On Thu, 21 Feb 2008 22:06:28 +0100, nicolas cellier wrote:

> Klaus D. Witzel a écrit :
>> Subject line says it all, check yourself,
>>  (Heap withAll: 'array') reject: [:x | x = $r]
>>  What's the rationale (there's no doc, no comment)? Archive shows that  
>> #species was changed to fix another (anonymous) bug but, this way the  
>> senders of #species can impossibly do what Smalltalk users expect from  
>> the collection hierarchy (and there is #asArray ...)
>>
>
> I have not the background to answer you.
> All I can tell is that this triggers another associativity failure:
>
>
> 	| anArray heap1 heap2 |
> 	anArray := #(1 2 3).
> 	heap1 := Heap withAll: (1 to: 3) sortBlock: [:a :b | a < b].
> 	heap2 := Heap withAll: (1 to: 3) sortBlock: [:a :b | b > a].
> 	(heap1 = anArray) & (heap2 = anArray) ==> (heap1 = heap2)
>
> Heaps do compare their sortBlock.

And no Array compares to a Heap, heap invariantly supers #= to  
SequenceableCollection.

> They are more relax when comparing with their species.

IC what you mean but, #species is invariant in their #= and their super #=.

> This is minor, but note nice.

It breaks all users of #species (#collect:, #reject:, #inject:, etc, all  
the powerful things in collection hierarchy); let's not call it minor then  
:)

> Nicolas
>
>> -  
>> http://lists.squeakfoundation.org/pipermail/squeak-dev/2000-March/011043.html  
>>   /Klaus
>>  P.S. Heap is used in not-so-uncritical parts, like  
>> #startTimerEventLoop and WorldState, so I don't want to "just" play  
>> with Heap's #species without any background info on what was fixed.
>>
>
>
>