MySQL 5.6.14 Source Code Document
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
unvis.c
1 /* $NetBSD: unvis.c,v 1.36 2011/03/18 09:07:20 martin Exp $ */
2 
3 /*-
4  * Copyright (c) 1989, 1993
5  * The Regents of the University of California. All rights reserved.
6  *
7  * Redistribution and use in source and binary forms, with or without
8  * modification, are permitted provided that the following conditions
9  * are met:
10  * 1. Redistributions of source code must retain the above copyright
11  * notice, this list of conditions and the following disclaimer.
12  * 2. Redistributions in binary form must reproduce the above copyright
13  * notice, this list of conditions and the following disclaimer in the
14  * documentation and/or other materials provided with the distribution.
15  * 3. Neither the name of the University nor the names of its contributors
16  * may be used to endorse or promote products derived from this software
17  * without specific prior written permission.
18  *
19  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22  * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29  * SUCH DAMAGE.
30  */
31 
32 #include "config.h"
33 
34 #if defined(LIBC_SCCS) && !defined(lint)
35 #if 0
36 static char sccsid[] = "@(#)unvis.c 8.1 (Berkeley) 6/4/93";
37 #else
38 #endif
39 #endif /* LIBC_SCCS and not lint */
40 
41 /* XXXMYSQL : Make compiler happy. */
42 #ifdef _LIBC
43 #include "namespace.h"
44 #endif
45 
46 #include <sys/types.h>
47 
48 #include <assert.h>
49 #include <ctype.h>
50 
51 /* XXXMYSQL : stdint.h might not be available on older Solaris platforms. */
52 #if defined(__sun) || defined(__sun__)
53 #include <sys/inttypes.h>
54 #else
55 #include <stdint.h>
56 #endif
57 
58 #include <stdio.h>
59 #include <errno.h>
60 /*
61  XXXMYSQL : Due to different versions of vis.h available,
62  use the one bundled with libedit.
63 */
64 #include "np/vis.h"
65 
66 #ifdef __weak_alias
67 __weak_alias(strnunvisx,_strnunvisx)
68 #endif
69 
70 #if !HAVE_VIS
71 /*
72  * decode driven by state machine
73  */
74 #define S_GROUND 0 /* haven't seen escape char */
75 #define S_START 1 /* start decoding special sequence */
76 #define S_META 2 /* metachar started (M) */
77 #define S_META1 3 /* metachar more, regular char (-) */
78 #define S_CTRL 4 /* control char started (^) */
79 #define S_OCTAL2 5 /* octal digit 2 */
80 #define S_OCTAL3 6 /* octal digit 3 */
81 #define S_HEX1 7 /* http hex digit */
82 #define S_HEX2 8 /* http hex digit 2 */
83 #define S_MIME1 9 /* mime hex digit 1 */
84 #define S_MIME2 10 /* mime hex digit 2 */
85 #define S_EATCRNL 11 /* mime eating CRNL */
86 #define S_AMP 12 /* seen & */
87 #define S_NUMBER 13 /* collecting number */
88 #define S_STRING 14 /* collecting string */
89 
90 #ifndef isoctal
91 #define isoctal(c) (((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
92 #endif
93 #define xtod(c) (isdigit(c) ? (c - '0') : ((tolower(c) - 'a') + 10))
94 #define XTOD(c) (isdigit(c) ? (c - '0') : ((c - 'A') + 10))
95 
96 /*
97  * RFC 1866
98  */
99 static const struct nv {
100  const char *name;
101  uint8_t value;
102 } nv[] = {
103  { "AElig", 198 }, /* capital AE diphthong (ligature) */
104  { "Aacute", 193 }, /* capital A, acute accent */
105  { "Acirc", 194 }, /* capital A, circumflex accent */
106  { "Agrave", 192 }, /* capital A, grave accent */
107  { "Aring", 197 }, /* capital A, ring */
108  { "Atilde", 195 }, /* capital A, tilde */
109  { "Auml", 196 }, /* capital A, dieresis or umlaut mark */
110  { "Ccedil", 199 }, /* capital C, cedilla */
111  { "ETH", 208 }, /* capital Eth, Icelandic */
112  { "Eacute", 201 }, /* capital E, acute accent */
113  { "Ecirc", 202 }, /* capital E, circumflex accent */
114  { "Egrave", 200 }, /* capital E, grave accent */
115  { "Euml", 203 }, /* capital E, dieresis or umlaut mark */
116  { "Iacute", 205 }, /* capital I, acute accent */
117  { "Icirc", 206 }, /* capital I, circumflex accent */
118  { "Igrave", 204 }, /* capital I, grave accent */
119  { "Iuml", 207 }, /* capital I, dieresis or umlaut mark */
120  { "Ntilde", 209 }, /* capital N, tilde */
121  { "Oacute", 211 }, /* capital O, acute accent */
122  { "Ocirc", 212 }, /* capital O, circumflex accent */
123  { "Ograve", 210 }, /* capital O, grave accent */
124  { "Oslash", 216 }, /* capital O, slash */
125  { "Otilde", 213 }, /* capital O, tilde */
126  { "Ouml", 214 }, /* capital O, dieresis or umlaut mark */
127  { "THORN", 222 }, /* capital THORN, Icelandic */
128  { "Uacute", 218 }, /* capital U, acute accent */
129  { "Ucirc", 219 }, /* capital U, circumflex accent */
130  { "Ugrave", 217 }, /* capital U, grave accent */
131  { "Uuml", 220 }, /* capital U, dieresis or umlaut mark */
132  { "Yacute", 221 }, /* capital Y, acute accent */
133  { "aacute", 225 }, /* small a, acute accent */
134  { "acirc", 226 }, /* small a, circumflex accent */
135  { "acute", 180 }, /* acute accent */
136  { "aelig", 230 }, /* small ae diphthong (ligature) */
137  { "agrave", 224 }, /* small a, grave accent */
138  { "amp", 38 }, /* ampersand */
139  { "aring", 229 }, /* small a, ring */
140  { "atilde", 227 }, /* small a, tilde */
141  { "auml", 228 }, /* small a, dieresis or umlaut mark */
142  { "brvbar", 166 }, /* broken (vertical) bar */
143  { "ccedil", 231 }, /* small c, cedilla */
144  { "cedil", 184 }, /* cedilla */
145  { "cent", 162 }, /* cent sign */
146  { "copy", 169 }, /* copyright sign */
147  { "curren", 164 }, /* general currency sign */
148  { "deg", 176 }, /* degree sign */
149  { "divide", 247 }, /* divide sign */
150  { "eacute", 233 }, /* small e, acute accent */
151  { "ecirc", 234 }, /* small e, circumflex accent */
152  { "egrave", 232 }, /* small e, grave accent */
153  { "eth", 240 }, /* small eth, Icelandic */
154  { "euml", 235 }, /* small e, dieresis or umlaut mark */
155  { "frac12", 189 }, /* fraction one-half */
156  { "frac14", 188 }, /* fraction one-quarter */
157  { "frac34", 190 }, /* fraction three-quarters */
158  { "gt", 62 }, /* greater than */
159  { "iacute", 237 }, /* small i, acute accent */
160  { "icirc", 238 }, /* small i, circumflex accent */
161  { "iexcl", 161 }, /* inverted exclamation mark */
162  { "igrave", 236 }, /* small i, grave accent */
163  { "iquest", 191 }, /* inverted question mark */
164  { "iuml", 239 }, /* small i, dieresis or umlaut mark */
165  { "laquo", 171 }, /* angle quotation mark, left */
166  { "lt", 60 }, /* less than */
167  { "macr", 175 }, /* macron */
168  { "micro", 181 }, /* micro sign */
169  { "middot", 183 }, /* middle dot */
170  { "nbsp", 160 }, /* no-break space */
171  { "not", 172 }, /* not sign */
172  { "ntilde", 241 }, /* small n, tilde */
173  { "oacute", 243 }, /* small o, acute accent */
174  { "ocirc", 244 }, /* small o, circumflex accent */
175  { "ograve", 242 }, /* small o, grave accent */
176  { "ordf", 170 }, /* ordinal indicator, feminine */
177  { "ordm", 186 }, /* ordinal indicator, masculine */
178  { "oslash", 248 }, /* small o, slash */
179  { "otilde", 245 }, /* small o, tilde */
180  { "ouml", 246 }, /* small o, dieresis or umlaut mark */
181  { "para", 182 }, /* pilcrow (paragraph sign) */
182  { "plusmn", 177 }, /* plus-or-minus sign */
183  { "pound", 163 }, /* pound sterling sign */
184  { "quot", 34 }, /* double quote */
185  { "raquo", 187 }, /* angle quotation mark, right */
186  { "reg", 174 }, /* registered sign */
187  { "sect", 167 }, /* section sign */
188  { "shy", 173 }, /* soft hyphen */
189  { "sup1", 185 }, /* superscript one */
190  { "sup2", 178 }, /* superscript two */
191  { "sup3", 179 }, /* superscript three */
192  { "szlig", 223 }, /* small sharp s, German (sz ligature) */
193  { "thorn", 254 }, /* small thorn, Icelandic */
194  { "times", 215 }, /* multiply sign */
195  { "uacute", 250 }, /* small u, acute accent */
196  { "ucirc", 251 }, /* small u, circumflex accent */
197  { "ugrave", 249 }, /* small u, grave accent */
198  { "uml", 168 }, /* umlaut (dieresis) */
199  { "uuml", 252 }, /* small u, dieresis or umlaut mark */
200  { "yacute", 253 }, /* small y, acute accent */
201  { "yen", 165 }, /* yen sign */
202  { "yuml", 255 }, /* small y, dieresis or umlaut mark */
203 };
204 
205 /*
206  * unvis - decode characters previously encoded by vis
207  */
208 int
209 unvis(char *cp, int c, int *astate, int flag)
210 {
211  unsigned char uc = (unsigned char)c;
212  unsigned char st, ia, is, lc;
213 
214 /*
215  * Bottom 8 bits of astate hold the state machine state.
216  * Top 8 bits hold the current character in the http 1866 nv string decoding
217  */
218 #define GS(a) ((a) & 0xff)
219 #define SS(a, b) (((uint32_t)(a) << 24) | (b))
220 #define GI(a) ((uint32_t)(a) >> 24)
221 
222  _DIAGASSERT(cp != NULL);
223  _DIAGASSERT(astate != NULL);
224  st = GS(*astate);
225 
226  if (flag & UNVIS_END) {
227  switch (st) {
228  case S_OCTAL2:
229  case S_OCTAL3:
230  case S_HEX2:
231  *astate = SS(0, S_GROUND);
232  return UNVIS_VALID;
233  case S_GROUND:
234  return UNVIS_NOCHAR;
235  default:
236  return UNVIS_SYNBAD;
237  }
238  }
239 
240  switch (st) {
241 
242  case S_GROUND:
243  *cp = 0;
244  if ((flag & VIS_NOESCAPE) == 0 && c == '\\') {
245  *astate = SS(0, S_START);
246  return UNVIS_NOCHAR;
247  }
248  if ((flag & VIS_HTTP1808) && c == '%') {
249  *astate = SS(0, S_HEX1);
250  return UNVIS_NOCHAR;
251  }
252  if ((flag & VIS_HTTP1866) && c == '&') {
253  *astate = SS(0, S_AMP);
254  return UNVIS_NOCHAR;
255  }
256  if ((flag & VIS_MIMESTYLE) && c == '=') {
257  *astate = SS(0, S_MIME1);
258  return UNVIS_NOCHAR;
259  }
260  *cp = c;
261  return UNVIS_VALID;
262 
263  case S_START:
264  switch(c) {
265  case '\\':
266  *cp = c;
267  *astate = SS(0, S_GROUND);
268  return UNVIS_VALID;
269  case '0': case '1': case '2': case '3':
270  case '4': case '5': case '6': case '7':
271  *cp = (c - '0');
272  *astate = SS(0, S_OCTAL2);
273  return UNVIS_NOCHAR;
274  case 'M':
275  *cp = (char)0200;
276  *astate = SS(0, S_META);
277  return UNVIS_NOCHAR;
278  case '^':
279  *astate = SS(0, S_CTRL);
280  return UNVIS_NOCHAR;
281  case 'n':
282  *cp = '\n';
283  *astate = SS(0, S_GROUND);
284  return UNVIS_VALID;
285  case 'r':
286  *cp = '\r';
287  *astate = SS(0, S_GROUND);
288  return UNVIS_VALID;
289  case 'b':
290  *cp = '\b';
291  *astate = SS(0, S_GROUND);
292  return UNVIS_VALID;
293  case 'a':
294  *cp = '\007';
295  *astate = SS(0, S_GROUND);
296  return UNVIS_VALID;
297  case 'v':
298  *cp = '\v';
299  *astate = SS(0, S_GROUND);
300  return UNVIS_VALID;
301  case 't':
302  *cp = '\t';
303  *astate = SS(0, S_GROUND);
304  return UNVIS_VALID;
305  case 'f':
306  *cp = '\f';
307  *astate = SS(0, S_GROUND);
308  return UNVIS_VALID;
309  case 's':
310  *cp = ' ';
311  *astate = SS(0, S_GROUND);
312  return UNVIS_VALID;
313  case 'E':
314  *cp = '\033';
315  *astate = SS(0, S_GROUND);
316  return UNVIS_VALID;
317  case '\n':
318  /*
319  * hidden newline
320  */
321  *astate = SS(0, S_GROUND);
322  return UNVIS_NOCHAR;
323  case '$':
324  /*
325  * hidden marker
326  */
327  *astate = SS(0, S_GROUND);
328  return UNVIS_NOCHAR;
329  }
330  goto bad;
331 
332  case S_META:
333  if (c == '-')
334  *astate = SS(0, S_META1);
335  else if (c == '^')
336  *astate = SS(0, S_CTRL);
337  else
338  goto bad;
339  return UNVIS_NOCHAR;
340 
341  case S_META1:
342  *astate = SS(0, S_GROUND);
343  *cp |= c;
344  return UNVIS_VALID;
345 
346  case S_CTRL:
347  if (c == '?')
348  *cp |= 0177;
349  else
350  *cp |= c & 037;
351  *astate = SS(0, S_GROUND);
352  return UNVIS_VALID;
353 
354  case S_OCTAL2: /* second possible octal digit */
355  if (isoctal(uc)) {
356  /*
357  * yes - and maybe a third
358  */
359  *cp = (*cp << 3) + (c - '0');
360  *astate = SS(0, S_OCTAL3);
361  return UNVIS_NOCHAR;
362  }
363  /*
364  * no - done with current sequence, push back passed char
365  */
366  *astate = SS(0, S_GROUND);
367  return UNVIS_VALIDPUSH;
368 
369  case S_OCTAL3: /* third possible octal digit */
370  *astate = SS(0, S_GROUND);
371  if (isoctal(uc)) {
372  *cp = (*cp << 3) + (c - '0');
373  return UNVIS_VALID;
374  }
375  /*
376  * we were done, push back passed char
377  */
378  return UNVIS_VALIDPUSH;
379 
380  case S_HEX1:
381  if (isxdigit(uc)) {
382  *cp = xtod(uc);
383  *astate = SS(0, S_HEX2);
384  return UNVIS_NOCHAR;
385  }
386  /*
387  * no - done with current sequence, push back passed char
388  */
389  *astate = SS(0, S_GROUND);
390  return UNVIS_VALIDPUSH;
391 
392  case S_HEX2:
393  *astate = S_GROUND;
394  if (isxdigit(uc)) {
395  *cp = xtod(uc) | (*cp << 4);
396  return UNVIS_VALID;
397  }
398  return UNVIS_VALIDPUSH;
399 
400  case S_MIME1:
401  if (uc == '\n' || uc == '\r') {
402  *astate = SS(0, S_EATCRNL);
403  return UNVIS_NOCHAR;
404  }
405  if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
406  *cp = XTOD(uc);
407  *astate = SS(0, S_MIME2);
408  return UNVIS_NOCHAR;
409  }
410  goto bad;
411 
412  case S_MIME2:
413  if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
414  *astate = SS(0, S_GROUND);
415  *cp = XTOD(uc) | (*cp << 4);
416  return UNVIS_VALID;
417  }
418  goto bad;
419 
420  case S_EATCRNL:
421  switch (uc) {
422  case '\r':
423  case '\n':
424  return UNVIS_NOCHAR;
425  case '=':
426  *astate = SS(0, S_MIME1);
427  return UNVIS_NOCHAR;
428  default:
429  *cp = uc;
430  *astate = SS(0, S_GROUND);
431  return UNVIS_VALID;
432  }
433 
434  case S_AMP:
435  *cp = 0;
436  if (uc == '#') {
437  *astate = SS(0, S_NUMBER);
438  return UNVIS_NOCHAR;
439  }
440  *astate = SS(0, S_STRING);
441  /*FALLTHROUGH*/
442 
443  case S_STRING:
444  ia = *cp; /* index in the array */
445  is = GI(*astate); /* index in the string */
446  lc = is == 0 ? 0 : nv[ia].name[is - 1]; /* last character */
447 
448  if (uc == ';')
449  uc = '\0';
450 
451  for (; ia < __arraycount(nv); ia++) {
452  if (is != 0 && nv[ia].name[is - 1] != lc)
453  goto bad;
454  if (nv[ia].name[is] == uc)
455  break;
456  }
457 
458  if (ia == __arraycount(nv))
459  goto bad;
460 
461  if (uc != 0) {
462  *cp = ia;
463  *astate = SS(is + 1, S_STRING);
464  return UNVIS_NOCHAR;
465  }
466 
467  *cp = nv[ia].value;
468  *astate = SS(0, S_GROUND);
469  return UNVIS_VALID;
470 
471  case S_NUMBER:
472  if (uc == ';')
473  return UNVIS_VALID;
474  if (!isdigit(uc))
475  goto bad;
476  *cp += (*cp * 10) + uc - '0';
477  return UNVIS_NOCHAR;
478 
479  default:
480  bad:
481  /*
482  * decoder in unknown state - (probably uninitialized)
483  */
484  *astate = SS(0, S_GROUND);
485  return UNVIS_SYNBAD;
486  }
487 }
488 
489 /*
490  * strnunvisx - decode src into dst
491  *
492  * Number of chars decoded into dst is returned, -1 on error.
493  * Dst is null terminated.
494  */
495 
496 int
497 strnunvisx(char *dst, size_t dlen, const char *src, int flag)
498 {
499  char c;
500  char t, *start = dst;
501  int state = 0;
502 
503  _DIAGASSERT(src != NULL);
504  _DIAGASSERT(dst != NULL);
505 #define CHECKSPACE() \
506  do { \
507  if (dlen-- == 0) { \
508  errno = ENOSPC; \
509  return -1; \
510  } \
511  } while (/*CONSTCOND*/0)
512 
513  while ((c = *src++) != '\0') {
514  again:
515  switch (unvis(&t, c, &state, flag)) {
516  case UNVIS_VALID:
517  CHECKSPACE();
518  *dst++ = t;
519  break;
520  case UNVIS_VALIDPUSH:
521  CHECKSPACE();
522  *dst++ = t;
523  goto again;
524  case 0:
525  case UNVIS_NOCHAR:
526  break;
527  case UNVIS_SYNBAD:
528  errno = EINVAL;
529  return -1;
530  default:
531  _DIAGASSERT(0);
532  errno = EINVAL;
533  return -1;
534  }
535  }
536  if (unvis(&t, c, &state, UNVIS_END) == UNVIS_VALID) {
537  CHECKSPACE();
538  *dst++ = t;
539  }
540  CHECKSPACE();
541  *dst = '\0';
542  return (int)(dst - start);
543 }
544 
545 int
546 strunvisx(char *dst, const char *src, int flag)
547 {
548  return strnunvisx(dst, (size_t)~0, src, flag);
549 }
550 
551 int
552 strunvis(char *dst, const char *src)
553 {
554  return strnunvisx(dst, (size_t)~0, src, 0);
555 }
556 
557 int
558 strnunvis(char *dst, size_t dlen, const char *src)
559 {
560  return strnunvisx(dst, dlen, src, 0);
561 }
562 #endif