Fix array_column() packed hash initialization

FILL assumes enough space is alredy allocated -- do that. Note that
the used size is an upper bound, albeit a likely one.

It's packed array

Optmized array_rand

Optimized array_pad

Folder marks

Optimized array_merge and cleanup

Optimized array_merge

Optimized array_column (thanks to Benjamin Coutu)

fixed IS_REFERENCE handling

Optimized array_reverse

Fix memory unsafety in array_walk()

Fixes bugs #61967, #62607, #69068, #70713.

The primary changes are:
a) Use the ht_iterator mechanism to ensure safety not only if the

Fix memory unsafety in array_walk()

Fixes bugs #61967, #62607, #69068, #70713.

The primary changes are:
a) Use the ht_iterator mechanism to ensure safety not only if the
iterated array itself changes, but also if it is replaced (and
potentially destroyed) entirely. We use the same semantics for
behavior under modification as foreach-by-reference. In
particular, we advance to the next element before processing it.
If the iterated entity is exchanged we iterate the new one from
the start. If it is not an array/object we warn and abort.
b) Always create a reference to the current value. Previously the
code kept the value as a non-reference and updated it to the
reference value produced by the user callback. However this is
unsafe, as the array may have been reallocated in the meantime,
so the previous value pointer is no longer value.
c) Around a recursive walk, incref the reference containing the
array. This ensures that the location where the currently
iterated value is stored cannot be freed.

One problem I was not able to solve is that we cannot decrement
the apply count if the array is exchanged during a recursive walk.

show more ...

Merge branch 'PHP-7.0' into PHP-7.1

Fix ref handling in array_merge/replace_recursive()

As usual, if the reference has rc=1 it really isn't a reference.

Merge branch 'PHP-7.0'

* PHP-7.0:
Fixed bug #72622 (array_walk + array_replace_recursive create references from nothing)

Fixed bug #72622 (array_walk + array_replace_recursive create references from nothing)

Increase array_rand() rehashing treshold

From 3/8 to 3/4. I was thinking in terms of nTableSize, where a
requirement > 1/2 is not tenable. However, we're actually working
with nNumUs

Increase array_rand() rehashing treshold

From 3/8 to 3/4. I was thinking in terms of nTableSize, where a
requirement > 1/2 is not tenable. However, we're actually working
with nNumUsed, in which case more than 1/4 tombstones should be
quite unusual.

show more ...

Optimize the n=1 case of array_rand()

Use zend_bitset

Improve array_rand distribution

Alias rand to mt_rand

Merge branch 'master' of git.php.net:/php-src

* 'master' of git.php.net:/php-src:
typo
Turn safe timeout handling into general interrupt handling ability.
Fixed compilation

Merge branch 'master' of git.php.net:/php-src

* 'master' of git.php.net:/php-src:
typo
Turn safe timeout handling into general interrupt handling ability.
Fixed compilation warnings
yet another test rewrite
Add additional LDAP SASL constants
Add tests for LDAP_OPT_* constants
Add LDAP TLS Protocol Min constants
Add tests for the LDAP CRL check option
Add missing LDAP constants

show more ...

Use zend_sort in array_multisort (tests change is expected)

Fixed compilation warnings

Added ZEND_ATTRIBUTE_FORMAT to some middind functions.
"%p" replaced by ZEND_LONG_FMT to avoid compilation warnings.
Fixed most incorrect use cases of format specifiers.

HANDLE_BLOCK/UNBLOCK_INTERRUPTIONS() protection on inter-process data is completely useless now.
Historicaly, these macros were introduced as a protection from execution timeout handling, but we

HANDLE_BLOCK/UNBLOCK_INTERRUPTIONS() protection on inter-process data is completely useless now.
Historicaly, these macros were introduced as a protection from execution timeout handling, but we don't need them anymore after "safe execution timeout" implementation.
These macros are still useful to protect from termination during inner process data modification, because of OS signals (e.g. SIGTERM during OPcache SHM update).

show more ...