int
main(int argc, char **argv)
{
int len, i, flags, n;
char regex[50];
char *buf;
regex_t preg;
int status, seed;
seed = time(NULL);
seed = 1028358583;
printf("seed = %d\n", seed);
srand(seed);
n = 0;
for (n = 0; n < 0; n++)
rand();
while (1)
{
printf("*");
fflush(stdout);
printf("n = %d\n", n);
len = 1 + (int)(REGEXP_MAX_LEN * (rand() / (RAND_MAX + 1.0)));
n++;
for (i = 0; i < len; i++)
{
regex[i] = 1 + (int)(255 * (rand() / (RAND_MAX + 1.0)));
n++;
}
regex[i] = L'\0';
printf("len = %d, regexp = \"%s\"\n", len, regex);
for (flags = 0;
flags < (REG_EXTENDED | REG_ICASE | REG_NEWLINE | REG_NOSUB);
flags++)
{
buf = malloc(sizeof(*buf) * len);
strncpy(buf, regex, len - 1);
status = regncomp(&preg, buf, len, flags);
if (status == REG_OK)
regfree(&preg);
}
printf("\n");
}
return 0;
}
GLOBAL (void _p_CNewRegEx (RegExType *r, const char *Expression, int ExpressionLength,
Boolean ExtendedRegEx, Boolean CaseInsensitive, Boolean NewLines))
{
int result;
r->ErrorInternal = NULL;
r->Error = NULL;
r->From = 1;
r->Length = 0;
r->SubExpressions = 0;
r->RegMatch = 0;
r->RegEx = (regex_t *) malloc (sizeof (regex_t));
result = regncomp (r->RegEx, Expression, ExpressionLength,
(ExtendedRegEx ? REG_EXTENDED : 0) |
(CaseInsensitive ? REG_ICASE : 0) |
(NewLines ? REG_NEWLINE : 0));
if (result)
{
if (r->ErrorInternal)
{
free (r->ErrorInternal);
r->ErrorInternal = 0;
}
if (result)
{
size_t size = regerror (result, r->RegEx, 0, 0);
r->ErrorInternal = (char *) malloc (size);
regerror (result, r->RegEx, r->ErrorInternal, size);
}
free (r->RegEx);
r->RegEx = 0;
}
else
{
r->SubExpressions = r->RegEx->re_nsub - 1;
r->RegMatch = (regmatch_t *) malloc ((r->SubExpressions + 1) * sizeof (regmatch_t));
}
}
static void
compile_regex_1 (struct regex *new_regex, int needed_sub)
{
#ifdef REG_PERL
int errcode;
errcode = regncomp(&new_regex->pattern, new_regex->re, new_regex->sz,
(needed_sub ? 0 : REG_NOSUB)
| new_regex->flags
| extended_regexp_flags);
if (errcode)
{
char errorbuf[200];
regerror(errcode, NULL, errorbuf, 200);
bad_prog(gettext(errorbuf));
}
#else
const char *error;
int syntax = ((extended_regexp_flags & REG_EXTENDED)
? RE_SYNTAX_POSIX_EXTENDED
: RE_SYNTAX_POSIX_BASIC);
syntax &= ~RE_DOT_NOT_NULL;
syntax |= RE_NO_POSIX_BACKTRACKING;
switch (posixicity)
{
case POSIXLY_EXTENDED:
syntax &= ~RE_UNMATCHED_RIGHT_PAREN_ORD;
break;
case POSIXLY_CORRECT:
syntax |= RE_UNMATCHED_RIGHT_PAREN_ORD;
break;
case POSIXLY_BASIC:
syntax |= RE_UNMATCHED_RIGHT_PAREN_ORD | RE_LIMITED_OPS | RE_NO_GNU_OPS;
break;
}
#ifdef RE_ICASE
syntax |= (new_regex->flags & REG_ICASE) ? RE_ICASE : 0;
#endif
#ifdef RE_NO_SUB
syntax |= needed_sub ? 0 : RE_NO_SUB;
#endif
new_regex->pattern.fastmap = malloc (1 << (sizeof (char) * 8));
/* If REG_NEWLINE is set, newlines are treated differently. */
if (new_regex->flags & REG_NEWLINE)
{
/* REG_NEWLINE implies neither . nor [^...] match newline. */
syntax &= ~RE_DOT_NEWLINE;
syntax |= RE_HAT_LISTS_NOT_NEWLINE;
}
re_set_syntax (syntax);
error = re_compile_pattern (new_regex->re, new_regex->sz,
&new_regex->pattern);
new_regex->pattern.newline_anchor = (new_regex->flags & REG_NEWLINE) != 0;
new_regex->pattern.translate = NULL;
#ifndef RE_ICASE
if (new_regex->flags & REG_ICASE)
{
static char translate[1 << (sizeof(char) * 8)];
int i;
for (i = 0; i < sizeof(translate) / sizeof(char); i++)
translate[i] = tolower (i);
new_regex->pattern.translate = translate;
}
#endif
if (error)
bad_prog(error);
#endif
/* Just to be sure, I mark this as not POSIXLY_CORRECT behavior */
if (needed_sub
&& new_regex->pattern.re_nsub < needed_sub - 1
&& posixicity == POSIXLY_EXTENDED)
{
char buf[200];
sprintf(buf, _("invalid reference \\%d on `s' command's RHS"),
needed_sub - 1);
bad_prog(buf);
}
}
int extractSearchSpecData(struct scalpelState *state,
struct SearchSpecLine *s,
char **tokenarray) {
int err = 0;
// process one line from config file:
// token[0] = suffix
// token[1] = case sensitive?
// token[2] = maximum carve size
// token[3] = begintag
// token[4] = endtag
// token[5] = search type (optional)
s->suffix = (char *) malloc(MAX_SUFFIX_LENGTH * sizeof(char));
checkMemoryAllocation(state, s->suffix, __LINE__, __FILE__, "s->suffix");
s->begin = (char *) malloc(MAX_STRING_LENGTH * sizeof(char));
checkMemoryAllocation(state, s->begin, __LINE__, __FILE__, "s->begin");
s->end = (char *) malloc(MAX_STRING_LENGTH * sizeof(char));
checkMemoryAllocation(state, s->end, __LINE__, __FILE__, "s->end");
s->begintext = (char *) malloc(MAX_STRING_LENGTH * sizeof(char));
checkMemoryAllocation(state, s->begintext, __LINE__, __FILE__,
"s->begintext");
s->endtext = (char *) malloc(MAX_STRING_LENGTH * sizeof(char));
checkMemoryAllocation(state, s->endtext, __LINE__, __FILE__, "s->endtext");
if (!strncasecmp(tokenarray[0], SCALPEL_NOEXTENSION_SUFFIX,
strlen(SCALPEL_NOEXTENSION_SUFFIX))) {
s->suffix[0] = SCALPEL_NOEXTENSION;
s->suffix[1] = 0;
} else {
memcpy(s->suffix, tokenarray[0], MAX_SUFFIX_LENGTH);
}
// case sensitivity check
s->casesensitive = (!strncasecmp(tokenarray[1], "y", 1)
|| !strncasecmp(tokenarray[1], "yes", 3));
//#ifdef _WIN32
// s->length = _atoi64(tokenarray[2]);
//#else
// s->length = atoull(tokenarray[2]);
//#endif
char split[MAX_STRING_LENGTH];
char *maxcarvelength;
strcpy(split, tokenarray[2]);
maxcarvelength = strchr(split, ':');
if (!maxcarvelength) {
s->minlength = 0;
s->length = strtoull(split, 0, 10);
} else {
*maxcarvelength = 0;
maxcarvelength++;
s->minlength = strtoull(split, 0, 10);
s->length = strtoull(maxcarvelength, 0, 10);
}
// determine search type for this needle
s->searchtype = SEARCHTYPE_FORWARD;
if (!strncasecmp(tokenarray[5], "REVERSE", strlen("REVERSE"))) {
s->searchtype = SEARCHTYPE_REVERSE;
} else if (!strncasecmp(tokenarray[5], "NEXT", strlen("NEXT"))) {
s->searchtype = SEARCHTYPE_FORWARD_NEXT;
}
// FORWARD is the default, but OK if the user defines it explicitly
else if (!strncasecmp(tokenarray[5], "FORWARD", strlen("FORWARD"))) {
s->searchtype = SEARCHTYPE_FORWARD;
}
// regular expressions must be handled separately
if (isRegularExpression(tokenarray[3])) {
#ifdef GPU_THREADING
// GPU execution does not support regex needles.
std::stringstream ss;
ss << "ERROR: GPU search for regex headers is not supported!\n";
ss << "Please modify the config file for non-regex headers only.\n";
std::string msg = ss.str();
fprintf(stderr, msg.c_str());
throw std::runtime_error(msg);
#endif
// copy RE, zap leading/training '/' and prepare for regular expression compilation
s->beginisRE = 1;
strcpy(s->begin, tokenarray[3]);
strcpy(s->begintext, tokenarray[3]);
s->beginlength = strlen(tokenarray[3]);
s->begin[s->beginlength] = 0;
// compile regular expression
err = regncomp(&(s->beginstate.re), s->begin + 1, s->beginlength - 2,
REG_EXTENDED | (REG_ICASE * (!s->casesensitive)));
if (err) {
return SCALPEL_ERROR_BAD_HEADER_REGEX;
}
} else {
// non-regular expression header
s->beginisRE = 0;
strcpy(s->begintext, tokenarray[3]);
s->beginlength = translate(tokenarray[3]);
memcpy(s->begin, tokenarray[3], s->beginlength);
init_bm_table(s->begin, s->beginstate.bm_table, s->beginlength,
s->casesensitive);
}
if (isRegularExpression(tokenarray[4])) {
#ifdef GPU_THREADING
// GPU execution does not support regex needles.
std::stringstream ss;
ss << "ERROR: GPU search for regex footers is not supported!\n";
ss << "Please modify the config file for non-regex footers only.\n";
std::string msg = ss.str();
fprintf(stderr, msg.c_str());
throw std::runtime_error(msg);
#endif
// copy RE, zap leading/training '/' and prepare for for regular expression compilation
s->endisRE = 1;
strcpy(s->end, tokenarray[4]);
strcpy(s->endtext, tokenarray[4]);
s->endlength = strlen(tokenarray[4]);
s->end[s->endlength] = 0;
// compile regular expression
err = regncomp(&(s->endstate.re), s->end + 1, s->endlength - 2,
REG_EXTENDED | (REG_ICASE * (!s->casesensitive)));
if (err) {
return SCALPEL_ERROR_BAD_FOOTER_REGEX;
}
} else {
s->endisRE = 0;
strcpy(s->endtext, tokenarray[4]);
s->endlength = translate(tokenarray[4]);
memcpy(s->end, tokenarray[4], s->endlength);
init_bm_table(s->end, s->endstate.bm_table, s->endlength,
s->casesensitive);
}
return SCALPEL_OK;
}
