pcre2_code *compile(char *pattern) {
PCRE2_SPTR pcre2_pattern = (PCRE2_SPTR)pattern;
pcre2_code *re;
int errornumber;
PCRE2_SIZE erroroffset;
re = pcre2_compile(
pcre2_pattern, /* the pattern */
PCRE2_ZERO_TERMINATED, /* indicates pattern is zero-terminated */
0, /* default options */
&errornumber, /* for error number */
&erroroffset, /* for error offset */
NULL); /* use default compile context */
pcre2_jit_compile(re, PCRE2_JIT_COMPLETE);
if (re == NULL) {
PCRE2_UCHAR buffer[256];
pcre2_get_error_message(errornumber, buffer, sizeof(buffer));
printf("PCRE2 compilation failed at offset %d: %s\n for pattern: %s", (int)erroroffset,
buffer, pattern);
exit(1);
}
return re;
}
static void compile_pcre2_pattern(struct grep_pat *p, const struct grep_opt *opt)
{
int error;
PCRE2_UCHAR errbuf[256];
PCRE2_SIZE erroffset;
int options = PCRE2_MULTILINE;
const uint8_t *character_tables = NULL;
int jitret;
assert(opt->pcre2);
p->pcre2_compile_context = NULL;
if (opt->ignore_case) {
if (has_non_ascii(p->pattern)) {
character_tables = pcre2_maketables(NULL);
p->pcre2_compile_context = pcre2_compile_context_create(NULL);
pcre2_set_character_tables(p->pcre2_compile_context, character_tables);
}
options |= PCRE2_CASELESS;
}
if (is_utf8_locale() && has_non_ascii(p->pattern))
options |= PCRE2_UTF;
p->pcre2_pattern = pcre2_compile((PCRE2_SPTR)p->pattern,
p->patternlen, options, &error, &erroffset,
p->pcre2_compile_context);
if (p->pcre2_pattern) {
p->pcre2_match_data = pcre2_match_data_create_from_pattern(p->pcre2_pattern, NULL);
if (!p->pcre2_match_data)
die("Couldn't allocate PCRE2 match data");
} else {
pcre2_get_error_message(error, errbuf, sizeof(errbuf));
compile_regexp_failed(p, (const char *)&errbuf);
}
pcre2_config(PCRE2_CONFIG_JIT, &p->pcre2_jit_on);
if (p->pcre2_jit_on == 1) {
jitret = pcre2_jit_compile(p->pcre2_pattern, PCRE2_JIT_COMPLETE);
if (jitret)
die("Couldn't JIT the PCRE2 pattern '%s', got '%d'\n", p->pattern, jitret);
p->pcre2_jit_stack = pcre2_jit_stack_create(1, 1024 * 1024, NULL);
if (!p->pcre2_jit_stack)
die("Couldn't allocate PCRE2 JIT stack");
p->pcre2_match_context = pcre2_match_context_create(NULL);
if (!p->pcre2_match_context)
die("Couldn't allocate PCRE2 match context");
pcre2_jit_stack_assign(p->pcre2_match_context, NULL, p->pcre2_jit_stack);
} else if (p->pcre2_jit_on != 0) {
die("BUG: The pcre2_jit_on variable should be 0 or 1, not %d",
p->pcre1_jit_on);
}
}
/* Compile a regular expression to be used later
* Allowed flags are:
* - OS_CASE_SENSITIVE
* - OS_RETURN_SUBSTRING
* Returns 1 on success or 0 on error
* The error code is set on reg->error
*/
int OSRegex_Compile(const char *pattern, OSRegex *reg, int flags)
{
char *pattern_pcre2 = NULL;
int flags_compile = 0;
int error = 0;
PCRE2_SIZE erroroffset = 0;
size_t pattern_len = 0UL;
char first_char, last_char;
uint32_t count, i;
/* Check for references not initialized */
if (reg == NULL) {
return (0);
}
/* Initialize OSRegex structure */
reg->error = 0;
reg->sub_strings = NULL;
reg->regex = NULL;
reg->match_data = NULL;
reg->pattern_len = 0UL;
reg->pattern = NULL;
reg->exec_function = NULL;
/* The pattern can't be null */
if (pattern == NULL) {
reg->error = OS_REGEX_PATTERN_NULL;
goto compile_error;
}
/* Maximum size of the pattern */
pattern_len = strlen(pattern);
if (pattern_len > OS_PATTERN_MAXSIZE) {
reg->error = OS_REGEX_MAXSIZE;
goto compile_error;
}
if (OSRegex_CouldBeOptimized(pattern)) {
first_char = pattern[0];
last_char = pattern[pattern_len - 1];
if (first_char == '^') {
if (last_char == '$') {
reg->pattern = strdup(&pattern[1]);
reg->pattern_len = pattern_len - 2;
reg->pattern[reg->pattern_len] = '\0';
if (flags & OS_CASE_SENSITIVE) {
reg->exec_function = OSRegex_Execute_strcmp;
} else {
reg->exec_function = OSRegex_Execute_strcasecmp;
}
return (1);
} else {
reg->pattern = strdup(&pattern[1]);
reg->pattern_len = pattern_len - 1;
if (flags & OS_CASE_SENSITIVE) {
reg->exec_function = OSRegex_Execute_strncmp;
} else {
reg->exec_function = OSRegex_Execute_strncasecmp;
}
return (1);
}
} else {
if (last_char == '$') {
reg->pattern = strdup(pattern);
reg->pattern_len = pattern_len - 1;
reg->pattern[reg->pattern_len] = '\0';
if (flags & OS_CASE_SENSITIVE) {
reg->exec_function = OSRegex_Execute_strrcmp;
} else {
reg->exec_function = OSRegex_Execute_strrcasecmp;
}
return (1);
}
}
}
reg->exec_function = OSRegex_Execute_pcre2_match;
/* Ossec pattern conversion */
if (OSRegex_Convert(pattern, &pattern_pcre2, OS_CONVERT_REGEX) == 0) {
reg->error = OS_REGEX_BADREGEX;
goto compile_error;
}
flags_compile |= PCRE2_UTF;
flags_compile |= PCRE2_NO_UTF_CHECK;
flags_compile |= (flags & OS_CASE_SENSITIVE) ? 0 : PCRE2_CASELESS;
reg->regex = pcre2_compile((PCRE2_SPTR)pattern_pcre2, PCRE2_ZERO_TERMINATED, flags_compile,
&error, &erroroffset, NULL);
if (reg->regex == NULL) {
reg->error = OS_REGEX_BADREGEX;
goto compile_error;
}
reg->match_data = pcre2_match_data_create_from_pattern(reg->regex, NULL);
if (reg->match_data == NULL) {
reg->error = OS_REGEX_OUTOFMEMORY;
goto compile_error;
}
#ifdef USE_PCRE2_JIT
/* Just In Time compilation for faster execution */
if (pcre2_jit_compile(reg->regex, PCRE2_JIT_COMPLETE) != 0) {
reg->error = OS_REGEX_NO_JIT;
goto compile_error;
}
#endif
if (flags & OS_RETURN_SUBSTRING) {
pcre2_pattern_info(reg->regex, PCRE2_INFO_CAPTURECOUNT, (void *)&count);
count++; // to store NULL pointer at the end
reg->sub_strings = calloc(count, sizeof(char *));
if (reg->sub_strings == NULL) {
reg->error = OS_REGEX_OUTOFMEMORY;
goto compile_error;
}
for (i = 0; i < count; i++) {
reg->sub_strings[i] = NULL;
}
}
free(pattern_pcre2);
return (1);
compile_error:
/* Error handling */
if (pattern_pcre2) {
free(pattern_pcre2);
}
OSRegex_FreePattern(reg);
return (0);
}
int regex_compile(tvh_regex_t *regex, const char *re_str, int flags, int subsys)
{
#if ENABLE_PCRE || ENABLE_PCRE2
regex->is_posix = 0;
if (flags & TVHREGEX_POSIX) {
regex->is_posix = 1;
#endif
int options = REG_EXTENDED;
if (flags & TVHREGEX_CASELESS)
options |= REG_ICASE;
if (!regcomp(®ex->re_posix_code, re_str, options))
return 0;
tvherror(subsys, "Unable to compile regex '%s'", re_str);
return -1;
#if ENABLE_PCRE || ENABLE_PCRE2
} else {
#if ENABLE_PCRE
const char *estr;
int eoff;
int options = PCRE_UTF8;
if (flags & TVHREGEX_CASELESS)
options |= PCRE_CASELESS;
#if PCRE_STUDY_JIT_COMPILE
regex->re_jit_stack = NULL;
#endif
regex->re_extra = NULL;
regex->re_code = pcre_compile(re_str, options, &estr, &eoff, NULL);
if (regex->re_code == NULL) {
tvherror(subsys, "Unable to compile PCRE '%s': %s", re_str, estr);
} else {
regex->re_extra = pcre_study(regex->re_code,
PCRE_STUDY_JIT_COMPILE, &estr);
if (regex->re_extra == NULL && estr)
tvherror(subsys, "Unable to study PCRE '%s': %s", re_str, estr);
else {
#if PCRE_STUDY_JIT_COMPILE
regex->re_jit_stack = pcre_jit_stack_alloc(32*1024, 512*1024);
if (regex->re_jit_stack)
pcre_assign_jit_stack(regex->re_extra, NULL, regex->re_jit_stack);
#endif
return 0;
}
}
return -1;
#elif ENABLE_PCRE2
PCRE2_UCHAR8 ebuf[128];
int ecode;
PCRE2_SIZE eoff;
size_t jsz;
uint32_t options;
assert(regex->re_jit_stack == NULL);
regex->re_jit_stack = NULL;
regex->re_match = NULL;
regex->re_mcontext = pcre2_match_context_create(NULL);
options = PCRE2_UTF;
if (flags & TVHREGEX_CASELESS)
options |= PCRE2_CASELESS;
regex->re_code = pcre2_compile((PCRE2_SPTR8)re_str, -1, options,
&ecode, &eoff, NULL);
if (regex->re_code == NULL) {
(void)pcre2_get_error_message(ecode, ebuf, 120);
tvherror(subsys, "Unable to compile PCRE2 '%s': %s", re_str, ebuf);
} else {
regex->re_match = pcre2_match_data_create(TVHREGEX_MAX_MATCHES, NULL);
if (re_str[0] && pcre2_jit_compile(regex->re_code, PCRE2_JIT_COMPLETE) >= 0) {
jsz = 0;
if (pcre2_pattern_info(regex->re_code, PCRE2_INFO_JITSIZE, &jsz) >= 0 && jsz > 0) {
regex->re_jit_stack = pcre2_jit_stack_create(32 * 1024, 512 * 1024, NULL);
if (regex->re_jit_stack)
pcre2_jit_stack_assign(regex->re_mcontext, NULL, regex->re_jit_stack);
}
}
return 0;
}
return -1;
#endif
}
#endif
}
static void
rspamd_regexp_post_process (rspamd_regexp_t *r)
{
if (global_re_cache == NULL) {
rspamd_regexp_library_init (NULL);
}
#if defined(WITH_PCRE2)
gsize jsz;
guint jit_flags = PCRE2_JIT_COMPLETE;
/* Create match context */
r->mcontext = pcre2_match_context_create (NULL);
if (r->re != r->raw_re) {
r->raw_mcontext = pcre2_match_context_create (NULL);
}
else {
r->raw_mcontext = r->mcontext;
}
#ifdef HAVE_PCRE_JIT
if (pcre2_jit_compile (r->re, jit_flags) < 0) {
msg_err ("jit compilation of %s is not supported: %d", r->pattern, jit_flags);
r->flags |= RSPAMD_REGEXP_FLAG_DISABLE_JIT;
}
else {
if (!(pcre2_pattern_info (r->re, PCRE2_INFO_JITSIZE, &jsz) >= 0 && jsz > 0)) {
msg_err ("jit compilation of %s is not supported", r->pattern);
r->flags |= RSPAMD_REGEXP_FLAG_DISABLE_JIT;
}
}
if (!(r->flags & RSPAMD_REGEXP_FLAG_DISABLE_JIT)) {
pcre2_jit_stack_assign (r->mcontext, NULL, global_re_cache->jstack);
}
if (r->re != r->raw_re) {
if (pcre2_jit_compile (r->raw_re, jit_flags) < 0) {
msg_debug ("jit compilation of %s is not supported", r->pattern);
r->flags |= RSPAMD_REGEXP_FLAG_DISABLE_JIT;
}
if (!(pcre2_pattern_info (r->raw_re, PCRE2_INFO_JITSIZE, &jsz) >= 0 && jsz > 0)) {
msg_debug ("jit compilation of raw %s is not supported", r->pattern);
}
else if (!(r->flags & RSPAMD_REGEXP_FLAG_DISABLE_JIT)) {
pcre2_jit_stack_assign (r->raw_mcontext, NULL, global_re_cache->jstack);
}
}
#endif
#else
const gchar *err_str = "unknown";
gboolean try_jit = TRUE, try_raw_jit = TRUE;
gint study_flags = 0;
#if defined(HAVE_PCRE_JIT)
study_flags |= PCRE_STUDY_JIT_COMPILE;
#endif
/* Pcre 1 needs study */
if (r->re) {
r->extra = pcre_study (r->re, study_flags, &err_str);
if (r->extra == NULL) {
msg_debug ("cannot optimize regexp pattern: '%s': %s",
r->pattern, err_str);
try_jit = FALSE;
r->flags |= RSPAMD_REGEXP_FLAG_DISABLE_JIT;
}
}
else {
g_assert_not_reached ();
}
if (r->raw_re && r->raw_re != r->re) {
r->raw_extra = pcre_study (r->re, study_flags, &err_str);
}
else if (r->raw_re == r->re) {
r->raw_extra = r->extra;
}
if (r->raw_extra == NULL) {
msg_debug ("cannot optimize raw regexp pattern: '%s': %s",
r->pattern, err_str);
try_raw_jit = FALSE;
}
/* JIT path */
if (try_jit) {
#ifdef HAVE_PCRE_JIT
gint jit, n;
if (can_jit) {
jit = 0;
n = pcre_fullinfo (r->re, r->extra,
PCRE_INFO_JIT, &jit);
if (n != 0 || jit != 1) {
msg_debug ("jit compilation of %s is not supported", r->pattern);
r->flags |= RSPAMD_REGEXP_FLAG_DISABLE_JIT;
}
else {
pcre_assign_jit_stack (r->extra, NULL, global_re_cache->jstack);
}
}
#endif
}
else {
msg_debug ("cannot optimize regexp pattern: '%s': %s",
r->pattern, err_str);
r->flags |= RSPAMD_REGEXP_FLAG_DISABLE_JIT;
}
if (try_raw_jit) {
#ifdef HAVE_PCRE_JIT
gint jit, n;
if (can_jit) {
if (r->raw_re != r->re) {
jit = 0;
n = pcre_fullinfo (r->raw_re, r->raw_extra,
PCRE_INFO_JIT, &jit);
if (n != 0 || jit != 1) {
msg_debug ("jit compilation of %s is not supported", r->pattern);
r->flags |= RSPAMD_REGEXP_FLAG_DISABLE_JIT;
}
else {
pcre_assign_jit_stack (r->raw_extra, NULL,
global_re_cache->jstack);
}
}
}
#endif
}
#endif /* WITH_PCRE2 */
}
