CC to Christian Seiler http://www.haskell.org/pipermail/haskell-cafe/2008-June/044379.html http://www.haskell.org/pipermail/haskell-cafe/2008-June/thread.html On Wednesday 18 June 2008, Luke Palmer wrote:
* A closure must be able to call itself recursively (via a higher-order function typically)
I see two ways a closure might get a hold of itself, in order it can call itself: $f = function () { lexical $whatever; lexical $f; //get yourself by lexical scope return $f(); }; $f(); $g = function ($g) { //get yourself by parameter lexical $whatever; return $g($g); }; $g($g); Getting the first version to work is somewhat tricky in a non-lazy language, but it would be nice to have. The second should definately work. I guess I'll download the patch and try.
* I would recommend only saving $this in the op_array structure if the closure actually references $this -- if that is possible to deduce at the time. Otherwise you might run into unexpected poor memory performances in certain cases.
Agreed. A closure created inside an object should be able to outlive the object by not holding a reference to it. Since many PHP programmers put pretty much all of their functions into classes for style reasons, which would mean most closures are created in the context of an object, implicitly referencing $this might prevent a lot of objects from being garbage-collected. Also, doesn't that turn all lambdas defined inside an object into closures, which are heavier? Between always referencing $this in a lamda and requiring "lexical $this", the latter seems like the smaller evil to me. Gesundheit Wag