From: Tyson Andre <tysonandre775@hotmail.com>
Date: Sat, 7 Dec 2019 15:09:32 +0000 (-0500)
Subject: Speed up array_intersect/array_diff/array_filter
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=366713d250d1ee01ec20f3b71fcd1c720519edcd;p=php

Speed up array_intersect/array_diff/array_filter

Use zend_hash_update instead of zend_hash_add.

These are taking a subset of keys from an array with unique keys,
so the result should also have unique keys.
(this is already done for array_map())

Also, speed up array_intersect and array_diff slightly by
using ZEND_HASH_FOREACH macros.
This way, it doesn't need to load the same buckets and array counts
from memory every time (compiler previously couldn't infer they won't change)

```php
<?php
// $n=10000 now takes 0.095 seconds instead of 0.102
function test_bench(int $n) {
    $values = range(0,1000);
    $other = range(0,1000);
    unset($other[500]);
    unset($values[400]);

    $total = 0;
    for ($i = 0; $i < $n; $i++) {
        $total += count(array_intersect_key($values, $other));
    }
    return $total;
}
```
---

diff --git a/ext/standard/array.c b/ext/standard/array.c
index a96d3a47f6..a47aee5ee3 100644
--- a/ext/standard/array.c
+++ b/ext/standard/array.c
@@ -4681,8 +4681,6 @@ static int zval_user_compare(zval *a, zval *b) /* {{{ */
 
 static void php_array_intersect_key(INTERNAL_FUNCTION_PARAMETERS, int data_compare_type) /* {{{ */
 {
-    uint32_t idx;
-	Bucket *p;
 	int argc, i;
 	zval *args;
 	int (*intersect_data_compare_func)(zval *, zval *) = NULL;
@@ -4690,6 +4688,8 @@ static void php_array_intersect_key(INTERNAL_FUNCTION_PARAMETERS, int data_compa
 	zval *val, *data;
 	int req_args;
 	char *param_spec;
+	zend_string *key;
+	zend_ulong h;
 
 	/* Get the argument count */
 	argc = ZEND_NUM_ARGS();
@@ -4727,21 +4727,15 @@ static void php_array_intersect_key(INTERNAL_FUNCTION_PARAMETERS, int data_compa
 
 	array_init(return_value);
 
-	for (idx = 0; idx < Z_ARRVAL(args[0])->nNumUsed; idx++) {
-		p = Z_ARRVAL(args[0])->arData + idx;
-		val = &p->val;
-		if (Z_TYPE_P(val) == IS_UNDEF) continue;
-		if (UNEXPECTED(Z_TYPE_P(val) == IS_INDIRECT)) {
-			val = Z_INDIRECT_P(val);
-			if (Z_TYPE_P(val) == IS_UNDEF) continue;
-		}
+	/* Iterate over keys of the first array (handling possibility of indirects such as in $GLOBALS), to compute keys that are in all of the other arrays. */
+	ZEND_HASH_FOREACH_KEY_VAL_IND(Z_ARRVAL(args[0]), h, key, val) {
 		if (Z_ISREF_P(val) && Z_REFCOUNT_P(val) == 1) {
 			val = Z_REFVAL_P(val);
 		}
-		if (p->key == NULL) {
+		if (key == NULL) {
 			ok = 1;
 			for (i = 1; i < argc; i++) {
-				if ((data = zend_hash_index_find(Z_ARRVAL(args[i]), p->h)) == NULL ||
+				if ((data = zend_hash_index_find(Z_ARRVAL(args[i]), h)) == NULL ||
 					(intersect_data_compare_func &&
 					intersect_data_compare_func(val, data) != 0)
 				) {
@@ -4751,12 +4745,12 @@ static void php_array_intersect_key(INTERNAL_FUNCTION_PARAMETERS, int data_compa
 			}
 			if (ok) {
 				Z_TRY_ADDREF_P(val);
-				zend_hash_index_update(Z_ARRVAL_P(return_value), p->h, val);
+				zend_hash_index_add_new(Z_ARRVAL_P(return_value), h, val);
 			}
 		} else {
 			ok = 1;
 			for (i = 1; i < argc; i++) {
-				if ((data = zend_hash_find_ex_ind(Z_ARRVAL(args[i]), p->key, 1)) == NULL ||
+				if ((data = zend_hash_find_ex_ind(Z_ARRVAL(args[i]), key, 1)) == NULL ||
 					(intersect_data_compare_func &&
 					intersect_data_compare_func(val, data) != 0)
 				) {
@@ -4766,10 +4760,10 @@ static void php_array_intersect_key(INTERNAL_FUNCTION_PARAMETERS, int data_compa
 			}
 			if (ok) {
 				Z_TRY_ADDREF_P(val);
-				zend_hash_update(Z_ARRVAL_P(return_value), p->key, val);
+				zend_hash_add_new(Z_ARRVAL_P(return_value), key, val);
 			}
 		}
-	}
+	} ZEND_HASH_FOREACH_END();
 }
 /* }}} */
 
@@ -5097,13 +5091,13 @@ PHP_FUNCTION(array_uintersect_uassoc)
 
 static void php_array_diff_key(INTERNAL_FUNCTION_PARAMETERS, int data_compare_type) /* {{{ */
 {
-    uint32_t idx;
-	Bucket *p;
 	int argc, i;
 	zval *args;
 	int (*diff_data_compare_func)(zval *, zval *) = NULL;
 	zend_bool ok;
 	zval *val, *data;
+	zend_string *key;
+	zend_ulong h;
 
 	/* Get the argument count */
 	argc = ZEND_NUM_ARGS();
@@ -5138,21 +5132,15 @@ static void php_array_diff_key(INTERNAL_FUNCTION_PARAMETERS, int data_compare_ty
 
 	array_init(return_value);
 
-	for (idx = 0; idx < Z_ARRVAL(args[0])->nNumUsed; idx++) {
-		p = Z_ARRVAL(args[0])->arData + idx;
-		val = &p->val;
-		if (Z_TYPE_P(val) == IS_UNDEF) continue;
-		if (UNEXPECTED(Z_TYPE_P(val) == IS_INDIRECT)) {
-			val = Z_INDIRECT_P(val);
-			if (Z_TYPE_P(val) == IS_UNDEF) continue;
-		}
+	/* Iterate over keys of the first array (handling possibility of indirects such as in $GLOBALS), to compute keys that aren't in the other arrays. */
+	ZEND_HASH_FOREACH_KEY_VAL_IND(Z_ARRVAL(args[0]), h, key, val) {
 		if (Z_ISREF_P(val) && Z_REFCOUNT_P(val) == 1) {
 			val = Z_REFVAL_P(val);
 		}
-		if (p->key == NULL) {
+		if (key == NULL) {
 			ok = 1;
 			for (i = 1; i < argc; i++) {
-				if ((data = zend_hash_index_find(Z_ARRVAL(args[i]), p->h)) != NULL &&
+				if ((data = zend_hash_index_find(Z_ARRVAL(args[i]), h)) != NULL &&
 					(!diff_data_compare_func ||
 					diff_data_compare_func(val, data) == 0)
 				) {
@@ -5162,12 +5150,12 @@ static void php_array_diff_key(INTERNAL_FUNCTION_PARAMETERS, int data_compare_ty
 			}
 			if (ok) {
 				Z_TRY_ADDREF_P(val);
-				zend_hash_index_update(Z_ARRVAL_P(return_value), p->h, val);
+				zend_hash_index_add_new(Z_ARRVAL_P(return_value), h, val);
 			}
 		} else {
 			ok = 1;
 			for (i = 1; i < argc; i++) {
-				if ((data = zend_hash_find_ex_ind(Z_ARRVAL(args[i]), p->key, 1)) != NULL &&
+				if ((data = zend_hash_find_ex_ind(Z_ARRVAL(args[i]), key, 1)) != NULL &&
 					(!diff_data_compare_func ||
 					diff_data_compare_func(val, data) == 0)
 				) {
@@ -5177,10 +5165,10 @@ static void php_array_diff_key(INTERNAL_FUNCTION_PARAMETERS, int data_compare_ty
 			}
 			if (ok) {
 				Z_TRY_ADDREF_P(val);
-				zend_hash_update(Z_ARRVAL_P(return_value), p->key, val);
+				zend_hash_add_new(Z_ARRVAL_P(return_value), key, val);
 			}
 		}
-	}
+	} ZEND_HASH_FOREACH_END();
 }
 /* }}} */
 
@@ -6103,11 +6091,12 @@ PHP_FUNCTION(array_filter)
 		Z_PARAM_LONG(use_type)
 	ZEND_PARSE_PARAMETERS_END();
 
-	array_init(return_value);
 	if (zend_hash_num_elements(Z_ARRVAL_P(array)) == 0) {
+		RETVAL_EMPTY_ARRAY();
 		zend_release_fcall_info_cache(&fci_cache);
 		return;
 	}
+	array_init(return_value);
 
 	if (ZEND_NUM_ARGS() > 1) {
 		have_callback = 1;
@@ -6161,9 +6150,9 @@ PHP_FUNCTION(array_filter)
 		}
 
 		if (string_key) {
-			operand = zend_hash_update(Z_ARRVAL_P(return_value), string_key, operand);
+			operand = zend_hash_add_new(Z_ARRVAL_P(return_value), string_key, operand);
 		} else {
-			operand = zend_hash_index_update(Z_ARRVAL_P(return_value), num_key, operand);
+			operand = zend_hash_index_add_new(Z_ARRVAL_P(return_value), num_key, operand);
 		}
 		zval_add_ref(operand);
 	} ZEND_HASH_FOREACH_END();
@@ -6402,7 +6391,11 @@ PHP_FUNCTION(array_chunk)
 	num_in = zend_hash_num_elements(Z_ARRVAL_P(input));
 
 	if (size > num_in) {
-		size = num_in > 0 ? num_in : 1;
+		if (num_in == 0) {
+			RETVAL_EMPTY_ARRAY();
+			return;
+		}
+		size = num_in;
 	}
 
 	array_init_size(return_value, (uint32_t)(((num_in - 1) / size) + 1));
@@ -6418,9 +6411,9 @@ PHP_FUNCTION(array_chunk)
 		/* Add entry to the chunk, preserving keys if necessary. */
 		if (preserve_keys) {
 			if (str_key) {
-				entry = zend_hash_update(Z_ARRVAL(chunk), str_key, entry);
+				entry = zend_hash_add_new(Z_ARRVAL(chunk), str_key, entry);
 			} else {
-				entry = zend_hash_index_update(Z_ARRVAL(chunk), num_key, entry);
+				entry = zend_hash_index_add_new(Z_ARRVAL(chunk), num_key, entry);
 			}
 		} else {
 			entry = zend_hash_next_index_insert(Z_ARRVAL(chunk), entry);