Bug Static member about pthreads

[krakjoe]

I will assume that the example you posted was correct when you ran it.

I can see a leak and a segfault when I run a variation of it, which shouldn't happen. But static variables do not have the abilities or functionality you are trying to use here and should be avoided completely if at all possible, they cannot be used in a truly safe manner. They cannot be shared among threads safely, and your code will be more stable and safe if you do not use them.

EDIT: I thought you were someone else, have a look through previous bugs concerning static variables, it's been discussed before, I'm sorry it's not documented yet I just haven't had the time ...

[bride]

I did another test, error occurs using self, if access using

// Error
public static function get() {
if (!self::$instance)
self::$instance = new self;
return self::$instance;
}

//Works with access static property in thread

public function get() {
if (!$this::$instance)
$this::$instance = new self;
return $this::$instance;
}

//Works with access static property in thread

public static function get() {
if (!MyThread::$instance)
MyThread::$instance = new self;
return MyThread::$instance;
}

[krakjoe]

Could you please provide full, properly formatted examples, I think I understand what you're saying but it'd be nice to be certain :)

Thanks for taking the time to keep looking, it's very helpful ...

[bride]

I have a meeting now, will post later.
what I mean is:

• self :: $ propertyName does not work in the context of the thread (static properties).
• MyThread::$propertyName and $this::$propertyName work in the context of the thread (static properties).

[krakjoe]

Ah right yeah that makes sense, both $this and the entry are declared for the actual threading context explicitly, I guess I need to handle the self keyword somehow, I will look into it ... cheers