | ++here;
|
---|
167 | --len;
|
---|
168 | charlen = getutf8charlen ();
|
---|
169 |
|
---|
170 | } else if (utf8buflen < charlen) {
|
---|
171 | // assumes charlen is always less than MAXUTF8CHARLEN
|
---|
172 | utf8buf[utf8buflen++] = *here;
|
---|
173 | ++here;
|
---|
174 | --len;
|
---|
175 | }
|
---|
176 |
|
---|
177 | if (utf8buflen == charlen) {
|
---|
178 | // got a complete character
|
---|
179 | realcharlen = parse_utf8_char (utf8buf, &utf8buf[utf8buflen-1], &c);
|
---|
180 | output.push_back (c);
|
---|
181 |
|
---|
182 | // move any unparsed characters. If an error occurred some of
|
---|
183 | // the characters might be unused.
|
---|
184 | int i;
|
---|
185 | int diff = utf8buflen - realcharlen;
|
---|
186 | for (i=0; i < diff; ++i) utf8buf[i] = utf8buf[i+diff];
|
---|
187 | utf8buflen = diff;
|
---|
188 | charlen = getutf8charlen ();
|
---|
189 | }
|
---|
190 | }
|
---|
191 |
|
---|
192 | start = (char *)here; // save current position
|
---|
193 |
|
---|
194 | if (utf8buflen == 0) status = finished;
|
---|
195 | else status = stopped;
|
---|
196 | }
|
---|
197 |
|
---|
198 |
|
---|
199 | // returns the length that the current contents of the
|
---|
200 | // utf8buf should be
|
---|
201 | size_t utf8inconvertclass::getutf8charlen () {
|
---|
202 | if (utf8buflen == 0) return 0;
|
---|
203 |
|
---|
204 | // one byte character
|
---|
205 | if (utf8buf[0] < 0x80) return 1;
|
---|
206 |
|
---|
207 | // error, is not the start of a utf-8 character
|
---|
208 | if (utf8buf[0] < 0xc0) return 1;
|
---|
209 |
|
---|
210 | // two bute character
|
---|
211 | if (utf8buf[0] < 0xe0) return 2;
|
---|
212 |
|
---|
213 | // three byte character
|
---|
214 | if (utf8buf[0] < 0xf0) return 3;
|
---|
215 |
|
---|
216 | // error, character too long for unicode
|
---|
217 | return 1;
|
---|
218 | }
|
---|
219 |
|
---|
220 |
|
---|
221 | void utf8outconvertclass::reset () {
|
---|
222 | input = NULL;
|
---|
223 | outs = NULL;
|
---|
224 | utf8buflen = 0;
|
---|
225 | utf8bufhere = 0;
|
---|
226 | }
|
---|
227 |
|
---|
228 | // note that convert does not null-terminate the
|
---|
229 | // output array of characters
|
---|
230 | void utf8outconvertclass::convert (char *output, size_t maxlen,
|
---|
231 | size_t &len, status_t &status) {
|
---|
232 | if (input == NULL || output == NULL) {
|
---|
233 | if (utf8buflen == 0) status = finished;
|
---|
234 | else status = unfinished;
|
---|
235 | return;
|
---|
236 | }
|
---|
237 |
|
---|
238 | // don't want any funny sign conversions happening
|
---|
239 | unsigned char *uoutput = (unsigned char *)output;
|
---|
240 | text_t::iterator textend = input->end();
|
---|
241 | len = 0;
|
---|
242 | while (len < maxlen) {
|
---|
243 | // empty the contents of the internal buffer
|
---|
244 | if (utf8buflen > 0) {
|
---|
245 | while (len < maxlen && utf8bufhere < utf8buflen) {
|
---|
246 | *uoutput = utf8buf[utf8bufhere];
|
---|
247 | ++uoutput;
|
---|
248 | ++len;
|
---|
249 | ++utf8bufhere;
|
---|
250 | }
|
---|
251 |
|
---|
252 | if (utf8bufhere == utf8buflen) {
|
---|
253 | utf8bufhere = 0;
|
---|
254 | utf8buflen = 0;
|
---|
255 | }
|
---|
256 | }
|
---|
257 |
|
---|
258 | // fill up the buffer with the next character
|
---|
259 | if (utf8buflen == 0) {
|
---|
260 | if (texthere == textend) break; // finished!
|
---|
261 | if (!rzws || (*texthere != 0x200b))
|
---|
262 | utf8buflen = output_utf8_char (*texthere, utf8buf,
|
---|
263 | &utf8buf[MAXUTF8CHARLEN-1]);
|
---|
264 | ++texthere;
|
---|
265 | utf8bufhere = 0;
|
---|
266 | }
|
---|
267 | }
|
---|
268 |
|
---|
269 | if (texthere == textend && utf8buflen == 0) status = finished;
|
---|
270 | else status = unfinished;
|
---|
271 | }
|
---|
272 |
|
---|
273 |
|
---|
274 |
|
---|
275 |
|
---|
276 |
|
---|
277 |
|
---|
278 | mapdata_t::mapdata_t () {
|
---|
279 |
|
---|
280 | // reset all the map ptrs to be NULL
|
---|
281 | for (int i=0; i<256; ++i) {
|
---|
282 | ptrs[i] = (unsigned short *)NULL;
|
---|
283 | }
|
---|
284 |
|
---|
285 | // say nothing has been loaded
|
---|
286 | loaded = false;
|
---|
287 | }
|
---|
288 |
|
---|
289 |
|
---|
290 | mapconvert::mapconvert () {
|
---|
291 | absentc = 0;
|
---|
292 | }
|
---|
293 |
|
---|
294 | // setmapfile will cause loadmapfile to be called when conversion is
|
---|
295 | // needed
|
---|
296 | bool mapconvert::setmapfile (const text_t &themapfile, unsigned short theabsentc) {
|
---|
297 | // check to see if the mapfile has been already loaded
|
---|
298 | if (mapdata.loaded && mapfile == themapfile && absentc == theabsentc) return true;
|
---|
299 |
|
---|
300 | unloadmapfile ();
|
---|
301 | mapfile = themapfile;
|
---|
302 | absentc = theabsentc;
|
---|
303 |
|
---|
304 | return true;
|
---|
305 | }
|
---|
306 |
|
---|
307 |
|
---|
308 |
|
---|
309 | // loadmapfile should be called before any conversion is done
|
---|
310 | bool mapconvert::loadmapfile (const text_t &themapfile,
|
---|
311 | unsigned short theabsentc) {
|
---|
312 | FILE *mapfilein = (FILE *)NULL;
|
---|
313 |
|
---|
314 | // check to see if the mapfile has been already loaded
|
---|
315 | if (mapdata.loaded && mapfile == themapfile && absentc == theabsentc) return true;
|
---|
316 |
|
---|
317 | unloadmapfile ();
|
---|
318 | mapfile = themapfile;
|
---|
319 | absentc = theabsentc;
|
---|
320 |
|
---|
321 | // open the map file
|
---|
322 | char *cfilename = mapfile.getcstr();
|
---|
323 | if (cfilename == (char *)NULL) return false;
|
---|
324 | mapfilein = fopen(cfilename, "rb");
|
---|
325 | delete []cfilename; cfilename = NULL;
|
---|
326 |
|
---|
327 | if (mapfilein == (FILE *)NULL) return false;
|
---|
328 |
|
---|
329 | unsigned char c, n1, n2;
|
---|
330 | unsigned short *arrptr;
|
---|
331 | int i;
|
---|
332 | c = fgetc (mapfilein);
|
---|
333 | while (!feof (mapfilein)) {
|
---|
334 | if (mapdata.ptrs[c] == (unsigned short *)NULL) {
|
---|
335 | // allocate a new array
|
---|
336 | arrptr = new unsigned short[256];
|
---|
337 | mapdata.ptrs[c] = arrptr;
|
---|
338 | } else arrptr = mapdata.ptrs[c];
|
---|
339 |
|
---|
340 | // clear the array
|
---|
341 | for (i=0; i<256; ++i) arrptr[i] = 0;
|
---|
342 |
|
---|
343 | // read in this block
|
---|
344 | n1 = fgetc (mapfilein);
|
---|
345 | n2 = fgetc (mapfilein);
|
---|
346 | i=0;
|
---|
347 | while (!feof (mapfilein)) {
|
---|
348 | arrptr[i] = ((unsigned short)n1 << 8) | (unsigned short)n2;
|
---|
349 |
|
---|
350 | ++i;
|
---|
351 | if (i >= 256) break;
|
---|
352 | n1 = fgetc (mapfilein);
|
---|
353 | n2 = fgetc (mapfilein);
|
---|
354 | }
|
---|
355 |
|
---|
356 | c = fgetc (mapfilein);
|
---|
357 | }
|
---|
358 |
|
---|
359 | mapdata.loaded = true;
|
---|
360 |
|
---|
361 | return true;
|
---|
362 | }
|
---|
363 |
|
---|
364 | void mapconvert::unloadmapfile () {
|
---|
365 | if (!mapdata.loaded) return;
|
---|
366 |
|
---|
367 | for (int i=0; i<256; ++i) {
|
---|
368 | if (mapdata.ptrs[i] != (unsigned short *)NULL) {
|
---|
369 | delete [] mapdata.ptrs[i];
|
---|
370 | mapdata.ptrs[i] = (unsigned short *)NULL;
|
---|
371 | }
|
---|
372 | }
|
---|
373 |
|
---|
374 | mapdata.loaded = false;
|
---|
375 | }
|
---|
376 |
|
---|
377 |
|
---|
378 | unsigned short mapconvert::convert (unsigned short c) {
|
---|
379 | if (!mapdata.loaded) {
|
---|
380 | if (!mapfile.empty() && loadmapfile (mapfile, absentc)) {
|
---|
381 | // do nothing, successfully loaded database
|
---|
382 | } else return absentc;
|
---|
383 | }
|
---|
384 |
|
---|
385 | if (c == 0) return 0; // 0 always maps to 0...
|
---|
386 |
|
---|
387 | unsigned short n1 = c >> 8;
|
---|
388 | unsigned short n2 = c & 0xff;
|
---|
389 |
|
---|
390 | unsigned short *arrptr = mapdata.ptrs[n1];
|
---|
391 | if (arrptr == (unsigned short *)NULL) return absentc;
|
---|
392 |
|
---|
393 | if (arrptr[n2] == 0) return absentc;
|
---|
394 | return arrptr[n2];
|
---|
395 | }
|
---|
396 |
|
---|
397 | text_t mapconvert::convert (const text_t &instr) {
|
---|
398 | if (!mapdata.loaded) return absentc;
|
---|
399 |
|
---|
400 | text_t outstr;
|
---|
401 | text_t::const_iterator here = instr.begin();
|
---|
402 | text_t::const_iterator end = instr.end();
|
---|
403 |
|
---|
404 | while (here != end) {
|
---|
405 | outstr.push_back(this->convert(*here));
|
---|
406 | ++here;
|
---|
407 | }
|
---|
408 |
|
---|
409 | return outstr;
|
---|
410 | }
|
---|
411 |
|
---|
412 |
|
---|
413 |
|
---|
414 |
|
---|
415 | mapinconvertclass::mapinconvertclass () {
|
---|
416 | m_multibyte = 0;
|
---|
417 | mapbuflen = 0;
|
---|
418 | }
|
---|
419 |
|
---|
420 | void mapinconvertclass::reset () {
|
---|
421 | start = NULL;
|
---|
422 | len = 0;
|
---|
423 | mapbuflen=0;
|
---|
424 | }
|
---|
425 |
|
---|
426 | void mapinconvertclass::convert (text_t &output, status_t &status) {
|
---|
427 | output.clear();
|
---|
428 |
|
---|
429 | if (start == NULL || len == 0) {
|
---|
430 | if (mapbuflen == 0) status = finished;
|
---|
431 | else status = stopped;
|
---|
432 | return;
|
---|
433 | }
|
---|
434 |
|
---|
435 | // don't want any funny sign conversions happening
|
---|
436 | unsigned char *here = (unsigned char *)start;
|
---|
437 |
|
---|
438 | size_t charlen = getmapcharlen ();
|
---|
439 | while (len > 0) {
|
---|
440 | if (charlen == 0) {
|
---|
441 | // start parsing a new character
|
---|
442 | mapbuflen = 0;
|
---|
443 | mapbuf[mapbuflen++] = *here;
|
---|
444 | ++here;
|
---|
445 | --len;
|
---|
446 | charlen = getmapcharlen ();
|
---|
447 |
|
---|
448 | } else if (mapbuflen < charlen) {
|
---|
449 | // assumes charlen is always less than MAXMAPCHARLEN
|
---|
450 | mapbuf[mapbuflen++] = *here;
|
---|
451 | ++here;
|
---|
452 | --len;
|
---|
453 | }
|
---|
454 |
|
---|
455 | if (mapbuflen == charlen) {
|
---|
456 | // got a complete character
|
---|
457 | if (charlen == 1) {
|
---|
458 | if (mapbuf[0] < 0x80) {
|
---|
459 | // ascii character
|
---|
460 | output.push_back (mapbuf[0]);
|
---|
461 | } else {
|
---|
462 | output.push_back (converter.convert((unsigned short)mapbuf[0]));
|
---|
463 | }
|
---|
464 |
|
---|
465 | } else {
|
---|
466 | // two byte character
|
---|
467 | output.push_back (converter.convert(((unsigned short)mapbuf[0] << 8) |
|
---|
468 | (unsigned short)mapbuf[1]));
|
---|
469 | }
|
---|
470 |
|
---|
471 | mapbuflen = 0;
|
---|
472 | charlen = 0;
|
---|
473 | }
|
---|
474 | }
|
---|
475 |
|
---|
476 | start = (char *)here; // save current position
|
---|
477 |
|
---|
478 | if (mapbuflen == 0) status = finished;
|
---|
479 | else status = stopped;
|
---|
480 | }
|
---|
481 |
|
---|
482 |
|
---|
483 |
|
---|
484 | mapoutconvertclass::mapoutconvertclass () {
|
---|
485 | m_multibyte = 0;
|
---|
486 | mapbuflen=0;
|
---|
487 | mapbufhere=0;
|
---|
488 | }
|
---|
489 |
|
---|
490 | void mapoutconvertclass::reset () {
|
---|
491 | input = NULL;
|
---|
492 | outs = NULL;
|
---|
493 | mapbuflen = 0;
|
---|
494 | mapbufhere = 0;
|
---|
495 | }
|
---|
496 |
|
---|
497 | // note that convert does not null-terminate the
|
---|
498 | // output array of characters
|
---|
499 | void mapoutconvertclass::convert (char *output, size_t maxlen,
|
---|
500 | size_t &len, status_t &status) {
|
---|
501 | unsigned short outc;
|
---|
502 |
|
---|
503 | if (input == NULL || output == NULL) {
|
---|
504 | if (mapbuflen == 0) status = finished;
|
---|
505 | else status = unfinished;
|
---|
506 | return;
|
---|
507 | }
|
---|
508 |
|
---|
509 | // don't want any funny sign conversions happening
|
---|
510 | unsigned char *uoutput = (unsigned char *)output;
|
---|
511 | text_t::iterator textend = input->end();
|
---|
512 | len = 0;
|
---|
513 | while (len < maxlen) {
|
---|
514 | // empty the contents of the internal buffer
|
---|
515 | if (mapbuflen > 0) {
|
---|
516 | while (len < maxlen && mapbufhere < mapbuflen) {
|
---|
517 | *uoutput = mapbuf[mapbufhere];
|
---|
518 | ++uoutput;
|
---|
519 | ++len;
|
---|
520 | ++mapbufhere;
|
---|
521 | }
|
---|
522 |
|
---|
523 | if (mapbufhere == mapbuflen) {
|
---|
524 | mapbufhere = 0;
|
---|
525 | mapbuflen = 0;
|
---|
526 | }
|
---|
527 | }
|
---|
528 |
|
---|
529 | // fill up the buffer with the next character
|
---|
530 | if (mapbuflen == 0) {
|
---|
531 | if (texthere == textend) break; // finished!
|
---|
532 | if (!rzws || (*texthere != 0x200b)) {
|
---|
533 | if (*texthere < 0x80) {
|
---|
534 | mapbuf[0] = (unsigned char)*texthere;
|
---|
535 | mapbuflen = 1;
|
---|
536 | } else {
|
---|
537 | outc = converter.convert (*texthere);
|
---|
538 | if (m_multibyte) {
|
---|
539 | mapbuf[0] = (unsigned char)(outc >> 8);
|
---|
540 | mapbuf[1] = (unsigned char)(outc & 0xff);
|
---|
541 | mapbuflen = 2;
|
---|
542 | } else {
|
---|
543 | mapbuf[0] = outc;
|
---|
544 | mapbuflen = 1;
|
---|
545 | }
|
---|
546 | }
|
---|
547 | }
|
---|
548 |
|
---|
549 | ++texthere;
|
---|
550 | mapbufhere = 0;
|
---|
551 | }
|
---|
552 | }
|
---|
553 |
|
---|
554 | if (texthere == textend && mapbuflen == 0) status = finished;
|
---|
555 | else status = unfinished;
|
---|
556 | }
|
---|
557 |
|
---|
558 |
|
---|
559 | bool simplemapconvert::loadmapfile (bool in) {
|
---|
560 | if (loaded) return true;
|
---|
561 | if (mapfile.empty()) return false;
|
---|
562 |
|
---|
563 | char *cfilename = mapfile.getcstr();
|
---|
564 | #ifdef GSDL_USE_IOS_H
|
---|
565 | ifstream mapfilein (cfilename, ios::in | ios::nocreate);
|
---|
566 | #else
|
---|
567 | ifstream mapfilein (cfilename, ios::in);
|
---|
568 | #endif
|
---|
569 | delete []cfilename;
|
---|
570 | if (!mapfilein) return false;
|
---|
571 |
|
---|
572 | char cline[2048];
|
---|
573 | text_t line;
|
---|
574 |
|
---|
575 | while (!mapfilein.eof()) {
|
---|
576 | mapfilein.getline (cline, 2048);
|
---|
577 | line.clear();
|
---|
578 | line.appendcstr (cline);
|
---|
579 | if (line.empty()) continue;
|
---|
580 | // remove comments
|
---|
581 | text_t::iterator end = line.end();
|
---|
582 | text_t::iterator here = findchar (line.begin(), end, '#');
|
---|
583 | if (here != end) {
|
---|
584 | line.erase (here, end);
|
---|
585 | if (line.empty()) continue;
|
---|
586 | }
|
---|
587 |
|
---|
588 | text_tarray parts;
|
---|
589 | splitchar (line.begin(), line.end(), '\t', parts);
|
---|
590 |
|
---|
591 | // do some simple sanity checks
|
---|
592 | if (parts.size() < 2) continue;
|
---|
593 | text_t::iterator begin1 = parts[0].begin();
|
---|
594 | text_t::iterator begin2 = parts[1].begin();
|
---|
595 | if (*begin1 != '0' || *(begin1+1) != 'x') continue;
|
---|
596 | if (*begin2 != '0' || *(begin2+1) != 'x') continue;
|
---|
597 | char *from = parts[0].getcstr();
|
---|
598 | char *to = parts[1].getcstr();
|
---|
599 | unsigned int f = 0, t = 0;
|
---|
600 | sscanf (from, "%i", &f);
|
---|
601 | sscanf (to, "%i", &t);
|
---|
602 | delete []from;
|
---|
603 | delete []to;
|
---|
604 |
|
---|
605 | if (in) mapping[(unsigned short)f] = (unsigned short)t;
|
---|
606 | else mapping[(unsigned short)t] = (unsigned short)f;
|
---|
607 | }
|
---|
608 |
|
---|
609 | loaded = true;
|
---|
610 | return true;
|
---|
611 | }
|
---|
612 |
|
---|
613 | unsigned short simplemapconvert::convert (unsigned short c, bool in) {
|
---|
614 |
|
---|
615 | if (!loaded)
|
---|
616 | if (!loadmapfile(in)) return absentc;
|
---|
617 |
|
---|
618 | return mapping[c];
|
---|
619 | }
|
---|
620 |
|
---|
621 |
|
---|
622 | void simplemapinconvertclass::convert (text_t &output, status_t &status) {
|
---|
623 | output.clear();
|
---|
624 |
|
---|
625 | if (start == NULL || len == 0) {
|
---|
626 | status = finished;
|
---|
627 | return;
|
---|
628 | }
|
---|
629 |
|
---|
630 | // don't want any funny sign conversions happening
|
---|
631 | unsigned char *here = (unsigned char *)start;
|
---|
632 | while (len > 0) {
|
---|
633 |
|
---|
634 | if (*here < 0x80)
|
---|
635 | output.push_back (*here); // append this character
|
---|
636 | else
|
---|
637 | output.push_back (converter.convert(*here, true));
|
---|
638 |
|
---|
639 | ++here;
|
---|
640 | --len;
|
---|
641 | }
|
---|
642 |
|
---|
643 | start = (char *)here; // save current position
|
---|
644 | status = finished;
|
---|
645 | }
|
---|
646 |
|
---|
647 |
|
---|
648 | void simplemapoutconvertclass::convert (char *output, size_t maxlen,
|
---|
649 | size_t &len, status_t &status) {
|
---|
650 |
|
---|
651 | if (input == NULL || output == NULL) {
|
---|
652 | status = finished;
|
---|
653 | return;
|
---|
654 | }
|
---|
655 |
|
---|
656 | // don't want any funny sign conversions happening
|
---|
657 | unsigned char *uoutput = (unsigned char *)output;
|
---|
658 | text_t::iterator textend = input->end();
|
---|
659 | len = 0;
|
---|
660 | while ((len < maxlen) && (texthere != textend)) {
|
---|
661 |
|
---|
662 | if (*texthere < 0x80) *uoutput = (unsigned char)(*texthere);
|
---|
663 | else *uoutput = converter.convert (*texthere, false);
|
---|
664 |
|
---|
665 | ++uoutput;
|
---|
666 | ++len;
|
---|
667 | ++texthere;
|
---|
668 | }
|
---|
669 |
|
---|
670 | if (texthere == textend) status = finished;
|
---|
671 | else status = unfinished;
|
---|
672 | }
|
---|