>
> On Wed, Jun 13, 2012 at 5:29 PM, Levente Uzonyi <
leves@elte.hu> wrote:
>
> � � On Wed, 13 Jun 2012, Guillermo Polito wrote:
>
> � � � � soo, what do I do? :D
>
>
>
> � � Streaming into an OrderedCollection is unnecessary, because
> � � OrderedCollection is already a stream-like object. If I were
> � � implementing Smalltalk now, I would probably add the stream
> � � protocol to it (#nextPut:, #nextPutAll:, etc).
> � � A simple fix to your problem is to replace the line
>
> � � � � � �grownCollection := collection class new: newSize.
>
> � � with
>
> � � � � � �grownCollection := collection class ofSize: newSize.
>
> � � in WriteStream >> #growTo: (note that I didn't check current Pharo
> � � code, so this might be different there).
>
>
> � � Levente
>
>
>
>
> � � � � On 6/13/12, Camillo Bruni <
camillobruni@gmail.com> wrote:
>
> � � � � yup I know that :D
> � � � � � � And I provided a fix on e year ago, that got lost in a big
> � � � � � � refactoring...
> � � � � � � - I added an explicit #streamSpecies on the Collection
> � � � � � � classes.
> � � � � � � - By default it returns the same class
> � � � � � � - on Set / OrderedCollection / Symbol it returns the
> � � � � � � mutable types
> � � � � � � (LinkedList as well I think)
> � � � � � � - overwrote #streamContents: to convert from the
> � � � � � � #streamSpecies type back to
> � � � � � � the original class
>
>
>
> � � � � � � On 2012-06-13, at 14:56, Guillermo Polito wrote:
>
> � � � � � � � � � � � � Hi guys!
>
> � � � � � � � � I'm chasing a bug that appeared in glorp under pharo
> � � � � � � � � 1.4. �Now, the
> � � � � � � � � bug is due to some behavior changed in
> � � � � � � � � OrderedCollection I think. Look
> � � � � � � � � at this piece of code:
>
>
> � � � � � � � � oc := OrderedCollection new.
> � � � � � � � � ws := oc writeStream.
>
> � � � � � � � � "this explodes"
> � � � � � � � � ws nextPutAll: (OrderedCollection with: 1 with: 2
> � � � � � � � � with: 3).
>
> � � � � � � � � "this works"
> � � � � � � � � ws nextPutAll: {1.2.3}
>
>
> � � � � � � � � And I'm puzzled, why should one work and the other not
> � � � � � � � � from the pov of the
> � � � � � � � � user?
> � � � � � � � � And how should I replace that behavior if it's my bug?
>
> � � � � � � � � putting an asArray for each nextPutAll: does not look
> � � � � � � � � good for me... :S
>
> � � � � � � � � Tx,
> � � � � � � � � Guille
>
>
>
>
>
>
>
>
>
>
>
>