Trim trailing whitespace in source code files

year++

Set trailing zero

Fixed bug #74589 __DIR__ wrong for unicode character

Fixed another potential dangling pointer

Fixed bug #74546 (SIGILL in ZEND_FETCH_CLASS_CONSTANT_SPEC_CONST_CONST_HANDLER())

Improve accuracy of opline lineno information

If compile_var() was used instead of compile_expr() we did not
update the current lineno.

Fixed bug #74444: multiple catch freezes in some cases

zend_emit_jump() may reallocate, so reload the opline.

Fix lineno for AST_ZVAL nodes

Fix potential crash when setting invalid declare value

Using a non-literal expression in a declare value can cause the
compiler to crash trying to turn that AST node into a usable zval.

Fix potential crash when setting invalid declare value

Using a non-literal expression in a declare value can cause the
compiler to crash trying to turn that AST node into a usable zval.

There was an existing test for such values using 'encoding',
but that didn't crash because it's handled by the lexer
rather than being compiled.

Trying to use a non-literal with ticks reproduces the crash.

show more ...

Update copyright headers to 2017

Update copyright headers to 2017

Fix string extension in negate_num_string

Forgot to actually assign the reallocated string...

Revert "Reorder the oplines" (Break Zend/tests/return_types/bug71092.phpt)

This reverts commit a12f43ee2c3d27a7eb9a3c873901643df5f7197e.

Reorder the oplines

1. we should only do the return type checking when it is really about to
return
2. for 029.php, actually, the exception threw should be discard while it
jmp i

Reorder the oplines

1. we should only do the return type checking when it is really about to
return
2. for 029.php, actually, the exception threw should be discard while it
jmp into finally(it could be observed by change the return to return an array)
3. after this fix, the test 029.phpt behavior consistently with 7.0
4. good for optimizer too

show more ...

Fix T_NUM_STRING negation

T_NUM_STRING follows the rules of symtable numeric string
conversion. If the offset isn't an integer under those rules, it
is treated as a string. This shou

Fix T_NUM_STRING negation

T_NUM_STRING follows the rules of symtable numeric string
conversion. If the offset isn't an integer under those rules, it
is treated as a string. This should apply to negated T_NUM_STRINGs
as well.

show more ...

Fixed bug #73663 ("Invalid opcode 65/16/8" occurs with a variable created with list())

Fixed bug #73585 (Logging of "Internal Zend error - Missing class information" missing class name)

Actually, I doubt how thing bug could be triggered

Intriduced ZEND_ACC_INHERITED class flag.
It's going to be helpful for static optimisations, when "parent" is not known yet.

Fix bug #66773, #66862

This a partial backport of 8754b19. It
a) fixes the class/function/constant import table confusion in the
namespaced case, and
b) restricts conflict che

Fix bug #66773, #66862

This a partial backport of 8754b19. It
a) fixes the class/function/constant import table confusion in the
namespaced case, and
b) restricts conflict checks to a single file based on a filename
pointer comparison.

It does not fix the issues with filename reuse (e.g. due to eval)
and late-bound classes. This part of the change requires globals
changes.

show more ...

Handle resource keys in constexpr arrays

Fix the constant array case as well

Combine code for keyed/unkeyed list()

Fix segfault when empty entry in keyed array assignment

Use SEND_USER for CONST|TMP as well

Otherwise we're missing the "expected to be a reference, value
given" warning that appears for ordinary calls to call_user_func().

Also updat

Use SEND_USER for CONST|TMP as well

Otherwise we're missing the "expected to be a reference, value
given" warning that appears for ordinary calls to call_user_func().

Also update an UPGRADING note with recent changes wrt
call_user_func().

show more ...