1 /*-
2 * Copyright (c) 1990, 1993
3 * The Regents of the University of California. All rights reserved.
4 *
5 * This code is derived from software contributed to Berkeley by
6 * Chris Torek.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 * 3. Neither the name of the University nor the names of its contributors
17 * may be used to endorse or promote products derived from this software
18 * without specific prior written permission.
19 *
20 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30 * SUCH DAMAGE.
31 */
32
33 #include <ctype.h>
34 #include <stdlib.h>
35
36 #include "local.h"
37 #include "floatio.h"
38
39 #define BUF 513 /* Maximum length of numeric string. */
40
parsefloat(FILE * fp,char * buf,char * end)41 size_t parsefloat(FILE *fp, char *buf, char *end) {
42 char *commit, *p;
43 int infnanpos = 0;
44 enum {
45 S_START, S_GOTSIGN, S_INF, S_NAN, S_MAYBEHEX,
46 S_DIGITS, S_FRAC, S_EXP, S_EXPDIGITS
47 } state = S_START;
48 unsigned char c;
49 int gotmantdig = 0, ishex = 0;
50
51 /*
52 * We set commit = p whenever the string we have read so far
53 * constitutes a valid representation of a floating point
54 * number by itself. At some point, the parse will complete
55 * or fail, and we will ungetc() back to the last commit point.
56 * To ensure that the file offset gets updated properly, it is
57 * always necessary to read at least one character that doesn't
58 * match; thus, we can't short-circuit "infinity" or "nan(...)".
59 */
60 commit = buf - 1;
61 for (p = buf; p < end; ) {
62 c = *fp->_p;
63 reswitch:
64 switch (state) {
65 case S_START:
66 state = S_GOTSIGN;
67 if (c == '-' || c == '+')
68 break;
69 else
70 goto reswitch;
71 case S_GOTSIGN:
72 switch (c) {
73 case '0':
74 state = S_MAYBEHEX;
75 commit = p;
76 break;
77 case 'I':
78 case 'i':
79 state = S_INF;
80 break;
81 case 'N':
82 case 'n':
83 state = S_NAN;
84 break;
85 default:
86 state = S_DIGITS;
87 goto reswitch;
88 }
89 break;
90 case S_INF:
91 if (infnanpos > 6 ||
92 (c != "nfinity"[infnanpos] &&
93 c != "NFINITY"[infnanpos]))
94 goto parsedone;
95 if (infnanpos == 1 || infnanpos == 6)
96 commit = p; /* inf or infinity */
97 infnanpos++;
98 break;
99 case S_NAN:
100 switch (infnanpos) {
101 case -1: /* XXX kludge to deal with nan(...) */
102 goto parsedone;
103 case 0:
104 if (c != 'A' && c != 'a')
105 goto parsedone;
106 break;
107 case 1:
108 if (c != 'N' && c != 'n')
109 goto parsedone;
110 else
111 commit = p;
112 break;
113 case 2:
114 if (c != '(')
115 goto parsedone;
116 break;
117 default:
118 if (c == ')') {
119 commit = p;
120 infnanpos = -2;
121 } else if (!isalnum(c) && c != '_')
122 goto parsedone;
123 break;
124 }
125 infnanpos++;
126 break;
127 case S_MAYBEHEX:
128 state = S_DIGITS;
129 if (c == 'X' || c == 'x') {
130 ishex = 1;
131 break;
132 } else { /* we saw a '0', but no 'x' */
133 gotmantdig = 1;
134 goto reswitch;
135 }
136 case S_DIGITS:
137 if ((ishex && isxdigit(c)) || isdigit(c))
138 gotmantdig = 1;
139 else {
140 state = S_FRAC;
141 if (c != '.')
142 goto reswitch;
143 }
144 if (gotmantdig)
145 commit = p;
146 break;
147 case S_FRAC:
148 if (((c == 'E' || c == 'e') && !ishex) ||
149 ((c == 'P' || c == 'p') && ishex)) {
150 if (!gotmantdig)
151 goto parsedone;
152 else
153 state = S_EXP;
154 } else if ((ishex && isxdigit(c)) || isdigit(c)) {
155 commit = p;
156 gotmantdig = 1;
157 } else
158 goto parsedone;
159 break;
160 case S_EXP:
161 state = S_EXPDIGITS;
162 if (c == '-' || c == '+')
163 break;
164 else
165 goto reswitch;
166 case S_EXPDIGITS:
167 if (isdigit(c))
168 commit = p;
169 else
170 goto parsedone;
171 break;
172 default:
173 abort();
174 }
175 *p++ = c;
176 if (--fp->_r > 0)
177 fp->_p++;
178 else if (__srefill(fp))
179 break; /* EOF */
180 }
181
182 parsedone:
183 while (commit < --p)
184 (void)ungetc(*(unsigned char *)p, fp);
185 *++commit = '\0';
186 return commit - buf;
187 }
188
wparsefloat(FILE * fp,wchar_t * buf,wchar_t * end)189 size_t wparsefloat(FILE *fp, wchar_t *buf, wchar_t *end) {
190 wchar_t *commit, *p;
191 int infnanpos = 0;
192 enum {
193 S_START, S_GOTSIGN, S_INF, S_NAN, S_MAYBEHEX,
194 S_DIGITS, S_FRAC, S_EXP, S_EXPDIGITS
195 } state = S_START;
196 wint_t c;
197 int gotmantdig = 0, ishex = 0;
198
199 /*
200 * We set commit = p whenever the string we have read so far
201 * constitutes a valid representation of a floating point
202 * number by itself. At some point, the parse will complete
203 * or fail, and we will ungetc() back to the last commit point.
204 * To ensure that the file offset gets updated properly, it is
205 * always necessary to read at least one character that doesn't
206 * match; thus, we can't short-circuit "infinity" or "nan(...)".
207 */
208 commit = buf - 1;
209 c = WEOF;
210 for (p = buf; p < end; ) {
211 if ((c = __fgetwc_unlock(fp)) == WEOF)
212 break;
213 reswitch:
214 switch (state) {
215 case S_START:
216 state = S_GOTSIGN;
217 if (c == '-' || c == '+')
218 break;
219 else
220 goto reswitch;
221 case S_GOTSIGN:
222 switch (c) {
223 case '0':
224 state = S_MAYBEHEX;
225 commit = p;
226 break;
227 case 'I':
228 case 'i':
229 state = S_INF;
230 break;
231 case 'N':
232 case 'n':
233 state = S_NAN;
234 break;
235 default:
236 state = S_DIGITS;
237 goto reswitch;
238 }
239 break;
240 case S_INF:
241 if (infnanpos > 6 ||
242 (c != (wint_t)"nfinity"[infnanpos] &&
243 c != (wint_t)"NFINITY"[infnanpos]))
244 goto parsedone;
245 if (infnanpos == 1 || infnanpos == 6)
246 commit = p; /* inf or infinity */
247 infnanpos++;
248 break;
249 case S_NAN:
250 switch (infnanpos) {
251 case -1: /* XXX kludge to deal with nan(...) */
252 goto parsedone;
253 case 0:
254 if (c != 'A' && c != 'a')
255 goto parsedone;
256 break;
257 case 1:
258 if (c != 'N' && c != 'n')
259 goto parsedone;
260 else
261 commit = p;
262 break;
263 case 2:
264 if (c != '(')
265 goto parsedone;
266 break;
267 default:
268 if (c == ')') {
269 commit = p;
270 infnanpos = -2;
271 } else if (!iswalnum(c) && c != '_')
272 goto parsedone;
273 break;
274 }
275 infnanpos++;
276 break;
277 case S_MAYBEHEX:
278 state = S_DIGITS;
279 if (c == 'X' || c == 'x') {
280 ishex = 1;
281 break;
282 } else { /* we saw a '0', but no 'x' */
283 gotmantdig = 1;
284 goto reswitch;
285 }
286 case S_DIGITS:
287 if ((ishex && iswxdigit(c)) || iswdigit(c))
288 gotmantdig = 1;
289 else {
290 state = S_FRAC;
291 if (c != L'.')
292 goto reswitch;
293 }
294 if (gotmantdig)
295 commit = p;
296 break;
297 case S_FRAC:
298 if (((c == 'E' || c == 'e') && !ishex) ||
299 ((c == 'P' || c == 'p') && ishex)) {
300 if (!gotmantdig)
301 goto parsedone;
302 else
303 state = S_EXP;
304 } else if ((ishex && iswxdigit(c)) || iswdigit(c)) {
305 commit = p;
306 gotmantdig = 1;
307 } else
308 goto parsedone;
309 break;
310 case S_EXP:
311 state = S_EXPDIGITS;
312 if (c == '-' || c == '+')
313 break;
314 else
315 goto reswitch;
316 case S_EXPDIGITS:
317 if (iswdigit(c))
318 commit = p;
319 else
320 goto parsedone;
321 break;
322 default:
323 abort();
324 }
325 *p++ = c;
326 c = WEOF;
327 }
328
329 parsedone:
330 if (c != WEOF)
331 ungetwc(c, fp);
332 while (commit < --p)
333 ungetwc(*p, fp);
334 *++commit = '\0';
335 return (int)(commit - buf);
336 }
337