Skip to content

Add support for PCRE marks #609

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

Add support for PCRE marks #609

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 34 additions & 2 deletions ext/pcre/php_pcre.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -577,6 +577,8 @@ PHPAPI void php_pcre_match_impl(pcre_cache_entry *pce, char *subject, int subjec
int i, rc;
int subpats_order; /* Order of subpattern matches */
int offset_capture; /* Capture match offsets: yes/no */
unsigned char *mark; /* Target for MARK name */
zval *marks = NULL; /* Array of marks for PREG_PATTERN_ORDER */

/* Overwrite the passed-in value for subpatterns with an empty array. */
if (subpats != NULL) {
Expand Down Expand Up @@ -619,6 +621,8 @@ PHPAPI void php_pcre_match_impl(pcre_cache_entry *pce, char *subject, int subjec
}
extra->match_limit = PCRE_G(backtrack_limit);
extra->match_limit_recursion = PCRE_G(recursion_limit);
extra->mark = &mark;
extra->flags |= PCRE_EXTRA_MARK;

/* Calculate the size of the offsets array, and allocate memory for it. */
rc = pcre_fullinfo(pce->re, extra, PCRE_INFO_CAPTURECOUNT, &num_subpats);
Expand Down Expand Up @@ -695,6 +699,14 @@ PHPAPI void php_pcre_match_impl(pcre_cache_entry *pce, char *subject, int subjec
offsets[(i<<1)+1] - offsets[i<<1], 1);
}
}
/* Add MARK, if available */
if (mark) {
if (!marks) {
MAKE_STD_ZVAL(marks);
array_init(marks);
}
add_index_string(marks, matched - 1, (char *) mark, 1);
}
/*
* If the number of captured subpatterns on this run is
* less than the total possible number, pad the result
Expand Down Expand Up @@ -725,6 +737,10 @@ PHPAPI void php_pcre_match_impl(pcre_cache_entry *pce, char *subject, int subjec
offsets[(i<<1)+1] - offsets[i<<1], 1);
}
}
/* Add MARK, if available */
if (mark) {
add_assoc_string(result_set, "MARK", (char *) mark, 1);
}
/* And add it to the output array */
zend_hash_next_index_insert(Z_ARRVAL_P(subpats), &result_set, sizeof(zval *), NULL);
}
Expand All @@ -744,6 +760,10 @@ PHPAPI void php_pcre_match_impl(pcre_cache_entry *pce, char *subject, int subjec
offsets[(i<<1)+1] - offsets[i<<1], 1);
}
}
/* Add MARK, if available */
if (mark) {
add_assoc_string(subpats, "MARK", (char *) mark, 1);
}
}

pcre_free((void *) stringlist);
Expand Down Expand Up @@ -784,6 +804,10 @@ PHPAPI void php_pcre_match_impl(pcre_cache_entry *pce, char *subject, int subjec
zend_hash_next_index_insert(Z_ARRVAL_P(subpats), &match_sets[i], sizeof(zval *), NULL);
}
efree(match_sets);

if (marks) {
add_assoc_zval(subpats, "MARK", marks);
}
}

efree(offsets);
Expand Down Expand Up @@ -855,7 +879,7 @@ static int preg_get_backref(char **str, int *backref)

/* {{{ preg_do_repl_func
*/
static int preg_do_repl_func(zval *function, char *subject, int *offsets, char **subpat_names, int count, char **result TSRMLS_DC)
static int preg_do_repl_func(zval *function, char *subject, int *offsets, char **subpat_names, int count, unsigned char *mark, char **result TSRMLS_DC)
{
zval *retval_ptr; /* Function return value */
zval **args[1]; /* Argument to pass to function */
Expand All @@ -871,6 +895,9 @@ static int preg_do_repl_func(zval *function, char *subject, int *offsets, char *
}
add_next_index_stringl(subpats, &subject[offsets[i<<1]], offsets[(i<<1)+1] - offsets[i<<1], 1);
}
if (mark) {
add_assoc_string(subpats, "MARK", (char *) mark, 1);
}
args[0] = &subpats;

if (call_user_function_ex(EG(function_table), NULL, function, &retval_ptr, 1, args, 0, NULL TSRMLS_CC) == SUCCESS && retval_ptr) {
Expand Down Expand Up @@ -1032,13 +1059,16 @@ PHPAPI char *php_pcre_replace_impl(pcre_cache_entry *pce, char *subject, int sub
*eval_result, /* Result of eval or custom function */
walk_last; /* Last walked character */
int rc;
unsigned char *mark; /* Target for MARK */

if (extra == NULL) {
extra_data.flags = PCRE_EXTRA_MATCH_LIMIT | PCRE_EXTRA_MATCH_LIMIT_RECURSION;
extra = &extra_data;
}
extra->match_limit = PCRE_G(backtrack_limit);
extra->match_limit_recursion = PCRE_G(recursion_limit);
extra->mark = &mark;
extra->flags |= PCRE_EXTRA_MARK;

eval = pce->preg_options & PREG_REPLACE_EVAL;
if (is_callable_replace) {
Expand Down Expand Up @@ -1118,7 +1148,7 @@ PHPAPI char *php_pcre_replace_impl(pcre_cache_entry *pce, char *subject, int sub
new_len += eval_result_len;
} else if (is_callable_replace) {
/* Use custom function to get replacement string and its length. */
eval_result_len = preg_do_repl_func(replace_val, subject, offsets, subpat_names, count, &eval_result TSRMLS_CC);
eval_result_len = preg_do_repl_func(replace_val, subject, offsets, subpat_names, count, mark, &eval_result TSRMLS_CC);
new_len += eval_result_len;
} else { /* do regular substitution */
walk = replace;
Expand Down Expand Up @@ -1517,6 +1547,7 @@ PHPAPI void php_pcre_split_impl(pcre_cache_entry *pce, char *subject, int subjec
}
extra->match_limit = PCRE_G(backtrack_limit);
extra->match_limit_recursion = PCRE_G(recursion_limit);
extra->flags &= ~PCRE_EXTRA_MARK;

/* Initialize return value */
array_init(return_value);
Expand Down Expand Up @@ -1785,6 +1816,7 @@ PHPAPI void php_pcre_grep_impl(pcre_cache_entry *pce, zval *input, zval *return
}
extra->match_limit = PCRE_G(backtrack_limit);
extra->match_limit_recursion = PCRE_G(recursion_limit);
extra->flags &= ~PCRE_EXTRA_MARK;

/* Calculate the size of the offsets array, and allocate memory for it. */
rc = pcre_fullinfo(pce->re, extra, PCRE_INFO_CAPTURECOUNT, &size_offsets);
Expand Down
196 changes: 196 additions & 0 deletions ext/pcre/tests/marks.phpt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,196 @@
--TEST--
Test support for PCRE marks
--FILE--
<?php

$regex = <<<'REGEX'
/
_ (a) (*MARK:A_MARK) _
| _ (b) _
| _ (c) (*MARK:C_MARK) _
| _ (d) _
/x
REGEX;

var_dump(preg_match($regex, '_c_', $matches));
var_dump($matches);

var_dump(preg_match_all($regex, '_a__b__c__d_', $matches, PREG_PATTERN_ORDER));
var_dump($matches);

var_dump(preg_match_all($regex, '_a__b__c__d_', $matches, PREG_SET_ORDER));
var_dump($matches);

var_dump(preg_replace_callback($regex, function($matches) {
var_dump($matches);
return $matches[0];
}, '_a__b__c__d_'));

?>
--EXPECTF--
int(1)
array(5) {
[0]=>
string(3) "_c_"
[1]=>
string(0) ""
[2]=>
string(0) ""
[3]=>
string(1) "c"
["MARK"]=>
string(6) "C_MARK"
}
int(4)
array(6) {
[0]=>
array(4) {
[0]=>
string(3) "_a_"
[1]=>
string(3) "_b_"
[2]=>
string(3) "_c_"
[3]=>
string(3) "_d_"
}
[1]=>
array(4) {
[0]=>
string(1) "a"
[1]=>
string(0) ""
[2]=>
string(0) ""
[3]=>
string(0) ""
}
[2]=>
array(4) {
[0]=>
string(0) ""
[1]=>
string(1) "b"
[2]=>
string(0) ""
[3]=>
string(0) ""
}
[3]=>
array(4) {
[0]=>
string(0) ""
[1]=>
string(0) ""
[2]=>
string(1) "c"
[3]=>
string(0) ""
}
[4]=>
array(4) {
[0]=>
string(0) ""
[1]=>
string(0) ""
[2]=>
string(0) ""
[3]=>
string(1) "d"
}
["MARK"]=>
array(2) {
[0]=>
string(6) "A_MARK"
[2]=>
string(6) "C_MARK"
}
}
int(4)
array(4) {
[0]=>
array(3) {
[0]=>
string(3) "_a_"
[1]=>
string(1) "a"
["MARK"]=>
string(6) "A_MARK"
}
[1]=>
array(3) {
[0]=>
string(3) "_b_"
[1]=>
string(0) ""
[2]=>
string(1) "b"
}
[2]=>
array(5) {
[0]=>
string(3) "_c_"
[1]=>
string(0) ""
[2]=>
string(0) ""
[3]=>
string(1) "c"
["MARK"]=>
string(6) "C_MARK"
}
[3]=>
array(5) {
[0]=>
string(3) "_d_"
[1]=>
string(0) ""
[2]=>
string(0) ""
[3]=>
string(0) ""
[4]=>
string(1) "d"
}
}
array(3) {
[0]=>
string(3) "_a_"
[1]=>
string(1) "a"
["MARK"]=>
string(6) "A_MARK"
}
array(3) {
[0]=>
string(3) "_b_"
[1]=>
string(0) ""
[2]=>
string(1) "b"
}
array(5) {
[0]=>
string(3) "_c_"
[1]=>
string(0) ""
[2]=>
string(0) ""
[3]=>
string(1) "c"
["MARK"]=>
string(6) "C_MARK"
}
array(5) {
[0]=>
string(3) "_d_"
[1]=>
string(0) ""
[2]=>
string(0) ""
[3]=>
string(0) ""
[4]=>
string(1) "d"
}
string(12) "_a__b__c__d_"