This isn't really workable since PHP doesn't support intercepting
accesses to global variables (nor to static class properties, so
we can't wrap C/C++ global variables that way either).
The _get() and _set() function wrappers actually work and have
been generated for a very long time.
Mark the parameter as "byref" and write back through the reference
after the call.
Adjust testcase argout to uncomment the part that's meant to test this,
and to remove lingering traces of PHP's old call-time pass-by-reference
(which was completely removed before PHP 7).
Fixes#1457
This was used to store custom properties, but we can just ask the PHP
object to store them like it normally would, after checking for our
custom pseudo-properties.
It existed to work around const-correctness issues in older versions of
PHP's C API. It's conceivable user code might be using it, but unlikely
and the switch to creating classes via the API is a natural time for a
compatibility break.
long long[] map was using JNI type as C/C++ type. General spirit of
JAVA_ARRAYS_TYPEMAPS is to cast individual array elements, not to
rely on potentially incompatible pointer casts.
As for __int64 definition. __int64 is a non-standard Visual-C-specific
type used in win32/jni_md.h. It is defined by other Win32 compilers in
one way or another, obviously for compatibility. It's more appropriate
to give the compiler a chance to make necessary arrangements instead
of reinventing the wheel. This, giving a chance, can be achieved by
including virtually any standard header. Since jni.h includes stdio.h,
defining __int64 in javahead.swg is redundant. Since doing so actually
triggers compilation errors on MinGW if a system header is included
in the %begin section, it's arguably appropriate to omit it.
As for #undef _LP64 removal. Undefining a pre-defined macro, which
_LP64 is, is bad style, and if followed by inclusion of systems
headers, it's actually error-prone. Log suggests that it was added
to resolve a warning. I'm inclined to believe that it rather was a
misunderstanding of some kind. Or a bug in warning subsystem of
some particular compiler version, in which case it would have been
more appropriate to advise users to ignore the warning.
Reportedly the code we were using in the directorin case gave segfaults
in PHP 7.2 and later - we've been unable to reproduce these, but the new
approach is also simpler and should be bit faster too.
Fixes#1527, #1975
PHP already provides all the wrapped constants and all the wrapped
functions except fabs() (PHP provides abs() instead). Rewrapping
the constants causes warnings or errors (depending on PHP version)
and the rewrapped functions seem to be hidden by the built-in
versions, so only wrap fabs() for PHP.
(Even a wrapper for fabs() seems of little use since abs() is already
provided, but really math.i seems of little use more generally since
any general purpose programming language will provide its own maths
functions and constants - the key motivation here is to eliminate
warnings and errors from running the testsuite.)
Add testcase for previous commit.
Add missing assert for future type conversions support that will use
the heap instead of pointer casts (for smart pointer type conversions).
Closes#1963
