| 1 |
/*************************************************
|
| 2 |
* Perl-Compatible Regular Expressions *
|
| 3 |
*************************************************/
|
| 4 |
|
| 5 |
/*
|
| 6 |
This is a library of functions to support regular expressions whose syntax
|
| 7 |
and semantics are as close as possible to those of the Perl 5 language. See
|
| 8 |
the file Tech.Notes for some information on the internals.
|
| 9 |
|
| 10 |
Written by: Philip Hazel <ph10@cam.ac.uk>
|
| 11 |
|
| 12 |
Copyright (c) 1997-2004 University of Cambridge
|
| 13 |
|
| 14 |
-----------------------------------------------------------------------------
|
| 15 |
Redistribution and use in source and binary forms, with or without
|
| 16 |
modification, are permitted provided that the following conditions are met:
|
| 17 |
|
| 18 |
* Redistributions of source code must retain the above copyright notice,
|
| 19 |
this list of conditions and the following disclaimer.
|
| 20 |
|
| 21 |
* Redistributions in binary form must reproduce the above copyright
|
| 22 |
notice, this list of conditions and the following disclaimer in the
|
| 23 |
documentation and/or other materials provided with the distribution.
|
| 24 |
|
| 25 |
* Neither the name of the University of Cambridge nor the names of its
|
| 26 |
contributors may be used to endorse or promote products derived from
|
| 27 |
this software without specific prior written permission.
|
| 28 |
|
| 29 |
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
| 30 |
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
| 31 |
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
| 32 |
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
|
| 33 |
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
|
| 34 |
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
|
| 35 |
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
|
| 36 |
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
|
| 37 |
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
|
| 38 |
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
| 39 |
POSSIBILITY OF SUCH DAMAGE.
|
| 40 |
-----------------------------------------------------------------------------
|
| 41 |
*/
|
| 42 |
|
| 43 |
|
| 44 |
/* This module contains a debugging function for printing out the internal form
|
| 45 |
of a compiled regular expression. It is kept in a separate file so that it can
|
| 46 |
be #included both in the pcretest program, and in the library itself when
|
| 47 |
compiled with the debugging switch. */
|
| 48 |
|
| 49 |
|
| 50 |
static const char *OP_names[] = { OP_NAME_LIST };
|
| 51 |
|
| 52 |
|
| 53 |
/*************************************************
|
| 54 |
* Print single- or multi-byte character *
|
| 55 |
*************************************************/
|
| 56 |
|
| 57 |
/* These tables are actually copies of ones in pcre.c. If we compile the
|
| 58 |
library with debugging, they are included twice, but that isn't really a
|
| 59 |
problem - compiling with debugging is pretty rare and these are very small. */
|
| 60 |
|
| 61 |
static const int utf8_t3[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01};
|
| 62 |
|
| 63 |
static const uschar utf8_t4[] = {
|
| 64 |
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
|
| 65 |
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
|
| 66 |
2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
|
| 67 |
3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 };
|
| 68 |
|
| 69 |
static int
|
| 70 |
print_char(FILE *f, uschar *ptr, BOOL utf8)
|
| 71 |
{
|
| 72 |
int c = *ptr;
|
| 73 |
|
| 74 |
if (!utf8 || (c & 0xc0) != 0xc0)
|
| 75 |
{
|
| 76 |
if (isprint(c)) fprintf(f, "%c", c); else fprintf(f, "\\x%02x", c);
|
| 77 |
return 0;
|
| 78 |
}
|
| 79 |
else
|
| 80 |
{
|
| 81 |
int i;
|
| 82 |
int a = utf8_t4[c & 0x3f]; /* Number of additional bytes */
|
| 83 |
int s = 6*a;
|
| 84 |
c = (c & utf8_t3[a]) << s;
|
| 85 |
for (i = 1; i <= a; i++)
|
| 86 |
{
|
| 87 |
/* This is a check for malformed UTF-8; it should only occur if the sanity
|
| 88 |
check has been turned off. Rather than swallow random bytes, just stop if
|
| 89 |
we hit a bad one. Print it with \X instead of \x as an indication. */
|
| 90 |
|
| 91 |
if ((ptr[i] & 0xc0) != 0x80)
|
| 92 |
{
|
| 93 |
fprintf(f, "\\X{%x}", c);
|
| 94 |
return i - 1;
|
| 95 |
}
|
| 96 |
|
| 97 |
/* The byte is OK */
|
| 98 |
|
| 99 |
s -= 6;
|
| 100 |
c |= (ptr[i] & 0x3f) << s;
|
| 101 |
}
|
| 102 |
if (c < 128) fprintf(f, "\\x%02x", c); else fprintf(f, "\\x{%x}", c);
|
| 103 |
return a;
|
| 104 |
}
|
| 105 |
}
|
| 106 |
|
| 107 |
|
| 108 |
|
| 109 |
|
| 110 |
/*************************************************
|
| 111 |
* Find Unicode property name *
|
| 112 |
*************************************************/
|
| 113 |
|
| 114 |
static const char *
|
| 115 |
get_ucpname(int property)
|
| 116 |
{
|
| 117 |
int i;
|
| 118 |
for (i = sizeof(utt)/sizeof(ucp_type_table); i >= 0; i--)
|
| 119 |
{
|
| 120 |
if (property == utt[i].value) break;
|
| 121 |
}
|
| 122 |
return (i >= 0)? utt[i].name : "??";
|
| 123 |
}
|
| 124 |
|
| 125 |
|
| 126 |
|
| 127 |
/*************************************************
|
| 128 |
* Print compiled regex *
|
| 129 |
*************************************************/
|
| 130 |
|
| 131 |
/* Make this function work for a regex with integers either byte order.
|
| 132 |
However, we assume that what we are passed is a compiled regex. */
|
| 133 |
|
| 134 |
static void
|
| 135 |
print_internals(pcre *external_re, FILE *f)
|
| 136 |
{
|
| 137 |
real_pcre *re = (real_pcre *)external_re;
|
| 138 |
uschar *codestart, *code;
|
| 139 |
BOOL utf8;
|
| 140 |
|
| 141 |
unsigned int options = re->options;
|
| 142 |
int offset = re->name_table_offset;
|
| 143 |
int count = re->name_count;
|
| 144 |
int size = re->name_entry_size;
|
| 145 |
|
| 146 |
if (re->magic_number != MAGIC_NUMBER)
|
| 147 |
{
|
| 148 |
offset = ((offset << 8) & 0xff00) | ((offset >> 8) & 0xff);
|
| 149 |
count = ((count << 8) & 0xff00) | ((count >> 8) & 0xff);
|
| 150 |
size = ((size << 8) & 0xff00) | ((size >> 8) & 0xff);
|
| 151 |
options = ((options << 24) & 0xff000000) |
|
| 152 |
((options << 8) & 0x00ff0000) |
|
| 153 |
((options >> 8) & 0x0000ff00) |
|
| 154 |
((options >> 24) & 0x000000ff);
|
| 155 |
}
|
| 156 |
|
| 157 |
code = codestart = (uschar *)re + offset + count * size;
|
| 158 |
utf8 = (options & PCRE_UTF8) != 0;
|
| 159 |
|
| 160 |
for(;;)
|
| 161 |
{
|
| 162 |
uschar *ccode;
|
| 163 |
int c;
|
| 164 |
int extra = 0;
|
| 165 |
|
| 166 |
fprintf(f, "%3d ", code - codestart);
|
| 167 |
|
| 168 |
if (*code >= OP_BRA)
|
| 169 |
{
|
| 170 |
if (*code - OP_BRA > EXTRACT_BASIC_MAX)
|
| 171 |
fprintf(f, "%3d Bra extra\n", GET(code, 1));
|
| 172 |
else
|
| 173 |
fprintf(f, "%3d Bra %d\n", GET(code, 1), *code - OP_BRA);
|
| 174 |
code += OP_lengths[OP_BRA];
|
| 175 |
continue;
|
| 176 |
}
|
| 177 |
|
| 178 |
switch(*code)
|
| 179 |
{
|
| 180 |
case OP_END:
|
| 181 |
fprintf(f, " %s\n", OP_names[*code]);
|
| 182 |
fprintf(f, "------------------------------------------------------------------\n");
|
| 183 |
return;
|
| 184 |
|
| 185 |
case OP_OPT:
|
| 186 |
fprintf(f, " %.2x %s", code[1], OP_names[*code]);
|
| 187 |
break;
|
| 188 |
|
| 189 |
case OP_CHAR:
|
| 190 |
{
|
| 191 |
fprintf(f, " ");
|
| 192 |
do
|
| 193 |
{
|
| 194 |
code++;
|
| 195 |
code += 1 + print_char(f, code, utf8);
|
| 196 |
}
|
| 197 |
while (*code == OP_CHAR);
|
| 198 |
fprintf(f, "\n");
|
| 199 |
continue;
|
| 200 |
}
|
| 201 |
break;
|
| 202 |
|
| 203 |
case OP_CHARNC:
|
| 204 |
{
|
| 205 |
fprintf(f, " NC ");
|
| 206 |
do
|
| 207 |
{
|
| 208 |
code++;
|
| 209 |
code += 1 + print_char(f, code, utf8);
|
| 210 |
}
|
| 211 |
while (*code == OP_CHARNC);
|
| 212 |
fprintf(f, "\n");
|
| 213 |
continue;
|
| 214 |
}
|
| 215 |
break;
|
| 216 |
|
| 217 |
case OP_KETRMAX:
|
| 218 |
case OP_KETRMIN:
|
| 219 |
case OP_ALT:
|
| 220 |
case OP_KET:
|
| 221 |
case OP_ASSERT:
|
| 222 |
case OP_ASSERT_NOT:
|
| 223 |
case OP_ASSERTBACK:
|
| 224 |
case OP_ASSERTBACK_NOT:
|
| 225 |
case OP_ONCE:
|
| 226 |
case OP_COND:
|
| 227 |
case OP_REVERSE:
|
| 228 |
fprintf(f, "%3d %s", GET(code, 1), OP_names[*code]);
|
| 229 |
break;
|
| 230 |
|
| 231 |
case OP_BRANUMBER:
|
| 232 |
printf("%3d %s", GET2(code, 1), OP_names[*code]);
|
| 233 |
break;
|
| 234 |
|
| 235 |
case OP_CREF:
|
| 236 |
if (GET2(code, 1) == CREF_RECURSE)
|
| 237 |
fprintf(f, " Cond recurse");
|
| 238 |
else
|
| 239 |
fprintf(f, "%3d %s", GET2(code,1), OP_names[*code]);
|
| 240 |
break;
|
| 241 |
|
| 242 |
case OP_STAR:
|
| 243 |
case OP_MINSTAR:
|
| 244 |
case OP_PLUS:
|
| 245 |
case OP_MINPLUS:
|
| 246 |
case OP_QUERY:
|
| 247 |
case OP_MINQUERY:
|
| 248 |
case OP_TYPESTAR:
|
| 249 |
case OP_TYPEMINSTAR:
|
| 250 |
case OP_TYPEPLUS:
|
| 251 |
case OP_TYPEMINPLUS:
|
| 252 |
case OP_TYPEQUERY:
|
| 253 |
case OP_TYPEMINQUERY:
|
| 254 |
fprintf(f, " ");
|
| 255 |
if (*code >= OP_TYPESTAR)
|
| 256 |
{
|
| 257 |
fprintf(f, "%s", OP_names[code[1]]);
|
| 258 |
if (code[1] == OP_PROP || code[1] == OP_NOTPROP)
|
| 259 |
{
|
| 260 |
fprintf(f, " %s ", get_ucpname(code[2]));
|
| 261 |
extra = 1;
|
| 262 |
}
|
| 263 |
}
|
| 264 |
else extra = print_char(f, code+1, utf8);
|
| 265 |
fprintf(f, "%s", OP_names[*code]);
|
| 266 |
break;
|
| 267 |
|
| 268 |
case OP_EXACT:
|
| 269 |
case OP_UPTO:
|
| 270 |
case OP_MINUPTO:
|
| 271 |
fprintf(f, " ");
|
| 272 |
extra = print_char(f, code+3, utf8);
|
| 273 |
fprintf(f, "{");
|
| 274 |
if (*code != OP_EXACT) fprintf(f, ",");
|
| 275 |
fprintf(f, "%d}", GET2(code,1));
|
| 276 |
if (*code == OP_MINUPTO) fprintf(f, "?");
|
| 277 |
break;
|
| 278 |
|
| 279 |
case OP_TYPEEXACT:
|
| 280 |
case OP_TYPEUPTO:
|
| 281 |
case OP_TYPEMINUPTO:
|
| 282 |
fprintf(f, " %s", OP_names[code[3]]);
|
| 283 |
if (code[3] == OP_PROP || code[3] == OP_NOTPROP)
|
| 284 |
{
|
| 285 |
fprintf(f, " %s ", get_ucpname(code[4]));
|
| 286 |
extra = 1;
|
| 287 |
}
|
| 288 |
fprintf(f, "{");
|
| 289 |
if (*code != OP_TYPEEXACT) fprintf(f, "0,");
|
| 290 |
fprintf(f, "%d}", GET2(code,1));
|
| 291 |
if (*code == OP_TYPEMINUPTO) fprintf(f, "?");
|
| 292 |
break;
|
| 293 |
|
| 294 |
case OP_NOT:
|
| 295 |
if (isprint(c = code[1])) fprintf(f, " [^%c]", c);
|
| 296 |
else fprintf(f, " [^\\x%02x]", c);
|
| 297 |
break;
|
| 298 |
|
| 299 |
case OP_NOTSTAR:
|
| 300 |
case OP_NOTMINSTAR:
|
| 301 |
case OP_NOTPLUS:
|
| 302 |
case OP_NOTMINPLUS:
|
| 303 |
case OP_NOTQUERY:
|
| 304 |
case OP_NOTMINQUERY:
|
| 305 |
if (isprint(c = code[1])) fprintf(f, " [^%c]", c);
|
| 306 |
else fprintf(f, " [^\\x%02x]", c);
|
| 307 |
fprintf(f, "%s", OP_names[*code]);
|
| 308 |
break;
|
| 309 |
|
| 310 |
case OP_NOTEXACT:
|
| 311 |
case OP_NOTUPTO:
|
| 312 |
case OP_NOTMINUPTO:
|
| 313 |
if (isprint(c = code[3])) fprintf(f, " [^%c]{", c);
|
| 314 |
else fprintf(f, " [^\\x%02x]{", c);
|
| 315 |
if (*code != OP_NOTEXACT) fprintf(f, ",");
|
| 316 |
fprintf(f, "%d}", GET2(code,1));
|
| 317 |
if (*code == OP_NOTMINUPTO) fprintf(f, "?");
|
| 318 |
break;
|
| 319 |
|
| 320 |
case OP_RECURSE:
|
| 321 |
fprintf(f, "%3d %s", GET(code, 1), OP_names[*code]);
|
| 322 |
break;
|
| 323 |
|
| 324 |
case OP_REF:
|
| 325 |
fprintf(f, " \\%d", GET2(code,1));
|
| 326 |
ccode = code + OP_lengths[*code];
|
| 327 |
goto CLASS_REF_REPEAT;
|
| 328 |
|
| 329 |
case OP_CALLOUT:
|
| 330 |
fprintf(f, " %s %d %d %d", OP_names[*code], code[1], GET(code,2),
|
| 331 |
GET(code, 2 + LINK_SIZE));
|
| 332 |
break;
|
| 333 |
|
| 334 |
case OP_PROP:
|
| 335 |
case OP_NOTPROP:
|
| 336 |
fprintf(f, " %s %s", OP_names[*code], get_ucpname(code[1]));
|
| 337 |
break;
|
| 338 |
|
| 339 |
/* OP_XCLASS can only occur in UTF-8 mode. However, there's no harm in
|
| 340 |
having this code always here, and it makes it less messy without all those
|
| 341 |
#ifdefs. */
|
| 342 |
|
| 343 |
case OP_CLASS:
|
| 344 |
case OP_NCLASS:
|
| 345 |
case OP_XCLASS:
|
| 346 |
{
|
| 347 |
int i, min, max;
|
| 348 |
BOOL printmap;
|
| 349 |
|
| 350 |
fprintf(f, " [");
|
| 351 |
|
| 352 |
if (*code == OP_XCLASS)
|
| 353 |
{
|
| 354 |
extra = GET(code, 1);
|
| 355 |
ccode = code + LINK_SIZE + 1;
|
| 356 |
printmap = (*ccode & XCL_MAP) != 0;
|
| 357 |
if ((*ccode++ & XCL_NOT) != 0) fprintf(f, "^");
|
| 358 |
}
|
| 359 |
else
|
| 360 |
{
|
| 361 |
printmap = TRUE;
|
| 362 |
ccode = code + 1;
|
| 363 |
}
|
| 364 |
|
| 365 |
/* Print a bit map */
|
| 366 |
|
| 367 |
if (printmap)
|
| 368 |
{
|
| 369 |
for (i = 0; i < 256; i++)
|
| 370 |
{
|
| 371 |
if ((ccode[i/8] & (1 << (i&7))) != 0)
|
| 372 |
{
|
| 373 |
int j;
|
| 374 |
for (j = i+1; j < 256; j++)
|
| 375 |
if ((ccode[j/8] & (1 << (j&7))) == 0) break;
|
| 376 |
if (i == '-' || i == ']') fprintf(f, "\\");
|
| 377 |
if (isprint(i)) fprintf(f, "%c", i); else fprintf(f, "\\x%02x", i);
|
| 378 |
if (--j > i)
|
| 379 |
{
|
| 380 |
if (j != i + 1) fprintf(f, "-");
|
| 381 |
if (j == '-' || j == ']') fprintf(f, "\\");
|
| 382 |
if (isprint(j)) fprintf(f, "%c", j); else fprintf(f, "\\x%02x", j);
|
| 383 |
}
|
| 384 |
i = j;
|
| 385 |
}
|
| 386 |
}
|
| 387 |
ccode += 32;
|
| 388 |
}
|
| 389 |
|
| 390 |
/* For an XCLASS there is always some additional data */
|
| 391 |
|
| 392 |
if (*code == OP_XCLASS)
|
| 393 |
{
|
| 394 |
int ch;
|
| 395 |
while ((ch = *ccode++) != XCL_END)
|
| 396 |
{
|
| 397 |
if (ch == XCL_PROP)
|
| 398 |
{
|
| 399 |
fprintf(f, "\\p{%s}", get_ucpname(*ccode++));
|
| 400 |
}
|
| 401 |
else if (ch == XCL_NOTPROP)
|
| 402 |
{
|
| 403 |
fprintf(f, "\\P{%s}", get_ucpname(*ccode++));
|
| 404 |
}
|
| 405 |
else
|
| 406 |
{
|
| 407 |
ccode += 1 + print_char(f, ccode, TRUE);
|
| 408 |
if (ch == XCL_RANGE)
|
| 409 |
{
|
| 410 |
fprintf(f, "-");
|
| 411 |
ccode += 1 + print_char(f, ccode, TRUE);
|
| 412 |
}
|
| 413 |
}
|
| 414 |
}
|
| 415 |
}
|
| 416 |
|
| 417 |
/* Indicate a non-UTF8 class which was created by negation */
|
| 418 |
|
| 419 |
fprintf(f, "]%s", (*code == OP_NCLASS)? " (neg)" : "");
|
| 420 |
|
| 421 |
/* Handle repeats after a class or a back reference */
|
| 422 |
|
| 423 |
CLASS_REF_REPEAT:
|
| 424 |
switch(*ccode)
|
| 425 |
{
|
| 426 |
case OP_CRSTAR:
|
| 427 |
case OP_CRMINSTAR:
|
| 428 |
case OP_CRPLUS:
|
| 429 |
case OP_CRMINPLUS:
|
| 430 |
case OP_CRQUERY:
|
| 431 |
case OP_CRMINQUERY:
|
| 432 |
fprintf(f, "%s", OP_names[*ccode]);
|
| 433 |
extra += OP_lengths[*ccode];
|
| 434 |
break;
|
| 435 |
|
| 436 |
case OP_CRRANGE:
|
| 437 |
case OP_CRMINRANGE:
|
| 438 |
min = GET2(ccode,1);
|
| 439 |
max = GET2(ccode,3);
|
| 440 |
if (max == 0) fprintf(f, "{%d,}", min);
|
| 441 |
else fprintf(f, "{%d,%d}", min, max);
|
| 442 |
if (*ccode == OP_CRMINRANGE) fprintf(f, "?");
|
| 443 |
extra += OP_lengths[*ccode];
|
| 444 |
break;
|
| 445 |
}
|
| 446 |
}
|
| 447 |
break;
|
| 448 |
|
| 449 |
/* Anything else is just an item with no data*/
|
| 450 |
|
| 451 |
default:
|
| 452 |
fprintf(f, " %s", OP_names[*code]);
|
| 453 |
break;
|
| 454 |
}
|
| 455 |
|
| 456 |
code += OP_lengths[*code] + extra;
|
| 457 |
fprintf(f, "\n");
|
| 458 |
}
|
| 459 |
}
|
| 460 |
|
| 461 |
/* End of printint.c */
|
Parent Directory
| 
Revision Log