1 /*-
2 * Copyright (C) 2009, 2010 Gabor Kovesdan <[email protected]>
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 * notice, this list of conditions and the following disclaimer in the
12 * documentation and/or other materials provided with the distribution.
13 *
14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
18 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
24 * SUCH DAMAGE.
25 */
26
27 #include <sys/cdefs.h>
28 __FBSDID("$FreeBSD$");
29
30 #include <sys/endian.h>
31 #include <sys/types.h>
32
33 #include <err.h>
34 #include <errno.h>
35 #include <getopt.h>
36 #include <iconv.h>
37 #include <stdbool.h>
38 #include <stdio.h>
39 #include <stdlib.h>
40
41 #define UC_TO_MB_FLAG 1
42 #define MB_TO_WC_FLAG 2
43 #define MB_TO_UC_FLAG 4
44 #define WC_TO_MB_FLAG 8
45
46 #define MAX(a,b) ((a) < (b) ? (b) : (a))
47
48 extern char *__progname;
49
50 static const char *optstr = "cdilrt";
51 static const char *citrus_common = "SRC_ZONE\t0x0000-0xFFFF\n"
52 "OOB_MODE\tILSEQ\n"
53 "DST_ILSEQ\t0xFFFE\n"
54 "DST_UNIT_BITS\t32\n\n"
55 "BEGIN_MAP\n"
56 "#\n# Generated with Citrus iconv (FreeBSD)\n#\n";
57 bool cflag;
58 bool dflag;
59 bool iflag;
60 bool lflag;
61 bool tflag;
62 bool rflag;
63 int fb_flags;
64
65 static void do_conv(iconv_t, bool);
66 void mb_to_uc_fb(const char*, size_t,
67 void (*write_replacement)(const unsigned int *,
68 size_t, void *), void *, void *);
69 void mb_to_wc_fb(const char*, size_t,
70 void (*write_replacement) (const wchar_t *, size_t, void *),
71 void *, void *);
72 void uc_to_mb_fb(unsigned int,
73 void (*write_replacement) (const char *, size_t, void *), void *,
74 void *);
75 void wc_to_mb_fb(wchar_t,
76 void (*write_replacement)(const char *,
77 size_t, void *), void *, void *);
78
79 struct option long_options[] =
80 {
81 {"citrus", no_argument, NULL, 'c'},
82 {"diagnostic", no_argument, NULL, 'd'},
83 {"ignore", no_argument, NULL, 'i'},
84 {"long", no_argument, NULL, 'l'},
85 {"reverse", no_argument, NULL, 'r'},
86 {"translit", no_argument, NULL, 't'},
87 {NULL, no_argument, NULL, 0}
88 };
89
90 static void
usage(void)91 usage(void) {
92
93 fprintf(stderr, "Usage: %s [-cdilrt] ENCODING\n", __progname);
94 exit(EXIT_FAILURE);
95 }
96
97 static void
format_diag(int errcode)98 format_diag(int errcode)
99 {
100 const char *errstr;
101 const char *u2m, *m2u, *m2w, *w2m;
102
103 switch (errcode) {
104 case EINVAL:
105 errstr = "EINVAL ";
106 break;
107 case EILSEQ:
108 errstr = "EILSEQ ";
109 break;
110 case E2BIG:
111 errstr = "E2BIG ";
112 break;
113 default:
114 errstr = "UNKNOWN ";
115 break;
116 }
117
118 u2m = (fb_flags & UC_TO_MB_FLAG) ? "U2M " : "";
119 m2w = (fb_flags & MB_TO_WC_FLAG) ? "M2W " : "";
120 m2u = (fb_flags & MB_TO_UC_FLAG) ? "M2U " : "";
121 w2m = (fb_flags & WC_TO_MB_FLAG) ? "W2M " : "";
122
123 printf("%s%s%s%s%s", errstr, u2m, m2w, m2u, w2m);
124 }
125
126 static int
magnitude(const uint32_t p)127 magnitude(const uint32_t p)
128 {
129
130 if (p >> 8 == 0)
131 return (1);
132 else if (p >> 16 == 0)
133 return (2);
134 else
135 return (p >> 24 == 0 ? 3 : 4);
136 }
137
138 static void
format(const uint32_t data)139 format(const uint32_t data)
140 {
141
142 /* XXX: could be simpler, something like this but with leading 0s?
143
144 printf("0x%.*X", magnitude(data), data);
145 */
146
147 switch (magnitude(data)) {
148 default:
149 case 2:
150 printf("0x%04X", data);
151 break;
152 case 3:
153 printf("0x%06X", data);
154 break;
155 case 4:
156 printf("0x%08X", data);
157 break;
158 }
159 }
160
161 void
uc_to_mb_fb(unsigned int code,void (* write_replacement)(const char * buf,size_t buflen,void * callback_arg),void * callback_arg,void * data)162 uc_to_mb_fb(unsigned int code,
163 void (*write_replacement)(const char *buf, size_t buflen,
164 void* callback_arg), void* callback_arg, void* data)
165 {
166
167 fb_flags |= UC_TO_MB_FLAG;
168 }
169
170 void
mb_to_wc_fb(const char * inbuf,size_t inbufsize,void (* write_replacement)(const wchar_t * buf,size_t buflen,void * callback_arg),void * callback_arg,void * data)171 mb_to_wc_fb(const char* inbuf, size_t inbufsize,
172 void (*write_replacement)(const wchar_t *buf, size_t buflen,
173 void* callback_arg), void* callback_arg, void* data)
174 {
175
176 fb_flags |= MB_TO_WC_FLAG;
177 }
178
179 void
mb_to_uc_fb(const char * inbuf,size_t inbufsize,void (* write_replacement)(const unsigned int * buf,size_t buflen,void * callback_arg),void * callback_arg,void * data)180 mb_to_uc_fb(const char* inbuf, size_t inbufsize,
181 void (*write_replacement)(const unsigned int *buf, size_t buflen,
182 void* callback_arg), void* callback_arg, void* data)
183 {
184
185 fb_flags |= MB_TO_UC_FLAG;
186 }
187
188 void
wc_to_mb_fb(wchar_t wc,void (* write_replacement)(const char * buf,size_t buflen,void * callback_arg),void * callback_arg,void * data)189 wc_to_mb_fb(wchar_t wc,
190 void (*write_replacement)(const char *buf, size_t buflen,
191 void* callback_arg), void* callback_arg, void* data)
192 {
193
194 fb_flags |= WC_TO_MB_FLAG;
195 }
196
197 int
main(int argc,char * argv[])198 main (int argc, char *argv[])
199 {
200 struct iconv_fallbacks fbs;
201 iconv_t cd;
202 char *tocode;
203 char c;
204
205 while (((c = getopt_long(argc, argv, optstr, long_options, NULL)) != -1)) {
206 switch (c) {
207 case 'c':
208 cflag = true;
209 break;
210 case 'd':
211 dflag = true;
212 break;
213 case 'i':
214 iflag = true;
215 break;
216 case 'l':
217 lflag = true;
218 break;
219 case 'r':
220 rflag = true;
221 break;
222 case 't':
223 tflag = true;
224 break;
225 }
226 }
227 argc -= optind;
228 argv += optind;
229
230 if (argc < 1)
231 usage();
232
233 fbs.uc_to_mb_fallback = uc_to_mb_fb;
234 fbs.mb_to_wc_fallback = mb_to_wc_fb;
235 fbs.mb_to_uc_fallback = mb_to_uc_fb;
236 fbs.wc_to_mb_fallback = wc_to_mb_fb;
237 fbs.data = NULL;
238
239 if (argc == 2) {
240 asprintf(&tocode, "%s%s%s", argv[1], tflag ? "//TRASNLIT" : "",
241 iflag ? "//IGNORE" : "");
242
243 if ((cd = iconv_open(tocode, argv[0])) == (iconv_t)-1)
244 err(1, NULL);
245 if (dflag) {
246 if (iconvctl(cd, ICONV_SET_FALLBACKS, &fbs) != 0)
247 err(1, NULL);
248 }
249 do_conv(cd, false);
250 } else if (rflag) {
251 asprintf(&tocode, "%s%s%s", argv[0], tflag ? "//TRANSLIT" : "",
252 iflag ? "//IGNORE" : "");
253
254 if ((cd = iconv_open(tocode, "UTF-32LE")) == (iconv_t)-1)
255 err(1, NULL);
256 if (dflag && iconvctl(cd, ICONV_SET_FALLBACKS, &fbs) != 0)
257 err(1, NULL);
258 if (cflag) {
259 printf("# $FreeBSD$\n\n");
260 printf("TYPE\t\tROWCOL\n");
261 printf("NAME\t\tUCS/%s\n", argv[0]);
262 printf("%s", citrus_common);
263 }
264 do_conv(cd, true);
265 } else {
266 if ((cd = iconv_open("UTF-32LE//TRANSLIT", argv[0])) == (iconv_t)-1)
267 err(1, NULL);
268 if (dflag && (iconvctl(cd, ICONV_SET_FALLBACKS, &fbs) != 0))
269 err(1, NULL);
270 if (cflag) {
271 printf("# $FreeBSD$\n\n");
272 printf("TYPE\t\tROWCOL\n");
273 printf("NAME\t\t%s/UCS\n", argv[0]);
274 printf("%s", citrus_common);
275 }
276 do_conv(cd, false);
277 }
278
279 if (iconv_close(cd) != 0)
280 err(1, NULL);
281
282 return (EXIT_SUCCESS);
283 }
284
285 static void
do_conv(iconv_t cd,bool uniinput)286 do_conv(iconv_t cd, bool uniinput) {
287 size_t inbytesleft, outbytesleft, ret;
288 uint32_t outbuf;
289 uint32_t inbuf;
290 const char *inbuf_;
291 char *outbuf_;
292
293 for (inbuf = 0; inbuf < (lflag ? 0x100000 : 0x10000); inbuf += 1) {
294 if (uniinput && (inbuf >= 0xD800) && (inbuf <= 0xDF00))
295 continue;
296 inbytesleft = uniinput ? 4 : magnitude(inbuf);
297 outbytesleft = 4;
298 outbuf = 0x00000000;
299 outbuf_ = (char *)&outbuf;
300 inbuf_ = (const char *)&inbuf;
301 iconv(cd, NULL, NULL, NULL, NULL);
302 fb_flags = 0;
303 errno = 0;
304 ret = iconv(cd, &inbuf_, &inbytesleft, &outbuf_, &outbytesleft);
305 if (ret == (size_t)-1) {
306 if (dflag) {
307 format(inbuf);
308 printf(" = ");
309 format_diag(errno);
310 printf("\n");
311 }
312 continue;
313 }
314 format(inbuf);
315 printf(" = ");
316 format(outbuf);
317 printf("\n");
318 }
319 if (cflag)
320 printf("END_MAP\n");
321 }
322