Final tweaks

More locking work

*** empty log message ***

More stuff

Step 4:
Move to a 7-bit counter (not fully implemented yet)

Phase 3:
Use a single bit to mark IS_REF variables

Step 2:
Rename is_ref to EA

Support isset()/empty() for string offsets

* Use to_string() instead of __print()
* Support boolean casts ((bool) and (boolean))

* Make the output handling of variables much, much cooler.
Uses zend_make_printable_zval() instead of convert_to_string() now:

$foo = true;
print "\$foo is $foo";
will now prin

* Make the output handling of variables much, much cooler.
Uses zend_make_printable_zval() instead of convert_to_string() now:

$foo = true;
print "\$foo is $foo";
will now print
$foo is true
(instead of "$foo is 1", earlier).

Also, with objects, it automatically tries to call __print() and use it as a printing
function.

For example:

class foo {
function __print() { return "Foo Object"; }
};

$foo = new foo;
print $foo;

will print "Foo Object".

show more ...

* Fix cases where you assign an array element to the parent array (the array was
being erased before the assignment, so the element was being smashed).

* Fix foreach() that receives a non array argument
* Clean up some C++ comments

* Fix a by-name call/method call bug
* Clean and optimize the whole function call process

- Fixed Karl's bug report. It's not really a thorough fix, we really need to rethink the INIT_FCALL/DO_FCALL issue.
- Fixed numerous AiCount problems

New $GLOBALS init

- We weren't counting newlines in heredocs. The only place which is still questionable
is when there's a \ followed by a newline but it seems we have a parse error in this
case anyways.
-

- We weren't counting newlines in heredocs. The only place which is still questionable
is when there's a \ followed by a newline but it seems we have a parse error in this
case anyways.
- Fixed the alloca() macros so that the alloca() #define in win32 mode won't clash
with the real win32 alloca().

show more ...

- Make execute() use less stack in thread-safe win32 due to Microsoft's shitty 256kb stack.

Fixes

Correct fix

Fix a leak

* Support getThis() for internal functions.
* Fix 'new object or die' and AiCount issue thoroughly (earlier fix didn't
work with the optimizer).
* Add new macros for standardized defini

* Support getThis() for internal functions.
* Fix 'new object or die' and AiCount issue thoroughly (earlier fix didn't
work with the optimizer).
* Add new macros for standardized definition of classes.
* Only report AiCount problems if shutdown was not silent.

show more ...

Fix the AiCount issue with objects

Sigh, another leak bites the dust. FREE_OP missing in case of a SEND_VAR.

* Add struct name to all typedef's so that they can be debugged with MSVC
* Fix an AiCount bug - list(...) = $var was using $var multiple times, and thus
causing AiCount to be decreased multi

* Add struct name to all typedef's so that they can be debugged with MSVC
* Fix an AiCount bug - list(...) = $var was using $var multiple times, and thus
causing AiCount to be decreased multiple times even though it was increased only
once for $var. Mark all FETCH_DIM's so that they won't decrease AiCount, and only
decrease AiCount on the last FETCH_DIM.
* Fix a stupid bug - forgot to pass CLS_C to some compiler function. For some reason
MSVC doesn't report these :I

show more ...

Optimize allocations into uninitialized_zval assignments