| 4 |
|
|
| 5 |
/* This program was hacked up as a tester for PCRE. I really should have |
/* This program was hacked up as a tester for PCRE. I really should have |
| 6 |
written it more tidily in the first place. Will I ever learn? It has grown and |
written it more tidily in the first place. Will I ever learn? It has grown and |
| 7 |
been extended and consequently is now rather, er, *very* untidy in places. |
been extended and consequently is now rather, er, *very* untidy in places. The |
| 8 |
|
addition of 16-bit support has made it even worse. :-( |
| 9 |
|
|
| 10 |
----------------------------------------------------------------------------- |
----------------------------------------------------------------------------- |
| 11 |
Redistribution and use in source and binary forms, with or without |
Redistribution and use in source and binary forms, with or without |
| 36 |
----------------------------------------------------------------------------- |
----------------------------------------------------------------------------- |
| 37 |
*/ |
*/ |
| 38 |
|
|
| 39 |
|
/* This program now supports the testing of both the 8-bit and 16-bit PCRE |
| 40 |
|
libraries in a single program. This is different from the modules such as |
| 41 |
|
pcre_compile.c in the library itself, which are compiled separately for each |
| 42 |
|
mode. If both modes are enabled, for example, pcre_compile.c is compiled twice |
| 43 |
|
(the second time with COMPILE_PCRE16 defined). By contrast, pcretest.c is |
| 44 |
|
compiled only once. Therefore, it must not make use of any of the macros from |
| 45 |
|
pcre_internal.h that depend on COMPILE_PCRE8 or COMPILE_PCRE16. It does, |
| 46 |
|
however, make use of SUPPORT_PCRE8 and SUPPORT_PCRE16 to ensure that it calls |
| 47 |
|
only supported library functions. */ |
| 48 |
|
|
| 49 |
|
|
| 50 |
#ifdef HAVE_CONFIG_H |
#ifdef HAVE_CONFIG_H |
| 51 |
#include "config.h" |
#include "config.h" |
| 116 |
appropriately for an application, not for building PCRE. */ |
appropriately for an application, not for building PCRE. */ |
| 117 |
|
|
| 118 |
#include "pcre.h" |
#include "pcre.h" |
| 119 |
|
|
| 120 |
|
#if defined SUPPORT_PCRE16 && !defined SUPPORT_PCRE8 |
| 121 |
|
/* Configure internal macros to 16 bit mode. */ |
| 122 |
|
#define COMPILE_PCRE16 |
| 123 |
|
#endif |
| 124 |
|
|
| 125 |
#include "pcre_internal.h" |
#include "pcre_internal.h" |
| 126 |
|
|
| 127 |
|
/* The pcre_printint() function, which prints the internal form of a compiled |
| 128 |
|
regex, is held in a separate file so that (a) it can be compiled in either |
| 129 |
|
8-bit or 16-bit mode, and (b) it can be #included directly in pcre_compile.c |
| 130 |
|
when that is compiled in debug mode. */ |
| 131 |
|
|
| 132 |
|
#ifdef SUPPORT_PCRE8 |
| 133 |
|
void pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths); |
| 134 |
|
#endif |
| 135 |
|
#ifdef SUPPORT_PCRE16 |
| 136 |
|
void pcre16_printint(pcre *external_re, FILE *f, BOOL print_lengths); |
| 137 |
|
#endif |
| 138 |
|
|
| 139 |
/* We need access to some of the data tables that PCRE uses. So as not to have |
/* We need access to some of the data tables that PCRE uses. So as not to have |
| 140 |
to keep two copies, we include the source file here, changing the names of the |
to keep two copies, we include the source file here, changing the names of the |
| 141 |
external symbols to prevent clashes. */ |
external symbols to prevent clashes. */ |
| 142 |
|
|
| 143 |
#define _pcre_ucp_gentype ucp_gentype |
#define PCRE_INCLUDED |
| 144 |
#define _pcre_ucp_typerange ucp_typerange |
#undef PRIV |
| 145 |
#define _pcre_utf8_table1 utf8_table1 |
#define PRIV(name) name |
|
#define _pcre_utf8_table1_size utf8_table1_size |
|
|
#define _pcre_utf8_table2 utf8_table2 |
|
|
#define _pcre_utf8_table3 utf8_table3 |
|
|
#define _pcre_utf8_table4 utf8_table4 |
|
|
#define _pcre_utf8_char_sizes utf8_char_sizes |
|
|
#define _pcre_utt utt |
|
|
#define _pcre_utt_size utt_size |
|
|
#define _pcre_utt_names utt_names |
|
|
#define _pcre_OP_lengths OP_lengths |
|
| 146 |
|
|
| 147 |
#include "pcre_tables.c" |
#include "pcre_tables.c" |
| 148 |
|
|
|
/* We also need the pcre_printint() function for printing out compiled |
|
|
patterns. This function is in a separate file so that it can be included in |
|
|
pcre_compile.c when that module is compiled with debugging enabled. It needs to |
|
|
know which case is being compiled. */ |
|
|
|
|
|
#define COMPILING_PCRETEST |
|
|
#include "pcre_printint.src" |
|
|
|
|
| 149 |
/* The definition of the macro PRINTABLE, which determines whether to print an |
/* The definition of the macro PRINTABLE, which determines whether to print an |
| 150 |
output character as-is or as a hex value when showing compiled patterns, is |
output character as-is or as a hex value when showing compiled patterns, is |
| 151 |
contained in the printint.src file. We uses it here also, in cases when the |
the same as in the printint.src file. We uses it here in cases when the locale |
| 152 |
locale has not been explicitly changed, so as to get consistent output from |
has not been explicitly changed, so as to get consistent output from systems |
| 153 |
systems that differ in their output from isprint() even in the "C" locale. */ |
that differ in their output from isprint() even in the "C" locale. */ |
| 154 |
|
|
| 155 |
|
#ifdef EBCDIC |
| 156 |
|
#define PRINTABLE(c) ((c) >= 64 && (c) < 255) |
| 157 |
|
#else |
| 158 |
|
#define PRINTABLE(c) ((c) >= 32 && (c) < 127) |
| 159 |
|
#endif |
| 160 |
|
|
| 161 |
|
#define PRINTOK(c) (locale_set? isprint(c) : PRINTABLE(c)) |
| 162 |
|
|
| 163 |
#define PRINTHEX(c) (locale_set? isprint(c) : PRINTABLE(c)) |
/* Posix support is disabled in 16 bit only mode. */ |
| 164 |
|
#if defined SUPPORT_PCRE16 && !defined SUPPORT_PCRE8 && !defined NOPOSIX |
| 165 |
|
#define NOPOSIX |
| 166 |
|
#endif |
| 167 |
|
|
| 168 |
/* It is possible to compile this test program without including support for |
/* It is possible to compile this test program without including support for |
| 169 |
testing the POSIX interface, though this is not available via the standard |
testing the POSIX interface, though this is not available via the standard |
| 173 |
#include "pcreposix.h" |
#include "pcreposix.h" |
| 174 |
#endif |
#endif |
| 175 |
|
|
| 176 |
/* It is also possible, for the benefit of the version currently imported into |
/* It is also possible, originally for the benefit of a version that was |
| 177 |
Exim, to build pcretest without support for UTF8 (define NOUTF8), without the |
imported into Exim, to build pcretest without support for UTF8 or UTF16 (define |
| 178 |
interface to the DFA matcher (NODFA), and without the doublecheck of the old |
NOUTF), without the interface to the DFA matcher (NODFA). In fact, we |
| 179 |
"info" function (define NOINFOCHECK). In fact, we automatically cut out the |
automatically cut out the UTF support if PCRE is built without it. */ |
| 180 |
UTF8 support if PCRE is built without it. */ |
|
| 181 |
|
#ifndef SUPPORT_UTF |
| 182 |
#ifndef SUPPORT_UTF8 |
#ifndef NOUTF |
| 183 |
#ifndef NOUTF8 |
#define NOUTF |
|
#define NOUTF8 |
|
| 184 |
#endif |
#endif |
| 185 |
#endif |
#endif |
| 186 |
|
|
| 187 |
|
/* To make the code a bit tidier for 8-bit and 16-bit support, we define macros |
| 188 |
|
for all the pcre[16]_xxx functions (except pcre16_fullinfo, which is called |
| 189 |
|
only from one place and is handled differently). I couldn't dream up any way of |
| 190 |
|
using a single macro to do this in a generic way, because of the many different |
| 191 |
|
argument requirements. We know that at least one of SUPPORT_PCRE8 and |
| 192 |
|
SUPPORT_PCRE16 must be set. First define macros for each individual mode; then |
| 193 |
|
use these in the definitions of generic macros. |
| 194 |
|
|
| 195 |
|
**** Special note about the PCHARSxxx macros: the address of the string to be |
| 196 |
|
printed is always given as two arguments: a base address followed by an offset. |
| 197 |
|
The base address is cast to the correct data size for 8 or 16 bit data; the |
| 198 |
|
offset is in units of this size. If the string were given as base+offset in one |
| 199 |
|
argument, the casting might be incorrectly applied. */ |
| 200 |
|
|
| 201 |
|
#ifdef SUPPORT_PCRE8 |
| 202 |
|
|
| 203 |
|
#define PCHARS8(lv, p, offset, len, f) \ |
| 204 |
|
lv = pchars((pcre_uint8 *)(p) + offset, len, f) |
| 205 |
|
|
| 206 |
|
#define PCHARSV8(p, offset, len, f) \ |
| 207 |
|
(void)pchars((pcre_uint8 *)(p) + offset, len, f) |
| 208 |
|
|
| 209 |
|
#define READ_CAPTURE_NAME8(p, cn8, cn16, re) \ |
| 210 |
|
p = read_capture_name8(p, cn8, re) |
| 211 |
|
|
| 212 |
|
#define SET_PCRE_CALLOUT8(callout) \ |
| 213 |
|
pcre_callout = callout |
| 214 |
|
|
| 215 |
|
#define STRLEN8(p) ((int)strlen((char *)p)) |
| 216 |
|
|
| 217 |
|
|
| 218 |
|
#define PCRE_COMPILE8(re, pat, options, error, erroffset, tables) \ |
| 219 |
|
re = pcre_compile((char *)pat, options, error, erroffset, tables) |
| 220 |
|
|
| 221 |
|
#define PCRE_COPY_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \ |
| 222 |
|
namesptr, cbuffer, size) \ |
| 223 |
|
rc = pcre_copy_named_substring(re, (char *)bptr, offsets, count, \ |
| 224 |
|
(char *)namesptr, cbuffer, size) |
| 225 |
|
|
| 226 |
|
#define PCRE_COPY_SUBSTRING8(rc, bptr, offsets, count, i, cbuffer, size) \ |
| 227 |
|
rc = pcre_copy_substring((char *)bptr, offsets, count, i, cbuffer, size) |
| 228 |
|
|
| 229 |
|
#define PCRE_DFA_EXEC8(count, re, extra, bptr, len, start_offset, options, \ |
| 230 |
|
offsets, size_offsets, workspace, size_workspace) \ |
| 231 |
|
count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset, options, \ |
| 232 |
|
offsets, size_offsets, workspace, size_workspace) |
| 233 |
|
|
| 234 |
|
#define PCRE_EXEC8(count, re, extra, bptr, len, start_offset, options, \ |
| 235 |
|
offsets, size_offsets) \ |
| 236 |
|
count = pcre_exec(re, extra, (char *)bptr, len, start_offset, options, \ |
| 237 |
|
offsets, size_offsets) |
| 238 |
|
|
| 239 |
|
#define PCRE_FREE_STUDY8(extra) \ |
| 240 |
|
pcre_free_study(extra) |
| 241 |
|
|
| 242 |
|
#define PCRE_FREE_SUBSTRING8(substring) \ |
| 243 |
|
pcre_free_substring(substring) |
| 244 |
|
|
| 245 |
|
#define PCRE_FREE_SUBSTRING_LIST8(listptr) \ |
| 246 |
|
pcre_free_substring_list(listptr) |
| 247 |
|
|
| 248 |
|
#define PCRE_GET_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \ |
| 249 |
|
getnamesptr, subsptr) \ |
| 250 |
|
rc = pcre_get_named_substring(re, (char *)bptr, offsets, count, \ |
| 251 |
|
(char *)getnamesptr, subsptr) |
| 252 |
|
|
| 253 |
|
#define PCRE_GET_STRINGNUMBER8(n, rc, ptr) \ |
| 254 |
|
n = pcre_get_stringnumber(re, (char *)ptr) |
| 255 |
|
|
| 256 |
|
#define PCRE_GET_SUBSTRING8(rc, bptr, offsets, count, i, subsptr) \ |
| 257 |
|
rc = pcre_get_substring((char *)bptr, offsets, count, i, subsptr) |
| 258 |
|
|
| 259 |
|
#define PCRE_GET_SUBSTRING_LIST8(rc, bptr, offsets, count, listptr) \ |
| 260 |
|
rc = pcre_get_substring_list((const char *)bptr, offsets, count, listptr) |
| 261 |
|
|
| 262 |
|
#define PCRE_PATTERN_TO_HOST_BYTE_ORDER8(re, extra, tables) \ |
| 263 |
|
pcre_pattern_to_host_byte_order(re, extra, tables) |
| 264 |
|
|
| 265 |
|
#define PCRE_PRINTINT8(re, outfile, debug_lengths) \ |
| 266 |
|
pcre_printint(re, outfile, debug_lengths) |
| 267 |
|
|
| 268 |
|
#define PCRE_STUDY8(extra, re, options, error) \ |
| 269 |
|
extra = pcre_study(re, options, error) |
| 270 |
|
|
| 271 |
|
#endif /* SUPPORT_PCRE8 */ |
| 272 |
|
|
| 273 |
|
/* -----------------------------------------------------------*/ |
| 274 |
|
|
| 275 |
|
#ifdef SUPPORT_PCRE16 |
| 276 |
|
|
| 277 |
|
#define PCHARS16(lv, p, offset, len, f) \ |
| 278 |
|
lv = pchars16((PCRE_SPTR16)(p) + offset, len, f) |
| 279 |
|
|
| 280 |
|
#define PCHARSV16(p, offset, len, f) \ |
| 281 |
|
(void)pchars16((PCRE_SPTR16)(p) + offset, len, f) |
| 282 |
|
|
| 283 |
|
#define READ_CAPTURE_NAME16(p, cn8, cn16, re) \ |
| 284 |
|
p = read_capture_name16(p, cn16, re) |
| 285 |
|
|
| 286 |
|
#define STRLEN16(p) ((int)strlen16((PCRE_SPTR16)p)) |
| 287 |
|
|
| 288 |
|
#define SET_PCRE_CALLOUT16(callout) \ |
| 289 |
|
pcre16_callout = callout |
| 290 |
|
|
| 291 |
|
|
| 292 |
|
#define PCRE_COMPILE16(re, pat, options, error, erroffset, tables) \ |
| 293 |
|
re = pcre16_compile((PCRE_SPTR16)pat, options, error, erroffset, tables) |
| 294 |
|
|
| 295 |
|
#define PCRE_COPY_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \ |
| 296 |
|
namesptr, cbuffer, size) \ |
| 297 |
|
rc = pcre16_copy_named_substring(re, (PCRE_SPTR16)bptr, offsets, count, \ |
| 298 |
|
(PCRE_SPTR16)namesptr, (PCRE_SCHAR16 *)cbuffer, size/2) |
| 299 |
|
|
| 300 |
|
#define PCRE_COPY_SUBSTRING16(rc, bptr, offsets, count, i, cbuffer, size) \ |
| 301 |
|
rc = pcre16_copy_substring((PCRE_SPTR16)bptr, offsets, count, i, \ |
| 302 |
|
(PCRE_SCHAR16 *)cbuffer, size/2) |
| 303 |
|
|
| 304 |
|
#define PCRE_DFA_EXEC16(count, re, extra, bptr, len, start_offset, options, \ |
| 305 |
|
offsets, size_offsets, workspace, size_workspace) \ |
| 306 |
|
count = pcre16_dfa_exec(re, extra, (PCRE_SPTR16)bptr, len, start_offset, \ |
| 307 |
|
options, offsets, size_offsets, workspace, size_workspace) |
| 308 |
|
|
| 309 |
|
#define PCRE_EXEC16(count, re, extra, bptr, len, start_offset, options, \ |
| 310 |
|
offsets, size_offsets) \ |
| 311 |
|
count = pcre16_exec(re, extra, (PCRE_SPTR16)bptr, len, start_offset, \ |
| 312 |
|
options, offsets, size_offsets) |
| 313 |
|
|
| 314 |
|
#define PCRE_FREE_STUDY16(extra) \ |
| 315 |
|
pcre16_free_study(extra) |
| 316 |
|
|
| 317 |
|
#define PCRE_FREE_SUBSTRING16(substring) \ |
| 318 |
|
pcre16_free_substring((PCRE_SPTR16)substring) |
| 319 |
|
|
| 320 |
|
#define PCRE_FREE_SUBSTRING_LIST16(listptr) \ |
| 321 |
|
pcre16_free_substring_list((PCRE_SPTR16 *)listptr) |
| 322 |
|
|
| 323 |
|
#define PCRE_GET_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \ |
| 324 |
|
getnamesptr, subsptr) \ |
| 325 |
|
rc = pcre16_get_named_substring(re, (PCRE_SPTR16)bptr, offsets, count, \ |
| 326 |
|
(PCRE_SPTR16)getnamesptr, (PCRE_SPTR16 *)(void*)subsptr) |
| 327 |
|
|
| 328 |
|
#define PCRE_GET_STRINGNUMBER16(n, rc, ptr) \ |
| 329 |
|
n = pcre16_get_stringnumber(re, (PCRE_SPTR16)ptr) |
| 330 |
|
|
| 331 |
|
#define PCRE_GET_SUBSTRING16(rc, bptr, offsets, count, i, subsptr) \ |
| 332 |
|
rc = pcre16_get_substring((PCRE_SPTR16)bptr, offsets, count, i, \ |
| 333 |
|
(PCRE_SPTR16 *)(void*)subsptr) |
| 334 |
|
|
| 335 |
|
#define PCRE_GET_SUBSTRING_LIST16(rc, bptr, offsets, count, listptr) \ |
| 336 |
|
rc = pcre16_get_substring_list((PCRE_SPTR16)bptr, offsets, count, \ |
| 337 |
|
(PCRE_SPTR16 **)(void*)listptr) |
| 338 |
|
|
| 339 |
|
#define PCRE_PATTERN_TO_HOST_BYTE_ORDER16(re, extra, tables) \ |
| 340 |
|
pcre16_pattern_to_host_byte_order(re, extra, tables) |
| 341 |
|
|
| 342 |
|
#define PCRE_PRINTINT16(re, outfile, debug_lengths) \ |
| 343 |
|
pcre16_printint(re, outfile, debug_lengths) |
| 344 |
|
|
| 345 |
|
#define PCRE_STUDY16(extra, re, options, error) \ |
| 346 |
|
extra = pcre16_study(re, options, error) |
| 347 |
|
|
| 348 |
|
#endif /* SUPPORT_PCRE16 */ |
| 349 |
|
|
| 350 |
|
|
| 351 |
|
/* ----- Both modes are supported; a runtime test is needed, except for |
| 352 |
|
pcre_config(), and the JIT stack functions, when it doesn't matter which |
| 353 |
|
version is called. ----- */ |
| 354 |
|
|
| 355 |
|
#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE16 |
| 356 |
|
|
| 357 |
|
#define CHAR_SIZE (use_pcre16? 2:1) |
| 358 |
|
|
| 359 |
|
#define PCHARS(lv, p, offset, len, f) \ |
| 360 |
|
if (use_pcre16) \ |
| 361 |
|
PCHARS16(lv, p, offset, len, f); \ |
| 362 |
|
else \ |
| 363 |
|
PCHARS8(lv, p, offset, len, f) |
| 364 |
|
|
| 365 |
|
#define PCHARSV(p, offset, len, f) \ |
| 366 |
|
if (use_pcre16) \ |
| 367 |
|
PCHARSV16(p, offset, len, f); \ |
| 368 |
|
else \ |
| 369 |
|
PCHARSV8(p, offset, len, f) |
| 370 |
|
|
| 371 |
|
#define READ_CAPTURE_NAME(p, cn8, cn16, re) \ |
| 372 |
|
if (use_pcre16) \ |
| 373 |
|
READ_CAPTURE_NAME16(p, cn8, cn16, re); \ |
| 374 |
|
else \ |
| 375 |
|
READ_CAPTURE_NAME8(p, cn8, cn16, re) |
| 376 |
|
|
| 377 |
|
#define SET_PCRE_CALLOUT(callout) \ |
| 378 |
|
if (use_pcre16) \ |
| 379 |
|
SET_PCRE_CALLOUT16(callout); \ |
| 380 |
|
else \ |
| 381 |
|
SET_PCRE_CALLOUT8(callout) |
| 382 |
|
|
| 383 |
|
#define STRLEN(p) (use_pcre16? STRLEN16(p) : STRLEN8(p)) |
| 384 |
|
|
| 385 |
|
#define PCRE_ASSIGN_JIT_STACK pcre_assign_jit_stack |
| 386 |
|
|
| 387 |
|
#define PCRE_COMPILE(re, pat, options, error, erroffset, tables) \ |
| 388 |
|
if (use_pcre16) \ |
| 389 |
|
PCRE_COMPILE16(re, pat, options, error, erroffset, tables); \ |
| 390 |
|
else \ |
| 391 |
|
PCRE_COMPILE8(re, pat, options, error, erroffset, tables) |
| 392 |
|
|
| 393 |
|
#define PCRE_CONFIG pcre_config |
| 394 |
|
|
| 395 |
|
#define PCRE_COPY_NAMED_SUBSTRING(rc, re, bptr, offsets, count, \ |
| 396 |
|
namesptr, cbuffer, size) \ |
| 397 |
|
if (use_pcre16) \ |
| 398 |
|
PCRE_COPY_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \ |
| 399 |
|
namesptr, cbuffer, size); \ |
| 400 |
|
else \ |
| 401 |
|
PCRE_COPY_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \ |
| 402 |
|
namesptr, cbuffer, size) |
| 403 |
|
|
| 404 |
|
#define PCRE_COPY_SUBSTRING(rc, bptr, offsets, count, i, cbuffer, size) \ |
| 405 |
|
if (use_pcre16) \ |
| 406 |
|
PCRE_COPY_SUBSTRING16(rc, bptr, offsets, count, i, cbuffer, size); \ |
| 407 |
|
else \ |
| 408 |
|
PCRE_COPY_SUBSTRING8(rc, bptr, offsets, count, i, cbuffer, size) |
| 409 |
|
|
| 410 |
|
#define PCRE_DFA_EXEC(count, re, extra, bptr, len, start_offset, options, \ |
| 411 |
|
offsets, size_offsets, workspace, size_workspace) \ |
| 412 |
|
if (use_pcre16) \ |
| 413 |
|
PCRE_DFA_EXEC16(count, re, extra, bptr, len, start_offset, options, \ |
| 414 |
|
offsets, size_offsets, workspace, size_workspace); \ |
| 415 |
|
else \ |
| 416 |
|
PCRE_DFA_EXEC8(count, re, extra, bptr, len, start_offset, options, \ |
| 417 |
|
offsets, size_offsets, workspace, size_workspace) |
| 418 |
|
|
| 419 |
|
#define PCRE_EXEC(count, re, extra, bptr, len, start_offset, options, \ |
| 420 |
|
offsets, size_offsets) \ |
| 421 |
|
if (use_pcre16) \ |
| 422 |
|
PCRE_EXEC16(count, re, extra, bptr, len, start_offset, options, \ |
| 423 |
|
offsets, size_offsets); \ |
| 424 |
|
else \ |
| 425 |
|
PCRE_EXEC8(count, re, extra, bptr, len, start_offset, options, \ |
| 426 |
|
offsets, size_offsets) |
| 427 |
|
|
| 428 |
|
#define PCRE_FREE_STUDY(extra) \ |
| 429 |
|
if (use_pcre16) \ |
| 430 |
|
PCRE_FREE_STUDY16(extra); \ |
| 431 |
|
else \ |
| 432 |
|
PCRE_FREE_STUDY8(extra) |
| 433 |
|
|
| 434 |
|
#define PCRE_FREE_SUBSTRING(substring) \ |
| 435 |
|
if (use_pcre16) \ |
| 436 |
|
PCRE_FREE_SUBSTRING16(substring); \ |
| 437 |
|
else \ |
| 438 |
|
PCRE_FREE_SUBSTRING8(substring) |
| 439 |
|
|
| 440 |
|
#define PCRE_FREE_SUBSTRING_LIST(listptr) \ |
| 441 |
|
if (use_pcre16) \ |
| 442 |
|
PCRE_FREE_SUBSTRING_LIST16(listptr); \ |
| 443 |
|
else \ |
| 444 |
|
PCRE_FREE_SUBSTRING_LIST8(listptr) |
| 445 |
|
|
| 446 |
|
#define PCRE_GET_NAMED_SUBSTRING(rc, re, bptr, offsets, count, \ |
| 447 |
|
getnamesptr, subsptr) \ |
| 448 |
|
if (use_pcre16) \ |
| 449 |
|
PCRE_GET_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \ |
| 450 |
|
getnamesptr, subsptr); \ |
| 451 |
|
else \ |
| 452 |
|
PCRE_GET_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \ |
| 453 |
|
getnamesptr, subsptr) |
| 454 |
|
|
| 455 |
|
#define PCRE_GET_STRINGNUMBER(n, rc, ptr) \ |
| 456 |
|
if (use_pcre16) \ |
| 457 |
|
PCRE_GET_STRINGNUMBER16(n, rc, ptr); \ |
| 458 |
|
else \ |
| 459 |
|
PCRE_GET_STRINGNUMBER8(n, rc, ptr) |
| 460 |
|
|
| 461 |
|
#define PCRE_GET_SUBSTRING(rc, bptr, use_offsets, count, i, subsptr) \ |
| 462 |
|
if (use_pcre16) \ |
| 463 |
|
PCRE_GET_SUBSTRING16(rc, bptr, use_offsets, count, i, subsptr); \ |
| 464 |
|
else \ |
| 465 |
|
PCRE_GET_SUBSTRING8(rc, bptr, use_offsets, count, i, subsptr) |
| 466 |
|
|
| 467 |
|
#define PCRE_GET_SUBSTRING_LIST(rc, bptr, offsets, count, listptr) \ |
| 468 |
|
if (use_pcre16) \ |
| 469 |
|
PCRE_GET_SUBSTRING_LIST16(rc, bptr, offsets, count, listptr); \ |
| 470 |
|
else \ |
| 471 |
|
PCRE_GET_SUBSTRING_LIST8(rc, bptr, offsets, count, listptr) |
| 472 |
|
|
| 473 |
|
#define PCRE_JIT_STACK_ALLOC pcre_jit_stack_alloc |
| 474 |
|
#define PCRE_JIT_STACK_FREE pcre_jit_stack_free |
| 475 |
|
|
| 476 |
|
#define PCRE_MAKETABLES \ |
| 477 |
|
(use_pcre16? pcre16_maketables() : pcre_maketables()) |
| 478 |
|
|
| 479 |
|
#define PCRE_PATTERN_TO_HOST_BYTE_ORDER(re, extra, tables) \ |
| 480 |
|
if (use_pcre16) \ |
| 481 |
|
PCRE_PATTERN_TO_HOST_BYTE_ORDER16(re, extra, tables); \ |
| 482 |
|
else \ |
| 483 |
|
PCRE_PATTERN_TO_HOST_BYTE_ORDER8(re, extra, tables) |
| 484 |
|
|
| 485 |
|
#define PCRE_PRINTINT(re, outfile, debug_lengths) \ |
| 486 |
|
if (use_pcre16) \ |
| 487 |
|
PCRE_PRINTINT16(re, outfile, debug_lengths); \ |
| 488 |
|
else \ |
| 489 |
|
PCRE_PRINTINT8(re, outfile, debug_lengths) |
| 490 |
|
|
| 491 |
|
#define PCRE_STUDY(extra, re, options, error) \ |
| 492 |
|
if (use_pcre16) \ |
| 493 |
|
PCRE_STUDY16(extra, re, options, error); \ |
| 494 |
|
else \ |
| 495 |
|
PCRE_STUDY8(extra, re, options, error) |
| 496 |
|
|
| 497 |
|
/* ----- Only 8-bit mode is supported ----- */ |
| 498 |
|
|
| 499 |
|
#elif defined SUPPORT_PCRE8 |
| 500 |
|
#define CHAR_SIZE 1 |
| 501 |
|
#define PCHARS PCHARS8 |
| 502 |
|
#define PCHARSV PCHARSV8 |
| 503 |
|
#define READ_CAPTURE_NAME READ_CAPTURE_NAME8 |
| 504 |
|
#define SET_PCRE_CALLOUT SET_PCRE_CALLOUT8 |
| 505 |
|
#define STRLEN STRLEN8 |
| 506 |
|
#define PCRE_ASSIGN_JIT_STACK pcre_assign_jit_stack |
| 507 |
|
#define PCRE_COMPILE PCRE_COMPILE8 |
| 508 |
|
#define PCRE_CONFIG pcre_config |
| 509 |
|
#define PCRE_COPY_NAMED_SUBSTRING PCRE_COPY_NAMED_SUBSTRING8 |
| 510 |
|
#define PCRE_COPY_SUBSTRING PCRE_COPY_SUBSTRING8 |
| 511 |
|
#define PCRE_DFA_EXEC PCRE_DFA_EXEC8 |
| 512 |
|
#define PCRE_EXEC PCRE_EXEC8 |
| 513 |
|
#define PCRE_FREE_STUDY PCRE_FREE_STUDY8 |
| 514 |
|
#define PCRE_FREE_SUBSTRING PCRE_FREE_SUBSTRING8 |
| 515 |
|
#define PCRE_FREE_SUBSTRING_LIST PCRE_FREE_SUBSTRING_LIST8 |
| 516 |
|
#define PCRE_GET_NAMED_SUBSTRING PCRE_GET_NAMED_SUBSTRING8 |
| 517 |
|
#define PCRE_GET_STRINGNUMBER PCRE_GET_STRINGNUMBER8 |
| 518 |
|
#define PCRE_GET_SUBSTRING PCRE_GET_SUBSTRING8 |
| 519 |
|
#define PCRE_GET_SUBSTRING_LIST PCRE_GET_SUBSTRING_LIST8 |
| 520 |
|
#define PCRE_JIT_STACK_ALLOC pcre_jit_stack_alloc |
| 521 |
|
#define PCRE_JIT_STACK_FREE pcre_jit_stack_free |
| 522 |
|
#define PCRE_MAKETABLES pcre_maketables() |
| 523 |
|
#define PCRE_PATTERN_TO_HOST_BYTE_ORDER PCRE_PATTERN_TO_HOST_BYTE_ORDER8 |
| 524 |
|
#define PCRE_PRINTINT PCRE_PRINTINT8 |
| 525 |
|
#define PCRE_STUDY PCRE_STUDY8 |
| 526 |
|
|
| 527 |
|
/* ----- Only 16-bit mode is supported ----- */ |
| 528 |
|
|
| 529 |
|
#else |
| 530 |
|
#define CHAR_SIZE 2 |
| 531 |
|
#define PCHARS PCHARS16 |
| 532 |
|
#define PCHARSV PCHARSV16 |
| 533 |
|
#define READ_CAPTURE_NAME READ_CAPTURE_NAME16 |
| 534 |
|
#define SET_PCRE_CALLOUT SET_PCRE_CALLOUT16 |
| 535 |
|
#define STRLEN STRLEN16 |
| 536 |
|
#define PCRE_ASSIGN_JIT_STACK pcre16_assign_jit_stack |
| 537 |
|
#define PCRE_COMPILE PCRE_COMPILE16 |
| 538 |
|
#define PCRE_CONFIG pcre16_config |
| 539 |
|
#define PCRE_COPY_NAMED_SUBSTRING PCRE_COPY_NAMED_SUBSTRING16 |
| 540 |
|
#define PCRE_COPY_SUBSTRING PCRE_COPY_SUBSTRING16 |
| 541 |
|
#define PCRE_DFA_EXEC PCRE_DFA_EXEC16 |
| 542 |
|
#define PCRE_EXEC PCRE_EXEC16 |
| 543 |
|
#define PCRE_FREE_STUDY PCRE_FREE_STUDY16 |
| 544 |
|
#define PCRE_FREE_SUBSTRING PCRE_FREE_SUBSTRING16 |
| 545 |
|
#define PCRE_FREE_SUBSTRING_LIST PCRE_FREE_SUBSTRING_LIST16 |
| 546 |
|
#define PCRE_GET_NAMED_SUBSTRING PCRE_GET_NAMED_SUBSTRING16 |
| 547 |
|
#define PCRE_GET_STRINGNUMBER PCRE_GET_STRINGNUMBER16 |
| 548 |
|
#define PCRE_GET_SUBSTRING PCRE_GET_SUBSTRING16 |
| 549 |
|
#define PCRE_GET_SUBSTRING_LIST PCRE_GET_SUBSTRING_LIST16 |
| 550 |
|
#define PCRE_JIT_STACK_ALLOC pcre16_jit_stack_alloc |
| 551 |
|
#define PCRE_JIT_STACK_FREE pcre16_jit_stack_free |
| 552 |
|
#define PCRE_MAKETABLES pcre16_maketables() |
| 553 |
|
#define PCRE_PATTERN_TO_HOST_BYTE_ORDER PCRE_PATTERN_TO_HOST_BYTE_ORDER16 |
| 554 |
|
#define PCRE_PRINTINT PCRE_PRINTINT16 |
| 555 |
|
#define PCRE_STUDY PCRE_STUDY16 |
| 556 |
|
#endif |
| 557 |
|
|
| 558 |
|
/* ----- End of mode-specific function call macros ----- */ |
| 559 |
|
|
| 560 |
|
|
| 561 |
/* Other parameters */ |
/* Other parameters */ |
| 562 |
|
|
| 584 |
static int first_callout; |
static int first_callout; |
| 585 |
static int locale_set = 0; |
static int locale_set = 0; |
| 586 |
static int show_malloc; |
static int show_malloc; |
| 587 |
static int use_utf8; |
static int use_utf; |
| 588 |
static size_t gotten_store; |
static size_t gotten_store; |
| 589 |
|
static size_t first_gotten_store = 0; |
| 590 |
static const unsigned char *last_callout_mark = NULL; |
static const unsigned char *last_callout_mark = NULL; |
| 591 |
|
|
| 592 |
/* The buffers grow automatically if very long input lines are encountered. */ |
/* The buffers grow automatically if very long input lines are encountered. */ |
| 593 |
|
|
| 594 |
static int buffer_size = 50000; |
static int buffer_size = 50000; |
| 595 |
static uschar *buffer = NULL; |
static pcre_uint8 *buffer = NULL; |
| 596 |
static uschar *dbuffer = NULL; |
static pcre_uint8 *dbuffer = NULL; |
| 597 |
static uschar *pbuffer = NULL; |
static pcre_uint8 *pbuffer = NULL; |
| 598 |
|
|
| 599 |
|
/* Another buffer is needed translation to 16-bit character strings. It will |
| 600 |
|
obtained and extended as required. */ |
| 601 |
|
|
| 602 |
|
#ifdef SUPPORT_PCRE16 |
| 603 |
|
static int buffer16_size = 0; |
| 604 |
|
static pcre_uint16 *buffer16 = NULL; |
| 605 |
|
|
| 606 |
|
#ifdef SUPPORT_PCRE8 |
| 607 |
|
|
| 608 |
|
/* We need the table of operator lengths that is used for 16-bit compiling, in |
| 609 |
|
order to swap bytes in a pattern for saving/reloading testing. Luckily, the |
| 610 |
|
data is defined as a macro. However, we must ensure that LINK_SIZE is adjusted |
| 611 |
|
appropriately for the 16-bit world. Just as a safety check, make sure that |
| 612 |
|
COMPILE_PCRE16 is *not* set. */ |
| 613 |
|
|
| 614 |
|
#ifdef COMPILE_PCRE16 |
| 615 |
|
#error COMPILE_PCRE16 must not be set when compiling pcretest.c |
| 616 |
|
#endif |
| 617 |
|
|
| 618 |
|
#if LINK_SIZE == 2 |
| 619 |
|
#undef LINK_SIZE |
| 620 |
|
#define LINK_SIZE 1 |
| 621 |
|
#elif LINK_SIZE == 3 || LINK_SIZE == 4 |
| 622 |
|
#undef LINK_SIZE |
| 623 |
|
#define LINK_SIZE 2 |
| 624 |
|
#else |
| 625 |
|
#error LINK_SIZE must be either 2, 3, or 4 |
| 626 |
|
#endif |
| 627 |
|
|
| 628 |
|
#endif /* SUPPORT_PCRE8 */ |
| 629 |
|
|
| 630 |
|
static const pcre_uint16 OP_lengths16[] = { OP_LENGTHS }; |
| 631 |
|
#endif /* SUPPORT_PCRE16 */ |
| 632 |
|
|
| 633 |
|
/* If we have 8-bit support, default use_pcre16 to false; if there is also |
| 634 |
|
16-bit support, it can be changed by an option. If there is no 8-bit support, |
| 635 |
|
there must be 16-bit support, so default it to 1. */ |
| 636 |
|
|
| 637 |
|
#ifdef SUPPORT_PCRE8 |
| 638 |
|
static int use_pcre16 = 0; |
| 639 |
|
#else |
| 640 |
|
static int use_pcre16 = 1; |
| 641 |
|
#endif |
| 642 |
|
|
| 643 |
/* Textual explanations for runtime error codes */ |
/* Textual explanations for runtime error codes */ |
| 644 |
|
|
| 653 |
NULL, /* never returned by pcre_exec() or pcre_dfa_exec() */ |
NULL, /* never returned by pcre_exec() or pcre_dfa_exec() */ |
| 654 |
"match limit exceeded", |
"match limit exceeded", |
| 655 |
"callout error code", |
"callout error code", |
| 656 |
NULL, /* BADUTF8 is handled specially */ |
NULL, /* BADUTF8/16 is handled specially */ |
| 657 |
"bad UTF-8 offset", |
NULL, /* BADUTF8/16 offset is handled specially */ |
| 658 |
NULL, /* PARTIAL is handled specially */ |
NULL, /* PARTIAL is handled specially */ |
| 659 |
"not used - internal error", |
"not used - internal error", |
| 660 |
"internal error - pattern overwritten?", |
"internal error - pattern overwritten?", |
| 668 |
"not used - internal error", |
"not used - internal error", |
| 669 |
"invalid combination of newline options", |
"invalid combination of newline options", |
| 670 |
"bad offset value", |
"bad offset value", |
| 671 |
NULL, /* SHORTUTF8 is handled specially */ |
NULL, /* SHORTUTF8/16 is handled specially */ |
| 672 |
"nested recursion at the same subject position", |
"nested recursion at the same subject position", |
| 673 |
"JIT stack limit reached" |
"JIT stack limit reached", |
| 674 |
|
"pattern compiled in wrong mode: 8-bit/16-bit error" |
| 675 |
}; |
}; |
| 676 |
|
|
| 677 |
|
|
| 687 |
/* This is the set of tables distributed as default with PCRE. It recognizes |
/* This is the set of tables distributed as default with PCRE. It recognizes |
| 688 |
only ASCII characters. */ |
only ASCII characters. */ |
| 689 |
|
|
| 690 |
static const unsigned char tables0[] = { |
static const pcre_uint8 tables0[] = { |
| 691 |
|
|
| 692 |
/* This table is a lower casing table. */ |
/* This table is a lower casing table. */ |
| 693 |
|
|
| 860 |
be at least an approximation of ISO 8859. In particular, there are characters |
be at least an approximation of ISO 8859. In particular, there are characters |
| 861 |
greater than 128 that are marked as spaces, letters, etc. */ |
greater than 128 that are marked as spaces, letters, etc. */ |
| 862 |
|
|
| 863 |
static const unsigned char tables1[] = { |
static const pcre_uint8 tables1[] = { |
| 864 |
0,1,2,3,4,5,6,7, |
0,1,2,3,4,5,6,7, |
| 865 |
8,9,10,11,12,13,14,15, |
8,9,10,11,12,13,14,15, |
| 866 |
16,17,18,19,20,21,22,23, |
16,17,18,19,20,21,22,23, |
| 1033 |
} |
} |
| 1034 |
|
|
| 1035 |
|
|
| 1036 |
|
#if !defined NOUTF || defined SUPPORT_PCRE16 |
| 1037 |
|
/************************************************* |
| 1038 |
|
* Convert UTF-8 string to value * |
| 1039 |
|
*************************************************/ |
| 1040 |
|
|
| 1041 |
|
/* This function takes one or more bytes that represents a UTF-8 character, |
| 1042 |
|
and returns the value of the character. |
| 1043 |
|
|
| 1044 |
|
Argument: |
| 1045 |
|
utf8bytes a pointer to the byte vector |
| 1046 |
|
vptr a pointer to an int to receive the value |
| 1047 |
|
|
| 1048 |
|
Returns: > 0 => the number of bytes consumed |
| 1049 |
|
-6 to 0 => malformed UTF-8 character at offset = (-return) |
| 1050 |
|
*/ |
| 1051 |
|
|
| 1052 |
|
static int |
| 1053 |
|
utf82ord(pcre_uint8 *utf8bytes, int *vptr) |
| 1054 |
|
{ |
| 1055 |
|
int c = *utf8bytes++; |
| 1056 |
|
int d = c; |
| 1057 |
|
int i, j, s; |
| 1058 |
|
|
| 1059 |
|
for (i = -1; i < 6; i++) /* i is number of additional bytes */ |
| 1060 |
|
{ |
| 1061 |
|
if ((d & 0x80) == 0) break; |
| 1062 |
|
d <<= 1; |
| 1063 |
|
} |
| 1064 |
|
|
| 1065 |
|
if (i == -1) { *vptr = c; return 1; } /* ascii character */ |
| 1066 |
|
if (i == 0 || i == 6) return 0; /* invalid UTF-8 */ |
| 1067 |
|
|
| 1068 |
|
/* i now has a value in the range 1-5 */ |
| 1069 |
|
|
| 1070 |
|
s = 6*i; |
| 1071 |
|
d = (c & utf8_table3[i]) << s; |
| 1072 |
|
|
| 1073 |
|
for (j = 0; j < i; j++) |
| 1074 |
|
{ |
| 1075 |
|
c = *utf8bytes++; |
| 1076 |
|
if ((c & 0xc0) != 0x80) return -(j+1); |
| 1077 |
|
s -= 6; |
| 1078 |
|
d |= (c & 0x3f) << s; |
| 1079 |
|
} |
| 1080 |
|
|
| 1081 |
|
/* Check that encoding was the correct unique one */ |
| 1082 |
|
|
| 1083 |
|
for (j = 0; j < utf8_table1_size; j++) |
| 1084 |
|
if (d <= utf8_table1[j]) break; |
| 1085 |
|
if (j != i) return -(i+1); |
| 1086 |
|
|
| 1087 |
|
/* Valid value */ |
| 1088 |
|
|
| 1089 |
|
*vptr = d; |
| 1090 |
|
return i+1; |
| 1091 |
|
} |
| 1092 |
|
#endif /* NOUTF || SUPPORT_PCRE16 */ |
| 1093 |
|
|
| 1094 |
|
|
| 1095 |
|
|
| 1096 |
|
#if !defined NOUTF || defined SUPPORT_PCRE16 |
| 1097 |
|
/************************************************* |
| 1098 |
|
* Convert character value to UTF-8 * |
| 1099 |
|
*************************************************/ |
| 1100 |
|
|
| 1101 |
|
/* This function takes an integer value in the range 0 - 0x7fffffff |
| 1102 |
|
and encodes it as a UTF-8 character in 0 to 6 bytes. |
| 1103 |
|
|
| 1104 |
|
Arguments: |
| 1105 |
|
cvalue the character value |
| 1106 |
|
utf8bytes pointer to buffer for result - at least 6 bytes long |
| 1107 |
|
|
| 1108 |
|
Returns: number of characters placed in the buffer |
| 1109 |
|
*/ |
| 1110 |
|
|
| 1111 |
|
static int |
| 1112 |
|
ord2utf8(int cvalue, pcre_uint8 *utf8bytes) |
| 1113 |
|
{ |
| 1114 |
|
register int i, j; |
| 1115 |
|
for (i = 0; i < utf8_table1_size; i++) |
| 1116 |
|
if (cvalue <= utf8_table1[i]) break; |
| 1117 |
|
utf8bytes += i; |
| 1118 |
|
for (j = i; j > 0; j--) |
| 1119 |
|
{ |
| 1120 |
|
*utf8bytes-- = 0x80 | (cvalue & 0x3f); |
| 1121 |
|
cvalue >>= 6; |
| 1122 |
|
} |
| 1123 |
|
*utf8bytes = utf8_table2[i] | cvalue; |
| 1124 |
|
return i + 1; |
| 1125 |
|
} |
| 1126 |
|
#endif /* NOUTF || SUPPORT_PCRE16 */ |
| 1127 |
|
|
| 1128 |
|
|
| 1129 |
|
|
| 1130 |
|
#ifdef SUPPORT_PCRE16 |
| 1131 |
|
/************************************************* |
| 1132 |
|
* Convert a string to 16-bit * |
| 1133 |
|
*************************************************/ |
| 1134 |
|
|
| 1135 |
|
/* In non-UTF mode, the space needed for a 16-bit string is exactly double the |
| 1136 |
|
8-bit size. For a UTF-8 string, the size needed for UTF-16 is no more than |
| 1137 |
|
double, because up to 0xffff uses no more than 3 bytes in UTF-8 but possibly 4 |
| 1138 |
|
in UTF-16. Higher values use 4 bytes in UTF-8 and up to 4 bytes in UTF-16. The |
| 1139 |
|
result is always left in buffer16. |
| 1140 |
|
|
| 1141 |
|
Note that this function does not object to surrogate values. This is |
| 1142 |
|
deliberate; it makes it possible to construct UTF-16 strings that are invalid, |
| 1143 |
|
for the purpose of testing that they are correctly faulted. |
| 1144 |
|
|
| 1145 |
|
Patterns to be converted are either plain ASCII or UTF-8; data lines are always |
| 1146 |
|
in UTF-8 so that values greater than 255 can be handled. |
| 1147 |
|
|
| 1148 |
|
Arguments: |
| 1149 |
|
data TRUE if converting a data line; FALSE for a regex |
| 1150 |
|
p points to a byte string |
| 1151 |
|
utf true if UTF-8 (to be converted to UTF-16) |
| 1152 |
|
len number of bytes in the string (excluding trailing zero) |
| 1153 |
|
|
| 1154 |
|
Returns: number of 16-bit data items used (excluding trailing zero) |
| 1155 |
|
OR -1 if a UTF-8 string is malformed |
| 1156 |
|
OR -2 if a value > 0x10ffff is encountered |
| 1157 |
|
OR -3 if a value > 0xffff is encountered when not in UTF mode |
| 1158 |
|
*/ |
| 1159 |
|
|
| 1160 |
|
static int |
| 1161 |
|
to16(int data, pcre_uint8 *p, int utf, int len) |
| 1162 |
|
{ |
| 1163 |
|
pcre_uint16 *pp; |
| 1164 |
|
|
| 1165 |
|
if (buffer16_size < 2*len + 2) |
| 1166 |
|
{ |
| 1167 |
|
if (buffer16 != NULL) free(buffer16); |
| 1168 |
|
buffer16_size = 2*len + 2; |
| 1169 |
|
buffer16 = (pcre_uint16 *)malloc(buffer16_size); |
| 1170 |
|
if (buffer16 == NULL) |
| 1171 |
|
{ |
| 1172 |
|
fprintf(stderr, "pcretest: malloc(%d) failed for buffer16\n", buffer16_size); |
| 1173 |
|
exit(1); |
| 1174 |
|
} |
| 1175 |
|
} |
| 1176 |
|
|
| 1177 |
|
pp = buffer16; |
| 1178 |
|
|
| 1179 |
|
if (!utf && !data) |
| 1180 |
|
{ |
| 1181 |
|
while (len-- > 0) *pp++ = *p++; |
| 1182 |
|
} |
| 1183 |
|
|
| 1184 |
|
else |
| 1185 |
|
{ |
| 1186 |
|
int c = 0; |
| 1187 |
|
while (len > 0) |
| 1188 |
|
{ |
| 1189 |
|
int chlen = utf82ord(p, &c); |
| 1190 |
|
if (chlen <= 0) return -1; |
| 1191 |
|
if (c > 0x10ffff) return -2; |
| 1192 |
|
p += chlen; |
| 1193 |
|
len -= chlen; |
| 1194 |
|
if (c < 0x10000) *pp++ = c; else |
| 1195 |
|
{ |
| 1196 |
|
if (!utf) return -3; |
| 1197 |
|
c -= 0x10000; |
| 1198 |
|
*pp++ = 0xD800 | (c >> 10); |
| 1199 |
|
*pp++ = 0xDC00 | (c & 0x3ff); |
| 1200 |
|
} |
| 1201 |
|
} |
| 1202 |
|
} |
| 1203 |
|
|
| 1204 |
|
*pp = 0; |
| 1205 |
|
return pp - buffer16; |
| 1206 |
|
} |
| 1207 |
|
#endif |
| 1208 |
|
|
| 1209 |
|
|
| 1210 |
/************************************************* |
/************************************************* |
| 1211 |
* Read or extend an input line * |
* Read or extend an input line * |
| 1212 |
*************************************************/ |
*************************************************/ |
| 1230 |
NULL if no data read and EOF reached |
NULL if no data read and EOF reached |
| 1231 |
*/ |
*/ |
| 1232 |
|
|
| 1233 |
static uschar * |
static pcre_uint8 * |
| 1234 |
extend_inputline(FILE *f, uschar *start, const char *prompt) |
extend_inputline(FILE *f, pcre_uint8 *start, const char *prompt) |
| 1235 |
{ |
{ |
| 1236 |
uschar *here = start; |
pcre_uint8 *here = start; |
| 1237 |
|
|
| 1238 |
for (;;) |
for (;;) |
| 1239 |
{ |
{ |
| 1280 |
else |
else |
| 1281 |
{ |
{ |
| 1282 |
int new_buffer_size = 2*buffer_size; |
int new_buffer_size = 2*buffer_size; |
| 1283 |
uschar *new_buffer = (unsigned char *)malloc(new_buffer_size); |
pcre_uint8 *new_buffer = (pcre_uint8 *)malloc(new_buffer_size); |
| 1284 |
uschar *new_dbuffer = (unsigned char *)malloc(new_buffer_size); |
pcre_uint8 *new_dbuffer = (pcre_uint8 *)malloc(new_buffer_size); |
| 1285 |
uschar *new_pbuffer = (unsigned char *)malloc(new_buffer_size); |
pcre_uint8 *new_pbuffer = (pcre_uint8 *)malloc(new_buffer_size); |
| 1286 |
|
|
| 1287 |
if (new_buffer == NULL || new_dbuffer == NULL || new_pbuffer == NULL) |
if (new_buffer == NULL || new_dbuffer == NULL || new_pbuffer == NULL) |
| 1288 |
{ |
{ |
| 1313 |
|
|
| 1314 |
|
|
| 1315 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1316 |
/************************************************* |
/************************************************* |
| 1317 |
* Read number from string * |
* Read number from string * |
| 1318 |
*************************************************/ |
*************************************************/ |
| 1329 |
*/ |
*/ |
| 1330 |
|
|
| 1331 |
static int |
static int |
| 1332 |
get_value(unsigned char *str, unsigned char **endptr) |
get_value(pcre_uint8 *str, pcre_uint8 **endptr) |
| 1333 |
{ |
{ |
| 1334 |
int result = 0; |
int result = 0; |
| 1335 |
while(*str != 0 && isspace(*str)) str++; |
while(*str != 0 && isspace(*str)) str++; |
| 1340 |
|
|
| 1341 |
|
|
| 1342 |
|
|
|
|
|
| 1343 |
/************************************************* |
/************************************************* |
| 1344 |
* Convert UTF-8 string to value * |
* Print one character * |
| 1345 |
*************************************************/ |
*************************************************/ |
| 1346 |
|
|
| 1347 |
/* This function takes one or more bytes that represents a UTF-8 character, |
/* Print a single character either literally, or as a hex escape. */ |
|
and returns the value of the character. |
|
| 1348 |
|
|
| 1349 |
Argument: |
static int pchar(int c, FILE *f) |
|
utf8bytes a pointer to the byte vector |
|
|
vptr a pointer to an int to receive the value |
|
|
|
|
|
Returns: > 0 => the number of bytes consumed |
|
|
-6 to 0 => malformed UTF-8 character at offset = (-return) |
|
|
*/ |
|
|
|
|
|
#if !defined NOUTF8 |
|
|
|
|
|
static int |
|
|
utf82ord(unsigned char *utf8bytes, int *vptr) |
|
| 1350 |
{ |
{ |
| 1351 |
int c = *utf8bytes++; |
if (PRINTOK(c)) |
| 1352 |
int d = c; |
{ |
| 1353 |
int i, j, s; |
if (f != NULL) fprintf(f, "%c", c); |
| 1354 |
|
return 1; |
| 1355 |
|
} |
| 1356 |
|
|
| 1357 |
for (i = -1; i < 6; i++) /* i is number of additional bytes */ |
if (c < 0x100) |
| 1358 |
{ |
{ |
| 1359 |
if ((d & 0x80) == 0) break; |
if (use_utf) |
| 1360 |
d <<= 1; |
{ |
| 1361 |
|
if (f != NULL) fprintf(f, "\\x{%02x}", c); |
| 1362 |
|
return 6; |
| 1363 |
|
} |
| 1364 |
|
else |
| 1365 |
|
{ |
| 1366 |
|
if (f != NULL) fprintf(f, "\\x%02x", c); |
| 1367 |
|
return 4; |
| 1368 |
|
} |
| 1369 |
} |
} |
| 1370 |
|
|
| 1371 |
if (i == -1) { *vptr = c; return 1; } /* ascii character */ |
if (f != NULL) fprintf(f, "\\x{%02x}", c); |
| 1372 |
if (i == 0 || i == 6) return 0; /* invalid UTF-8 */ |
return (c <= 0x000000ff)? 6 : |
| 1373 |
|
(c <= 0x00000fff)? 7 : |
| 1374 |
|
(c <= 0x0000ffff)? 8 : |
| 1375 |
|
(c <= 0x000fffff)? 9 : 10; |
| 1376 |
|
} |
| 1377 |
|
|
|
/* i now has a value in the range 1-5 */ |
|
| 1378 |
|
|
|
s = 6*i; |
|
|
d = (c & utf8_table3[i]) << s; |
|
| 1379 |
|
|
| 1380 |
for (j = 0; j < i; j++) |
#ifdef SUPPORT_PCRE8 |
| 1381 |
{ |
/************************************************* |
| 1382 |
c = *utf8bytes++; |
* Print 8-bit character string * |
| 1383 |
if ((c & 0xc0) != 0x80) return -(j+1); |
*************************************************/ |
|
s -= 6; |
|
|
d |= (c & 0x3f) << s; |
|
|
} |
|
| 1384 |
|
|
| 1385 |
/* Check that encoding was the correct unique one */ |
/* Must handle UTF-8 strings in utf8 mode. Yields number of characters printed. |
| 1386 |
|
If handed a NULL file, just counts chars without printing. */ |
| 1387 |
|
|
| 1388 |
for (j = 0; j < utf8_table1_size; j++) |
static int pchars(pcre_uint8 *p, int length, FILE *f) |
| 1389 |
if (d <= utf8_table1[j]) break; |
{ |
| 1390 |
if (j != i) return -(i+1); |
int c = 0; |
| 1391 |
|
int yield = 0; |
| 1392 |
|
|
| 1393 |
/* Valid value */ |
if (length < 0) |
| 1394 |
|
length = strlen((char *)p); |
| 1395 |
|
|
| 1396 |
*vptr = d; |
while (length-- > 0) |
| 1397 |
return i+1; |
{ |
| 1398 |
} |
#if !defined NOUTF |
| 1399 |
|
if (use_utf) |
| 1400 |
|
{ |
| 1401 |
|
int rc = utf82ord(p, &c); |
| 1402 |
|
if (rc > 0 && rc <= length + 1) /* Mustn't run over the end */ |
| 1403 |
|
{ |
| 1404 |
|
length -= rc - 1; |
| 1405 |
|
p += rc; |
| 1406 |
|
yield += pchar(c, f); |
| 1407 |
|
continue; |
| 1408 |
|
} |
| 1409 |
|
} |
| 1410 |
|
#endif |
| 1411 |
|
c = *p++; |
| 1412 |
|
yield += pchar(c, f); |
| 1413 |
|
} |
| 1414 |
|
|
| 1415 |
|
return yield; |
| 1416 |
|
} |
| 1417 |
#endif |
#endif |
| 1418 |
|
|
| 1419 |
|
|
| 1420 |
|
|
| 1421 |
|
#ifdef SUPPORT_PCRE16 |
| 1422 |
/************************************************* |
/************************************************* |
| 1423 |
* Convert character value to UTF-8 * |
* Find length of 0-terminated 16-bit string * |
| 1424 |
*************************************************/ |
*************************************************/ |
| 1425 |
|
|
| 1426 |
/* This function takes an integer value in the range 0 - 0x7fffffff |
static int strlen16(PCRE_SPTR16 p) |
|
and encodes it as a UTF-8 character in 0 to 6 bytes. |
|
|
|
|
|
Arguments: |
|
|
cvalue the character value |
|
|
utf8bytes pointer to buffer for result - at least 6 bytes long |
|
|
|
|
|
Returns: number of characters placed in the buffer |
|
|
*/ |
|
|
|
|
|
#if !defined NOUTF8 |
|
|
|
|
|
static int |
|
|
ord2utf8(int cvalue, uschar *utf8bytes) |
|
| 1427 |
{ |
{ |
| 1428 |
register int i, j; |
int len = 0; |
| 1429 |
for (i = 0; i < utf8_table1_size; i++) |
while (*p++ != 0) len++; |
| 1430 |
if (cvalue <= utf8_table1[i]) break; |
return len; |
|
utf8bytes += i; |
|
|
for (j = i; j > 0; j--) |
|
|
{ |
|
|
*utf8bytes-- = 0x80 | (cvalue & 0x3f); |
|
|
cvalue >>= 6; |
|
|
} |
|
|
*utf8bytes = utf8_table2[i] | cvalue; |
|
|
return i + 1; |
|
| 1431 |
} |
} |
| 1432 |
|
#endif /* SUPPORT_PCRE16 */ |
|
#endif |
|
|
|
|
| 1433 |
|
|
| 1434 |
|
|
| 1435 |
|
#ifdef SUPPORT_PCRE16 |
| 1436 |
/************************************************* |
/************************************************* |
| 1437 |
* Print character string * |
* Print 16-bit character string * |
| 1438 |
*************************************************/ |
*************************************************/ |
| 1439 |
|
|
| 1440 |
/* Character string printing function. Must handle UTF-8 strings in utf8 |
/* Must handle UTF-16 strings in utf mode. Yields number of characters printed. |
| 1441 |
mode. Yields number of characters printed. If handed a NULL file, just counts |
If handed a NULL file, just counts chars without printing. */ |
|
chars without printing. */ |
|
| 1442 |
|
|
| 1443 |
static int pchars(unsigned char *p, int length, FILE *f) |
static int pchars16(PCRE_SPTR16 p, int length, FILE *f) |
| 1444 |
{ |
{ |
|
int c = 0; |
|
| 1445 |
int yield = 0; |
int yield = 0; |
| 1446 |
|
|
| 1447 |
|
if (length < 0) |
| 1448 |
|
length = strlen16(p); |
| 1449 |
|
|
| 1450 |
while (length-- > 0) |
while (length-- > 0) |
| 1451 |
{ |
{ |
| 1452 |
#if !defined NOUTF8 |
int c = *p++ & 0xffff; |
| 1453 |
if (use_utf8) |
#if !defined NOUTF |
| 1454 |
|
if (use_utf && c >= 0xD800 && c < 0xDC00 && length > 0) |
| 1455 |
{ |
{ |
| 1456 |
int rc = utf82ord(p, &c); |
int d = *p & 0xffff; |
| 1457 |
|
if (d >= 0xDC00 && d < 0xDFFF) |
|
if (rc > 0 && rc <= length + 1) /* Mustn't run over the end */ |
|
| 1458 |
{ |
{ |
| 1459 |
length -= rc - 1; |
c = ((c & 0x3ff) << 10) + (d & 0x3ff) + 0x10000; |
| 1460 |
p += rc; |
length--; |
| 1461 |
if (PRINTHEX(c)) |
p++; |
|
{ |
|
|
if (f != NULL) fprintf(f, "%c", c); |
|
|
yield++; |
|
|
} |
|
|
else |
|
|
{ |
|
|
int n = 4; |
|
|
if (f != NULL) fprintf(f, "\\x{%02x}", c); |
|
|
yield += (n <= 0x000000ff)? 2 : |
|
|
(n <= 0x00000fff)? 3 : |
|
|
(n <= 0x0000ffff)? 4 : |
|
|
(n <= 0x000fffff)? 5 : 6; |
|
|
} |
|
|
continue; |
|
| 1462 |
} |
} |
| 1463 |
} |
} |
| 1464 |
#endif |
#endif |
| 1465 |
|
yield += pchar(c, f); |
| 1466 |
|
} |
| 1467 |
|
|
| 1468 |
/* Not UTF-8, or malformed UTF-8 */ |
return yield; |
| 1469 |
|
} |
| 1470 |
|
#endif /* SUPPORT_PCRE16 */ |
| 1471 |
|
|
| 1472 |
c = *p++; |
|
| 1473 |
if (PRINTHEX(c)) |
|
| 1474 |
{ |
#ifdef SUPPORT_PCRE8 |
| 1475 |
if (f != NULL) fprintf(f, "%c", c); |
/************************************************* |
| 1476 |
yield++; |
* Read a capture name (8-bit) and check it * |
| 1477 |
} |
*************************************************/ |
| 1478 |
else |
|
| 1479 |
{ |
static pcre_uint8 * |
| 1480 |
if (f != NULL) fprintf(f, "\\x%02x", c); |
read_capture_name8(pcre_uint8 *p, pcre_uint8 **pp, pcre *re) |
| 1481 |
yield += 4; |
{ |
| 1482 |
} |
pcre_uint8 *npp = *pp; |
| 1483 |
|
while (isalnum(*p)) *npp++ = *p++; |
| 1484 |
|
*npp++ = 0; |
| 1485 |
|
*npp = 0; |
| 1486 |
|
if (pcre_get_stringnumber(re, (char *)(*pp)) < 0) |
| 1487 |
|
{ |
| 1488 |
|
fprintf(outfile, "no parentheses with name \""); |
| 1489 |
|
PCHARSV(*pp, 0, -1, outfile); |
| 1490 |
|
fprintf(outfile, "\"\n"); |
| 1491 |
} |
} |
| 1492 |
|
|
| 1493 |
return yield; |
*pp = npp; |
| 1494 |
|
return p; |
| 1495 |
} |
} |
| 1496 |
|
#endif /* SUPPORT_PCRE8 */ |
| 1497 |
|
|
| 1498 |
|
|
| 1499 |
|
|
| 1500 |
|
#ifdef SUPPORT_PCRE16 |
| 1501 |
|
/************************************************* |
| 1502 |
|
* Read a capture name (16-bit) and check it * |
| 1503 |
|
*************************************************/ |
| 1504 |
|
|
| 1505 |
|
/* Note that the text being read is 8-bit. */ |
| 1506 |
|
|
| 1507 |
|
static pcre_uint8 * |
| 1508 |
|
read_capture_name16(pcre_uint8 *p, pcre_uint16 **pp, pcre *re) |
| 1509 |
|
{ |
| 1510 |
|
pcre_uint16 *npp = *pp; |
| 1511 |
|
while (isalnum(*p)) *npp++ = *p++; |
| 1512 |
|
*npp++ = 0; |
| 1513 |
|
*npp = 0; |
| 1514 |
|
if (pcre16_get_stringnumber(re, (PCRE_SPTR16)(*pp)) < 0) |
| 1515 |
|
{ |
| 1516 |
|
fprintf(outfile, "no parentheses with name \""); |
| 1517 |
|
PCHARSV(*pp, 0, -1, outfile); |
| 1518 |
|
fprintf(outfile, "\"\n"); |
| 1519 |
|
} |
| 1520 |
|
*pp = npp; |
| 1521 |
|
return p; |
| 1522 |
|
} |
| 1523 |
|
#endif /* SUPPORT_PCRE16 */ |
| 1524 |
|
|
| 1525 |
|
|
| 1526 |
|
|
| 1549 |
else |
else |
| 1550 |
{ |
{ |
| 1551 |
fprintf(f, "%2d: ", i/2); |
fprintf(f, "%2d: ", i/2); |
| 1552 |
(void)pchars((unsigned char *)cb->subject + cb->offset_vector[i], |
PCHARSV(cb->subject, cb->offset_vector[i], |
| 1553 |
cb->offset_vector[i+1] - cb->offset_vector[i], f); |
cb->offset_vector[i+1] - cb->offset_vector[i], f); |
| 1554 |
fprintf(f, "\n"); |
fprintf(f, "\n"); |
| 1555 |
} |
} |
| 1562 |
|
|
| 1563 |
if (f != NULL) fprintf(f, "--->"); |
if (f != NULL) fprintf(f, "--->"); |
| 1564 |
|
|
| 1565 |
pre_start = pchars((unsigned char *)cb->subject, cb->start_match, f); |
PCHARS(pre_start, cb->subject, 0, cb->start_match, f); |
| 1566 |
post_start = pchars((unsigned char *)(cb->subject + cb->start_match), |
PCHARS(post_start, cb->subject, cb->start_match, |
| 1567 |
cb->current_position - cb->start_match, f); |
cb->current_position - cb->start_match, f); |
| 1568 |
|
|
| 1569 |
subject_length = pchars((unsigned char *)cb->subject, cb->subject_length, NULL); |
PCHARS(subject_length, cb->subject, 0, cb->subject_length, NULL); |
| 1570 |
|
|
| 1571 |
(void)pchars((unsigned char *)(cb->subject + cb->current_position), |
PCHARSV(cb->subject, cb->current_position, |
| 1572 |
cb->subject_length - cb->current_position, f); |
cb->subject_length - cb->current_position, f); |
| 1573 |
|
|
| 1574 |
if (f != NULL) fprintf(f, "\n"); |
if (f != NULL) fprintf(f, "\n"); |
| 1607 |
|
|
| 1608 |
if (cb->mark != last_callout_mark) |
if (cb->mark != last_callout_mark) |
| 1609 |
{ |
{ |
| 1610 |
fprintf(outfile, "Latest Mark: %s\n", |
if (cb->mark == NULL) |
| 1611 |
(cb->mark == NULL)? "<unset>" : (char *)(cb->mark)); |
fprintf(outfile, "Latest Mark: <unset>\n"); |
| 1612 |
|
else |
| 1613 |
|
{ |
| 1614 |
|
fprintf(outfile, "Latest Mark: "); |
| 1615 |
|
PCHARSV(cb->mark, 0, -1, outfile); |
| 1616 |
|
putc('\n', outfile); |
| 1617 |
|
} |
| 1618 |
last_callout_mark = cb->mark; |
last_callout_mark = cb->mark; |
| 1619 |
} |
} |
| 1620 |
|
|
| 1638 |
*************************************************/ |
*************************************************/ |
| 1639 |
|
|
| 1640 |
/* Alternative malloc function, to test functionality and save the size of a |
/* Alternative malloc function, to test functionality and save the size of a |
| 1641 |
compiled re. The show_malloc variable is set only during matching. */ |
compiled re, which is the first store request that pcre_compile() makes. The |
| 1642 |
|
show_malloc variable is set only during matching. */ |
| 1643 |
|
|
| 1644 |
static void *new_malloc(size_t size) |
static void *new_malloc(size_t size) |
| 1645 |
{ |
{ |
| 1646 |
void *block = malloc(size); |
void *block = malloc(size); |
| 1647 |
gotten_store = size; |
gotten_store = size; |
| 1648 |
|
if (first_gotten_store == 0) first_gotten_store = size; |
| 1649 |
if (show_malloc) |
if (show_malloc) |
| 1650 |
fprintf(outfile, "malloc %3d %p\n", (int)size, block); |
fprintf(outfile, "malloc %3d %p\n", (int)size, block); |
| 1651 |
return block; |
return block; |
| 1677 |
|
|
| 1678 |
|
|
| 1679 |
/************************************************* |
/************************************************* |
| 1680 |
* Call pcre_fullinfo() * |
* Call pcre_fullinfo() * |
| 1681 |
|
*************************************************/ |
| 1682 |
|
|
| 1683 |
|
/* Get one piece of information from the pcre_fullinfo() function. When only |
| 1684 |
|
one of 8-bit or 16-bit is supported, use_pcre16 should always have the correct |
| 1685 |
|
value, but the code is defensive. |
| 1686 |
|
|
| 1687 |
|
Arguments: |
| 1688 |
|
re compiled regex |
| 1689 |
|
study study data |
| 1690 |
|
option PCRE_INFO_xxx option |
| 1691 |
|
ptr where to put the data |
| 1692 |
|
|
| 1693 |
|
Returns: 0 when OK, < 0 on error |
| 1694 |
|
*/ |
| 1695 |
|
|
| 1696 |
|
static int |
| 1697 |
|
new_info(pcre *re, pcre_extra *study, int option, void *ptr) |
| 1698 |
|
{ |
| 1699 |
|
int rc; |
| 1700 |
|
|
| 1701 |
|
if (use_pcre16) |
| 1702 |
|
#ifdef SUPPORT_PCRE16 |
| 1703 |
|
rc = pcre16_fullinfo(re, study, option, ptr); |
| 1704 |
|
#else |
| 1705 |
|
rc = PCRE_ERROR_BADMODE; |
| 1706 |
|
#endif |
| 1707 |
|
else |
| 1708 |
|
#ifdef SUPPORT_PCRE8 |
| 1709 |
|
rc = pcre_fullinfo(re, study, option, ptr); |
| 1710 |
|
#else |
| 1711 |
|
rc = PCRE_ERROR_BADMODE; |
| 1712 |
|
#endif |
| 1713 |
|
|
| 1714 |
|
if (rc < 0) |
| 1715 |
|
{ |
| 1716 |
|
fprintf(outfile, "Error %d from pcre%s_fullinfo(%d)\n", rc, |
| 1717 |
|
use_pcre16? "16" : "", option); |
| 1718 |
|
if (rc == PCRE_ERROR_BADMODE) |
| 1719 |
|
fprintf(outfile, "Running in %s-bit mode but pattern was compiled in " |
| 1720 |
|
"%s-bit mode\n", use_pcre16? "16":"8", use_pcre16? "8":"16"); |
| 1721 |
|
} |
| 1722 |
|
|
| 1723 |
|
return rc; |
| 1724 |
|
} |
| 1725 |
|
|
| 1726 |
|
|
| 1727 |
|
|
| 1728 |
|
/************************************************* |
| 1729 |
|
* Swap byte functions * |
| 1730 |
|
*************************************************/ |
| 1731 |
|
|
| 1732 |
|
/* The following functions swap the bytes of a pcre_uint16 and pcre_uint32 |
| 1733 |
|
value, respectively. |
| 1734 |
|
|
| 1735 |
|
Arguments: |
| 1736 |
|
value any number |
| 1737 |
|
|
| 1738 |
|
Returns: the byte swapped value |
| 1739 |
|
*/ |
| 1740 |
|
|
| 1741 |
|
static pcre_uint32 |
| 1742 |
|
swap_uint32(pcre_uint32 value) |
| 1743 |
|
{ |
| 1744 |
|
return ((value & 0x000000ff) << 24) | |
| 1745 |
|
((value & 0x0000ff00) << 8) | |
| 1746 |
|
((value & 0x00ff0000) >> 8) | |
| 1747 |
|
(value >> 24); |
| 1748 |
|
} |
| 1749 |
|
|
| 1750 |
|
static pcre_uint16 |
| 1751 |
|
swap_uint16(pcre_uint16 value) |
| 1752 |
|
{ |
| 1753 |
|
return (value >> 8) | (value << 8); |
| 1754 |
|
} |
| 1755 |
|
|
| 1756 |
|
|
| 1757 |
|
|
| 1758 |
|
/************************************************* |
| 1759 |
|
* Flip bytes in a compiled pattern * |
| 1760 |
*************************************************/ |
*************************************************/ |
| 1761 |
|
|
| 1762 |
/* Get one piece of information from the pcre_fullinfo() function */ |
/* This function is called if the 'F' option was present on a pattern that is |
| 1763 |
|
to be written to a file. We flip the bytes of all the integer fields in the |
| 1764 |
|
regex data block and the study block. In 16-bit mode this also flips relevant |
| 1765 |
|
bytes in the pattern itself. This is to make it possible to test PCRE's |
| 1766 |
|
ability to reload byte-flipped patterns, e.g. those compiled on a different |
| 1767 |
|
architecture. */ |
| 1768 |
|
|
| 1769 |
static void new_info(pcre *re, pcre_extra *study, int option, void *ptr) |
static void |
| 1770 |
|
regexflip(pcre *ere, pcre_extra *extra) |
| 1771 |
{ |
{ |
| 1772 |
int rc; |
real_pcre *re = (real_pcre *)ere; |
| 1773 |
if ((rc = pcre_fullinfo(re, study, option, ptr)) < 0) |
#ifdef SUPPORT_PCRE16 |
| 1774 |
fprintf(outfile, "Error %d from pcre_fullinfo(%d)\n", rc, option); |
int op; |
| 1775 |
} |
pcre_uint16 *ptr = (pcre_uint16 *)re + re->name_table_offset; |
| 1776 |
|
int length = re->name_count * re->name_entry_size; |
| 1777 |
|
#ifdef SUPPORT_UTF |
| 1778 |
|
BOOL utf = (re->options & PCRE_UTF16) != 0; |
| 1779 |
|
BOOL utf16_char = FALSE; |
| 1780 |
|
#endif /* SUPPORT_UTF */ |
| 1781 |
|
#endif /* SUPPORT_PCRE16 */ |
| 1782 |
|
|
| 1783 |
|
/* Always flip the bytes in the main data block and study blocks. */ |
| 1784 |
|
|
| 1785 |
|
re->magic_number = REVERSED_MAGIC_NUMBER; |
| 1786 |
|
re->size = swap_uint32(re->size); |
| 1787 |
|
re->options = swap_uint32(re->options); |
| 1788 |
|
re->flags = swap_uint16(re->flags); |
| 1789 |
|
re->top_bracket = swap_uint16(re->top_bracket); |
| 1790 |
|
re->top_backref = swap_uint16(re->top_backref); |
| 1791 |
|
re->first_char = swap_uint16(re->first_char); |
| 1792 |
|
re->req_char = swap_uint16(re->req_char); |
| 1793 |
|
re->name_table_offset = swap_uint16(re->name_table_offset); |
| 1794 |
|
re->name_entry_size = swap_uint16(re->name_entry_size); |
| 1795 |
|
re->name_count = swap_uint16(re->name_count); |
| 1796 |
|
|
| 1797 |
|
if (extra != NULL) |
| 1798 |
|
{ |
| 1799 |
|
pcre_study_data *rsd = (pcre_study_data *)(extra->study_data); |
| 1800 |
|
rsd->size = swap_uint32(rsd->size); |
| 1801 |
|
rsd->flags = swap_uint32(rsd->flags); |
| 1802 |
|
rsd->minlength = swap_uint32(rsd->minlength); |
| 1803 |
|
} |
| 1804 |
|
|
| 1805 |
|
/* In 8-bit mode, that is all we need to do. In 16-bit mode we must swap bytes |
| 1806 |
|
in the name table, if present, and then in the pattern itself. */ |
| 1807 |
|
|
| 1808 |
/************************************************* |
#ifdef SUPPORT_PCRE16 |
| 1809 |
* Byte flipping function * |
if (!use_pcre16) return; |
|
*************************************************/ |
|
| 1810 |
|
|
| 1811 |
static unsigned long int |
while(TRUE) |
| 1812 |
byteflip(unsigned long int value, int n) |
{ |
| 1813 |
{ |
/* Swap previous characters. */ |
| 1814 |
if (n == 2) return ((value & 0x00ff) << 8) | ((value & 0xff00) >> 8); |
while (length-- > 0) |
| 1815 |
return ((value & 0x000000ff) << 24) | |
{ |
| 1816 |
((value & 0x0000ff00) << 8) | |
*ptr = swap_uint16(*ptr); |
| 1817 |
((value & 0x00ff0000) >> 8) | |
ptr++; |
| 1818 |
((value & 0xff000000) >> 24); |
} |
| 1819 |
} |
#ifdef SUPPORT_UTF |
| 1820 |
|
if (utf16_char) |
| 1821 |
|
{ |
| 1822 |
|
if ((ptr[-1] & 0xfc00) == 0xd800) |
| 1823 |
|
{ |
| 1824 |
|
/* We know that there is only one extra character in UTF-16. */ |
| 1825 |
|
*ptr = swap_uint16(*ptr); |
| 1826 |
|
ptr++; |
| 1827 |
|
} |
| 1828 |
|
} |
| 1829 |
|
utf16_char = FALSE; |
| 1830 |
|
#endif /* SUPPORT_UTF */ |
| 1831 |
|
|
| 1832 |
|
/* Get next opcode. */ |
| 1833 |
|
|
| 1834 |
|
length = 0; |
| 1835 |
|
op = *ptr; |
| 1836 |
|
*ptr++ = swap_uint16(op); |
| 1837 |
|
|
| 1838 |
|
switch (op) |
| 1839 |
|
{ |
| 1840 |
|
case OP_END: |
| 1841 |
|
return; |
| 1842 |
|
|
| 1843 |
|
#ifdef SUPPORT_UTF |
| 1844 |
|
case OP_CHAR: |
| 1845 |
|
case OP_CHARI: |
| 1846 |
|
case OP_NOT: |
| 1847 |
|
case OP_NOTI: |
| 1848 |
|
case OP_STAR: |
| 1849 |
|
case OP_MINSTAR: |
| 1850 |
|
case OP_PLUS: |
| 1851 |
|
case OP_MINPLUS: |
| 1852 |
|
case OP_QUERY: |
| 1853 |
|
case OP_MINQUERY: |
| 1854 |
|
case OP_UPTO: |
| 1855 |
|
case OP_MINUPTO: |
| 1856 |
|
case OP_EXACT: |
| 1857 |
|
case OP_POSSTAR: |
| 1858 |
|
case OP_POSPLUS: |
| 1859 |
|
case OP_POSQUERY: |
| 1860 |
|
case OP_POSUPTO: |
| 1861 |
|
case OP_STARI: |
| 1862 |
|
case OP_MINSTARI: |
| 1863 |
|
case OP_PLUSI: |
| 1864 |
|
case OP_MINPLUSI: |
| 1865 |
|
case OP_QUERYI: |
| 1866 |
|
case OP_MINQUERYI: |
| 1867 |
|
case OP_UPTOI: |
| 1868 |
|
case OP_MINUPTOI: |
| 1869 |
|
case OP_EXACTI: |
| 1870 |
|
case OP_POSSTARI: |
| 1871 |
|
case OP_POSPLUSI: |
| 1872 |
|
case OP_POSQUERYI: |
| 1873 |
|
case OP_POSUPTOI: |
| 1874 |
|
case OP_NOTSTAR: |
| 1875 |
|
case OP_NOTMINSTAR: |
| 1876 |
|
case OP_NOTPLUS: |
| 1877 |
|
case OP_NOTMINPLUS: |
| 1878 |
|
case OP_NOTQUERY: |
| 1879 |
|
case OP_NOTMINQUERY: |
| 1880 |
|
case OP_NOTUPTO: |
| 1881 |
|
case OP_NOTMINUPTO: |
| 1882 |
|
case OP_NOTEXACT: |
| 1883 |
|
case OP_NOTPOSSTAR: |
| 1884 |
|
case OP_NOTPOSPLUS: |
| 1885 |
|
case OP_NOTPOSQUERY: |
| 1886 |
|
case OP_NOTPOSUPTO: |
| 1887 |
|
case OP_NOTSTARI: |
| 1888 |
|
case OP_NOTMINSTARI: |
| 1889 |
|
case OP_NOTPLUSI: |
| 1890 |
|
case OP_NOTMINPLUSI: |
| 1891 |
|
case OP_NOTQUERYI: |
| 1892 |
|
case OP_NOTMINQUERYI: |
| 1893 |
|
case OP_NOTUPTOI: |
| 1894 |
|
case OP_NOTMINUPTOI: |
| 1895 |
|
case OP_NOTEXACTI: |
| 1896 |
|
case OP_NOTPOSSTARI: |
| 1897 |
|
case OP_NOTPOSPLUSI: |
| 1898 |
|
case OP_NOTPOSQUERYI: |
| 1899 |
|
case OP_NOTPOSUPTOI: |
| 1900 |
|
if (utf) utf16_char = TRUE; |
| 1901 |
|
#endif |
| 1902 |
|
/* Fall through. */ |
| 1903 |
|
|
| 1904 |
|
default: |
| 1905 |
|
length = OP_lengths16[op] - 1; |
| 1906 |
|
break; |
| 1907 |
|
|
| 1908 |
|
case OP_CLASS: |
| 1909 |
|
case OP_NCLASS: |
| 1910 |
|
/* Skip the character bit map. */ |
| 1911 |
|
ptr += 32/sizeof(pcre_uint16); |
| 1912 |
|
length = 0; |
| 1913 |
|
break; |
| 1914 |
|
|
| 1915 |
|
case OP_XCLASS: |
| 1916 |
|
/* Reverse the size of the XCLASS instance. */ |
| 1917 |
|
ptr++; |
| 1918 |
|
*ptr = swap_uint16(*ptr); |
| 1919 |
|
if (LINK_SIZE > 1) |
| 1920 |
|
{ |
| 1921 |
|
/* LINK_SIZE can be 1 or 2 in 16 bit mode. */ |
| 1922 |
|
ptr++; |
| 1923 |
|
*ptr = swap_uint16(*ptr); |
| 1924 |
|
} |
| 1925 |
|
ptr++; |
| 1926 |
|
|
| 1927 |
|
if (LINK_SIZE > 1) |
| 1928 |
|
length = ((ptr[-LINK_SIZE] << 16) | ptr[-LINK_SIZE + 1]) - |
| 1929 |
|
(1 + LINK_SIZE + 1); |
| 1930 |
|
else |
| 1931 |
|
length = ptr[-LINK_SIZE] - (1 + LINK_SIZE + 1); |
| 1932 |
|
|
| 1933 |
|
op = *ptr; |
| 1934 |
|
*ptr = swap_uint16(op); |
| 1935 |
|
if ((op & XCL_MAP) != 0) |
| 1936 |
|
{ |
| 1937 |
|
/* Skip the character bit map. */ |
| 1938 |
|
ptr += 32/sizeof(pcre_uint16); |
| 1939 |
|
length -= 32/sizeof(pcre_uint16); |
| 1940 |
|
} |
| 1941 |
|
break; |
| 1942 |
|
} |
| 1943 |
|
} |
| 1944 |
|
/* Control should never reach here in 16 bit mode. */ |
| 1945 |
|
#endif /* SUPPORT_PCRE16 */ |
| 1946 |
|
} |
| 1947 |
|
|
| 1948 |
|
|
| 1949 |
|
|
| 1952 |
*************************************************/ |
*************************************************/ |
| 1953 |
|
|
| 1954 |
static int |
static int |
| 1955 |
check_match_limit(pcre *re, pcre_extra *extra, uschar *bptr, int len, |
check_match_limit(pcre *re, pcre_extra *extra, pcre_uint8 *bptr, int len, |
| 1956 |
int start_offset, int options, int *use_offsets, int use_size_offsets, |
int start_offset, int options, int *use_offsets, int use_size_offsets, |
| 1957 |
int flag, unsigned long int *limit, int errnumber, const char *msg) |
int flag, unsigned long int *limit, int errnumber, const char *msg) |
| 1958 |
{ |
{ |
| 1967 |
{ |
{ |
| 1968 |
*limit = mid; |
*limit = mid; |
| 1969 |
|
|
| 1970 |
count = pcre_exec(re, extra, (char *)bptr, len, start_offset, options, |
PCRE_EXEC(count, re, extra, bptr, len, start_offset, options, |
| 1971 |
use_offsets, use_size_offsets); |
use_offsets, use_size_offsets); |
| 1972 |
|
|
| 1973 |
if (count == errnumber) |
if (count == errnumber) |
| 2012 |
*/ |
*/ |
| 2013 |
|
|
| 2014 |
static int |
static int |
| 2015 |
strncmpic(uschar *s, uschar *t, int n) |
strncmpic(pcre_uint8 *s, pcre_uint8 *t, int n) |
| 2016 |
{ |
{ |
| 2017 |
while (n--) |
while (n--) |
| 2018 |
{ |
{ |
| 2039 |
*/ |
*/ |
| 2040 |
|
|
| 2041 |
static int |
static int |
| 2042 |
check_newline(uschar *p, FILE *f) |
check_newline(pcre_uint8 *p, FILE *f) |
| 2043 |
{ |
{ |
| 2044 |
if (strncmpic(p, (uschar *)"cr>", 3) == 0) return PCRE_NEWLINE_CR; |
if (strncmpic(p, (pcre_uint8 *)"cr>", 3) == 0) return PCRE_NEWLINE_CR; |
| 2045 |
if (strncmpic(p, (uschar *)"lf>", 3) == 0) return PCRE_NEWLINE_LF; |
if (strncmpic(p, (pcre_uint8 *)"lf>", 3) == 0) return PCRE_NEWLINE_LF; |
| 2046 |
if (strncmpic(p, (uschar *)"crlf>", 5) == 0) return PCRE_NEWLINE_CRLF; |
if (strncmpic(p, (pcre_uint8 *)"crlf>", 5) == 0) return PCRE_NEWLINE_CRLF; |
| 2047 |
if (strncmpic(p, (uschar *)"anycrlf>", 8) == 0) return PCRE_NEWLINE_ANYCRLF; |
if (strncmpic(p, (pcre_uint8 *)"anycrlf>", 8) == 0) return PCRE_NEWLINE_ANYCRLF; |
| 2048 |
if (strncmpic(p, (uschar *)"any>", 4) == 0) return PCRE_NEWLINE_ANY; |
if (strncmpic(p, (pcre_uint8 *)"any>", 4) == 0) return PCRE_NEWLINE_ANY; |
| 2049 |
if (strncmpic(p, (uschar *)"bsr_anycrlf>", 12) == 0) return PCRE_BSR_ANYCRLF; |
if (strncmpic(p, (pcre_uint8 *)"bsr_anycrlf>", 12) == 0) return PCRE_BSR_ANYCRLF; |
| 2050 |
if (strncmpic(p, (uschar *)"bsr_unicode>", 12) == 0) return PCRE_BSR_UNICODE; |
if (strncmpic(p, (pcre_uint8 *)"bsr_unicode>", 12) == 0) return PCRE_BSR_UNICODE; |
| 2051 |
fprintf(f, "Unknown newline type at: <%s\n", p); |
fprintf(f, "Unknown newline type at: <%s\n", p); |
| 2052 |
return 0; |
return 0; |
| 2053 |
} |
} |
| 2069 |
printf("This version of pcretest is not linked with readline().\n"); |
printf("This version of pcretest is not linked with readline().\n"); |
| 2070 |
#endif |
#endif |
| 2071 |
printf("\nOptions:\n"); |
printf("\nOptions:\n"); |
| 2072 |
|
#ifdef SUPPORT_PCRE16 |
| 2073 |
|
printf(" -16 use 16-bit interface\n"); |
| 2074 |
|
#endif |
| 2075 |
printf(" -b show compiled code (bytecode)\n"); |
printf(" -b show compiled code (bytecode)\n"); |
| 2076 |
printf(" -C show PCRE compile-time options and exit\n"); |
printf(" -C show PCRE compile-time options and exit\n"); |
| 2077 |
|
printf(" -C arg show a specific compile-time option\n"); |
| 2078 |
|
printf(" and exit with its value. The arg can be:\n"); |
| 2079 |
|
printf(" linksize internal link size [2, 3, 4]\n"); |
| 2080 |
|
printf(" pcre8 8 bit library support enabled [0, 1]\n"); |
| 2081 |
|
printf(" pcre16 16 bit library support enabled [0, 1]\n"); |
| 2082 |
|
printf(" utf Unicode Transformation Format supported [0, 1]\n"); |
| 2083 |
|
printf(" ucp Unicode Properties supported [0, 1]\n"); |
| 2084 |
|
printf(" jit Just-in-time compiler supported [0, 1]\n"); |
| 2085 |
printf(" -d debug: show compiled code and information (-b and -i)\n"); |
printf(" -d debug: show compiled code and information (-b and -i)\n"); |
| 2086 |
#if !defined NODFA |
#if !defined NODFA |
| 2087 |
printf(" -dfa force DFA matching for all subjects\n"); |
printf(" -dfa force DFA matching for all subjects\n"); |
| 2117 |
int main(int argc, char **argv) |
int main(int argc, char **argv) |
| 2118 |
{ |
{ |
| 2119 |
FILE *infile = stdin; |
FILE *infile = stdin; |
| 2120 |
|
const char *version; |
| 2121 |
int options = 0; |
int options = 0; |
| 2122 |
int study_options = 0; |
int study_options = 0; |
| 2123 |
int default_find_match_limit = FALSE; |
int default_find_match_limit = FALSE; |
| 2143 |
|
|
| 2144 |
pcre_jit_stack *jit_stack = NULL; |
pcre_jit_stack *jit_stack = NULL; |
| 2145 |
|
|
| 2146 |
|
/* These vectors store, end-to-end, a list of zero-terminated captured |
| 2147 |
/* These vectors store, end-to-end, a list of captured substring names. Assume |
substring names, each list itself being terminated by an empty name. Assume |
| 2148 |
that 1024 is plenty long enough for the few names we'll be testing. */ |
that 1024 is plenty long enough for the few names we'll be testing. It is |
| 2149 |
|
easiest to keep separate 8-bit and 16-bit versions, using the 16-bit version |
| 2150 |
uschar copynames[1024]; |
for the actual memory, to ensure alignment. By defining these variables always |
| 2151 |
uschar getnames[1024]; |
(whether or not 8-bit or 16-bit is supported), we avoid too much mess with |
| 2152 |
|
#ifdefs in the code. */ |
| 2153 |
uschar *copynamesptr; |
|
| 2154 |
uschar *getnamesptr; |
pcre_uint16 copynames[1024]; |
| 2155 |
|
pcre_uint16 getnames[1024]; |
| 2156 |
/* Get buffers from malloc() so that Electric Fence will check their misuse |
|
| 2157 |
when I am debugging. They grow automatically when very long lines are read. */ |
pcre_uint16 *cn16ptr; |
| 2158 |
|
pcre_uint16 *gn16ptr; |
| 2159 |
buffer = (unsigned char *)malloc(buffer_size); |
|
| 2160 |
dbuffer = (unsigned char *)malloc(buffer_size); |
pcre_uint8 *copynames8 = (pcre_uint8 *)copynames; |
| 2161 |
pbuffer = (unsigned char *)malloc(buffer_size); |
pcre_uint8 *getnames8 = (pcre_uint8 *)getnames; |
| 2162 |
|
pcre_uint8 *cn8ptr; |
| 2163 |
|
pcre_uint8 *gn8ptr; |
| 2164 |
|
|
| 2165 |
|
/* Get buffers from malloc() so that valgrind will check their misuse when |
| 2166 |
|
debugging. They grow automatically when very long lines are read. The 16-bit |
| 2167 |
|
buffer (buffer16) is obtained only if needed. */ |
| 2168 |
|
|
| 2169 |
|
buffer = (pcre_uint8 *)malloc(buffer_size); |
| 2170 |
|
dbuffer = (pcre_uint8 *)malloc(buffer_size); |
| 2171 |
|
pbuffer = (pcre_uint8 *)malloc(buffer_size); |
| 2172 |
|
|
| 2173 |
/* The outfile variable is static so that new_malloc can use it. */ |
/* The outfile variable is static so that new_malloc can use it. */ |
| 2174 |
|
|
| 2183 |
_setmode( _fileno( stdout ), _O_BINARY ); |
_setmode( _fileno( stdout ), _O_BINARY ); |
| 2184 |
#endif |
#endif |
| 2185 |
|
|
| 2186 |
|
/* Get the version number: both pcre_version() and pcre16_version() give the |
| 2187 |
|
same answer. We just need to ensure that we call one that is available. */ |
| 2188 |
|
|
| 2189 |
|
#ifdef SUPPORT_PCRE8 |
| 2190 |
|
version = pcre_version(); |
| 2191 |
|
#else |
| 2192 |
|
version = pcre16_version(); |
| 2193 |
|
#endif |
| 2194 |
|
|
| 2195 |
/* Scan options */ |
/* Scan options */ |
| 2196 |
|
|
| 2197 |
while (argc > 1 && argv[op][0] == '-') |
while (argc > 1 && argv[op][0] == '-') |
| 2198 |
{ |
{ |
| 2199 |
unsigned char *endptr; |
pcre_uint8 *endptr; |
| 2200 |
|
|
| 2201 |
if (strcmp(argv[op], "-m") == 0) showstore = 1; |
if (strcmp(argv[op], "-m") == 0) showstore = 1; |
| 2202 |
else if (strcmp(argv[op], "-s") == 0) force_study = 0; |
else if (strcmp(argv[op], "-s") == 0) force_study = 0; |
| 2205 |
force_study = 1; |
force_study = 1; |
| 2206 |
force_study_options = PCRE_STUDY_JIT_COMPILE; |
force_study_options = PCRE_STUDY_JIT_COMPILE; |
| 2207 |
} |
} |
| 2208 |
|
else if (strcmp(argv[op], "-16") == 0) |
| 2209 |
|
{ |
| 2210 |
|
#ifdef SUPPORT_PCRE16 |
| 2211 |
|
use_pcre16 = 1; |
| 2212 |
|
#else |
| 2213 |
|
printf("** This version of PCRE was built without 16-bit support\n"); |
| 2214 |
|
exit(1); |
| 2215 |
|
#endif |
| 2216 |
|
} |
| 2217 |
else if (strcmp(argv[op], "-q") == 0) quiet = 1; |
else if (strcmp(argv[op], "-q") == 0) quiet = 1; |
| 2218 |
else if (strcmp(argv[op], "-b") == 0) debug = 1; |
else if (strcmp(argv[op], "-b") == 0) debug = 1; |
| 2219 |
else if (strcmp(argv[op], "-i") == 0) showinfo = 1; |
else if (strcmp(argv[op], "-i") == 0) showinfo = 1; |
| 2223 |
else if (strcmp(argv[op], "-dfa") == 0) all_use_dfa = 1; |
else if (strcmp(argv[op], "-dfa") == 0) all_use_dfa = 1; |
| 2224 |
#endif |
#endif |
| 2225 |
else if (strcmp(argv[op], "-o") == 0 && argc > 2 && |
else if (strcmp(argv[op], "-o") == 0 && argc > 2 && |
| 2226 |
((size_offsets = get_value((unsigned char *)argv[op+1], &endptr)), |
((size_offsets = get_value((pcre_uint8 *)argv[op+1], &endptr)), |
| 2227 |
*endptr == 0)) |
*endptr == 0)) |
| 2228 |
{ |
{ |
| 2229 |
op++; |
op++; |
| 2233 |
{ |
{ |
| 2234 |
int both = argv[op][2] == 0; |
int both = argv[op][2] == 0; |
| 2235 |
int temp; |
int temp; |
| 2236 |
if (argc > 2 && (temp = get_value((unsigned char *)argv[op+1], &endptr), |
if (argc > 2 && (temp = get_value((pcre_uint8 *)argv[op+1], &endptr), |
| 2237 |
*endptr == 0)) |
*endptr == 0)) |
| 2238 |
{ |
{ |
| 2239 |
timeitm = temp; |
timeitm = temp; |
| 2244 |
if (both) timeit = timeitm; |
if (both) timeit = timeitm; |
| 2245 |
} |
} |
| 2246 |
else if (strcmp(argv[op], "-S") == 0 && argc > 2 && |
else if (strcmp(argv[op], "-S") == 0 && argc > 2 && |
| 2247 |
((stack_size = get_value((unsigned char *)argv[op+1], &endptr)), |
((stack_size = get_value((pcre_uint8 *)argv[op+1], &endptr)), |
| 2248 |
*endptr == 0)) |
*endptr == 0)) |
| 2249 |
{ |
{ |
| 2250 |
#if defined(_WIN32) || defined(WIN32) || defined(__minix) |
#if defined(_WIN32) || defined(WIN32) || defined(__minix) |
| 2272 |
{ |
{ |
| 2273 |
int rc; |
int rc; |
| 2274 |
unsigned long int lrc; |
unsigned long int lrc; |
| 2275 |
printf("PCRE version %s\n", pcre_version()); |
|
| 2276 |
|
if (argc > 2) |
| 2277 |
|
{ |
| 2278 |
|
if (strcmp(argv[op + 1], "linksize") == 0) |
| 2279 |
|
{ |
| 2280 |
|
(void)PCRE_CONFIG(PCRE_CONFIG_LINK_SIZE, &rc); |
| 2281 |
|
printf("%d\n", rc); |
| 2282 |
|
yield = rc; |
| 2283 |
|
goto EXIT; |
| 2284 |
|
} |
| 2285 |
|
if (strcmp(argv[op + 1], "pcre8") == 0) |
| 2286 |
|
{ |
| 2287 |
|
#ifdef SUPPORT_PCRE8 |
| 2288 |
|
printf("1\n"); |
| 2289 |
|
yield = 1; |
| 2290 |
|
#else |
| 2291 |
|
printf("0\n"); |
| 2292 |
|
yield = 0; |
| 2293 |
|
#endif |
| 2294 |
|
goto EXIT; |
| 2295 |
|
} |
| 2296 |
|
if (strcmp(argv[op + 1], "pcre16") == 0) |
| 2297 |
|
{ |
| 2298 |
|
#ifdef SUPPORT_PCRE16 |
| 2299 |
|
printf("1\n"); |
| 2300 |
|
yield = 1; |
| 2301 |
|
#else |
| 2302 |
|
printf("0\n"); |
| 2303 |
|
yield = 0; |
| 2304 |
|
#endif |
| 2305 |
|
goto EXIT; |
| 2306 |
|
} |
| 2307 |
|
if (strcmp(argv[op + 1], "utf") == 0) |
| 2308 |
|
{ |
| 2309 |
|
#ifdef SUPPORT_PCRE8 |
| 2310 |
|
(void)pcre_config(PCRE_CONFIG_UTF8, &rc); |
| 2311 |
|
printf("%d\n", rc); |
| 2312 |
|
yield = rc; |
| 2313 |
|
#else |
| 2314 |
|
(void)pcre16_config(PCRE_CONFIG_UTF16, &rc); |
| 2315 |
|
printf("%d\n", rc); |
| 2316 |
|
yield = rc; |
| 2317 |
|
#endif |
| 2318 |
|
goto EXIT; |
| 2319 |
|
} |
| 2320 |
|
if (strcmp(argv[op + 1], "ucp") == 0) |
| 2321 |
|
{ |
| 2322 |
|
(void)PCRE_CONFIG(PCRE_CONFIG_UNICODE_PROPERTIES, &rc); |
| 2323 |
|
printf("%d\n", rc); |
| 2324 |
|
yield = rc; |
| 2325 |
|
goto EXIT; |
| 2326 |
|
} |
| 2327 |
|
if (strcmp(argv[op + 1], "jit") == 0) |
| 2328 |
|
{ |
| 2329 |
|
(void)PCRE_CONFIG(PCRE_CONFIG_JIT, &rc); |
| 2330 |
|
printf("%d\n", rc); |
| 2331 |
|
yield = rc; |
| 2332 |
|
goto EXIT; |
| 2333 |
|
} |
| 2334 |
|
printf("Unknown option: %s\n", argv[op + 1]); |
| 2335 |
|
goto EXIT; |
| 2336 |
|
} |
| 2337 |
|
|
| 2338 |
|
printf("PCRE version %s\n", version); |
| 2339 |
printf("Compiled with\n"); |
printf("Compiled with\n"); |
| 2340 |
|
|
| 2341 |
|
/* At least one of SUPPORT_PCRE8 and SUPPORT_PCRE16 will be set. If both |
| 2342 |
|
are set, either both UTFs are supported or both are not supported. */ |
| 2343 |
|
|
| 2344 |
|
#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE16 |
| 2345 |
|
printf(" 8-bit and 16-bit support\n"); |
| 2346 |
|
(void)pcre_config(PCRE_CONFIG_UTF8, &rc); |
| 2347 |
|
if (rc) |
| 2348 |
|
printf(" UTF-8 and UTF-16 support\n"); |
| 2349 |
|
else |
| 2350 |
|
printf(" No UTF-8 or UTF-16 support\n"); |
| 2351 |
|
#elif defined SUPPORT_PCRE8 |
| 2352 |
|
printf(" 8-bit support only\n"); |
| 2353 |
(void)pcre_config(PCRE_CONFIG_UTF8, &rc); |
(void)pcre_config(PCRE_CONFIG_UTF8, &rc); |
| 2354 |
printf(" %sUTF-8 support\n", rc? "" : "No "); |
printf(" %sUTF-8 support\n", rc? "" : "No "); |
| 2355 |
(void)pcre_config(PCRE_CONFIG_UNICODE_PROPERTIES, &rc); |
#else |
| 2356 |
|
printf(" 16-bit support only\n"); |
| 2357 |
|
(void)pcre16_config(PCRE_CONFIG_UTF16, &rc); |
| 2358 |
|
printf(" %sUTF-16 support\n", rc? "" : "No "); |
| 2359 |
|
#endif |
| 2360 |
|
|
| 2361 |
|
(void)PCRE_CONFIG(PCRE_CONFIG_UNICODE_PROPERTIES, &rc); |
| 2362 |
printf(" %sUnicode properties support\n", rc? "" : "No "); |
printf(" %sUnicode properties support\n", rc? "" : "No "); |
| 2363 |
(void)pcre_config(PCRE_CONFIG_JIT, &rc); |
(void)PCRE_CONFIG(PCRE_CONFIG_JIT, &rc); |
| 2364 |
if (rc) |
if (rc) |
| 2365 |
printf(" Just-in-time compiler support\n"); |
printf(" Just-in-time compiler support\n"); |
| 2366 |
else |
else |
| 2367 |
printf(" No just-in-time compiler support\n"); |
printf(" No just-in-time compiler support\n"); |
| 2368 |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &rc); |
(void)PCRE_CONFIG(PCRE_CONFIG_NEWLINE, &rc); |
| 2369 |
/* Note that these values are always the ASCII values, even |
/* Note that these values are always the ASCII values, even |
| 2370 |
in EBCDIC environments. CR is 13 and NL is 10. */ |
in EBCDIC environments. CR is 13 and NL is 10. */ |
| 2371 |
printf(" Newline sequence is %s\n", (rc == 13)? "CR" : |
printf(" Newline sequence is %s\n", (rc == 13)? "CR" : |
| 2372 |
(rc == 10)? "LF" : (rc == (13<<8 | 10))? "CRLF" : |
(rc == 10)? "LF" : (rc == (13<<8 | 10))? "CRLF" : |
| 2373 |
(rc == -2)? "ANYCRLF" : |
(rc == -2)? "ANYCRLF" : |
| 2374 |
(rc == -1)? "ANY" : "???"); |
(rc == -1)? "ANY" : "???"); |
| 2375 |
(void)pcre_config(PCRE_CONFIG_BSR, &rc); |
(void)PCRE_CONFIG(PCRE_CONFIG_BSR, &rc); |
| 2376 |
printf(" \\R matches %s\n", rc? "CR, LF, or CRLF only" : |
printf(" \\R matches %s\n", rc? "CR, LF, or CRLF only" : |
| 2377 |
"all Unicode newlines"); |
"all Unicode newlines"); |
| 2378 |
(void)pcre_config(PCRE_CONFIG_LINK_SIZE, &rc); |
(void)PCRE_CONFIG(PCRE_CONFIG_LINK_SIZE, &rc); |
| 2379 |
printf(" Internal link size = %d\n", rc); |
printf(" Internal link size = %d\n", rc); |
| 2380 |
(void)pcre_config(PCRE_CONFIG_POSIX_MALLOC_THRESHOLD, &rc); |
(void)PCRE_CONFIG(PCRE_CONFIG_POSIX_MALLOC_THRESHOLD, &rc); |
| 2381 |
printf(" POSIX malloc threshold = %d\n", rc); |
printf(" POSIX malloc threshold = %d\n", rc); |
| 2382 |
(void)pcre_config(PCRE_CONFIG_MATCH_LIMIT, &lrc); |
(void)PCRE_CONFIG(PCRE_CONFIG_MATCH_LIMIT, &lrc); |
| 2383 |
printf(" Default match limit = %ld\n", lrc); |
printf(" Default match limit = %ld\n", lrc); |
| 2384 |
(void)pcre_config(PCRE_CONFIG_MATCH_LIMIT_RECURSION, &lrc); |
(void)PCRE_CONFIG(PCRE_CONFIG_MATCH_LIMIT_RECURSION, &lrc); |
| 2385 |
printf(" Default recursion depth limit = %ld\n", lrc); |
printf(" Default recursion depth limit = %ld\n", lrc); |
| 2386 |
(void)pcre_config(PCRE_CONFIG_STACKRECURSE, &rc); |
(void)PCRE_CONFIG(PCRE_CONFIG_STACKRECURSE, &rc); |
| 2387 |
printf(" Match recursion uses %s\n", rc? "stack" : "heap"); |
printf(" Match recursion uses %s\n", rc? "stack" : "heap"); |
| 2388 |
goto EXIT; |
goto EXIT; |
| 2389 |
} |
} |
| 2442 |
|
|
| 2443 |
/* Set alternative malloc function */ |
/* Set alternative malloc function */ |
| 2444 |
|
|
| 2445 |
|
#ifdef SUPPORT_PCRE8 |
| 2446 |
pcre_malloc = new_malloc; |
pcre_malloc = new_malloc; |
| 2447 |
pcre_free = new_free; |
pcre_free = new_free; |
| 2448 |
pcre_stack_malloc = stack_malloc; |
pcre_stack_malloc = stack_malloc; |
| 2449 |
pcre_stack_free = stack_free; |
pcre_stack_free = stack_free; |
| 2450 |
|
#endif |
| 2451 |
|
|
| 2452 |
|
#ifdef SUPPORT_PCRE16 |
| 2453 |
|
pcre16_malloc = new_malloc; |
| 2454 |
|
pcre16_free = new_free; |
| 2455 |
|
pcre16_stack_malloc = stack_malloc; |
| 2456 |
|
pcre16_stack_free = stack_free; |
| 2457 |
|
#endif |
| 2458 |
|
|
| 2459 |
/* Heading line unless quiet, then prompt for first regex if stdin */ |
/* Heading line unless quiet, then prompt for first regex if stdin */ |
| 2460 |
|
|
| 2461 |
if (!quiet) fprintf(outfile, "PCRE version %s\n\n", pcre_version()); |
if (!quiet) fprintf(outfile, "PCRE version %s\n\n", version); |
| 2462 |
|
|
| 2463 |
/* Main loop */ |
/* Main loop */ |
| 2464 |
|
|
| 2473 |
#endif |
#endif |
| 2474 |
|
|
| 2475 |
const char *error; |
const char *error; |
| 2476 |
unsigned char *markptr; |
pcre_uint8 *markptr; |
| 2477 |
unsigned char *p, *pp, *ppp; |
pcre_uint8 *p, *pp, *ppp; |
| 2478 |
unsigned char *to_file = NULL; |
pcre_uint8 *to_file = NULL; |
| 2479 |
const unsigned char *tables = NULL; |
const pcre_uint8 *tables = NULL; |
| 2480 |
unsigned long int true_size, true_study_size = 0; |
unsigned long int true_size, true_study_size = 0; |
| 2481 |
size_t size, regex_gotten_store; |
size_t size, regex_gotten_store; |
| 2482 |
int do_allcaps = 0; |
int do_allcaps = 0; |
| 2492 |
int do_flip = 0; |
int do_flip = 0; |
| 2493 |
int erroroffset, len, delimiter, poffset; |
int erroroffset, len, delimiter, poffset; |
| 2494 |
|
|
| 2495 |
use_utf8 = 0; |
use_utf = 0; |
| 2496 |
debug_lengths = 1; |
debug_lengths = 1; |
| 2497 |
|
|
| 2498 |
if (extend_inputline(infile, buffer, " re> ") == NULL) break; |
if (extend_inputline(infile, buffer, " re> ") == NULL) break; |
| 2508 |
if (*p == '<' && strchr((char *)(p+1), '<') == NULL) |
if (*p == '<' && strchr((char *)(p+1), '<') == NULL) |
| 2509 |
{ |
{ |
| 2510 |
unsigned long int magic, get_options; |
unsigned long int magic, get_options; |
| 2511 |
uschar sbuf[8]; |
pcre_uint8 sbuf[8]; |
| 2512 |
FILE *f; |
FILE *f; |
| 2513 |
|
|
| 2514 |
p++; |
p++; |
| 2531 |
(sbuf[4] << 24) | (sbuf[5] << 16) | (sbuf[6] << 8) | sbuf[7]; |
(sbuf[4] << 24) | (sbuf[5] << 16) | (sbuf[6] << 8) | sbuf[7]; |
| 2532 |
|
|
| 2533 |
re = (real_pcre *)new_malloc(true_size); |
re = (real_pcre *)new_malloc(true_size); |
| 2534 |
regex_gotten_store = gotten_store; |
regex_gotten_store = first_gotten_store; |
| 2535 |
|
|
| 2536 |
if (fread(re, 1, true_size, f) != true_size) goto FAIL_READ; |
if (fread(re, 1, true_size, f) != true_size) goto FAIL_READ; |
| 2537 |
|
|
| 2538 |
magic = ((real_pcre *)re)->magic_number; |
magic = ((real_pcre *)re)->magic_number; |
| 2539 |
if (magic != MAGIC_NUMBER) |
if (magic != MAGIC_NUMBER) |
| 2540 |
{ |
{ |
| 2541 |
if (byteflip(magic, sizeof(magic)) == MAGIC_NUMBER) |
if (swap_uint32(magic) == MAGIC_NUMBER) |
| 2542 |
{ |
{ |
| 2543 |
do_flip = 1; |
do_flip = 1; |
| 2544 |
} |
} |
| 2553 |
fprintf(outfile, "Compiled pattern%s loaded from %s\n", |
fprintf(outfile, "Compiled pattern%s loaded from %s\n", |
| 2554 |
do_flip? " (byte-inverted)" : "", p); |
do_flip? " (byte-inverted)" : "", p); |
| 2555 |
|
|
|
/* Need to know if UTF-8 for printing data strings */ |
|
|
|
|
|
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
|
|
use_utf8 = (get_options & PCRE_UTF8) != 0; |
|
|
|
|
| 2556 |
/* Now see if there is any following study data. */ |
/* Now see if there is any following study data. */ |
| 2557 |
|
|
| 2558 |
if (true_study_size != 0) |
if (true_study_size != 0) |
| 2569 |
{ |
{ |
| 2570 |
FAIL_READ: |
FAIL_READ: |
| 2571 |
fprintf(outfile, "Failed to read data from %s\n", p); |
fprintf(outfile, "Failed to read data from %s\n", p); |
| 2572 |
if (extra != NULL) pcre_free_study(extra); |
if (extra != NULL) |
| 2573 |
|
{ |
| 2574 |
|
PCRE_FREE_STUDY(extra); |
| 2575 |
|
} |
| 2576 |
if (re != NULL) new_free(re); |
if (re != NULL) new_free(re); |
| 2577 |
fclose(f); |
fclose(f); |
| 2578 |
continue; |
continue; |
| 2582 |
} |
} |
| 2583 |
else fprintf(outfile, "No study data\n"); |
else fprintf(outfile, "No study data\n"); |
| 2584 |
|
|
| 2585 |
|
/* Flip the necessary bytes. */ |
| 2586 |
|
if (do_flip) |
| 2587 |
|
{ |
| 2588 |
|
PCRE_PATTERN_TO_HOST_BYTE_ORDER(re, extra, NULL); |
| 2589 |
|
} |
| 2590 |
|
|
| 2591 |
|
/* Need to know if UTF-8 for printing data strings. */ |
| 2592 |
|
|
| 2593 |
|
if (new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options) < 0) continue; |
| 2594 |
|
use_utf = (get_options & PCRE_UTF8) != 0; |
| 2595 |
|
|
| 2596 |
fclose(f); |
fclose(f); |
| 2597 |
goto SHOW_INFO; |
goto SHOW_INFO; |
| 2598 |
} |
} |
| 2599 |
|
|
| 2600 |
/* In-line pattern (the usual case). Get the delimiter and seek the end of |
/* In-line pattern (the usual case). Get the delimiter and seek the end of |
| 2601 |
the pattern; if is isn't complete, read more. */ |
the pattern; if it isn't complete, read more. */ |
| 2602 |
|
|
| 2603 |
delimiter = *p++; |
delimiter = *p++; |
| 2604 |
|
|
| 2649 |
/* Look for options after final delimiter */ |
/* Look for options after final delimiter */ |
| 2650 |
|
|
| 2651 |
options = 0; |
options = 0; |
| 2652 |
|
study_options = 0; |
| 2653 |
log_store = showstore; /* default from command line */ |
log_store = showstore; /* default from command line */ |
| 2654 |
|
|
| 2655 |
while (*pp != 0) |
while (*pp != 0) |
| 2707 |
case 'X': options |= PCRE_EXTRA; break; |
case 'X': options |= PCRE_EXTRA; break; |
| 2708 |
case 'Y': options |= PCRE_NO_START_OPTIMISE; break; |
case 'Y': options |= PCRE_NO_START_OPTIMISE; break; |
| 2709 |
case 'Z': debug_lengths = 0; break; |
case 'Z': debug_lengths = 0; break; |
| 2710 |
case '8': options |= PCRE_UTF8; use_utf8 = 1; break; |
case '8': options |= PCRE_UTF8; use_utf = 1; break; |
| 2711 |
case '?': options |= PCRE_NO_UTF8_CHECK; break; |
case '?': options |= PCRE_NO_UTF8_CHECK; break; |
| 2712 |
|
|
| 2713 |
case 'T': |
case 'T': |
| 2741 |
goto SKIP_DATA; |
goto SKIP_DATA; |
| 2742 |
} |
} |
| 2743 |
locale_set = 1; |
locale_set = 1; |
| 2744 |
tables = pcre_maketables(); |
tables = PCRE_MAKETABLES; |
| 2745 |
pp = ppp; |
pp = ppp; |
| 2746 |
break; |
break; |
| 2747 |
|
|
| 2754 |
|
|
| 2755 |
case '<': |
case '<': |
| 2756 |
{ |
{ |
| 2757 |
if (strncmpic(pp, (uschar *)"JS>", 3) == 0) |
if (strncmpic(pp, (pcre_uint8 *)"JS>", 3) == 0) |
| 2758 |
{ |
{ |
| 2759 |
options |= PCRE_JAVASCRIPT_COMPAT; |
options |= PCRE_JAVASCRIPT_COMPAT; |
| 2760 |
pp += 3; |
pp += 3; |
| 2782 |
|
|
| 2783 |
/* Handle compiling via the POSIX interface, which doesn't support the |
/* Handle compiling via the POSIX interface, which doesn't support the |
| 2784 |
timing, showing, or debugging options, nor the ability to pass over |
timing, showing, or debugging options, nor the ability to pass over |
| 2785 |
local character tables. */ |
local character tables. Neither does it have 16-bit support. */ |
| 2786 |
|
|
| 2787 |
#if !defined NOPOSIX |
#if !defined NOPOSIX |
| 2788 |
if (posix || do_posix) |
if (posix || do_posix) |
| 2798 |
if ((options & PCRE_UCP) != 0) cflags |= REG_UCP; |
if ((options & PCRE_UCP) != 0) cflags |= REG_UCP; |
| 2799 |
if ((options & PCRE_UNGREEDY) != 0) cflags |= REG_UNGREEDY; |
if ((options & PCRE_UNGREEDY) != 0) cflags |= REG_UNGREEDY; |
| 2800 |
|
|
| 2801 |
|
first_gotten_store = 0; |
| 2802 |
rc = regcomp(&preg, (char *)p, cflags); |
rc = regcomp(&preg, (char *)p, cflags); |
| 2803 |
|
|
| 2804 |
/* Compilation failed; go back for another re, skipping to blank line |
/* Compilation failed; go back for another re, skipping to blank line |
| 2820 |
{ |
{ |
| 2821 |
unsigned long int get_options; |
unsigned long int get_options; |
| 2822 |
|
|
| 2823 |
|
/* In 16-bit mode, convert the input. */ |
| 2824 |
|
|
| 2825 |
|
#ifdef SUPPORT_PCRE16 |
| 2826 |
|
if (use_pcre16) |
| 2827 |
|
{ |
| 2828 |
|
switch(to16(FALSE, p, options & PCRE_UTF8, (int)strlen((char *)p))) |
| 2829 |
|
{ |
| 2830 |
|
case -1: |
| 2831 |
|
fprintf(outfile, "**Failed: invalid UTF-8 string cannot be " |
| 2832 |
|
"converted to UTF-16\n"); |
| 2833 |
|
goto SKIP_DATA; |
| 2834 |
|
|
| 2835 |
|
case -2: |
| 2836 |
|
fprintf(outfile, "**Failed: character value greater than 0x10ffff " |
| 2837 |
|
"cannot be converted to UTF-16\n"); |
| 2838 |
|
goto SKIP_DATA; |
| 2839 |
|
|
| 2840 |
|
case -3: /* "Impossible error" when to16 is called arg1 FALSE */ |
| 2841 |
|
fprintf(outfile, "**Failed: character value greater than 0xffff " |
| 2842 |
|
"cannot be converted to 16-bit in non-UTF mode\n"); |
| 2843 |
|
goto SKIP_DATA; |
| 2844 |
|
|
| 2845 |
|
default: |
| 2846 |
|
break; |
| 2847 |
|
} |
| 2848 |
|
p = (pcre_uint8 *)buffer16; |
| 2849 |
|
} |
| 2850 |
|
#endif |
| 2851 |
|
|
| 2852 |
|
/* Compile many times when timing */ |
| 2853 |
|
|
| 2854 |
if (timeit > 0) |
if (timeit > 0) |
| 2855 |
{ |
{ |
| 2856 |
register int i; |
register int i; |
| 2858 |
clock_t start_time = clock(); |
clock_t start_time = clock(); |
| 2859 |
for (i = 0; i < timeit; i++) |
for (i = 0; i < timeit; i++) |
| 2860 |
{ |
{ |
| 2861 |
re = pcre_compile((char *)p, options, &error, &erroroffset, tables); |
PCRE_COMPILE(re, p, options, &error, &erroroffset, tables); |
| 2862 |
if (re != NULL) free(re); |
if (re != NULL) free(re); |
| 2863 |
} |
} |
| 2864 |
time_taken = clock() - start_time; |
time_taken = clock() - start_time; |
| 2867 |
(double)CLOCKS_PER_SEC); |
(double)CLOCKS_PER_SEC); |
| 2868 |
} |
} |
| 2869 |
|
|
| 2870 |
re = pcre_compile((char *)p, options, &error, &erroroffset, tables); |
first_gotten_store = 0; |
| 2871 |
|
PCRE_COMPILE(re, p, options, &error, &erroroffset, tables); |
| 2872 |
|
|
| 2873 |
/* Compilation failed; go back for another re, skipping to blank line |
/* Compilation failed; go back for another re, skipping to blank line |
| 2874 |
if non-interactive. */ |
if non-interactive. */ |
| 2899 |
within the regex; check for this so that we know how to process the data |
within the regex; check for this so that we know how to process the data |
| 2900 |
lines. */ |
lines. */ |
| 2901 |
|
|
| 2902 |
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
if (new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options) < 0) |
| 2903 |
if ((get_options & PCRE_UTF8) != 0) use_utf8 = 1; |
goto SKIP_DATA; |
| 2904 |
|
if ((get_options & PCRE_UTF8) != 0) use_utf = 1; |
| 2905 |
|
|
| 2906 |
|
/* Extract the size for possible writing before possibly flipping it, |
| 2907 |
|
and remember the store that was got. */ |
| 2908 |
|
|
| 2909 |
|
true_size = ((real_pcre *)re)->size; |
| 2910 |
|
regex_gotten_store = first_gotten_store; |
| 2911 |
|
|
| 2912 |
/* Print information if required. There are now two info-returning |
/* Output code size information if requested */ |
|
functions. The old one has a limited interface and returns only limited |
|
|
data. Check that it agrees with the newer one. */ |
|
| 2913 |
|
|
| 2914 |
if (log_store) |
if (log_store) |
| 2915 |
fprintf(outfile, "Memory allocation (code space): %d\n", |
fprintf(outfile, "Memory allocation (code space): %d\n", |
| 2916 |
(int)(gotten_store - |
(int)(first_gotten_store - |
| 2917 |
sizeof(real_pcre) - |
sizeof(real_pcre) - |
| 2918 |
((real_pcre *)re)->name_count * ((real_pcre *)re)->name_entry_size)); |
((real_pcre *)re)->name_count * ((real_pcre *)re)->name_entry_size)); |
| 2919 |
|
|
|
/* Extract the size for possible writing before possibly flipping it, |
|
|
and remember the store that was got. */ |
|
|
|
|
|
true_size = ((real_pcre *)re)->size; |
|
|
regex_gotten_store = gotten_store; |
|
|
|
|
| 2920 |
/* If -s or /S was present, study the regex to generate additional info to |
/* If -s or /S was present, study the regex to generate additional info to |
| 2921 |
help with the matching, unless the pattern has the SS option, which |
help with the matching, unless the pattern has the SS option, which |
| 2922 |
suppresses the effect of /S (used for a few test patterns where studying is |
suppresses the effect of /S (used for a few test patterns where studying is |
| 2930 |
clock_t time_taken; |
clock_t time_taken; |
| 2931 |
clock_t start_time = clock(); |
clock_t start_time = clock(); |
| 2932 |
for (i = 0; i < timeit; i++) |
for (i = 0; i < timeit; i++) |
| 2933 |
extra = pcre_study(re, study_options | force_study_options, &error); |
{ |
| 2934 |
|
PCRE_STUDY(extra, re, study_options | force_study_options, &error); |
| 2935 |
|
} |
| 2936 |
time_taken = clock() - start_time; |
time_taken = clock() - start_time; |
| 2937 |
if (extra != NULL) pcre_free_study(extra); |
if (extra != NULL) |
| 2938 |
|
{ |
| 2939 |
|
PCRE_FREE_STUDY(extra); |
| 2940 |
|
} |
| 2941 |
fprintf(outfile, " Study time %.4f milliseconds\n", |
fprintf(outfile, " Study time %.4f milliseconds\n", |
| 2942 |
(((double)time_taken * 1000.0) / (double)timeit) / |
(((double)time_taken * 1000.0) / (double)timeit) / |
| 2943 |
(double)CLOCKS_PER_SEC); |
(double)CLOCKS_PER_SEC); |
| 2944 |
} |
} |
| 2945 |
extra = pcre_study(re, study_options | force_study_options, &error); |
PCRE_STUDY(extra, re, study_options | force_study_options, &error); |
| 2946 |
if (error != NULL) |
if (error != NULL) |
| 2947 |
fprintf(outfile, "Failed to study: %s\n", error); |
fprintf(outfile, "Failed to study: %s\n", error); |
| 2948 |
else if (extra != NULL) |
else if (extra != NULL) |
| 2949 |
|
{ |
| 2950 |
true_study_size = ((pcre_study_data *)(extra->study_data))->size; |
true_study_size = ((pcre_study_data *)(extra->study_data))->size; |
| 2951 |
|
if (log_store) |
| 2952 |
|
{ |
| 2953 |
|
size_t jitsize; |
| 2954 |
|
if (new_info(re, extra, PCRE_INFO_JITSIZE, &jitsize) == 0 && |
| 2955 |
|
jitsize != 0) |
| 2956 |
|
fprintf(outfile, "Memory allocation (JIT code): %d\n", (int)jitsize); |
| 2957 |
|
} |
| 2958 |
|
} |
| 2959 |
} |
} |
| 2960 |
|
|
| 2961 |
/* If /K was present, we set up for handling MARK data. */ |
/* If /K was present, we set up for handling MARK data. */ |
| 2971 |
extra->flags |= PCRE_EXTRA_MARK; |
extra->flags |= PCRE_EXTRA_MARK; |
| 2972 |
} |
} |
| 2973 |
|
|
| 2974 |
/* If the 'F' option was present, we flip the bytes of all the integer |
/* Extract and display information from the compiled data if required. */ |
|
fields in the regex data block and the study block. This is to make it |
|
|
possible to test PCRE's handling of byte-flipped patterns, e.g. those |
|
|
compiled on a different architecture. */ |
|
|
|
|
|
if (do_flip) |
|
|
{ |
|
|
real_pcre *rre = (real_pcre *)re; |
|
|
rre->magic_number = |
|
|
byteflip(rre->magic_number, sizeof(rre->magic_number)); |
|
|
rre->size = byteflip(rre->size, sizeof(rre->size)); |
|
|
rre->options = byteflip(rre->options, sizeof(rre->options)); |
|
|
rre->flags = (pcre_uint16)byteflip(rre->flags, sizeof(rre->flags)); |
|
|
rre->top_bracket = |
|
|
(pcre_uint16)byteflip(rre->top_bracket, sizeof(rre->top_bracket)); |
|
|
rre->top_backref = |
|
|
(pcre_uint16)byteflip(rre->top_backref, sizeof(rre->top_backref)); |
|
|
rre->first_byte = |
|
|
(pcre_uint16)byteflip(rre->first_byte, sizeof(rre->first_byte)); |
|
|
rre->req_byte = |
|
|
(pcre_uint16)byteflip(rre->req_byte, sizeof(rre->req_byte)); |
|
|
rre->name_table_offset = (pcre_uint16)byteflip(rre->name_table_offset, |
|
|
sizeof(rre->name_table_offset)); |
|
|
rre->name_entry_size = (pcre_uint16)byteflip(rre->name_entry_size, |
|
|
sizeof(rre->name_entry_size)); |
|
|
rre->name_count = (pcre_uint16)byteflip(rre->name_count, |
|
|
sizeof(rre->name_count)); |
|
|
|
|
|
if (extra != NULL) |
|
|
{ |
|
|
pcre_study_data *rsd = (pcre_study_data *)(extra->study_data); |
|
|
rsd->size = byteflip(rsd->size, sizeof(rsd->size)); |
|
|
rsd->flags = byteflip(rsd->flags, sizeof(rsd->flags)); |
|
|
rsd->minlength = byteflip(rsd->minlength, sizeof(rsd->minlength)); |
|
|
} |
|
|
} |
|
|
|
|
|
/* Extract information from the compiled data if required */ |
|
| 2975 |
|
|
| 2976 |
SHOW_INFO: |
SHOW_INFO: |
| 2977 |
|
|
| 2978 |
if (do_debug) |
if (do_debug) |
| 2979 |
{ |
{ |
| 2980 |
fprintf(outfile, "------------------------------------------------------------------\n"); |
fprintf(outfile, "------------------------------------------------------------------\n"); |
| 2981 |
pcre_printint(re, outfile, debug_lengths); |
PCRE_PRINTINT(re, outfile, debug_lengths); |
| 2982 |
} |
} |
| 2983 |
|
|
| 2984 |
/* We already have the options in get_options (see above) */ |
/* We already have the options in get_options (see above) */ |
| 2986 |
if (do_showinfo) |
if (do_showinfo) |
| 2987 |
{ |
{ |
| 2988 |
unsigned long int all_options; |
unsigned long int all_options; |
|
#if !defined NOINFOCHECK |
|
|
int old_first_char, old_options, old_count; |
|
|
#endif |
|
| 2989 |
int count, backrefmax, first_char, need_char, okpartial, jchanged, |
int count, backrefmax, first_char, need_char, okpartial, jchanged, |
| 2990 |
hascrorlf; |
hascrorlf; |
| 2991 |
int nameentrysize, namecount; |
int nameentrysize, namecount; |
| 2992 |
const uschar *nametable; |
const pcre_uint8 *nametable; |
| 2993 |
|
|
| 2994 |
new_info(re, NULL, PCRE_INFO_SIZE, &size); |
if (new_info(re, NULL, PCRE_INFO_SIZE, &size) + |
| 2995 |
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count); |
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count) + |
| 2996 |
new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax); |
new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax) + |
| 2997 |
new_info(re, NULL, PCRE_INFO_FIRSTBYTE, &first_char); |
new_info(re, NULL, PCRE_INFO_FIRSTBYTE, &first_char) + |
| 2998 |
new_info(re, NULL, PCRE_INFO_LASTLITERAL, &need_char); |
new_info(re, NULL, PCRE_INFO_LASTLITERAL, &need_char) + |
| 2999 |
new_info(re, NULL, PCRE_INFO_NAMEENTRYSIZE, &nameentrysize); |
new_info(re, NULL, PCRE_INFO_NAMEENTRYSIZE, &nameentrysize) + |
| 3000 |
new_info(re, NULL, PCRE_INFO_NAMECOUNT, &namecount); |
new_info(re, NULL, PCRE_INFO_NAMECOUNT, &namecount) + |
| 3001 |
new_info(re, NULL, PCRE_INFO_NAMETABLE, (void *)&nametable); |
new_info(re, NULL, PCRE_INFO_NAMETABLE, (void *)&nametable) + |
| 3002 |
new_info(re, NULL, PCRE_INFO_OKPARTIAL, &okpartial); |
new_info(re, NULL, PCRE_INFO_OKPARTIAL, &okpartial) + |
| 3003 |
new_info(re, NULL, PCRE_INFO_JCHANGED, &jchanged); |
new_info(re, NULL, PCRE_INFO_JCHANGED, &jchanged) + |
| 3004 |
new_info(re, NULL, PCRE_INFO_HASCRORLF, &hascrorlf); |
new_info(re, NULL, PCRE_INFO_HASCRORLF, &hascrorlf) |
| 3005 |
|
!= 0) |
| 3006 |
#if !defined NOINFOCHECK |
goto SKIP_DATA; |
|
old_count = pcre_info(re, &old_options, &old_first_char); |
|
|
if (count < 0) fprintf(outfile, |
|
|
"Error %d from pcre_info()\n", count); |
|
|
else |
|
|
{ |
|
|
if (old_count != count) fprintf(outfile, |
|
|
"Count disagreement: pcre_fullinfo=%d pcre_info=%d\n", count, |
|
|
old_count); |
|
|
|
|
|
if (old_first_char != first_char) fprintf(outfile, |
|
|
"First char disagreement: pcre_fullinfo=%d pcre_info=%d\n", |
|
|
first_char, old_first_char); |
|
|
|
|
|
if (old_options != (int)get_options) fprintf(outfile, |
|
|
"Options disagreement: pcre_fullinfo=%ld pcre_info=%d\n", |
|
|
get_options, old_options); |
|
|
} |
|
|
#endif |
|
| 3007 |
|
|
| 3008 |
if (size != regex_gotten_store) fprintf(outfile, |
if (size != regex_gotten_store) fprintf(outfile, |
| 3009 |
"Size disagreement: pcre_fullinfo=%d call to malloc for %d\n", |
"Size disagreement: pcre_fullinfo=%d call to malloc for %d\n", |
| 3018 |
fprintf(outfile, "Named capturing subpatterns:\n"); |
fprintf(outfile, "Named capturing subpatterns:\n"); |
| 3019 |
while (namecount-- > 0) |
while (namecount-- > 0) |
| 3020 |
{ |
{ |
| 3021 |
fprintf(outfile, " %s %*s%3d\n", nametable + 2, |
#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE16 |
| 3022 |
nameentrysize - 3 - (int)strlen((char *)nametable + 2), "", |
int imm2_size = use_pcre16 ? 1 : 2; |
| 3023 |
GET2(nametable, 0)); |
#else |
| 3024 |
|
int imm2_size = IMM2_SIZE; |
| 3025 |
|
#endif |
| 3026 |
|
int length = (int)STRLEN(nametable + imm2_size); |
| 3027 |
|
fprintf(outfile, " "); |
| 3028 |
|
PCHARSV(nametable, imm2_size, length, outfile); |
| 3029 |
|
while (length++ < nameentrysize - imm2_size) putc(' ', outfile); |
| 3030 |
|
#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE16 |
| 3031 |
|
fprintf(outfile, "%3d\n", use_pcre16? |
| 3032 |
|
(int)(((PCRE_SPTR16)nametable)[0]) |
| 3033 |
|
:((int)nametable[0] << 8) | (int)nametable[1]); |
| 3034 |
|
nametable += nameentrysize * (use_pcre16 ? 2 : 1); |
| 3035 |
|
#else |
| 3036 |
|
fprintf(outfile, "%3d\n", GET2(nametable, 0)); |
| 3037 |
|
#ifdef SUPPORT_PCRE8 |
| 3038 |
nametable += nameentrysize; |
nametable += nameentrysize; |
| 3039 |
|
#else |
| 3040 |
|
nametable += nameentrysize * 2; |
| 3041 |
|
#endif |
| 3042 |
|
#endif |
| 3043 |
} |
} |
| 3044 |
} |
} |
| 3045 |
|
|
| 3047 |
if (hascrorlf) fprintf(outfile, "Contains explicit CR or LF match\n"); |
if (hascrorlf) fprintf(outfile, "Contains explicit CR or LF match\n"); |
| 3048 |
|
|
| 3049 |
all_options = ((real_pcre *)re)->options; |
all_options = ((real_pcre *)re)->options; |
| 3050 |
if (do_flip) all_options = byteflip(all_options, sizeof(all_options)); |
if (do_flip) all_options = swap_uint32(all_options); |
| 3051 |
|
|
| 3052 |
if (get_options == 0) fprintf(outfile, "No options\n"); |
if (get_options == 0) fprintf(outfile, "No options\n"); |
| 3053 |
else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s\n", |
else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s\n", |
| 3063 |
((get_options & PCRE_EXTRA) != 0)? " extra" : "", |
((get_options & PCRE_EXTRA) != 0)? " extra" : "", |
| 3064 |
((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "", |
((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "", |
| 3065 |
((get_options & PCRE_NO_AUTO_CAPTURE) != 0)? " no_auto_capture" : "", |
((get_options & PCRE_NO_AUTO_CAPTURE) != 0)? " no_auto_capture" : "", |
| 3066 |
((get_options & PCRE_UTF8) != 0)? " utf8" : "", |
((get_options & PCRE_UTF8) != 0)? " utf" : "", |
| 3067 |
((get_options & PCRE_UCP) != 0)? " ucp" : "", |
((get_options & PCRE_UCP) != 0)? " ucp" : "", |
| 3068 |
((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "", |
((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf_check" : "", |
| 3069 |
((get_options & PCRE_NO_START_OPTIMIZE) != 0)? " no_start_optimize" : "", |
((get_options & PCRE_NO_START_OPTIMIZE) != 0)? " no_start_optimize" : "", |
| 3070 |
((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : ""); |
((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : ""); |
| 3071 |
|
|
| 3107 |
} |
} |
| 3108 |
else |
else |
| 3109 |
{ |
{ |
| 3110 |
int ch = first_char & 255; |
const char *caseless = |
| 3111 |
const char *caseless = ((first_char & REQ_CASELESS) == 0)? |
((((real_pcre *)re)->flags & PCRE_FCH_CASELESS) == 0)? |
| 3112 |
"" : " (caseless)"; |
"" : " (caseless)"; |
| 3113 |
if (PRINTHEX(ch)) |
|
| 3114 |
fprintf(outfile, "First char = \'%c\'%s\n", ch, caseless); |
if (PRINTOK(first_char)) |
| 3115 |
|
fprintf(outfile, "First char = \'%c\'%s\n", first_char, caseless); |
| 3116 |
else |
else |
| 3117 |
fprintf(outfile, "First char = %d%s\n", ch, caseless); |
{ |
| 3118 |
|
fprintf(outfile, "First char = "); |
| 3119 |
|
pchar(first_char, outfile); |
| 3120 |
|
fprintf(outfile, "%s\n", caseless); |
| 3121 |
|
} |
| 3122 |
} |
} |
| 3123 |
|
|
| 3124 |
if (need_char < 0) |
if (need_char < 0) |
| 3127 |
} |
} |
| 3128 |
else |
else |
| 3129 |
{ |
{ |
| 3130 |
int ch = need_char & 255; |
const char *caseless = |
| 3131 |
const char *caseless = ((need_char & REQ_CASELESS) == 0)? |
((((real_pcre *)re)->flags & PCRE_RCH_CASELESS) == 0)? |
| 3132 |
"" : " (caseless)"; |
"" : " (caseless)"; |
| 3133 |
if (PRINTHEX(ch)) |
|
| 3134 |
fprintf(outfile, "Need char = \'%c\'%s\n", ch, caseless); |
if (PRINTOK(need_char)) |
| 3135 |
|
fprintf(outfile, "Need char = \'%c\'%s\n", need_char, caseless); |
| 3136 |
else |
else |
| 3137 |
fprintf(outfile, "Need char = %d%s\n", ch, caseless); |
{ |
| 3138 |
|
fprintf(outfile, "Need char = "); |
| 3139 |
|
pchar(need_char, outfile); |
| 3140 |
|
fprintf(outfile, "%s\n", caseless); |
| 3141 |
|
} |
| 3142 |
} |
} |
| 3143 |
|
|
| 3144 |
/* Don't output study size; at present it is in any case a fixed |
/* Don't output study size; at present it is in any case a fixed |
| 3155 |
fprintf(outfile, "Study returned NULL\n"); |
fprintf(outfile, "Study returned NULL\n"); |
| 3156 |
else |
else |
| 3157 |
{ |
{ |
| 3158 |
uschar *start_bits = NULL; |
pcre_uint8 *start_bits = NULL; |
| 3159 |
int minlength; |
int minlength; |
| 3160 |
|
|
| 3161 |
new_info(re, extra, PCRE_INFO_MINLENGTH, &minlength); |
if (new_info(re, extra, PCRE_INFO_MINLENGTH, &minlength) == 0) |
| 3162 |
fprintf(outfile, "Subject length lower bound = %d\n", minlength); |
fprintf(outfile, "Subject length lower bound = %d\n", minlength); |
| 3163 |
|
|
| 3164 |
new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits); |
if (new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits) == 0) |
|
if (start_bits == NULL) |
|
|
fprintf(outfile, "No set of starting bytes\n"); |
|
|
else |
|
| 3165 |
{ |
{ |
| 3166 |
int i; |
if (start_bits == NULL) |
| 3167 |
int c = 24; |
fprintf(outfile, "No set of starting bytes\n"); |
| 3168 |
fprintf(outfile, "Starting byte set: "); |
else |
|
for (i = 0; i < 256; i++) |
|
| 3169 |
{ |
{ |
| 3170 |
if ((start_bits[i/8] & (1<<(i&7))) != 0) |
int i; |
| 3171 |
|
int c = 24; |
| 3172 |
|
fprintf(outfile, "Starting byte set: "); |
| 3173 |
|
for (i = 0; i < 256; i++) |
| 3174 |
{ |
{ |
| 3175 |
if (c > 75) |
if ((start_bits[i/8] & (1<<(i&7))) != 0) |
|
{ |
|
|
fprintf(outfile, "\n "); |
|
|
c = 2; |
|
|
} |
|
|
if (PRINTHEX(i) && i != ' ') |
|
|
{ |
|
|
fprintf(outfile, "%c ", i); |
|
|
c += 2; |
|
|
} |
|
|
else |
|
| 3176 |
{ |
{ |
| 3177 |
fprintf(outfile, "\\x%02x ", i); |
if (c > 75) |
| 3178 |
c += 5; |
{ |
| 3179 |
|
fprintf(outfile, "\n "); |
| 3180 |
|
c = 2; |
| 3181 |
|
} |
| 3182 |
|
if (PRINTOK(i) && i != ' ') |
| 3183 |
|
{ |
| 3184 |
|
fprintf(outfile, "%c ", i); |
| 3185 |
|
c += 2; |
| 3186 |
|
} |
| 3187 |
|
else |
| 3188 |
|
{ |
| 3189 |
|
fprintf(outfile, "\\x%02x ", i); |
| 3190 |
|
c += 5; |
| 3191 |
|
} |
| 3192 |
} |
} |
| 3193 |
} |
} |
| 3194 |
|
fprintf(outfile, "\n"); |
| 3195 |
} |
} |
|
fprintf(outfile, "\n"); |
|
| 3196 |
} |
} |
| 3197 |
} |
} |
| 3198 |
|
|
| 3201 |
if ((study_options & PCRE_STUDY_JIT_COMPILE) != 0) |
if ((study_options & PCRE_STUDY_JIT_COMPILE) != 0) |
| 3202 |
{ |
{ |
| 3203 |
int jit; |
int jit; |
| 3204 |
new_info(re, extra, PCRE_INFO_JIT, &jit); |
if (new_info(re, extra, PCRE_INFO_JIT, &jit) == 0) |
| 3205 |
if (jit) |
{ |
| 3206 |
fprintf(outfile, "JIT study was successful\n"); |
if (jit) |
| 3207 |
else |
fprintf(outfile, "JIT study was successful\n"); |
| 3208 |
|
else |
| 3209 |
#ifdef SUPPORT_JIT |
#ifdef SUPPORT_JIT |
| 3210 |
fprintf(outfile, "JIT study was not successful\n"); |
fprintf(outfile, "JIT study was not successful\n"); |
| 3211 |
#else |
#else |
| 3212 |
fprintf(outfile, "JIT support is not available in this version of PCRE\n"); |
fprintf(outfile, "JIT support is not available in this version of PCRE\n"); |
| 3213 |
#endif |
#endif |
| 3214 |
|
} |
| 3215 |
} |
} |
| 3216 |
} |
} |
| 3217 |
} |
} |
| 3229 |
} |
} |
| 3230 |
else |
else |
| 3231 |
{ |
{ |
| 3232 |
uschar sbuf[8]; |
pcre_uint8 sbuf[8]; |
| 3233 |
sbuf[0] = (uschar)((true_size >> 24) & 255); |
|
| 3234 |
sbuf[1] = (uschar)((true_size >> 16) & 255); |
if (do_flip) regexflip(re, extra); |
| 3235 |
sbuf[2] = (uschar)((true_size >> 8) & 255); |
sbuf[0] = (pcre_uint8)((true_size >> 24) & 255); |
| 3236 |
sbuf[3] = (uschar)((true_size) & 255); |
sbuf[1] = (pcre_uint8)((true_size >> 16) & 255); |
| 3237 |
|
sbuf[2] = (pcre_uint8)((true_size >> 8) & 255); |
| 3238 |
sbuf[4] = (uschar)((true_study_size >> 24) & 255); |
sbuf[3] = (pcre_uint8)((true_size) & 255); |
| 3239 |
sbuf[5] = (uschar)((true_study_size >> 16) & 255); |
sbuf[4] = (pcre_uint8)((true_study_size >> 24) & 255); |
| 3240 |
sbuf[6] = (uschar)((true_study_size >> 8) & 255); |
sbuf[5] = (pcre_uint8)((true_study_size >> 16) & 255); |
| 3241 |
sbuf[7] = (uschar)((true_study_size) & 255); |
sbuf[6] = (pcre_uint8)((true_study_size >> 8) & 255); |
| 3242 |
|
sbuf[7] = (pcre_uint8)((true_study_size) & 255); |
| 3243 |
|
|
| 3244 |
if (fwrite(sbuf, 1, 8, f) < 8 || |
if (fwrite(sbuf, 1, 8, f) < 8 || |
| 3245 |
fwrite(re, 1, true_size, f) < true_size) |
fwrite(re, 1, true_size, f) < true_size) |
| 3267 |
} |
} |
| 3268 |
|
|
| 3269 |
new_free(re); |
new_free(re); |
| 3270 |
if (extra != NULL) pcre_free_study(extra); |
if (extra != NULL) |
| 3271 |
|
{ |
| 3272 |
|
PCRE_FREE_STUDY(extra); |
| 3273 |
|
} |
| 3274 |
if (locale_set) |
if (locale_set) |
| 3275 |
{ |
{ |
| 3276 |
new_free((void *)tables); |
new_free((void *)tables); |
| 3285 |
|
|
| 3286 |
for (;;) |
for (;;) |
| 3287 |
{ |
{ |
| 3288 |
uschar *q; |
pcre_uint8 *q; |
| 3289 |
uschar *bptr; |
pcre_uint8 *bptr; |
| 3290 |
int *use_offsets = offsets; |
int *use_offsets = offsets; |
| 3291 |
int use_size_offsets = size_offsets; |
int use_size_offsets = size_offsets; |
| 3292 |
int callout_data = 0; |
int callout_data = 0; |
| 3302 |
int g_notempty = 0; |
int g_notempty = 0; |
| 3303 |
int use_dfa = 0; |
int use_dfa = 0; |
| 3304 |
|
|
|
options = 0; |
|
|
|
|
| 3305 |
*copynames = 0; |
*copynames = 0; |
| 3306 |
*getnames = 0; |
*getnames = 0; |
| 3307 |
|
|
| 3308 |
copynamesptr = copynames; |
cn16ptr = copynames; |
| 3309 |
getnamesptr = getnames; |
gn16ptr = getnames; |
| 3310 |
|
cn8ptr = copynames8; |
| 3311 |
|
gn8ptr = getnames8; |
| 3312 |
|
|
| 3313 |
pcre_callout = callout; |
SET_PCRE_CALLOUT(callout); |
| 3314 |
first_callout = 1; |
first_callout = 1; |
| 3315 |
last_callout_mark = NULL; |
last_callout_mark = NULL; |
| 3316 |
callout_extra = 0; |
callout_extra = 0; |
| 3318 |
callout_fail_count = 999999; |
callout_fail_count = 999999; |
| 3319 |
callout_fail_id = -1; |
callout_fail_id = -1; |
| 3320 |
show_malloc = 0; |
show_malloc = 0; |
| 3321 |
|
options = 0; |
| 3322 |
|
|
| 3323 |
if (extra != NULL) extra->flags &= |
if (extra != NULL) extra->flags &= |
| 3324 |
~(PCRE_EXTRA_MATCH_LIMIT|PCRE_EXTRA_MATCH_LIMIT_RECURSION); |
~(PCRE_EXTRA_MATCH_LIMIT|PCRE_EXTRA_MATCH_LIMIT_RECURSION); |
| 3353 |
{ |
{ |
| 3354 |
int i = 0; |
int i = 0; |
| 3355 |
int n = 0; |
int n = 0; |
| 3356 |
|
|
| 3357 |
if (c == '\\') switch ((c = *p++)) |
/* In UTF mode, input can be UTF-8, so just copy all non-backslash bytes. |
| 3358 |
|
In non-UTF mode, allow the value of the byte to fall through to later, |
| 3359 |
|
where values greater than 127 are turned into UTF-8 when running in |
| 3360 |
|
16-bit mode. */ |
| 3361 |
|
|
| 3362 |
|
if (c != '\\') |
| 3363 |
|
{ |
| 3364 |
|
if (use_utf) |
| 3365 |
|
{ |
| 3366 |
|
*q++ = c; |
| 3367 |
|
continue; |
| 3368 |
|
} |
| 3369 |
|
} |
| 3370 |
|
|
| 3371 |
|
/* Handle backslash escapes */ |
| 3372 |
|
|
| 3373 |
|
else switch ((c = *p++)) |
| 3374 |
{ |
{ |
| 3375 |
case 'a': c = 7; break; |
case 'a': c = 7; break; |
| 3376 |
case 'b': c = '\b'; break; |
case 'b': c = '\b'; break; |
| 3386 |
c -= '0'; |
c -= '0'; |
| 3387 |
while (i++ < 2 && isdigit(*p) && *p != '8' && *p != '9') |
while (i++ < 2 && isdigit(*p) && *p != '8' && *p != '9') |
| 3388 |
c = c * 8 + *p++ - '0'; |
c = c * 8 + *p++ - '0'; |
|
|
|
|
#if !defined NOUTF8 |
|
|
if (use_utf8 && c > 255) |
|
|
{ |
|
|
unsigned char buff8[8]; |
|
|
int ii, utn; |
|
|
utn = ord2utf8(c, buff8); |
|
|
for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii]; |
|
|
c = buff8[ii]; /* Last byte */ |
|
|
} |
|
|
#endif |
|
| 3389 |
break; |
break; |
| 3390 |
|
|
| 3391 |
case 'x': |
case 'x': |
|
|
|
|
/* Handle \x{..} specially - new Perl thing for utf8 */ |
|
|
|
|
|
#if !defined NOUTF8 |
|
| 3392 |
if (*p == '{') |
if (*p == '{') |
| 3393 |
{ |
{ |
| 3394 |
unsigned char *pt = p; |
pcre_uint8 *pt = p; |
| 3395 |
c = 0; |
c = 0; |
| 3396 |
|
|
| 3397 |
/* We used to have "while (isxdigit(*(++pt)))" here, but it fails |
/* We used to have "while (isxdigit(*(++pt)))" here, but it fails |
| 3403 |
c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'a' - 10); |
c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'a' - 10); |
| 3404 |
if (*pt == '}') |
if (*pt == '}') |
| 3405 |
{ |
{ |
|
unsigned char buff8[8]; |
|
|
int ii, utn; |
|
|
if (use_utf8) |
|
|
{ |
|
|
utn = ord2utf8(c, buff8); |
|
|
for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii]; |
|
|
c = buff8[ii]; /* Last byte */ |
|
|
} |
|
|
else |
|
|
{ |
|
|
if (c > 255) |
|
|
fprintf(outfile, "** Character \\x{%x} is greater than 255 and " |
|
|
"UTF-8 mode is not enabled.\n" |
|
|
"** Truncation will probably give the wrong result.\n", c); |
|
|
} |
|
| 3406 |
p = pt + 1; |
p = pt + 1; |
| 3407 |
break; |
break; |
| 3408 |
} |
} |
| 3409 |
/* Not correct form; fall through */ |
/* Not correct form for \x{...}; fall through */ |
| 3410 |
} |
} |
|
#endif |
|
| 3411 |
|
|
| 3412 |
/* Ordinary \x */ |
/* \x without {} always defines just one byte in 8-bit mode. This |
| 3413 |
|
allows UTF-8 characters to be constructed byte by byte, and also allows |
| 3414 |
|
invalid UTF-8 sequences to be made. Just copy the byte in UTF mode. |
| 3415 |
|
Otherwise, pass it down to later code so that it can be turned into |
| 3416 |
|
UTF-8 when running in 16-bit mode. */ |
| 3417 |
|
|
| 3418 |
c = 0; |
c = 0; |
| 3419 |
while (i++ < 2 && isxdigit(*p)) |
while (i++ < 2 && isxdigit(*p)) |
| 3421 |
c = c * 16 + tolower(*p) - ((isdigit(*p))? '0' : 'a' - 10); |
c = c * 16 + tolower(*p) - ((isdigit(*p))? '0' : 'a' - 10); |
| 3422 |
p++; |
p++; |
| 3423 |
} |
} |
| 3424 |
|
if (use_utf) |
| 3425 |
|
{ |
| 3426 |
|
*q++ = c; |
| 3427 |
|
continue; |
| 3428 |
|
} |
| 3429 |
break; |
break; |
| 3430 |
|
|
| 3431 |
case 0: /* \ followed by EOF allows for an empty line */ |
case 0: /* \ followed by EOF allows for an empty line */ |
| 3458 |
} |
} |
| 3459 |
else if (isalnum(*p)) |
else if (isalnum(*p)) |
| 3460 |
{ |
{ |
| 3461 |
uschar *npp = copynamesptr; |
READ_CAPTURE_NAME(p, &cn8ptr, &cn16ptr, re); |
|
while (isalnum(*p)) *npp++ = *p++; |
|
|
*npp++ = 0; |
|
|
*npp = 0; |
|
|
n = pcre_get_stringnumber(re, (char *)copynamesptr); |
|
|
if (n < 0) |
|
|
fprintf(outfile, "no parentheses with name \"%s\"\n", copynamesptr); |
|
|
copynamesptr = npp; |
|
| 3462 |
} |
} |
| 3463 |
else if (*p == '+') |
else if (*p == '+') |
| 3464 |
{ |
{ |
| 3467 |
} |
} |
| 3468 |
else if (*p == '-') |
else if (*p == '-') |
| 3469 |
{ |
{ |
| 3470 |
pcre_callout = NULL; |
SET_PCRE_CALLOUT(NULL); |
| 3471 |
p++; |
p++; |
| 3472 |
} |
} |
| 3473 |
else if (*p == '!') |
else if (*p == '!') |
| 3521 |
} |
} |
| 3522 |
else if (isalnum(*p)) |
else if (isalnum(*p)) |
| 3523 |
{ |
{ |
| 3524 |
uschar *npp = getnamesptr; |
READ_CAPTURE_NAME(p, &gn8ptr, &gn16ptr, re); |
|
while (isalnum(*p)) *npp++ = *p++; |
|
|
*npp++ = 0; |
|
|
*npp = 0; |
|
|
n = pcre_get_stringnumber(re, (char *)getnamesptr); |
|
|
if (n < 0) |
|
|
fprintf(outfile, "no parentheses with name \"%s\"\n", getnamesptr); |
|
|
getnamesptr = npp; |
|
| 3525 |
} |
} |
| 3526 |
continue; |
continue; |
| 3527 |
|
|
| 3531 |
&& (extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 |
&& (extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 |
| 3532 |
&& extra->executable_jit != NULL) |
&& extra->executable_jit != NULL) |
| 3533 |
{ |
{ |
| 3534 |
if (jit_stack != NULL) pcre_jit_stack_free(jit_stack); |
if (jit_stack != NULL) PCRE_JIT_STACK_FREE(jit_stack); |
| 3535 |
jit_stack = pcre_jit_stack_alloc(1, n * 1024); |
jit_stack = PCRE_JIT_STACK_ALLOC(1, n * 1024); |
| 3536 |
pcre_assign_jit_stack(extra, jit_callback, jit_stack); |
PCRE_ASSIGN_JIT_STACK(extra, jit_callback, jit_stack); |
| 3537 |
} |
} |
| 3538 |
continue; |
continue; |
| 3539 |
|
|
| 3629 |
} |
} |
| 3630 |
continue; |
continue; |
| 3631 |
} |
} |
| 3632 |
*q++ = c; |
|
| 3633 |
|
/* We now have a character value in c that may be greater than 255. In |
| 3634 |
|
16-bit mode, we always convert characters to UTF-8 so that values greater |
| 3635 |
|
than 255 can be passed to non-UTF 16-bit strings. In 8-bit mode we |
| 3636 |
|
convert to UTF-8 if we are in UTF mode. Values greater than 127 in UTF |
| 3637 |
|
mode must have come from \x{...} or octal constructs because values from |
| 3638 |
|
\x.. get this far only in non-UTF mode. */ |
| 3639 |
|
|
| 3640 |
|
if (use_pcre16 || use_utf) |
| 3641 |
|
{ |
| 3642 |
|
pcre_uint8 buff8[8]; |
| 3643 |
|
int ii, utn; |
| 3644 |
|
utn = ord2utf8(c, buff8); |
| 3645 |
|
for (ii = 0; ii < utn; ii++) *q++ = buff8[ii]; |
| 3646 |
|
} |
| 3647 |
|
else |
| 3648 |
|
{ |
| 3649 |
|
if (c > 255) |
| 3650 |
|
{ |
| 3651 |
|
fprintf(outfile, "** Character \\x{%x} is greater than 255 " |
| 3652 |
|
"and UTF-8 mode is not enabled.\n", c); |
| 3653 |
|
fprintf(outfile, "** Truncation will probably give the wrong " |
| 3654 |
|
"result.\n"); |
| 3655 |
|
} |
| 3656 |
|
*q++ = c; |
| 3657 |
|
} |
| 3658 |
} |
} |
| 3659 |
|
|
| 3660 |
|
/* Reached end of subject string */ |
| 3661 |
|
|
| 3662 |
*q = 0; |
*q = 0; |
| 3663 |
len = (int)(q - dbuffer); |
len = (int)(q - dbuffer); |
| 3664 |
|
|
| 3720 |
if (pmatch[i].rm_so >= 0) |
if (pmatch[i].rm_so >= 0) |
| 3721 |
{ |
{ |
| 3722 |
fprintf(outfile, "%2d: ", (int)i); |
fprintf(outfile, "%2d: ", (int)i); |
| 3723 |
(void)pchars(dbuffer + pmatch[i].rm_so, |
PCHARSV(dbuffer, pmatch[i].rm_so, |
| 3724 |
pmatch[i].rm_eo - pmatch[i].rm_so, outfile); |
pmatch[i].rm_eo - pmatch[i].rm_so, outfile); |
| 3725 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 3726 |
if (do_showcaprest || (i == 0 && do_showrest)) |
if (do_showcaprest || (i == 0 && do_showrest)) |
| 3727 |
{ |
{ |
| 3728 |
fprintf(outfile, "%2d+ ", (int)i); |
fprintf(outfile, "%2d+ ", (int)i); |
| 3729 |
(void)pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo, |
PCHARSV(dbuffer, pmatch[i].rm_eo, len - pmatch[i].rm_eo, |
| 3730 |
outfile); |
outfile); |
| 3731 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 3732 |
} |
} |
| 3734 |
} |
} |
| 3735 |
} |
} |
| 3736 |
free(pmatch); |
free(pmatch); |
| 3737 |
|
goto NEXT_DATA; |
| 3738 |
} |
} |
| 3739 |
|
|
| 3740 |
|
#endif /* !defined NOPOSIX */ |
| 3741 |
|
|
| 3742 |
/* Handle matching via the native interface - repeats for /g and /G */ |
/* Handle matching via the native interface - repeats for /g and /G */ |
| 3743 |
|
|
| 3744 |
else |
#ifdef SUPPORT_PCRE16 |
| 3745 |
#endif /* !defined NOPOSIX */ |
if (use_pcre16) |
| 3746 |
|
{ |
| 3747 |
|
len = to16(TRUE, bptr, (((real_pcre *)re)->options) & PCRE_UTF8, len); |
| 3748 |
|
switch(len) |
| 3749 |
|
{ |
| 3750 |
|
case -1: |
| 3751 |
|
fprintf(outfile, "**Failed: invalid UTF-8 string cannot be " |
| 3752 |
|
"converted to UTF-16\n"); |
| 3753 |
|
goto NEXT_DATA; |
| 3754 |
|
|
| 3755 |
|
case -2: |
| 3756 |
|
fprintf(outfile, "**Failed: character value greater than 0x10ffff " |
| 3757 |
|
"cannot be converted to UTF-16\n"); |
| 3758 |
|
goto NEXT_DATA; |
| 3759 |
|
|
| 3760 |
|
case -3: |
| 3761 |
|
fprintf(outfile, "**Failed: character value greater than 0xffff " |
| 3762 |
|
"cannot be converted to 16-bit in non-UTF mode\n"); |
| 3763 |
|
goto NEXT_DATA; |
| 3764 |
|
|
| 3765 |
|
default: |
| 3766 |
|
break; |
| 3767 |
|
} |
| 3768 |
|
bptr = (pcre_uint8 *)buffer16; |
| 3769 |
|
} |
| 3770 |
|
#endif |
| 3771 |
|
|
| 3772 |
for (;; gmatched++) /* Loop for /g or /G */ |
for (;; gmatched++) /* Loop for /g or /G */ |
| 3773 |
{ |
{ |
| 3784 |
{ |
{ |
| 3785 |
int workspace[1000]; |
int workspace[1000]; |
| 3786 |
for (i = 0; i < timeitm; i++) |
for (i = 0; i < timeitm; i++) |
| 3787 |
count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset, |
{ |
| 3788 |
options | g_notempty, use_offsets, use_size_offsets, workspace, |
PCRE_DFA_EXEC(count, re, extra, bptr, len, start_offset, |
| 3789 |
sizeof(workspace)/sizeof(int)); |
(options | g_notempty), use_offsets, use_size_offsets, workspace, |
| 3790 |
|
(sizeof(workspace)/sizeof(int))); |
| 3791 |
|
} |
| 3792 |
} |
} |
| 3793 |
else |
else |
| 3794 |
#endif |
#endif |
| 3795 |
|
|
| 3796 |
for (i = 0; i < timeitm; i++) |
for (i = 0; i < timeitm; i++) |
| 3797 |
count = pcre_exec(re, extra, (char *)bptr, len, |
{ |
| 3798 |
start_offset, options | g_notempty, use_offsets, use_size_offsets); |
PCRE_EXEC(count, re, extra, bptr, len, start_offset, |
| 3799 |
|
(options | g_notempty), use_offsets, use_size_offsets); |
| 3800 |
|
} |
| 3801 |
time_taken = clock() - start_time; |
time_taken = clock() - start_time; |
| 3802 |
fprintf(outfile, "Execute time %.4f milliseconds\n", |
fprintf(outfile, "Execute time %.4f milliseconds\n", |
| 3803 |
(((double)time_taken * 1000.0) / (double)timeitm) / |
(((double)time_taken * 1000.0) / (double)timeitm) / |
| 3842 |
} |
} |
| 3843 |
extra->flags |= PCRE_EXTRA_CALLOUT_DATA; |
extra->flags |= PCRE_EXTRA_CALLOUT_DATA; |
| 3844 |
extra->callout_data = &callout_data; |
extra->callout_data = &callout_data; |
| 3845 |
count = pcre_exec(re, extra, (char *)bptr, len, start_offset, |
PCRE_EXEC(count, re, extra, bptr, len, start_offset, |
| 3846 |
options | g_notempty, use_offsets, use_size_offsets); |
options | g_notempty, use_offsets, use_size_offsets); |
| 3847 |
extra->flags &= ~PCRE_EXTRA_CALLOUT_DATA; |
extra->flags &= ~PCRE_EXTRA_CALLOUT_DATA; |
| 3848 |
} |
} |
| 3854 |
else if (all_use_dfa || use_dfa) |
else if (all_use_dfa || use_dfa) |
| 3855 |
{ |
{ |
| 3856 |
int workspace[1000]; |
int workspace[1000]; |
| 3857 |
count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset, |
PCRE_DFA_EXEC(count, re, extra, bptr, len, start_offset, |
| 3858 |
options | g_notempty, use_offsets, use_size_offsets, workspace, |
(options | g_notempty), use_offsets, use_size_offsets, workspace, |
| 3859 |
sizeof(workspace)/sizeof(int)); |
(sizeof(workspace)/sizeof(int))); |
| 3860 |
if (count == 0) |
if (count == 0) |
| 3861 |
{ |
{ |
| 3862 |
fprintf(outfile, "Matched, but too many subsidiary matches\n"); |
fprintf(outfile, "Matched, but too many subsidiary matches\n"); |
| 3867 |
|
|
| 3868 |
else |
else |
| 3869 |
{ |
{ |
| 3870 |
count = pcre_exec(re, extra, (char *)bptr, len, |
PCRE_EXEC(count, re, extra, bptr, len, start_offset, |
| 3871 |
start_offset, options | g_notempty, use_offsets, use_size_offsets); |
options | g_notempty, use_offsets, use_size_offsets); |
| 3872 |
if (count == 0) |
if (count == 0) |
| 3873 |
{ |
{ |
| 3874 |
fprintf(outfile, "Matched, but too many substrings\n"); |
fprintf(outfile, "Matched, but too many substrings\n"); |
| 3881 |
if (count >= 0) |
if (count >= 0) |
| 3882 |
{ |
{ |
| 3883 |
int i, maxcount; |
int i, maxcount; |
| 3884 |
|
void *cnptr, *gnptr; |
| 3885 |
|
|
| 3886 |
#if !defined NODFA |
#if !defined NODFA |
| 3887 |
if (all_use_dfa || use_dfa) maxcount = use_size_offsets/2; else |
if (all_use_dfa || use_dfa) maxcount = use_size_offsets/2; else |
| 3908 |
|
|
| 3909 |
if (do_allcaps) |
if (do_allcaps) |
| 3910 |
{ |
{ |
| 3911 |
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count); |
if (new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count) < 0) |
| 3912 |
|
goto SKIP_DATA; |
| 3913 |
count++; /* Allow for full match */ |
count++; /* Allow for full match */ |
| 3914 |
if (count * 2 > use_size_offsets) count = use_size_offsets/2; |
if (count * 2 > use_size_offsets) count = use_size_offsets/2; |
| 3915 |
} |
} |
| 3931 |
else |
else |
| 3932 |
{ |
{ |
| 3933 |
fprintf(outfile, "%2d: ", i/2); |
fprintf(outfile, "%2d: ", i/2); |
| 3934 |
(void)pchars(bptr + use_offsets[i], |
PCHARSV(bptr, use_offsets[i], |
| 3935 |
use_offsets[i+1] - use_offsets[i], outfile); |
use_offsets[i+1] - use_offsets[i], outfile); |
| 3936 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 3937 |
if (do_showcaprest || (i == 0 && do_showrest)) |
if (do_showcaprest || (i == 0 && do_showrest)) |
| 3938 |
{ |
{ |
| 3939 |
fprintf(outfile, "%2d+ ", i/2); |
fprintf(outfile, "%2d+ ", i/2); |
| 3940 |
(void)pchars(bptr + use_offsets[i+1], len - use_offsets[i+1], |
PCHARSV(bptr, use_offsets[i+1], len - use_offsets[i+1], |
| 3941 |
outfile); |
outfile); |
| 3942 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 3943 |
} |
} |
| 3944 |
} |
} |
| 3945 |
} |
} |
| 3946 |
|
|
| 3947 |
if (markptr != NULL) fprintf(outfile, "MK: %s\n", markptr); |
if (markptr != NULL) |
| 3948 |
|
{ |
| 3949 |
|
fprintf(outfile, "MK: "); |
| 3950 |
|
PCHARSV(markptr, 0, -1, outfile); |
| 3951 |
|
fprintf(outfile, "\n"); |
| 3952 |
|
} |
| 3953 |
|
|
| 3954 |
for (i = 0; i < 32; i++) |
for (i = 0; i < 32; i++) |
| 3955 |
{ |
{ |
| 3956 |
if ((copystrings & (1 << i)) != 0) |
if ((copystrings & (1 << i)) != 0) |
| 3957 |
{ |
{ |
| 3958 |
|
int rc; |
| 3959 |
char copybuffer[256]; |
char copybuffer[256]; |
| 3960 |
int rc = pcre_copy_substring((char *)bptr, use_offsets, count, |
PCRE_COPY_SUBSTRING(rc, bptr, use_offsets, count, i, |
| 3961 |
i, copybuffer, sizeof(copybuffer)); |
copybuffer, sizeof(copybuffer)); |
| 3962 |
if (rc < 0) |
if (rc < 0) |
| 3963 |
fprintf(outfile, "copy substring %d failed %d\n", i, rc); |
fprintf(outfile, "copy substring %d failed %d\n", i, rc); |
| 3964 |
else |
else |
| 3965 |
fprintf(outfile, "%2dC %s (%d)\n", i, copybuffer, rc); |
{ |
| 3966 |
|
fprintf(outfile, "%2dC ", i); |
| 3967 |
|
PCHARSV(copybuffer, 0, rc, outfile); |
| 3968 |
|
fprintf(outfile, " (%d)\n", rc); |
| 3969 |
|
} |
| 3970 |
} |
} |
| 3971 |
} |
} |
| 3972 |
|
|
| 3973 |
for (copynamesptr = copynames; |
cnptr = copynames; |
| 3974 |
*copynamesptr != 0; |
for (;;) |
|
copynamesptr += (int)strlen((char*)copynamesptr) + 1) |
|
| 3975 |
{ |
{ |
| 3976 |
|
int rc; |
| 3977 |
char copybuffer[256]; |
char copybuffer[256]; |
| 3978 |
int rc = pcre_copy_named_substring(re, (char *)bptr, use_offsets, |
|
| 3979 |
count, (char *)copynamesptr, copybuffer, sizeof(copybuffer)); |
if (use_pcre16) |
| 3980 |
|
{ |
| 3981 |
|
if (*(pcre_uint16 *)cnptr == 0) break; |
| 3982 |
|
} |
| 3983 |
|
else |
| 3984 |
|
{ |
| 3985 |
|
if (*(pcre_uint8 *)cnptr == 0) break; |
| 3986 |
|
} |
| 3987 |
|
|
| 3988 |
|
PCRE_COPY_NAMED_SUBSTRING(rc, re, bptr, use_offsets, count, |
| 3989 |
|
cnptr, copybuffer, sizeof(copybuffer)); |
| 3990 |
|
|
| 3991 |
if (rc < 0) |
if (rc < 0) |
| 3992 |
fprintf(outfile, "copy substring %s failed %d\n", copynamesptr, rc); |
{ |
| 3993 |
|
fprintf(outfile, "copy substring "); |
| 3994 |
|
PCHARSV(cnptr, 0, -1, outfile); |
| 3995 |
|
fprintf(outfile, " failed %d\n", rc); |
| 3996 |
|
} |
| 3997 |
else |
else |
| 3998 |
fprintf(outfile, " C %s (%d) %s\n", copybuffer, rc, copynamesptr); |
{ |
| 3999 |
|
fprintf(outfile, " C "); |
| 4000 |
|
PCHARSV(copybuffer, 0, rc, outfile); |
| 4001 |
|
fprintf(outfile, " (%d) ", rc); |
| 4002 |
|
PCHARSV(cnptr, 0, -1, outfile); |
| 4003 |
|
putc('\n', outfile); |
| 4004 |
|
} |
| 4005 |
|
|
| 4006 |
|
cnptr = (char *)cnptr + (STRLEN(cnptr) + 1) * CHAR_SIZE; |
| 4007 |
} |
} |
| 4008 |
|
|
| 4009 |
for (i = 0; i < 32; i++) |
for (i = 0; i < 32; i++) |
| 4010 |
{ |
{ |
| 4011 |
if ((getstrings & (1 << i)) != 0) |
if ((getstrings & (1 << i)) != 0) |
| 4012 |
{ |
{ |
| 4013 |
|
int rc; |
| 4014 |
const char *substring; |
const char *substring; |
| 4015 |
int rc = pcre_get_substring((char *)bptr, use_offsets, count, |
PCRE_GET_SUBSTRING(rc, bptr, use_offsets, count, i, &substring); |
|
i, &substring); |
|
| 4016 |
if (rc < 0) |
if (rc < 0) |
| 4017 |
fprintf(outfile, "get substring %d failed %d\n", i, rc); |
fprintf(outfile, "get substring %d failed %d\n", i, rc); |
| 4018 |
else |
else |
| 4019 |
{ |
{ |
| 4020 |
fprintf(outfile, "%2dG %s (%d)\n", i, substring, rc); |
fprintf(outfile, "%2dG ", i); |
| 4021 |
pcre_free_substring(substring); |
PCHARSV(substring, 0, rc, outfile); |
| 4022 |
|
fprintf(outfile, " (%d)\n", rc); |
| 4023 |
|
PCRE_FREE_SUBSTRING(substring); |
| 4024 |
} |
} |
| 4025 |
} |
} |
| 4026 |
} |
} |
| 4027 |
|
|
| 4028 |
for (getnamesptr = getnames; |
gnptr = getnames; |
| 4029 |
*getnamesptr != 0; |
for (;;) |
|
getnamesptr += (int)strlen((char*)getnamesptr) + 1) |
|
| 4030 |
{ |
{ |
| 4031 |
|
int rc; |
| 4032 |
const char *substring; |
const char *substring; |
| 4033 |
int rc = pcre_get_named_substring(re, (char *)bptr, use_offsets, |
|
| 4034 |
count, (char *)getnamesptr, &substring); |
if (use_pcre16) |
| 4035 |
|
{ |
| 4036 |
|
if (*(pcre_uint16 *)gnptr == 0) break; |
| 4037 |
|
} |
| 4038 |
|
else |
| 4039 |
|
{ |
| 4040 |
|
if (*(pcre_uint8 *)gnptr == 0) break; |
| 4041 |
|
} |
| 4042 |
|
|
| 4043 |
|
PCRE_GET_NAMED_SUBSTRING(rc, re, bptr, use_offsets, count, |
| 4044 |
|
gnptr, &substring); |
| 4045 |
if (rc < 0) |
if (rc < 0) |
| 4046 |
fprintf(outfile, "copy substring %s failed %d\n", getnamesptr, rc); |
{ |
| 4047 |
|
fprintf(outfile, "get substring "); |
| 4048 |
|
PCHARSV(gnptr, 0, -1, outfile); |
| 4049 |
|
fprintf(outfile, " failed %d\n", rc); |
| 4050 |
|
} |
| 4051 |
else |
else |
| 4052 |
{ |
{ |
| 4053 |
fprintf(outfile, " G %s (%d) %s\n", substring, rc, getnamesptr); |
fprintf(outfile, " G "); |
| 4054 |
pcre_free_substring(substring); |
PCHARSV(substring, 0, rc, outfile); |
| 4055 |
|
fprintf(outfile, " (%d) ", rc); |
| 4056 |
|
PCHARSV(gnptr, 0, -1, outfile); |
| 4057 |
|
PCRE_FREE_SUBSTRING(substring); |
| 4058 |
|
putc('\n', outfile); |
| 4059 |
} |
} |
| 4060 |
|
|
| 4061 |
|
gnptr = (char *)gnptr + (STRLEN(gnptr) + 1) * CHAR_SIZE; |
| 4062 |
} |
} |
| 4063 |
|
|
| 4064 |
if (getlist) |
if (getlist) |
| 4065 |
{ |
{ |
| 4066 |
|
int rc; |
| 4067 |
const char **stringlist; |
const char **stringlist; |
| 4068 |
int rc = pcre_get_substring_list((char *)bptr, use_offsets, count, |
PCRE_GET_SUBSTRING_LIST(rc, bptr, use_offsets, count, &stringlist); |
|
&stringlist); |
|
| 4069 |
if (rc < 0) |
if (rc < 0) |
| 4070 |
fprintf(outfile, "get substring list failed %d\n", rc); |
fprintf(outfile, "get substring list failed %d\n", rc); |
| 4071 |
else |
else |
| 4072 |
{ |
{ |
| 4073 |
for (i = 0; i < count; i++) |
for (i = 0; i < count; i++) |
| 4074 |
fprintf(outfile, "%2dL %s\n", i, stringlist[i]); |
{ |
| 4075 |
|
fprintf(outfile, "%2dL ", i); |
| 4076 |
|
PCHARSV(stringlist[i], 0, -1, outfile); |
| 4077 |
|
putc('\n', outfile); |
| 4078 |
|
} |
| 4079 |
if (stringlist[i] != NULL) |
if (stringlist[i] != NULL) |
| 4080 |
fprintf(outfile, "string list not terminated by NULL\n"); |
fprintf(outfile, "string list not terminated by NULL\n"); |
| 4081 |
pcre_free_substring_list(stringlist); |
PCRE_FREE_SUBSTRING_LIST(stringlist); |
| 4082 |
} |
} |
| 4083 |
} |
} |
| 4084 |
} |
} |
| 4088 |
else if (count == PCRE_ERROR_PARTIAL) |
else if (count == PCRE_ERROR_PARTIAL) |
| 4089 |
{ |
{ |
| 4090 |
if (markptr == NULL) fprintf(outfile, "Partial match"); |
if (markptr == NULL) fprintf(outfile, "Partial match"); |
| 4091 |
else fprintf(outfile, "Partial match, mark=%s", markptr); |
else |
| 4092 |
|
{ |
| 4093 |
|
fprintf(outfile, "Partial match, mark="); |
| 4094 |
|
PCHARSV(markptr, 0, -1, outfile); |
| 4095 |
|
} |
| 4096 |
if (use_size_offsets > 1) |
if (use_size_offsets > 1) |
| 4097 |
{ |
{ |
| 4098 |
fprintf(outfile, ": "); |
fprintf(outfile, ": "); |
| 4099 |
pchars(bptr + use_offsets[0], use_offsets[1] - use_offsets[0], |
PCHARSV(bptr, use_offsets[0], use_offsets[1] - use_offsets[0], |
| 4100 |
outfile); |
outfile); |
| 4101 |
} |
} |
| 4102 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 4113 |
terminated by CRLF, an advance of one character just passes the \r, |
terminated by CRLF, an advance of one character just passes the \r, |
| 4114 |
whereas we should prefer the longer newline sequence, as does the code in |
whereas we should prefer the longer newline sequence, as does the code in |
| 4115 |
pcre_exec(). Fudge the offset value to achieve this. We check for a |
pcre_exec(). Fudge the offset value to achieve this. We check for a |
| 4116 |
newline setting in the pattern; if none was set, use pcre_config() to |
newline setting in the pattern; if none was set, use PCRE_CONFIG() to |
| 4117 |
find the default. |
find the default. |
| 4118 |
|
|
| 4119 |
Otherwise, in the case of UTF-8 matching, the advance must be one |
Otherwise, in the case of UTF-8 matching, the advance must be one |
| 4129 |
if ((obits & PCRE_NEWLINE_BITS) == 0) |
if ((obits & PCRE_NEWLINE_BITS) == 0) |
| 4130 |
{ |
{ |
| 4131 |
int d; |
int d; |
| 4132 |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &d); |
(void)PCRE_CONFIG(PCRE_CONFIG_NEWLINE, &d); |
| 4133 |
/* Note that these values are always the ASCII ones, even in |
/* Note that these values are always the ASCII ones, even in |
| 4134 |
EBCDIC environments. CR = 13, NL = 10. */ |
EBCDIC environments. CR = 13, NL = 10. */ |
| 4135 |
obits = (d == 13)? PCRE_NEWLINE_CR : |
obits = (d == 13)? PCRE_NEWLINE_CR : |
| 4143 |
(obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANYCRLF) |
(obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANYCRLF) |
| 4144 |
&& |
&& |
| 4145 |
start_offset < len - 1 && |
start_offset < len - 1 && |
| 4146 |
bptr[start_offset] == '\r' && |
#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE16 |
| 4147 |
bptr[start_offset+1] == '\n') |
(use_pcre16? |
| 4148 |
|
((PCRE_SPTR16)bptr)[start_offset] == '\r' |
| 4149 |
|
&& ((PCRE_SPTR16)bptr)[start_offset + 1] == '\n' |
| 4150 |
|
: |
| 4151 |
|
bptr[start_offset] == '\r' |
| 4152 |
|
&& bptr[start_offset + 1] == '\n') |
| 4153 |
|
#elif defined SUPPORT_PCRE16 |
| 4154 |
|
((PCRE_SPTR16)bptr)[start_offset] == '\r' |
| 4155 |
|
&& ((PCRE_SPTR16)bptr)[start_offset + 1] == '\n' |
| 4156 |
|
#else |
| 4157 |
|
bptr[start_offset] == '\r' |
| 4158 |
|
&& bptr[start_offset + 1] == '\n' |
| 4159 |
|
#endif |
| 4160 |
|
) |
| 4161 |
onechar++; |
onechar++; |
| 4162 |
else if (use_utf8) |
else if (use_utf) |
| 4163 |
{ |
{ |
| 4164 |
while (start_offset + onechar < len) |
while (start_offset + onechar < len) |
| 4165 |
{ |
{ |
| 4176 |
case PCRE_ERROR_NOMATCH: |
case PCRE_ERROR_NOMATCH: |
| 4177 |
if (gmatched == 0) |
if (gmatched == 0) |
| 4178 |
{ |
{ |
| 4179 |
if (markptr == NULL) fprintf(outfile, "No match\n"); |
if (markptr == NULL) |
| 4180 |
else fprintf(outfile, "No match, mark = %s\n", markptr); |
{ |
| 4181 |
|
fprintf(outfile, "No match\n"); |
| 4182 |
|
} |
| 4183 |
|
else |
| 4184 |
|
{ |
| 4185 |
|
fprintf(outfile, "No match, mark = "); |
| 4186 |
|
PCHARSV(markptr, 0, -1, outfile); |
| 4187 |
|
putc('\n', outfile); |
| 4188 |
|
} |
| 4189 |
} |
} |
| 4190 |
break; |
break; |
| 4191 |
|
|
| 4192 |
case PCRE_ERROR_BADUTF8: |
case PCRE_ERROR_BADUTF8: |
| 4193 |
case PCRE_ERROR_SHORTUTF8: |
case PCRE_ERROR_SHORTUTF8: |
| 4194 |
fprintf(outfile, "Error %d (%s UTF-8 string)", count, |
fprintf(outfile, "Error %d (%s UTF-%s string)", count, |
| 4195 |
(count == PCRE_ERROR_BADUTF8)? "bad" : "short"); |
(count == PCRE_ERROR_BADUTF8)? "bad" : "short", |
| 4196 |
|
use_pcre16? "16" : "8"); |
| 4197 |
if (use_size_offsets >= 2) |
if (use_size_offsets >= 2) |
| 4198 |
fprintf(outfile, " offset=%d reason=%d", use_offsets[0], |
fprintf(outfile, " offset=%d reason=%d", use_offsets[0], |
| 4199 |
use_offsets[1]); |
use_offsets[1]); |
| 4200 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 4201 |
break; |
break; |
| 4202 |
|
|
| 4203 |
|
case PCRE_ERROR_BADUTF8_OFFSET: |
| 4204 |
|
fprintf(outfile, "Error %d (bad UTF-%s offset)\n", count, |
| 4205 |
|
use_pcre16? "16" : "8"); |
| 4206 |
|
break; |
| 4207 |
|
|
| 4208 |
default: |
default: |
| 4209 |
if (count < 0 && (-count) < sizeof(errtexts)/sizeof(const char *)) |
if (count < 0 && (-count) < sizeof(errtexts)/sizeof(const char *)) |
| 4210 |
fprintf(outfile, "Error %d (%s)\n", count, errtexts[-count]); |
fprintf(outfile, "Error %d (%s)\n", count, errtexts[-count]); |
| 4244 |
|
|
| 4245 |
else |
else |
| 4246 |
{ |
{ |
| 4247 |
bptr += use_offsets[1]; |
bptr += use_offsets[1] * CHAR_SIZE; |
| 4248 |
len -= use_offsets[1]; |
len -= use_offsets[1]; |
| 4249 |
} |
} |
| 4250 |
} /* End of loop for /g and /G */ |
} /* End of loop for /g and /G */ |
| 4259 |
#endif |
#endif |
| 4260 |
|
|
| 4261 |
if (re != NULL) new_free(re); |
if (re != NULL) new_free(re); |
| 4262 |
if (extra != NULL) pcre_free_study(extra); |
if (extra != NULL) |
| 4263 |
|
{ |
| 4264 |
|
PCRE_FREE_STUDY(extra); |
| 4265 |
|
} |
| 4266 |
if (locale_set) |
if (locale_set) |
| 4267 |
{ |
{ |
| 4268 |
new_free((void *)tables); |
new_free((void *)tables); |
| 4271 |
} |
} |
| 4272 |
if (jit_stack != NULL) |
if (jit_stack != NULL) |
| 4273 |
{ |
{ |
| 4274 |
pcre_jit_stack_free(jit_stack); |
PCRE_JIT_STACK_FREE(jit_stack); |
| 4275 |
jit_stack = NULL; |
jit_stack = NULL; |
| 4276 |
} |
} |
| 4277 |
} |
} |
| 4288 |
free(pbuffer); |
free(pbuffer); |
| 4289 |
free(offsets); |
free(offsets); |
| 4290 |
|
|
| 4291 |
|
#ifdef SUPPORT_PCRE16 |
| 4292 |
|
if (buffer16 != NULL) free(buffer16); |
| 4293 |
|
#endif |
| 4294 |
|
|
| 4295 |
return yield; |
return yield; |
| 4296 |
} |
} |
| 4297 |
|
|