Re: Implementing NSFastEnumeration

On Oct 1, 2009, at 18:43, Gerd Knops wrote:

>
> On Oct 1, 2009, at 11:30 AM, Christiaan Hofman wrote:
>>
> [..]
>
>> While this works for small arrays, for larger arrays this gives a  
>> memory access crasher. The reason seems to be that NSArray can  
>> return more items than len (the size of stackbuf) in a single call.
>>
> Correct. The trick is to use a variable in the state to note you  
> count. state->state happens to be 0 when first called, and OS X will  
> not touch it. So you can use it to remember where you were at for  
> the next call. See my example code.

Well but that means that /I/ have to do the enumeration, while my  
question was really whether I could let NSArray do it. In other words,  
in my sample I first use NSFastEnumeration on the keys array. When I  
modify state->state myself I'm pretty sure that I cannot safely do  
that, because NSArray would want to use that field for its own  
enumeration.

>
>> And another question I have been unable to find an answer to in the  
>> docs. Foundation collections are implemented as class clusters. So  
>> is NSFastEnumeration implemented as a primitive or a derived  
>> method? In other words: when I subclass NSArray (or another  
>> collection class), does it already have a default implementation of  
>> NSFastEnumeration, or am I forced to always implement it when I  
>> want to use for(...in...)?
>>
> AFAIK subclassing class clusters doesn't work, which is why I think  
> they are a poor design.
>
> Gerd
>

That's definitely wrong: subclassing class clusters does work, and  
I've successfully done it many times. When done right it can actually  
be a pretty useful design, and apart from that it's important for toll- 
free bridging. The only thing to know is which methods are primitive,  
because you have to implement those. And that was exactly my question.

Christiaan