| 48 |
#include <locale.h> |
#include <locale.h> |
| 49 |
#include <errno.h> |
#include <errno.h> |
| 50 |
|
|
| 51 |
|
#ifdef SUPPORT_LIBREADLINE |
| 52 |
|
#ifdef HAVE_UNISTD_H |
| 53 |
|
#include <unistd.h> |
| 54 |
|
#endif |
| 55 |
|
#include <readline/readline.h> |
| 56 |
|
#include <readline/history.h> |
| 57 |
|
#endif |
| 58 |
|
|
| 59 |
|
|
| 60 |
/* A number of things vary for Windows builds. Originally, pcretest opened its |
/* A number of things vary for Windows builds. Originally, pcretest opened its |
| 61 |
input and output without "b"; then I was told that "b" was needed in some |
input and output without "b"; then I was told that "b" was needed in some |
| 71 |
#define INPUT_MODE "r" |
#define INPUT_MODE "r" |
| 72 |
#define OUTPUT_MODE "wb" |
#define OUTPUT_MODE "wb" |
| 73 |
|
|
| 74 |
|
#ifndef isatty |
| 75 |
|
#define isatty _isatty /* This is what Windows calls them, I'm told, */ |
| 76 |
|
#endif /* though in some environments they seem to */ |
| 77 |
|
/* be already defined, hence the #ifndefs. */ |
| 78 |
|
#ifndef fileno |
| 79 |
|
#define fileno _fileno |
| 80 |
|
#endif |
| 81 |
|
|
| 82 |
#else |
#else |
| 83 |
#include <sys/time.h> /* These two includes are needed */ |
#include <sys/time.h> /* These two includes are needed */ |
| 84 |
#include <sys/resource.h> /* for setrlimit(). */ |
#include <sys/resource.h> /* for setrlimit(). */ |
| 99 |
#include "pcre.h" |
#include "pcre.h" |
| 100 |
#include "pcre_internal.h" |
#include "pcre_internal.h" |
| 101 |
|
|
| 102 |
/* We need access to the data tables that PCRE uses. So as not to have to keep |
/* We need access to some of the data tables that PCRE uses. So as not to have |
| 103 |
two copies, we include the source file here, changing the names of the external |
to keep two copies, we include the source file here, changing the names of the |
| 104 |
symbols to prevent clashes. */ |
external symbols to prevent clashes. */ |
| 105 |
|
|
| 106 |
|
#define _pcre_ucp_gentype ucp_gentype |
| 107 |
#define _pcre_utf8_table1 utf8_table1 |
#define _pcre_utf8_table1 utf8_table1 |
| 108 |
#define _pcre_utf8_table1_size utf8_table1_size |
#define _pcre_utf8_table1_size utf8_table1_size |
| 109 |
#define _pcre_utf8_table2 utf8_table2 |
#define _pcre_utf8_table2 utf8_table2 |
| 206 |
Arguments: |
Arguments: |
| 207 |
f the file to read |
f the file to read |
| 208 |
start where in buffer to start (this *must* be within buffer) |
start where in buffer to start (this *must* be within buffer) |
| 209 |
|
prompt for stdin or readline() |
| 210 |
|
|
| 211 |
Returns: pointer to the start of new data |
Returns: pointer to the start of new data |
| 212 |
could be a copy of start, or could be moved |
could be a copy of start, or could be moved |
| 214 |
*/ |
*/ |
| 215 |
|
|
| 216 |
static uschar * |
static uschar * |
| 217 |
extend_inputline(FILE *f, uschar *start) |
extend_inputline(FILE *f, uschar *start, const char *prompt) |
| 218 |
{ |
{ |
| 219 |
uschar *here = start; |
uschar *here = start; |
| 220 |
|
|
| 225 |
if (rlen > 1000) |
if (rlen > 1000) |
| 226 |
{ |
{ |
| 227 |
int dlen; |
int dlen; |
| 228 |
if (fgets((char *)here, rlen, f) == NULL) |
|
| 229 |
return (here == start)? NULL : start; |
/* If libreadline support is required, use readline() to read a line if the |
| 230 |
|
input is a terminal. Note that readline() removes the trailing newline, so |
| 231 |
|
we must put it back again, to be compatible with fgets(). */ |
| 232 |
|
|
| 233 |
|
#ifdef SUPPORT_LIBREADLINE |
| 234 |
|
if (isatty(fileno(f))) |
| 235 |
|
{ |
| 236 |
|
size_t len; |
| 237 |
|
char *s = readline(prompt); |
| 238 |
|
if (s == NULL) return (here == start)? NULL : start; |
| 239 |
|
len = strlen(s); |
| 240 |
|
if (len > 0) add_history(s); |
| 241 |
|
if (len > rlen - 1) len = rlen - 1; |
| 242 |
|
memcpy(here, s, len); |
| 243 |
|
here[len] = '\n'; |
| 244 |
|
here[len+1] = 0; |
| 245 |
|
free(s); |
| 246 |
|
} |
| 247 |
|
else |
| 248 |
|
#endif |
| 249 |
|
|
| 250 |
|
/* Read the next line by normal means, prompting if the file is stdin. */ |
| 251 |
|
|
| 252 |
|
{ |
| 253 |
|
if (f == stdin) printf(prompt); |
| 254 |
|
if (fgets((char *)here, rlen, f) == NULL) |
| 255 |
|
return (here == start)? NULL : start; |
| 256 |
|
} |
| 257 |
|
|
| 258 |
dlen = (int)strlen((char *)here); |
dlen = (int)strlen((char *)here); |
| 259 |
if (dlen > 0 && here[dlen - 1] == '\n') return start; |
if (dlen > 0 && here[dlen - 1] == '\n') return start; |
| 260 |
here += dlen; |
here += dlen; |
| 774 |
static void |
static void |
| 775 |
usage(void) |
usage(void) |
| 776 |
{ |
{ |
| 777 |
printf("Usage: pcretest [options] [<input> [<output>]]\n"); |
printf("Usage: pcretest [options] [<input file> [<output file>]]\n\n"); |
| 778 |
|
printf("Input and output default to stdin and stdout.\n"); |
| 779 |
|
#ifdef SUPPORT_LIBREADLINE |
| 780 |
|
printf("If input is a terminal, readline() is used to read from it.\n"); |
| 781 |
|
#else |
| 782 |
|
printf("This version of pcretest is not linked with readline().\n"); |
| 783 |
|
#endif |
| 784 |
|
printf("\nOptions:\n"); |
| 785 |
printf(" -b show compiled code (bytecode)\n"); |
printf(" -b show compiled code (bytecode)\n"); |
| 786 |
printf(" -C show PCRE compile-time options and exit\n"); |
printf(" -C show PCRE compile-time options and exit\n"); |
| 787 |
printf(" -d debug: show compiled code and information (-b and -i)\n"); |
printf(" -d debug: show compiled code and information (-b and -i)\n"); |
| 790 |
#endif |
#endif |
| 791 |
printf(" -help show usage information\n"); |
printf(" -help show usage information\n"); |
| 792 |
printf(" -i show information about compiled patterns\n" |
printf(" -i show information about compiled patterns\n" |
| 793 |
|
" -M find MATCH_LIMIT minimum for each subject\n" |
| 794 |
" -m output memory used information\n" |
" -m output memory used information\n" |
| 795 |
" -o <n> set size of offsets vector to <n>\n"); |
" -o <n> set size of offsets vector to <n>\n"); |
| 796 |
#if !defined NOPOSIX |
#if !defined NOPOSIX |
| 820 |
FILE *infile = stdin; |
FILE *infile = stdin; |
| 821 |
int options = 0; |
int options = 0; |
| 822 |
int study_options = 0; |
int study_options = 0; |
| 823 |
|
int default_find_match_limit = FALSE; |
| 824 |
int op = 1; |
int op = 1; |
| 825 |
int timeit = 0; |
int timeit = 0; |
| 826 |
int timeitm = 0; |
int timeitm = 0; |
| 880 |
else if (strcmp(argv[op], "-b") == 0) debug = 1; |
else if (strcmp(argv[op], "-b") == 0) debug = 1; |
| 881 |
else if (strcmp(argv[op], "-i") == 0) showinfo = 1; |
else if (strcmp(argv[op], "-i") == 0) showinfo = 1; |
| 882 |
else if (strcmp(argv[op], "-d") == 0) showinfo = debug = 1; |
else if (strcmp(argv[op], "-d") == 0) showinfo = debug = 1; |
| 883 |
|
else if (strcmp(argv[op], "-M") == 0) default_find_match_limit = TRUE; |
| 884 |
#if !defined NODFA |
#if !defined NODFA |
| 885 |
else if (strcmp(argv[op], "-dfa") == 0) all_use_dfa = 1; |
else if (strcmp(argv[op], "-dfa") == 0) all_use_dfa = 1; |
| 886 |
#endif |
#endif |
| 933 |
else if (strcmp(argv[op], "-C") == 0) |
else if (strcmp(argv[op], "-C") == 0) |
| 934 |
{ |
{ |
| 935 |
int rc; |
int rc; |
| 936 |
|
unsigned long int lrc; |
| 937 |
printf("PCRE version %s\n", pcre_version()); |
printf("PCRE version %s\n", pcre_version()); |
| 938 |
printf("Compiled with\n"); |
printf("Compiled with\n"); |
| 939 |
(void)pcre_config(PCRE_CONFIG_UTF8, &rc); |
(void)pcre_config(PCRE_CONFIG_UTF8, &rc); |
| 941 |
(void)pcre_config(PCRE_CONFIG_UNICODE_PROPERTIES, &rc); |
(void)pcre_config(PCRE_CONFIG_UNICODE_PROPERTIES, &rc); |
| 942 |
printf(" %sUnicode properties support\n", rc? "" : "No "); |
printf(" %sUnicode properties support\n", rc? "" : "No "); |
| 943 |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &rc); |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &rc); |
| 944 |
printf(" Newline sequence is %s\n", (rc == '\r')? "CR" : |
/* Note that these values are always the ASCII values, even |
| 945 |
(rc == '\n')? "LF" : (rc == ('\r'<<8 | '\n'))? "CRLF" : |
in EBCDIC environments. CR is 13 and NL is 10. */ |
| 946 |
|
printf(" Newline sequence is %s\n", (rc == 13)? "CR" : |
| 947 |
|
(rc == 10)? "LF" : (rc == (13<<8 | 10))? "CRLF" : |
| 948 |
(rc == -2)? "ANYCRLF" : |
(rc == -2)? "ANYCRLF" : |
| 949 |
(rc == -1)? "ANY" : "???"); |
(rc == -1)? "ANY" : "???"); |
| 950 |
(void)pcre_config(PCRE_CONFIG_BSR, &rc); |
(void)pcre_config(PCRE_CONFIG_BSR, &rc); |
| 954 |
printf(" Internal link size = %d\n", rc); |
printf(" Internal link size = %d\n", rc); |
| 955 |
(void)pcre_config(PCRE_CONFIG_POSIX_MALLOC_THRESHOLD, &rc); |
(void)pcre_config(PCRE_CONFIG_POSIX_MALLOC_THRESHOLD, &rc); |
| 956 |
printf(" POSIX malloc threshold = %d\n", rc); |
printf(" POSIX malloc threshold = %d\n", rc); |
| 957 |
(void)pcre_config(PCRE_CONFIG_MATCH_LIMIT, &rc); |
(void)pcre_config(PCRE_CONFIG_MATCH_LIMIT, &lrc); |
| 958 |
printf(" Default match limit = %d\n", rc); |
printf(" Default match limit = %ld\n", lrc); |
| 959 |
(void)pcre_config(PCRE_CONFIG_MATCH_LIMIT_RECURSION, &rc); |
(void)pcre_config(PCRE_CONFIG_MATCH_LIMIT_RECURSION, &lrc); |
| 960 |
printf(" Default recursion depth limit = %d\n", rc); |
printf(" Default recursion depth limit = %ld\n", lrc); |
| 961 |
(void)pcre_config(PCRE_CONFIG_STACKRECURSE, &rc); |
(void)pcre_config(PCRE_CONFIG_STACKRECURSE, &rc); |
| 962 |
printf(" Match recursion uses %s\n", rc? "stack" : "heap"); |
printf(" Match recursion uses %s\n", rc? "stack" : "heap"); |
| 963 |
goto EXIT; |
goto EXIT; |
| 1056 |
use_utf8 = 0; |
use_utf8 = 0; |
| 1057 |
debug_lengths = 1; |
debug_lengths = 1; |
| 1058 |
|
|
| 1059 |
if (infile == stdin) printf(" re> "); |
if (extend_inputline(infile, buffer, " re> ") == NULL) break; |
|
if (extend_inputline(infile, buffer) == NULL) break; |
|
| 1060 |
if (infile != stdin) fprintf(outfile, "%s", (char *)buffer); |
if (infile != stdin) fprintf(outfile, "%s", (char *)buffer); |
| 1061 |
fflush(outfile); |
fflush(outfile); |
| 1062 |
|
|
| 1156 |
|
|
| 1157 |
if (isalnum(delimiter) || delimiter == '\\') |
if (isalnum(delimiter) || delimiter == '\\') |
| 1158 |
{ |
{ |
| 1159 |
fprintf(outfile, "** Delimiter must not be alphameric or \\\n"); |
fprintf(outfile, "** Delimiter must not be alphanumeric or \\\n"); |
| 1160 |
goto SKIP_DATA; |
goto SKIP_DATA; |
| 1161 |
} |
} |
| 1162 |
|
|
| 1172 |
pp++; |
pp++; |
| 1173 |
} |
} |
| 1174 |
if (*pp != 0) break; |
if (*pp != 0) break; |
| 1175 |
if (infile == stdin) printf(" > "); |
if ((pp = extend_inputline(infile, pp, " > ")) == NULL) |
|
if ((pp = extend_inputline(infile, pp)) == NULL) |
|
| 1176 |
{ |
{ |
| 1177 |
fprintf(outfile, "** Unexpected EOF\n"); |
fprintf(outfile, "** Unexpected EOF\n"); |
| 1178 |
done = 1; |
done = 1; |
| 1264 |
|
|
| 1265 |
case '<': |
case '<': |
| 1266 |
{ |
{ |
| 1267 |
int x = check_newline(pp, outfile); |
if (strncmp((char *)pp, "JS>", 3) == 0) |
| 1268 |
if (x == 0) goto SKIP_DATA; |
{ |
| 1269 |
options |= x; |
options |= PCRE_JAVASCRIPT_COMPAT; |
| 1270 |
while (*pp++ != '>'); |
pp += 3; |
| 1271 |
|
} |
| 1272 |
|
else |
| 1273 |
|
{ |
| 1274 |
|
int x = check_newline(pp, outfile); |
| 1275 |
|
if (x == 0) goto SKIP_DATA; |
| 1276 |
|
options |= x; |
| 1277 |
|
while (*pp++ != '>'); |
| 1278 |
|
} |
| 1279 |
} |
} |
| 1280 |
break; |
break; |
| 1281 |
|
|
| 1305 |
if ((options & PCRE_DOTALL) != 0) cflags |= REG_DOTALL; |
if ((options & PCRE_DOTALL) != 0) cflags |= REG_DOTALL; |
| 1306 |
if ((options & PCRE_NO_AUTO_CAPTURE) != 0) cflags |= REG_NOSUB; |
if ((options & PCRE_NO_AUTO_CAPTURE) != 0) cflags |= REG_NOSUB; |
| 1307 |
if ((options & PCRE_UTF8) != 0) cflags |= REG_UTF8; |
if ((options & PCRE_UTF8) != 0) cflags |= REG_UTF8; |
| 1308 |
|
if ((options & PCRE_UNGREEDY) != 0) cflags |= REG_UNGREEDY; |
| 1309 |
|
|
| 1310 |
rc = regcomp(&preg, (char *)p, cflags); |
rc = regcomp(&preg, (char *)p, cflags); |
| 1311 |
|
|
| 1326 |
#endif /* !defined NOPOSIX */ |
#endif /* !defined NOPOSIX */ |
| 1327 |
|
|
| 1328 |
{ |
{ |
| 1329 |
|
unsigned long int get_options; |
| 1330 |
|
|
| 1331 |
if (timeit > 0) |
if (timeit > 0) |
| 1332 |
{ |
{ |
| 1333 |
register int i; |
register int i; |
| 1357 |
{ |
{ |
| 1358 |
for (;;) |
for (;;) |
| 1359 |
{ |
{ |
| 1360 |
if (extend_inputline(infile, buffer) == NULL) |
if (extend_inputline(infile, buffer, NULL) == NULL) |
| 1361 |
{ |
{ |
| 1362 |
done = 1; |
done = 1; |
| 1363 |
goto CONTINUE; |
goto CONTINUE; |
| 1371 |
goto CONTINUE; |
goto CONTINUE; |
| 1372 |
} |
} |
| 1373 |
|
|
| 1374 |
/* Compilation succeeded; print data if required. There are now two |
/* Compilation succeeded. It is now possible to set the UTF-8 option from |
| 1375 |
info-returning functions. The old one has a limited interface and |
within the regex; check for this so that we know how to process the data |
| 1376 |
returns only limited data. Check that it agrees with the newer one. */ |
lines. */ |
| 1377 |
|
|
| 1378 |
|
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
| 1379 |
|
if ((get_options & PCRE_UTF8) != 0) use_utf8 = 1; |
| 1380 |
|
|
| 1381 |
|
/* Print information if required. There are now two info-returning |
| 1382 |
|
functions. The old one has a limited interface and returns only limited |
| 1383 |
|
data. Check that it agrees with the newer one. */ |
| 1384 |
|
|
| 1385 |
if (log_store) |
if (log_store) |
| 1386 |
fprintf(outfile, "Memory allocation (code space): %d\n", |
fprintf(outfile, "Memory allocation (code space): %d\n", |
| 1427 |
if (do_flip) |
if (do_flip) |
| 1428 |
{ |
{ |
| 1429 |
real_pcre *rre = (real_pcre *)re; |
real_pcre *rre = (real_pcre *)re; |
| 1430 |
rre->magic_number = byteflip(rre->magic_number, sizeof(rre->magic_number)); |
rre->magic_number = |
| 1431 |
|
byteflip(rre->magic_number, sizeof(rre->magic_number)); |
| 1432 |
rre->size = byteflip(rre->size, sizeof(rre->size)); |
rre->size = byteflip(rre->size, sizeof(rre->size)); |
| 1433 |
rre->options = byteflip(rre->options, sizeof(rre->options)); |
rre->options = byteflip(rre->options, sizeof(rre->options)); |
| 1434 |
rre->flags = byteflip(rre->flags, sizeof(rre->flags)); |
rre->flags = (pcre_uint16)byteflip(rre->flags, sizeof(rre->flags)); |
| 1435 |
rre->top_bracket = byteflip(rre->top_bracket, sizeof(rre->top_bracket)); |
rre->top_bracket = |
| 1436 |
rre->top_backref = byteflip(rre->top_backref, sizeof(rre->top_backref)); |
(pcre_uint16)byteflip(rre->top_bracket, sizeof(rre->top_bracket)); |
| 1437 |
rre->first_byte = byteflip(rre->first_byte, sizeof(rre->first_byte)); |
rre->top_backref = |
| 1438 |
rre->req_byte = byteflip(rre->req_byte, sizeof(rre->req_byte)); |
(pcre_uint16)byteflip(rre->top_backref, sizeof(rre->top_backref)); |
| 1439 |
rre->name_table_offset = byteflip(rre->name_table_offset, |
rre->first_byte = |
| 1440 |
|
(pcre_uint16)byteflip(rre->first_byte, sizeof(rre->first_byte)); |
| 1441 |
|
rre->req_byte = |
| 1442 |
|
(pcre_uint16)byteflip(rre->req_byte, sizeof(rre->req_byte)); |
| 1443 |
|
rre->name_table_offset = (pcre_uint16)byteflip(rre->name_table_offset, |
| 1444 |
sizeof(rre->name_table_offset)); |
sizeof(rre->name_table_offset)); |
| 1445 |
rre->name_entry_size = byteflip(rre->name_entry_size, |
rre->name_entry_size = (pcre_uint16)byteflip(rre->name_entry_size, |
| 1446 |
sizeof(rre->name_entry_size)); |
sizeof(rre->name_entry_size)); |
| 1447 |
rre->name_count = byteflip(rre->name_count, sizeof(rre->name_count)); |
rre->name_count = (pcre_uint16)byteflip(rre->name_count, |
| 1448 |
|
sizeof(rre->name_count)); |
| 1449 |
|
|
| 1450 |
if (extra != NULL) |
if (extra != NULL) |
| 1451 |
{ |
{ |
| 1465 |
pcre_printint(re, outfile, debug_lengths); |
pcre_printint(re, outfile, debug_lengths); |
| 1466 |
} |
} |
| 1467 |
|
|
| 1468 |
|
/* We already have the options in get_options (see above) */ |
| 1469 |
|
|
| 1470 |
if (do_showinfo) |
if (do_showinfo) |
| 1471 |
{ |
{ |
| 1472 |
unsigned long int get_options, all_options; |
unsigned long int all_options; |
| 1473 |
#if !defined NOINFOCHECK |
#if !defined NOINFOCHECK |
| 1474 |
int old_first_char, old_options, old_count; |
int old_first_char, old_options, old_count; |
| 1475 |
#endif |
#endif |
| 1478 |
int nameentrysize, namecount; |
int nameentrysize, namecount; |
| 1479 |
const uschar *nametable; |
const uschar *nametable; |
| 1480 |
|
|
|
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
|
| 1481 |
new_info(re, NULL, PCRE_INFO_SIZE, &size); |
new_info(re, NULL, PCRE_INFO_SIZE, &size); |
| 1482 |
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count); |
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count); |
| 1483 |
new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax); |
new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax); |
| 1678 |
else |
else |
| 1679 |
{ |
{ |
| 1680 |
uschar sbuf[8]; |
uschar sbuf[8]; |
| 1681 |
sbuf[0] = (true_size >> 24) & 255; |
sbuf[0] = (uschar)((true_size >> 24) & 255); |
| 1682 |
sbuf[1] = (true_size >> 16) & 255; |
sbuf[1] = (uschar)((true_size >> 16) & 255); |
| 1683 |
sbuf[2] = (true_size >> 8) & 255; |
sbuf[2] = (uschar)((true_size >> 8) & 255); |
| 1684 |
sbuf[3] = (true_size) & 255; |
sbuf[3] = (uschar)((true_size) & 255); |
| 1685 |
|
|
| 1686 |
sbuf[4] = (true_study_size >> 24) & 255; |
sbuf[4] = (uschar)((true_study_size >> 24) & 255); |
| 1687 |
sbuf[5] = (true_study_size >> 16) & 255; |
sbuf[5] = (uschar)((true_study_size >> 16) & 255); |
| 1688 |
sbuf[6] = (true_study_size >> 8) & 255; |
sbuf[6] = (uschar)((true_study_size >> 8) & 255); |
| 1689 |
sbuf[7] = (true_study_size) & 255; |
sbuf[7] = (uschar)((true_study_size) & 255); |
| 1690 |
|
|
| 1691 |
if (fwrite(sbuf, 1, 8, f) < 8 || |
if (fwrite(sbuf, 1, 8, f) < 8 || |
| 1692 |
fwrite(re, 1, true_size, f) < true_size) |
fwrite(re, 1, true_size, f) < true_size) |
| 1730 |
int callout_data_set = 0; |
int callout_data_set = 0; |
| 1731 |
int count, c; |
int count, c; |
| 1732 |
int copystrings = 0; |
int copystrings = 0; |
| 1733 |
int find_match_limit = 0; |
int find_match_limit = default_find_match_limit; |
| 1734 |
int getstrings = 0; |
int getstrings = 0; |
| 1735 |
int getlist = 0; |
int getlist = 0; |
| 1736 |
int gmatched = 0; |
int gmatched = 0; |
| 1760 |
len = 0; |
len = 0; |
| 1761 |
for (;;) |
for (;;) |
| 1762 |
{ |
{ |
| 1763 |
if (infile == stdin) printf("data> "); |
if (extend_inputline(infile, buffer + len, "data> ") == NULL) |
|
if (extend_inputline(infile, buffer + len) == NULL) |
|
| 1764 |
{ |
{ |
| 1765 |
if (len > 0) break; |
if (len > 0) break; |
| 1766 |
done = 1; |
done = 1; |
| 1828 |
{ |
{ |
| 1829 |
unsigned char buff8[8]; |
unsigned char buff8[8]; |
| 1830 |
int ii, utn; |
int ii, utn; |
| 1831 |
utn = ord2utf8(c, buff8); |
if (use_utf8) |
| 1832 |
for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii]; |
{ |
| 1833 |
c = buff8[ii]; /* Last byte */ |
utn = ord2utf8(c, buff8); |
| 1834 |
|
for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii]; |
| 1835 |
|
c = buff8[ii]; /* Last byte */ |
| 1836 |
|
} |
| 1837 |
|
else |
| 1838 |
|
{ |
| 1839 |
|
if (c > 255) |
| 1840 |
|
fprintf(outfile, "** Character \\x{%x} is greater than 255 and " |
| 1841 |
|
"UTF-8 mode is not enabled.\n" |
| 1842 |
|
"** Truncation will probably give the wrong result.\n", c); |
| 1843 |
|
} |
| 1844 |
p = pt + 1; |
p = pt + 1; |
| 1845 |
break; |
break; |
| 1846 |
} |
} |
| 1970 |
continue; |
continue; |
| 1971 |
|
|
| 1972 |
case 'N': |
case 'N': |
| 1973 |
options |= PCRE_NOTEMPTY; |
if ((options & PCRE_NOTEMPTY) != 0) |
| 1974 |
|
options = (options & ~PCRE_NOTEMPTY) | PCRE_NOTEMPTY_ATSTART; |
| 1975 |
|
else |
| 1976 |
|
options |= PCRE_NOTEMPTY; |
| 1977 |
continue; |
continue; |
| 1978 |
|
|
| 1979 |
case 'O': |
case 'O': |
| 1996 |
continue; |
continue; |
| 1997 |
|
|
| 1998 |
case 'P': |
case 'P': |
| 1999 |
options |= PCRE_PARTIAL; |
options |= ((options & PCRE_PARTIAL_SOFT) == 0)? |
| 2000 |
|
PCRE_PARTIAL_SOFT : PCRE_PARTIAL_HARD; |
| 2001 |
continue; |
continue; |
| 2002 |
|
|
| 2003 |
case 'Q': |
case 'Q': |
| 2032 |
show_malloc = 1; |
show_malloc = 1; |
| 2033 |
continue; |
continue; |
| 2034 |
|
|
| 2035 |
|
case 'Y': |
| 2036 |
|
options |= PCRE_NO_START_OPTIMIZE; |
| 2037 |
|
continue; |
| 2038 |
|
|
| 2039 |
case 'Z': |
case 'Z': |
| 2040 |
options |= PCRE_NOTEOL; |
options |= PCRE_NOTEOL; |
| 2041 |
continue; |
continue; |
| 2058 |
*q = 0; |
*q = 0; |
| 2059 |
len = q - dbuffer; |
len = q - dbuffer; |
| 2060 |
|
|
| 2061 |
|
/* Move the data to the end of the buffer so that a read over the end of |
| 2062 |
|
the buffer will be seen by valgrind, even if it doesn't cause a crash. If |
| 2063 |
|
we are using the POSIX interface, we must include the terminating zero. */ |
| 2064 |
|
|
| 2065 |
|
#if !defined NOPOSIX |
| 2066 |
|
if (posix || do_posix) |
| 2067 |
|
{ |
| 2068 |
|
memmove(bptr + buffer_size - len - 1, bptr, len + 1); |
| 2069 |
|
bptr += buffer_size - len - 1; |
| 2070 |
|
} |
| 2071 |
|
else |
| 2072 |
|
#endif |
| 2073 |
|
{ |
| 2074 |
|
memmove(bptr + buffer_size - len, bptr, len); |
| 2075 |
|
bptr += buffer_size - len; |
| 2076 |
|
} |
| 2077 |
|
|
| 2078 |
if ((all_use_dfa || use_dfa) && find_match_limit) |
if ((all_use_dfa || use_dfa) && find_match_limit) |
| 2079 |
{ |
{ |
| 2080 |
printf("**Match limit not relevant for DFA matching: ignored\n"); |
printf("**Match limit not relevant for DFA matching: ignored\n"); |
| 2094 |
pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets); |
pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets); |
| 2095 |
if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL; |
if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL; |
| 2096 |
if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL; |
if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL; |
| 2097 |
|
if ((options & PCRE_NOTEMPTY) != 0) eflags |= REG_NOTEMPTY; |
| 2098 |
|
|
| 2099 |
rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags); |
rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags); |
| 2100 |
|
|
| 2368 |
else if (count == PCRE_ERROR_PARTIAL) |
else if (count == PCRE_ERROR_PARTIAL) |
| 2369 |
{ |
{ |
| 2370 |
fprintf(outfile, "Partial match"); |
fprintf(outfile, "Partial match"); |
| 2371 |
#if !defined NODFA |
if (use_size_offsets > 1) |
| 2372 |
if ((all_use_dfa || use_dfa) && use_size_offsets > 2) |
{ |
| 2373 |
fprintf(outfile, ": %.*s", use_offsets[1] - use_offsets[0], |
fprintf(outfile, ": "); |
| 2374 |
bptr + use_offsets[0]); |
pchars(bptr + use_offsets[0], use_offsets[1] - use_offsets[0], |
| 2375 |
#endif |
outfile); |
| 2376 |
|
} |
| 2377 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 2378 |
break; /* Out of the /g loop */ |
break; /* Out of the /g loop */ |
| 2379 |
} |
} |
| 2403 |
{ |
{ |
| 2404 |
int d; |
int d; |
| 2405 |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &d); |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &d); |
| 2406 |
obits = (d == '\r')? PCRE_NEWLINE_CR : |
/* Note that these values are always the ASCII ones, even in |
| 2407 |
(d == '\n')? PCRE_NEWLINE_LF : |
EBCDIC environments. CR = 13, NL = 10. */ |
| 2408 |
(d == ('\r'<<8 | '\n'))? PCRE_NEWLINE_CRLF : |
obits = (d == 13)? PCRE_NEWLINE_CR : |
| 2409 |
|
(d == 10)? PCRE_NEWLINE_LF : |
| 2410 |
|
(d == (13<<8 | 10))? PCRE_NEWLINE_CRLF : |
| 2411 |
(d == -2)? PCRE_NEWLINE_ANYCRLF : |
(d == -2)? PCRE_NEWLINE_ANYCRLF : |
| 2412 |
(d == -1)? PCRE_NEWLINE_ANY : 0; |
(d == -1)? PCRE_NEWLINE_ANY : 0; |
| 2413 |
} |
} |
| 2446 |
if (!do_g && !do_G) break; |
if (!do_g && !do_G) break; |
| 2447 |
|
|
| 2448 |
/* If we have matched an empty string, first check to see if we are at |
/* If we have matched an empty string, first check to see if we are at |
| 2449 |
the end of the subject. If so, the /g loop is over. Otherwise, mimic |
the end of the subject. If so, the /g loop is over. Otherwise, mimic what |
| 2450 |
what Perl's /g options does. This turns out to be rather cunning. First |
Perl's /g options does. This turns out to be rather cunning. First we set |
| 2451 |
we set PCRE_NOTEMPTY and PCRE_ANCHORED and try the match again at the |
PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED and try the match again at the |
| 2452 |
same point. If this fails (picked up above) we advance to the next |
same point. If this fails (picked up above) we advance to the next |
| 2453 |
character. */ |
character. */ |
| 2454 |
|
|
| 2457 |
if (use_offsets[0] == use_offsets[1]) |
if (use_offsets[0] == use_offsets[1]) |
| 2458 |
{ |
{ |
| 2459 |
if (use_offsets[0] == len) break; |
if (use_offsets[0] == len) break; |
| 2460 |
g_notempty = PCRE_NOTEMPTY | PCRE_ANCHORED; |
g_notempty = PCRE_NOTEMPTY_ATSTART | PCRE_ANCHORED; |
| 2461 |
} |
} |
| 2462 |
|
|
| 2463 |
/* For /g, update the start offset, leaving the rest alone */ |
/* For /g, update the start offset, leaving the rest alone */ |