| 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 |
| 118 |
|
|
| 119 |
/* We also need the pcre_printint() function for printing out compiled |
/* We also need the pcre_printint() function for printing out compiled |
| 120 |
patterns. This function is in a separate file so that it can be included in |
patterns. This function is in a separate file so that it can be included in |
| 121 |
pcre_compile.c when that module is compiled with debugging enabled. |
pcre_compile.c when that module is compiled with debugging enabled. It needs to |
| 122 |
|
know which case is being compiled. */ |
|
The definition of the macro PRINTABLE, which determines whether to print an |
|
|
output character as-is or as a hex value when showing compiled patterns, is |
|
|
contained in this file. We uses it here also, in cases when the locale has not |
|
|
been explicitly changed, so as to get consistent output from systems that |
|
|
differ in their output from isprint() even in the "C" locale. */ |
|
| 123 |
|
|
| 124 |
|
#define COMPILING_PCRETEST |
| 125 |
#include "pcre_printint.src" |
#include "pcre_printint.src" |
| 126 |
|
|
| 127 |
#define PRINTHEX(c) (locale_set? isprint(c) : PRINTABLE(c)) |
/* The definition of the macro PRINTABLE, which determines whether to print an |
| 128 |
|
output character as-is or as a hex value when showing compiled patterns, is |
| 129 |
|
contained in the printint.src file. We uses it here also, in cases when the |
| 130 |
|
locale has not been explicitly changed, so as to get consistent output from |
| 131 |
|
systems that differ in their output from isprint() even in the "C" locale. */ |
| 132 |
|
|
| 133 |
|
#define PRINTHEX(c) (locale_set? isprint(c) : PRINTABLE(c)) |
| 134 |
|
|
| 135 |
/* It is possible to compile this test program without including support for |
/* It is possible to compile this test program without including support for |
| 136 |
testing the POSIX interface, though this is not available via the standard |
testing the POSIX interface, though this is not available via the standard |
| 207 |
Arguments: |
Arguments: |
| 208 |
f the file to read |
f the file to read |
| 209 |
start where in buffer to start (this *must* be within buffer) |
start where in buffer to start (this *must* be within buffer) |
| 210 |
|
prompt for stdin or readline() |
| 211 |
|
|
| 212 |
Returns: pointer to the start of new data |
Returns: pointer to the start of new data |
| 213 |
could be a copy of start, or could be moved |
could be a copy of start, or could be moved |
| 215 |
*/ |
*/ |
| 216 |
|
|
| 217 |
static uschar * |
static uschar * |
| 218 |
extend_inputline(FILE *f, uschar *start) |
extend_inputline(FILE *f, uschar *start, const char *prompt) |
| 219 |
{ |
{ |
| 220 |
uschar *here = start; |
uschar *here = start; |
| 221 |
|
|
| 222 |
for (;;) |
for (;;) |
| 223 |
{ |
{ |
| 224 |
int rlen = buffer_size - (here - buffer); |
int rlen = (int)(buffer_size - (here - buffer)); |
| 225 |
|
|
| 226 |
if (rlen > 1000) |
if (rlen > 1000) |
| 227 |
{ |
{ |
| 228 |
int dlen; |
int dlen; |
| 229 |
if (fgets((char *)here, rlen, f) == NULL) |
|
| 230 |
return (here == start)? NULL : start; |
/* If libreadline support is required, use readline() to read a line if the |
| 231 |
|
input is a terminal. Note that readline() removes the trailing newline, so |
| 232 |
|
we must put it back again, to be compatible with fgets(). */ |
| 233 |
|
|
| 234 |
|
#ifdef SUPPORT_LIBREADLINE |
| 235 |
|
if (isatty(fileno(f))) |
| 236 |
|
{ |
| 237 |
|
size_t len; |
| 238 |
|
char *s = readline(prompt); |
| 239 |
|
if (s == NULL) return (here == start)? NULL : start; |
| 240 |
|
len = strlen(s); |
| 241 |
|
if (len > 0) add_history(s); |
| 242 |
|
if (len > rlen - 1) len = rlen - 1; |
| 243 |
|
memcpy(here, s, len); |
| 244 |
|
here[len] = '\n'; |
| 245 |
|
here[len+1] = 0; |
| 246 |
|
free(s); |
| 247 |
|
} |
| 248 |
|
else |
| 249 |
|
#endif |
| 250 |
|
|
| 251 |
|
/* Read the next line by normal means, prompting if the file is stdin. */ |
| 252 |
|
|
| 253 |
|
{ |
| 254 |
|
if (f == stdin) printf("%s", prompt); |
| 255 |
|
if (fgets((char *)here, rlen, f) == NULL) |
| 256 |
|
return (here == start)? NULL : start; |
| 257 |
|
} |
| 258 |
|
|
| 259 |
dlen = (int)strlen((char *)here); |
dlen = (int)strlen((char *)here); |
| 260 |
if (dlen > 0 && here[dlen - 1] == '\n') return start; |
if (dlen > 0 && here[dlen - 1] == '\n') return start; |
| 261 |
here += dlen; |
here += dlen; |
| 741 |
* Check newline indicator * |
* Check newline indicator * |
| 742 |
*************************************************/ |
*************************************************/ |
| 743 |
|
|
| 744 |
/* This is used both at compile and run-time to check for <xxx> escapes, where |
/* This is used both at compile and run-time to check for <xxx> escapes. Print |
| 745 |
xxx is LF, CR, CRLF, ANYCRLF, or ANY. Print a message and return 0 if there is |
a message and return 0 if there is no match. |
|
no match. |
|
| 746 |
|
|
| 747 |
Arguments: |
Arguments: |
| 748 |
p points after the leading '<' |
p points after the leading '<' |
| 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; |
| 1039 |
#endif |
#endif |
| 1040 |
|
|
| 1041 |
const char *error; |
const char *error; |
| 1042 |
|
unsigned char *markptr; |
| 1043 |
unsigned char *p, *pp, *ppp; |
unsigned char *p, *pp, *ppp; |
| 1044 |
unsigned char *to_file = NULL; |
unsigned char *to_file = NULL; |
| 1045 |
const unsigned char *tables = NULL; |
const unsigned char *tables = NULL; |
| 1046 |
unsigned long int true_size, true_study_size = 0; |
unsigned long int true_size, true_study_size = 0; |
| 1047 |
size_t size, regex_gotten_store; |
size_t size, regex_gotten_store; |
| 1048 |
|
int do_mark = 0; |
| 1049 |
int do_study = 0; |
int do_study = 0; |
| 1050 |
int do_debug = debug; |
int do_debug = debug; |
| 1051 |
int do_G = 0; |
int do_G = 0; |
| 1058 |
use_utf8 = 0; |
use_utf8 = 0; |
| 1059 |
debug_lengths = 1; |
debug_lengths = 1; |
| 1060 |
|
|
| 1061 |
if (infile == stdin) printf(" re> "); |
if (extend_inputline(infile, buffer, " re> ") == NULL) break; |
|
if (extend_inputline(infile, buffer) == NULL) break; |
|
| 1062 |
if (infile != stdin) fprintf(outfile, "%s", (char *)buffer); |
if (infile != stdin) fprintf(outfile, "%s", (char *)buffer); |
| 1063 |
fflush(outfile); |
fflush(outfile); |
| 1064 |
|
|
| 1163 |
} |
} |
| 1164 |
|
|
| 1165 |
pp = p; |
pp = p; |
| 1166 |
poffset = p - buffer; |
poffset = (int)(p - buffer); |
| 1167 |
|
|
| 1168 |
for(;;) |
for(;;) |
| 1169 |
{ |
{ |
| 1174 |
pp++; |
pp++; |
| 1175 |
} |
} |
| 1176 |
if (*pp != 0) break; |
if (*pp != 0) break; |
| 1177 |
if (infile == stdin) printf(" > "); |
if ((pp = extend_inputline(infile, pp, " > ")) == NULL) |
|
if ((pp = extend_inputline(infile, pp)) == NULL) |
|
| 1178 |
{ |
{ |
| 1179 |
fprintf(outfile, "** Unexpected EOF\n"); |
fprintf(outfile, "** Unexpected EOF\n"); |
| 1180 |
done = 1; |
done = 1; |
| 1227 |
case 'G': do_G = 1; break; |
case 'G': do_G = 1; break; |
| 1228 |
case 'I': do_showinfo = 1; break; |
case 'I': do_showinfo = 1; break; |
| 1229 |
case 'J': options |= PCRE_DUPNAMES; break; |
case 'J': options |= PCRE_DUPNAMES; break; |
| 1230 |
|
case 'K': do_mark = 1; break; |
| 1231 |
case 'M': log_store = 1; break; |
case 'M': log_store = 1; break; |
| 1232 |
case 'N': options |= PCRE_NO_AUTO_CAPTURE; break; |
case 'N': options |= PCRE_NO_AUTO_CAPTURE; break; |
| 1233 |
|
|
| 1237 |
|
|
| 1238 |
case 'S': do_study = 1; break; |
case 'S': do_study = 1; break; |
| 1239 |
case 'U': options |= PCRE_UNGREEDY; break; |
case 'U': options |= PCRE_UNGREEDY; break; |
| 1240 |
|
case 'W': options |= PCRE_UCP; break; |
| 1241 |
case 'X': options |= PCRE_EXTRA; break; |
case 'X': options |= PCRE_EXTRA; break; |
| 1242 |
case 'Z': debug_lengths = 0; break; |
case 'Z': debug_lengths = 0; break; |
| 1243 |
case '8': options |= PCRE_UTF8; use_utf8 = 1; break; |
case '8': options |= PCRE_UTF8; use_utf8 = 1; break; |
| 1268 |
|
|
| 1269 |
case '<': |
case '<': |
| 1270 |
{ |
{ |
| 1271 |
int x = check_newline(pp, outfile); |
if (strncmpic(pp, (uschar *)"JS>", 3) == 0) |
| 1272 |
if (x == 0) goto SKIP_DATA; |
{ |
| 1273 |
options |= x; |
options |= PCRE_JAVASCRIPT_COMPAT; |
| 1274 |
while (*pp++ != '>'); |
pp += 3; |
| 1275 |
|
} |
| 1276 |
|
else |
| 1277 |
|
{ |
| 1278 |
|
int x = check_newline(pp, outfile); |
| 1279 |
|
if (x == 0) goto SKIP_DATA; |
| 1280 |
|
options |= x; |
| 1281 |
|
while (*pp++ != '>'); |
| 1282 |
|
} |
| 1283 |
} |
} |
| 1284 |
break; |
break; |
| 1285 |
|
|
| 1309 |
if ((options & PCRE_DOTALL) != 0) cflags |= REG_DOTALL; |
if ((options & PCRE_DOTALL) != 0) cflags |= REG_DOTALL; |
| 1310 |
if ((options & PCRE_NO_AUTO_CAPTURE) != 0) cflags |= REG_NOSUB; |
if ((options & PCRE_NO_AUTO_CAPTURE) != 0) cflags |= REG_NOSUB; |
| 1311 |
if ((options & PCRE_UTF8) != 0) cflags |= REG_UTF8; |
if ((options & PCRE_UTF8) != 0) cflags |= REG_UTF8; |
| 1312 |
|
if ((options & PCRE_UCP) != 0) cflags |= REG_UCP; |
| 1313 |
|
if ((options & PCRE_UNGREEDY) != 0) cflags |= REG_UNGREEDY; |
| 1314 |
|
|
| 1315 |
rc = regcomp(&preg, (char *)p, cflags); |
rc = regcomp(&preg, (char *)p, cflags); |
| 1316 |
|
|
| 1331 |
#endif /* !defined NOPOSIX */ |
#endif /* !defined NOPOSIX */ |
| 1332 |
|
|
| 1333 |
{ |
{ |
| 1334 |
|
unsigned long int get_options; |
| 1335 |
|
|
| 1336 |
if (timeit > 0) |
if (timeit > 0) |
| 1337 |
{ |
{ |
| 1338 |
register int i; |
register int i; |
| 1362 |
{ |
{ |
| 1363 |
for (;;) |
for (;;) |
| 1364 |
{ |
{ |
| 1365 |
if (extend_inputline(infile, buffer) == NULL) |
if (extend_inputline(infile, buffer, NULL) == NULL) |
| 1366 |
{ |
{ |
| 1367 |
done = 1; |
done = 1; |
| 1368 |
goto CONTINUE; |
goto CONTINUE; |
| 1376 |
goto CONTINUE; |
goto CONTINUE; |
| 1377 |
} |
} |
| 1378 |
|
|
| 1379 |
/* Compilation succeeded; print data if required. There are now two |
/* Compilation succeeded. It is now possible to set the UTF-8 option from |
| 1380 |
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 |
| 1381 |
returns only limited data. Check that it agrees with the newer one. */ |
lines. */ |
| 1382 |
|
|
| 1383 |
|
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
| 1384 |
|
if ((get_options & PCRE_UTF8) != 0) use_utf8 = 1; |
| 1385 |
|
|
| 1386 |
|
/* Print information if required. There are now two info-returning |
| 1387 |
|
functions. The old one has a limited interface and returns only limited |
| 1388 |
|
data. Check that it agrees with the newer one. */ |
| 1389 |
|
|
| 1390 |
if (log_store) |
if (log_store) |
| 1391 |
fprintf(outfile, "Memory allocation (code space): %d\n", |
fprintf(outfile, "Memory allocation (code space): %d\n", |
| 1424 |
true_study_size = ((pcre_study_data *)(extra->study_data))->size; |
true_study_size = ((pcre_study_data *)(extra->study_data))->size; |
| 1425 |
} |
} |
| 1426 |
|
|
| 1427 |
|
/* If /K was present, we set up for handling MARK data. */ |
| 1428 |
|
|
| 1429 |
|
if (do_mark) |
| 1430 |
|
{ |
| 1431 |
|
if (extra == NULL) |
| 1432 |
|
{ |
| 1433 |
|
extra = (pcre_extra *)malloc(sizeof(pcre_extra)); |
| 1434 |
|
extra->flags = 0; |
| 1435 |
|
} |
| 1436 |
|
extra->mark = &markptr; |
| 1437 |
|
extra->flags |= PCRE_EXTRA_MARK; |
| 1438 |
|
} |
| 1439 |
|
|
| 1440 |
/* If the 'F' option was present, we flip the bytes of all the integer |
/* If the 'F' option was present, we flip the bytes of all the integer |
| 1441 |
fields in the regex data block and the study block. This is to make it |
fields in the regex data block and the study block. This is to make it |
| 1442 |
possible to test PCRE's handling of byte-flipped patterns, e.g. those |
possible to test PCRE's handling of byte-flipped patterns, e.g. those |
| 1469 |
{ |
{ |
| 1470 |
pcre_study_data *rsd = (pcre_study_data *)(extra->study_data); |
pcre_study_data *rsd = (pcre_study_data *)(extra->study_data); |
| 1471 |
rsd->size = byteflip(rsd->size, sizeof(rsd->size)); |
rsd->size = byteflip(rsd->size, sizeof(rsd->size)); |
| 1472 |
rsd->options = byteflip(rsd->options, sizeof(rsd->options)); |
rsd->flags = byteflip(rsd->flags, sizeof(rsd->flags)); |
| 1473 |
|
rsd->minlength = byteflip(rsd->minlength, sizeof(rsd->minlength)); |
| 1474 |
} |
} |
| 1475 |
} |
} |
| 1476 |
|
|
| 1484 |
pcre_printint(re, outfile, debug_lengths); |
pcre_printint(re, outfile, debug_lengths); |
| 1485 |
} |
} |
| 1486 |
|
|
| 1487 |
|
/* We already have the options in get_options (see above) */ |
| 1488 |
|
|
| 1489 |
if (do_showinfo) |
if (do_showinfo) |
| 1490 |
{ |
{ |
| 1491 |
unsigned long int get_options, all_options; |
unsigned long int all_options; |
| 1492 |
#if !defined NOINFOCHECK |
#if !defined NOINFOCHECK |
| 1493 |
int old_first_char, old_options, old_count; |
int old_first_char, old_options, old_count; |
| 1494 |
#endif |
#endif |
| 1497 |
int nameentrysize, namecount; |
int nameentrysize, namecount; |
| 1498 |
const uschar *nametable; |
const uschar *nametable; |
| 1499 |
|
|
|
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
|
| 1500 |
new_info(re, NULL, PCRE_INFO_SIZE, &size); |
new_info(re, NULL, PCRE_INFO_SIZE, &size); |
| 1501 |
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count); |
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count); |
| 1502 |
new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax); |
new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax); |
| 1556 |
if (do_flip) all_options = byteflip(all_options, sizeof(all_options)); |
if (do_flip) all_options = byteflip(all_options, sizeof(all_options)); |
| 1557 |
|
|
| 1558 |
if (get_options == 0) fprintf(outfile, "No options\n"); |
if (get_options == 0) fprintf(outfile, "No options\n"); |
| 1559 |
else fprintf(outfile, "Options:%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\n", |
| 1560 |
((get_options & PCRE_ANCHORED) != 0)? " anchored" : "", |
((get_options & PCRE_ANCHORED) != 0)? " anchored" : "", |
| 1561 |
((get_options & PCRE_CASELESS) != 0)? " caseless" : "", |
((get_options & PCRE_CASELESS) != 0)? " caseless" : "", |
| 1562 |
((get_options & PCRE_EXTENDED) != 0)? " extended" : "", |
((get_options & PCRE_EXTENDED) != 0)? " extended" : "", |
| 1570 |
((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "", |
((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "", |
| 1571 |
((get_options & PCRE_NO_AUTO_CAPTURE) != 0)? " no_auto_capture" : "", |
((get_options & PCRE_NO_AUTO_CAPTURE) != 0)? " no_auto_capture" : "", |
| 1572 |
((get_options & PCRE_UTF8) != 0)? " utf8" : "", |
((get_options & PCRE_UTF8) != 0)? " utf8" : "", |
| 1573 |
|
((get_options & PCRE_UCP) != 0)? " ucp" : "", |
| 1574 |
((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "", |
((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "", |
| 1575 |
((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : ""); |
((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : ""); |
| 1576 |
|
|
| 1648 |
else |
else |
| 1649 |
{ |
{ |
| 1650 |
uschar *start_bits = NULL; |
uschar *start_bits = NULL; |
| 1651 |
new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits); |
int minlength; |
| 1652 |
|
|
| 1653 |
|
new_info(re, extra, PCRE_INFO_MINLENGTH, &minlength); |
| 1654 |
|
fprintf(outfile, "Subject length lower bound = %d\n", minlength); |
| 1655 |
|
|
| 1656 |
|
new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits); |
| 1657 |
if (start_bits == NULL) |
if (start_bits == NULL) |
| 1658 |
fprintf(outfile, "No starting byte set\n"); |
fprintf(outfile, "No set of starting bytes\n"); |
| 1659 |
else |
else |
| 1660 |
{ |
{ |
| 1661 |
int i; |
int i; |
| 1754 |
int callout_data_set = 0; |
int callout_data_set = 0; |
| 1755 |
int count, c; |
int count, c; |
| 1756 |
int copystrings = 0; |
int copystrings = 0; |
| 1757 |
int find_match_limit = 0; |
int find_match_limit = default_find_match_limit; |
| 1758 |
int getstrings = 0; |
int getstrings = 0; |
| 1759 |
int getlist = 0; |
int getlist = 0; |
| 1760 |
int gmatched = 0; |
int gmatched = 0; |
| 1784 |
len = 0; |
len = 0; |
| 1785 |
for (;;) |
for (;;) |
| 1786 |
{ |
{ |
| 1787 |
if (infile == stdin) printf("data> "); |
if (extend_inputline(infile, buffer + len, "data> ") == NULL) |
|
if (extend_inputline(infile, buffer + len) == NULL) |
|
| 1788 |
{ |
{ |
| 1789 |
if (len > 0) break; |
if (len > 0) break; |
| 1790 |
done = 1; |
done = 1; |
| 1852 |
{ |
{ |
| 1853 |
unsigned char buff8[8]; |
unsigned char buff8[8]; |
| 1854 |
int ii, utn; |
int ii, utn; |
| 1855 |
utn = ord2utf8(c, buff8); |
if (use_utf8) |
| 1856 |
for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii]; |
{ |
| 1857 |
c = buff8[ii]; /* Last byte */ |
utn = ord2utf8(c, buff8); |
| 1858 |
|
for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii]; |
| 1859 |
|
c = buff8[ii]; /* Last byte */ |
| 1860 |
|
} |
| 1861 |
|
else |
| 1862 |
|
{ |
| 1863 |
|
if (c > 255) |
| 1864 |
|
fprintf(outfile, "** Character \\x{%x} is greater than 255 and " |
| 1865 |
|
"UTF-8 mode is not enabled.\n" |
| 1866 |
|
"** Truncation will probably give the wrong result.\n", c); |
| 1867 |
|
} |
| 1868 |
p = pt + 1; |
p = pt + 1; |
| 1869 |
break; |
break; |
| 1870 |
} |
} |
| 1994 |
continue; |
continue; |
| 1995 |
|
|
| 1996 |
case 'N': |
case 'N': |
| 1997 |
options |= PCRE_NOTEMPTY; |
if ((options & PCRE_NOTEMPTY) != 0) |
| 1998 |
|
options = (options & ~PCRE_NOTEMPTY) | PCRE_NOTEMPTY_ATSTART; |
| 1999 |
|
else |
| 2000 |
|
options |= PCRE_NOTEMPTY; |
| 2001 |
continue; |
continue; |
| 2002 |
|
|
| 2003 |
case 'O': |
case 'O': |
| 2020 |
continue; |
continue; |
| 2021 |
|
|
| 2022 |
case 'P': |
case 'P': |
| 2023 |
options |= PCRE_PARTIAL; |
options |= ((options & PCRE_PARTIAL_SOFT) == 0)? |
| 2024 |
|
PCRE_PARTIAL_SOFT : PCRE_PARTIAL_HARD; |
| 2025 |
continue; |
continue; |
| 2026 |
|
|
| 2027 |
case 'Q': |
case 'Q': |
| 2056 |
show_malloc = 1; |
show_malloc = 1; |
| 2057 |
continue; |
continue; |
| 2058 |
|
|
| 2059 |
|
case 'Y': |
| 2060 |
|
options |= PCRE_NO_START_OPTIMIZE; |
| 2061 |
|
continue; |
| 2062 |
|
|
| 2063 |
case 'Z': |
case 'Z': |
| 2064 |
options |= PCRE_NOTEOL; |
options |= PCRE_NOTEOL; |
| 2065 |
continue; |
continue; |
| 2080 |
*q++ = c; |
*q++ = c; |
| 2081 |
} |
} |
| 2082 |
*q = 0; |
*q = 0; |
| 2083 |
len = q - dbuffer; |
len = (int)(q - dbuffer); |
| 2084 |
|
|
| 2085 |
|
/* Move the data to the end of the buffer so that a read over the end of |
| 2086 |
|
the buffer will be seen by valgrind, even if it doesn't cause a crash. If |
| 2087 |
|
we are using the POSIX interface, we must include the terminating zero. */ |
| 2088 |
|
|
| 2089 |
|
#if !defined NOPOSIX |
| 2090 |
|
if (posix || do_posix) |
| 2091 |
|
{ |
| 2092 |
|
memmove(bptr + buffer_size - len - 1, bptr, len + 1); |
| 2093 |
|
bptr += buffer_size - len - 1; |
| 2094 |
|
} |
| 2095 |
|
else |
| 2096 |
|
#endif |
| 2097 |
|
{ |
| 2098 |
|
memmove(bptr + buffer_size - len, bptr, len); |
| 2099 |
|
bptr += buffer_size - len; |
| 2100 |
|
} |
| 2101 |
|
|
| 2102 |
if ((all_use_dfa || use_dfa) && find_match_limit) |
if ((all_use_dfa || use_dfa) && find_match_limit) |
| 2103 |
{ |
{ |
| 2118 |
pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets); |
pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets); |
| 2119 |
if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL; |
if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL; |
| 2120 |
if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL; |
if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL; |
| 2121 |
|
if ((options & PCRE_NOTEMPTY) != 0) eflags |= REG_NOTEMPTY; |
| 2122 |
|
|
| 2123 |
rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags); |
rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags); |
| 2124 |
|
|
| 2163 |
|
|
| 2164 |
for (;; gmatched++) /* Loop for /g or /G */ |
for (;; gmatched++) /* Loop for /g or /G */ |
| 2165 |
{ |
{ |
| 2166 |
|
markptr = NULL; |
| 2167 |
|
|
| 2168 |
if (timeitm > 0) |
if (timeitm > 0) |
| 2169 |
{ |
{ |
| 2170 |
register int i; |
register int i; |
| 2176 |
{ |
{ |
| 2177 |
int workspace[1000]; |
int workspace[1000]; |
| 2178 |
for (i = 0; i < timeitm; i++) |
for (i = 0; i < timeitm; i++) |
| 2179 |
count = pcre_dfa_exec(re, NULL, (char *)bptr, len, start_offset, |
count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset, |
| 2180 |
options | g_notempty, use_offsets, use_size_offsets, workspace, |
options | g_notempty, use_offsets, use_size_offsets, workspace, |
| 2181 |
sizeof(workspace)/sizeof(int)); |
sizeof(workspace)/sizeof(int)); |
| 2182 |
} |
} |
| 2239 |
else if (all_use_dfa || use_dfa) |
else if (all_use_dfa || use_dfa) |
| 2240 |
{ |
{ |
| 2241 |
int workspace[1000]; |
int workspace[1000]; |
| 2242 |
count = pcre_dfa_exec(re, NULL, (char *)bptr, len, start_offset, |
count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset, |
| 2243 |
options | g_notempty, use_offsets, use_size_offsets, workspace, |
options | g_notempty, use_offsets, use_size_offsets, workspace, |
| 2244 |
sizeof(workspace)/sizeof(int)); |
sizeof(workspace)/sizeof(int)); |
| 2245 |
if (count == 0) |
if (count == 0) |
| 2310 |
} |
} |
| 2311 |
} |
} |
| 2312 |
|
|
| 2313 |
|
if (markptr != NULL) fprintf(outfile, "MK: %s\n", markptr); |
| 2314 |
|
|
| 2315 |
for (i = 0; i < 32; i++) |
for (i = 0; i < 32; i++) |
| 2316 |
{ |
{ |
| 2317 |
if ((copystrings & (1 << i)) != 0) |
if ((copystrings & (1 << i)) != 0) |
| 2395 |
|
|
| 2396 |
else if (count == PCRE_ERROR_PARTIAL) |
else if (count == PCRE_ERROR_PARTIAL) |
| 2397 |
{ |
{ |
| 2398 |
fprintf(outfile, "Partial match"); |
if (markptr == NULL) fprintf(outfile, "Partial match"); |
| 2399 |
#if !defined NODFA |
else fprintf(outfile, "Partial match, mark=%s", markptr); |
| 2400 |
if ((all_use_dfa || use_dfa) && use_size_offsets > 2) |
if (use_size_offsets > 1) |
| 2401 |
fprintf(outfile, ": %.*s", use_offsets[1] - use_offsets[0], |
{ |
| 2402 |
bptr + use_offsets[0]); |
fprintf(outfile, ": "); |
| 2403 |
#endif |
pchars(bptr + use_offsets[0], use_offsets[1] - use_offsets[0], |
| 2404 |
|
outfile); |
| 2405 |
|
} |
| 2406 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 2407 |
break; /* Out of the /g loop */ |
break; /* Out of the /g loop */ |
| 2408 |
} |
} |
| 2432 |
{ |
{ |
| 2433 |
int d; |
int d; |
| 2434 |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &d); |
(void)pcre_config(PCRE_CONFIG_NEWLINE, &d); |
| 2435 |
obits = (d == '\r')? PCRE_NEWLINE_CR : |
/* Note that these values are always the ASCII ones, even in |
| 2436 |
(d == '\n')? PCRE_NEWLINE_LF : |
EBCDIC environments. CR = 13, NL = 10. */ |
| 2437 |
(d == ('\r'<<8 | '\n'))? PCRE_NEWLINE_CRLF : |
obits = (d == 13)? PCRE_NEWLINE_CR : |
| 2438 |
|
(d == 10)? PCRE_NEWLINE_LF : |
| 2439 |
|
(d == (13<<8 | 10))? PCRE_NEWLINE_CRLF : |
| 2440 |
(d == -2)? PCRE_NEWLINE_ANYCRLF : |
(d == -2)? PCRE_NEWLINE_ANYCRLF : |
| 2441 |
(d == -1)? PCRE_NEWLINE_ANY : 0; |
(d == -1)? PCRE_NEWLINE_ANY : 0; |
| 2442 |
} |
} |
| 2463 |
{ |
{ |
| 2464 |
if (count == PCRE_ERROR_NOMATCH) |
if (count == PCRE_ERROR_NOMATCH) |
| 2465 |
{ |
{ |
| 2466 |
if (gmatched == 0) fprintf(outfile, "No match\n"); |
if (gmatched == 0) |
| 2467 |
|
{ |
| 2468 |
|
if (markptr == NULL) fprintf(outfile, "No match\n"); |
| 2469 |
|
else fprintf(outfile, "No match, mark = %s\n", markptr); |
| 2470 |
|
} |
| 2471 |
} |
} |
| 2472 |
else fprintf(outfile, "Error %d\n", count); |
else fprintf(outfile, "Error %d\n", count); |
| 2473 |
break; /* Out of the /g loop */ |
break; /* Out of the /g loop */ |
| 2479 |
if (!do_g && !do_G) break; |
if (!do_g && !do_G) break; |
| 2480 |
|
|
| 2481 |
/* 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 |
| 2482 |
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 |
| 2483 |
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 |
| 2484 |
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 |
| 2485 |
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 |
| 2486 |
character. */ |
character. */ |
| 2487 |
|
|
| 2490 |
if (use_offsets[0] == use_offsets[1]) |
if (use_offsets[0] == use_offsets[1]) |
| 2491 |
{ |
{ |
| 2492 |
if (use_offsets[0] == len) break; |
if (use_offsets[0] == len) break; |
| 2493 |
g_notempty = PCRE_NOTEMPTY | PCRE_ANCHORED; |
g_notempty = PCRE_NOTEMPTY_ATSTART | PCRE_ANCHORED; |
| 2494 |
} |
} |
| 2495 |
|
|
| 2496 |
/* For /g, update the start offset, leaving the rest alone */ |
/* For /g, update the start offset, leaving the rest alone */ |