| 79 |
#define fileno _fileno |
#define fileno _fileno |
| 80 |
#endif |
#endif |
| 81 |
|
|
| 82 |
|
/* A user sent this fix for Borland Builder 5 under Windows. */ |
| 83 |
|
|
| 84 |
|
#ifdef __BORLANDC__ |
| 85 |
|
#define _setmode(handle, mode) setmode(handle, mode) |
| 86 |
|
#endif |
| 87 |
|
|
| 88 |
|
/* Not Windows */ |
| 89 |
|
|
| 90 |
#else |
#else |
| 91 |
#include <sys/time.h> /* These two includes are needed */ |
#include <sys/time.h> /* These two includes are needed */ |
| 92 |
#include <sys/resource.h> /* for setrlimit(). */ |
#include <sys/resource.h> /* for setrlimit(). */ |
| 197 |
static uschar *dbuffer = NULL; |
static uschar *dbuffer = NULL; |
| 198 |
static uschar *pbuffer = NULL; |
static uschar *pbuffer = NULL; |
| 199 |
|
|
| 200 |
|
/* Textual explanations for runtime error codes */ |
| 201 |
|
|
| 202 |
|
static const char *errtexts[] = { |
| 203 |
|
NULL, /* 0 is no error */ |
| 204 |
|
NULL, /* NOMATCH is handled specially */ |
| 205 |
|
"NULL argument passed", |
| 206 |
|
"bad option value", |
| 207 |
|
"magic number missing", |
| 208 |
|
"unknown opcode - pattern overwritten?", |
| 209 |
|
"no more memory", |
| 210 |
|
NULL, /* never returned by pcre_exec() or pcre_dfa_exec() */ |
| 211 |
|
"match limit exceeded", |
| 212 |
|
"callout error code", |
| 213 |
|
NULL, /* BADUTF8 is handled specially */ |
| 214 |
|
"bad UTF-8 offset", |
| 215 |
|
NULL, /* PARTIAL is handled specially */ |
| 216 |
|
"not used - internal error", |
| 217 |
|
"internal error - pattern overwritten?", |
| 218 |
|
"bad count value", |
| 219 |
|
"item unsupported for DFA matching", |
| 220 |
|
"backreference condition or recursion test not supported for DFA matching", |
| 221 |
|
"match limit not supported for DFA matching", |
| 222 |
|
"workspace size exceeded in DFA matching", |
| 223 |
|
"too much recursion for DFA matching", |
| 224 |
|
"recursion limit exceeded", |
| 225 |
|
"not used - internal error", |
| 226 |
|
"invalid combination of newline options", |
| 227 |
|
"bad offset value", |
| 228 |
|
NULL /* SHORTUTF8 is handled specially */ |
| 229 |
|
}; |
| 230 |
|
|
| 231 |
|
|
| 232 |
/************************************************* |
/************************************************* |
| 233 |
* Alternate character tables * |
* Alternate character tables * |
| 1186 |
#endif |
#endif |
| 1187 |
printf(" -q quiet: do not output PCRE version number at start\n"); |
printf(" -q quiet: do not output PCRE version number at start\n"); |
| 1188 |
printf(" -S <n> set stack size to <n> megabytes\n"); |
printf(" -S <n> set stack size to <n> megabytes\n"); |
| 1189 |
printf(" -s output store (memory) used information\n" |
printf(" -s force each pattern to be studied\n" |
| 1190 |
" -t time compilation and execution\n"); |
" -t time compilation and execution\n"); |
| 1191 |
printf(" -t <n> time compilation and execution, repeating <n> times\n"); |
printf(" -t <n> time compilation and execution, repeating <n> times\n"); |
| 1192 |
printf(" -tm time execution (matching) only\n"); |
printf(" -tm time execution (matching) only\n"); |
| 1214 |
int timeitm = 0; |
int timeitm = 0; |
| 1215 |
int showinfo = 0; |
int showinfo = 0; |
| 1216 |
int showstore = 0; |
int showstore = 0; |
| 1217 |
|
int force_study = 0; |
| 1218 |
int quiet = 0; |
int quiet = 0; |
| 1219 |
int size_offsets = 45; |
int size_offsets = 45; |
| 1220 |
int size_offsets_max; |
int size_offsets_max; |
| 1263 |
{ |
{ |
| 1264 |
unsigned char *endptr; |
unsigned char *endptr; |
| 1265 |
|
|
| 1266 |
if (strcmp(argv[op], "-s") == 0 || strcmp(argv[op], "-m") == 0) |
if (strcmp(argv[op], "-m") == 0) showstore = 1; |
| 1267 |
showstore = 1; |
else if (strcmp(argv[op], "-s") == 0) force_study = 1; |
| 1268 |
else if (strcmp(argv[op], "-q") == 0) quiet = 1; |
else if (strcmp(argv[op], "-q") == 0) quiet = 1; |
| 1269 |
else if (strcmp(argv[op], "-b") == 0) debug = 1; |
else if (strcmp(argv[op], "-b") == 0) debug = 1; |
| 1270 |
else if (strcmp(argv[op], "-i") == 0) showinfo = 1; |
else if (strcmp(argv[op], "-i") == 0) showinfo = 1; |
| 1436 |
size_t size, regex_gotten_store; |
size_t size, regex_gotten_store; |
| 1437 |
int do_mark = 0; |
int do_mark = 0; |
| 1438 |
int do_study = 0; |
int do_study = 0; |
| 1439 |
|
int no_force_study = 0; |
| 1440 |
int do_debug = debug; |
int do_debug = debug; |
| 1441 |
int do_G = 0; |
int do_G = 0; |
| 1442 |
int do_g = 0; |
int do_g = 0; |
| 1443 |
int do_showinfo = showinfo; |
int do_showinfo = showinfo; |
| 1444 |
int do_showrest = 0; |
int do_showrest = 0; |
| 1445 |
|
int do_showcaprest = 0; |
| 1446 |
int do_flip = 0; |
int do_flip = 0; |
| 1447 |
int erroroffset, len, delimiter, poffset; |
int erroroffset, len, delimiter, poffset; |
| 1448 |
|
|
| 1504 |
} |
} |
| 1505 |
} |
} |
| 1506 |
|
|
| 1507 |
fprintf(outfile, "Compiled regex%s loaded from %s\n", |
fprintf(outfile, "Compiled pattern%s loaded from %s\n", |
| 1508 |
do_flip? " (byte-inverted)" : "", p); |
do_flip? " (byte-inverted)" : "", p); |
| 1509 |
|
|
| 1510 |
/* Need to know if UTF-8 for printing data strings */ |
/* Need to know if UTF-8 for printing data strings */ |
| 1512 |
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options); |
| 1513 |
use_utf8 = (get_options & PCRE_UTF8) != 0; |
use_utf8 = (get_options & PCRE_UTF8) != 0; |
| 1514 |
|
|
| 1515 |
/* Now see if there is any following study data */ |
/* Now see if there is any following study data. */ |
| 1516 |
|
|
| 1517 |
if (true_study_size != 0) |
if (true_study_size != 0) |
| 1518 |
{ |
{ |
| 1608 |
case 's': options |= PCRE_DOTALL; break; |
case 's': options |= PCRE_DOTALL; break; |
| 1609 |
case 'x': options |= PCRE_EXTENDED; break; |
case 'x': options |= PCRE_EXTENDED; break; |
| 1610 |
|
|
| 1611 |
case '+': do_showrest = 1; break; |
case '+': |
| 1612 |
|
if (do_showrest) do_showcaprest = 1; else do_showrest = 1; |
| 1613 |
|
break; |
| 1614 |
|
|
| 1615 |
case 'A': options |= PCRE_ANCHORED; break; |
case 'A': options |= PCRE_ANCHORED; break; |
| 1616 |
case 'B': do_debug = 1; break; |
case 'B': do_debug = 1; break; |
| 1617 |
case 'C': options |= PCRE_AUTO_CALLOUT; break; |
case 'C': options |= PCRE_AUTO_CALLOUT; break; |
| 1629 |
case 'P': do_posix = 1; break; |
case 'P': do_posix = 1; break; |
| 1630 |
#endif |
#endif |
| 1631 |
|
|
| 1632 |
case 'S': do_study = 1; break; |
case 'S': |
| 1633 |
|
if (do_study == 0) do_study = 1; else |
| 1634 |
|
{ |
| 1635 |
|
do_study = 0; |
| 1636 |
|
no_force_study = 1; |
| 1637 |
|
} |
| 1638 |
|
break; |
| 1639 |
|
|
| 1640 |
case 'U': options |= PCRE_UNGREEDY; break; |
case 'U': options |= PCRE_UNGREEDY; break; |
| 1641 |
case 'W': options |= PCRE_UCP; break; |
case 'W': options |= PCRE_UCP; break; |
| 1642 |
case 'X': options |= PCRE_EXTRA; break; |
case 'X': options |= PCRE_EXTRA; break; |
| 1643 |
|
case 'Y': options |= PCRE_NO_START_OPTIMISE; break; |
| 1644 |
case 'Z': debug_lengths = 0; break; |
case 'Z': debug_lengths = 0; break; |
| 1645 |
case '8': options |= PCRE_UTF8; use_utf8 = 1; break; |
case '8': options |= PCRE_UTF8; use_utf8 = 1; break; |
| 1646 |
case '?': options |= PCRE_NO_UTF8_CHECK; break; |
case '?': options |= PCRE_NO_UTF8_CHECK; break; |
| 1820 |
true_size = ((real_pcre *)re)->size; |
true_size = ((real_pcre *)re)->size; |
| 1821 |
regex_gotten_store = gotten_store; |
regex_gotten_store = gotten_store; |
| 1822 |
|
|
| 1823 |
/* If /S was present, study the regexp to generate additional info to |
/* If -s or /S was present, study the regex to generate additional info to |
| 1824 |
help with the matching. */ |
help with the matching, unless the pattern has the SS option, which |
| 1825 |
|
suppresses the effect of /S (used for a few test patterns where studying is |
| 1826 |
|
never sensible). */ |
| 1827 |
|
|
| 1828 |
if (do_study) |
if (do_study || (force_study && !no_force_study)) |
| 1829 |
{ |
{ |
| 1830 |
if (timeit > 0) |
if (timeit > 0) |
| 1831 |
{ |
{ |
| 1979 |
if (do_flip) all_options = byteflip(all_options, sizeof(all_options)); |
if (do_flip) all_options = byteflip(all_options, sizeof(all_options)); |
| 1980 |
|
|
| 1981 |
if (get_options == 0) fprintf(outfile, "No options\n"); |
if (get_options == 0) fprintf(outfile, "No options\n"); |
| 1982 |
else fprintf(outfile, "Options:%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", |
| 1983 |
((get_options & PCRE_ANCHORED) != 0)? " anchored" : "", |
((get_options & PCRE_ANCHORED) != 0)? " anchored" : "", |
| 1984 |
((get_options & PCRE_CASELESS) != 0)? " caseless" : "", |
((get_options & PCRE_CASELESS) != 0)? " caseless" : "", |
| 1985 |
((get_options & PCRE_EXTENDED) != 0)? " extended" : "", |
((get_options & PCRE_EXTENDED) != 0)? " extended" : "", |
| 1995 |
((get_options & PCRE_UTF8) != 0)? " utf8" : "", |
((get_options & PCRE_UTF8) != 0)? " utf8" : "", |
| 1996 |
((get_options & PCRE_UCP) != 0)? " ucp" : "", |
((get_options & PCRE_UCP) != 0)? " ucp" : "", |
| 1997 |
((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "", |
((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "", |
| 1998 |
|
((get_options & PCRE_NO_START_OPTIMIZE) != 0)? " no_start_optimize" : "", |
| 1999 |
((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : ""); |
((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : ""); |
| 2000 |
|
|
| 2001 |
if (jchanged) fprintf(outfile, "Duplicate name status changes\n"); |
if (jchanged) fprintf(outfile, "Duplicate name status changes\n"); |
| 2063 |
/* 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 |
| 2064 |
value, but it varies, depending on the computer architecture, and |
value, but it varies, depending on the computer architecture, and |
| 2065 |
so messes up the test suite. (And with the /F option, it might be |
so messes up the test suite. (And with the /F option, it might be |
| 2066 |
flipped.) */ |
flipped.) If study was forced by an external -s, don't show this |
| 2067 |
|
information unless -i or -d was also present. This means that, except |
| 2068 |
|
when auto-callouts are involved, the output from runs with and without |
| 2069 |
|
-s should be identical. */ |
| 2070 |
|
|
| 2071 |
if (do_study) |
if (do_study || (force_study && showinfo && !no_force_study)) |
| 2072 |
{ |
{ |
| 2073 |
if (extra == NULL) |
if (extra == NULL) |
| 2074 |
fprintf(outfile, "Study returned NULL\n"); |
fprintf(outfile, "Study returned NULL\n"); |
| 2146 |
} |
} |
| 2147 |
else |
else |
| 2148 |
{ |
{ |
| 2149 |
fprintf(outfile, "Compiled regex written to %s\n", to_file); |
fprintf(outfile, "Compiled pattern written to %s\n", to_file); |
| 2150 |
|
|
| 2151 |
|
/* If there is study data, write it, but verify the writing only |
| 2152 |
|
if the studying was requested by /S, not just by -s. */ |
| 2153 |
|
|
| 2154 |
if (extra != NULL) |
if (extra != NULL) |
| 2155 |
{ |
{ |
| 2156 |
if (fwrite(extra->study_data, 1, true_study_size, f) < |
if (fwrite(extra->study_data, 1, true_study_size, f) < |
| 2160 |
strerror(errno)); |
strerror(errno)); |
| 2161 |
} |
} |
| 2162 |
else fprintf(outfile, "Study data written to %s\n", to_file); |
else fprintf(outfile, "Study data written to %s\n", to_file); |
|
|
|
| 2163 |
} |
} |
| 2164 |
} |
} |
| 2165 |
fclose(f); |
fclose(f); |
| 2194 |
int getlist = 0; |
int getlist = 0; |
| 2195 |
int gmatched = 0; |
int gmatched = 0; |
| 2196 |
int start_offset = 0; |
int start_offset = 0; |
| 2197 |
|
int start_offset_sign = 1; |
| 2198 |
int g_notempty = 0; |
int g_notempty = 0; |
| 2199 |
int use_dfa = 0; |
int use_dfa = 0; |
| 2200 |
|
|
| 2327 |
continue; |
continue; |
| 2328 |
|
|
| 2329 |
case '>': |
case '>': |
| 2330 |
|
if (*p == '-') |
| 2331 |
|
{ |
| 2332 |
|
start_offset_sign = -1; |
| 2333 |
|
p++; |
| 2334 |
|
} |
| 2335 |
while(isdigit(*p)) start_offset = start_offset * 10 + *p++ - '0'; |
while(isdigit(*p)) start_offset = start_offset * 10 + *p++ - '0'; |
| 2336 |
|
start_offset *= start_offset_sign; |
| 2337 |
continue; |
continue; |
| 2338 |
|
|
| 2339 |
case 'A': /* Option setting */ |
case 'A': /* Option setting */ |
| 2591 |
(void)pchars(dbuffer + pmatch[i].rm_so, |
(void)pchars(dbuffer + pmatch[i].rm_so, |
| 2592 |
pmatch[i].rm_eo - pmatch[i].rm_so, outfile); |
pmatch[i].rm_eo - pmatch[i].rm_so, outfile); |
| 2593 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 2594 |
if (i == 0 && do_showrest) |
if (do_showcaprest || (i == 0 && do_showrest)) |
| 2595 |
{ |
{ |
| 2596 |
fprintf(outfile, " 0+ "); |
fprintf(outfile, "%2d+ ", (int)i); |
| 2597 |
(void)pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo, |
(void)pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo, |
| 2598 |
outfile); |
outfile); |
| 2599 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 2745 |
(void)pchars(bptr + use_offsets[i], |
(void)pchars(bptr + use_offsets[i], |
| 2746 |
use_offsets[i+1] - use_offsets[i], outfile); |
use_offsets[i+1] - use_offsets[i], outfile); |
| 2747 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
| 2748 |
if (i == 0) |
if (do_showcaprest || (i == 0 && do_showrest)) |
| 2749 |
{ |
{ |
| 2750 |
if (do_showrest) |
fprintf(outfile, "%2d+ ", i/2); |
| 2751 |
{ |
(void)pchars(bptr + use_offsets[i+1], len - use_offsets[i+1], |
| 2752 |
fprintf(outfile, " 0+ "); |
outfile); |
| 2753 |
(void)pchars(bptr + use_offsets[i+1], len - use_offsets[i+1], |
fprintf(outfile, "\n"); |
|
outfile); |
|
|
fprintf(outfile, "\n"); |
|
|
} |
|
| 2754 |
} |
} |
| 2755 |
} |
} |
| 2756 |
} |
} |
| 2857 |
to advance the start offset, and continue. We won't be at the end of the |
to advance the start offset, and continue. We won't be at the end of the |
| 2858 |
string - that was checked before setting g_notempty. |
string - that was checked before setting g_notempty. |
| 2859 |
|
|
| 2860 |
Complication arises in the case when the newline option is "any" or |
Complication arises in the case when the newline convention is "any", |
| 2861 |
"anycrlf". If the previous match was at the end of a line terminated by |
"crlf", or "anycrlf". If the previous match was at the end of a line |
| 2862 |
CRLF, an advance of one character just passes the \r, whereas we should |
terminated by CRLF, an advance of one character just passes the \r, |
| 2863 |
prefer the longer newline sequence, as does the code in pcre_exec(). |
whereas we should prefer the longer newline sequence, as does the code in |
| 2864 |
Fudge the offset value to achieve this. |
pcre_exec(). Fudge the offset value to achieve this. We check for a |
| 2865 |
|
newline setting in the pattern; if none was set, use pcre_config() to |
| 2866 |
|
find the default. |
| 2867 |
|
|
| 2868 |
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 |
| 2869 |
character, not one byte. */ |
character, not one byte. */ |
| 2888 |
(d == -1)? PCRE_NEWLINE_ANY : 0; |
(d == -1)? PCRE_NEWLINE_ANY : 0; |
| 2889 |
} |
} |
| 2890 |
if (((obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANY || |
if (((obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANY || |
| 2891 |
|
(obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_CRLF || |
| 2892 |
(obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANYCRLF) |
(obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANYCRLF) |
| 2893 |
&& |
&& |
| 2894 |
start_offset < len - 1 && |
start_offset < len - 1 && |
| 2899 |
{ |
{ |
| 2900 |
while (start_offset + onechar < len) |
while (start_offset + onechar < len) |
| 2901 |
{ |
{ |
| 2902 |
int tb = bptr[start_offset+onechar]; |
if ((bptr[start_offset+onechar] & 0xc0) != 0x80) break; |
| 2903 |
if (tb <= 127) break; |
onechar++; |
|
tb &= 0xc0; |
|
|
if (tb != 0 && tb != 0xc0) onechar++; |
|
| 2904 |
} |
} |
| 2905 |
} |
} |
| 2906 |
use_offsets[1] = start_offset + onechar; |
use_offsets[1] = start_offset + onechar; |
| 2907 |
} |
} |
| 2908 |
else |
else |
| 2909 |
{ |
{ |
| 2910 |
if (count == PCRE_ERROR_NOMATCH) |
switch(count) |
| 2911 |
{ |
{ |
| 2912 |
|
case PCRE_ERROR_NOMATCH: |
| 2913 |
if (gmatched == 0) |
if (gmatched == 0) |
| 2914 |
{ |
{ |
| 2915 |
if (markptr == NULL) fprintf(outfile, "No match\n"); |
if (markptr == NULL) fprintf(outfile, "No match\n"); |
| 2916 |
else fprintf(outfile, "No match, mark = %s\n", markptr); |
else fprintf(outfile, "No match, mark = %s\n", markptr); |
| 2917 |
} |
} |
| 2918 |
|
break; |
| 2919 |
|
|
| 2920 |
|
case PCRE_ERROR_BADUTF8: |
| 2921 |
|
case PCRE_ERROR_SHORTUTF8: |
| 2922 |
|
fprintf(outfile, "Error %d (%s UTF-8 string)", count, |
| 2923 |
|
(count == PCRE_ERROR_BADUTF8)? "bad" : "short"); |
| 2924 |
|
if (use_size_offsets >= 2) |
| 2925 |
|
fprintf(outfile, " offset=%d reason=%d", use_offsets[0], |
| 2926 |
|
use_offsets[1]); |
| 2927 |
|
fprintf(outfile, "\n"); |
| 2928 |
|
break; |
| 2929 |
|
|
| 2930 |
|
default: |
| 2931 |
|
if (count < 0 && (-count) < sizeof(errtexts)/sizeof(const char *)) |
| 2932 |
|
fprintf(outfile, "Error %d (%s)\n", count, errtexts[-count]); |
| 2933 |
|
else |
| 2934 |
|
fprintf(outfile, "Error %d (Unexpected value)\n", count); |
| 2935 |
|
break; |
| 2936 |
} |
} |
| 2937 |
else fprintf(outfile, "Error %d\n", count); |
|
| 2938 |
break; /* Out of the /g loop */ |
break; /* Out of the /g loop */ |
| 2939 |
} |
} |
| 2940 |
} |
} |