This content has been marked as final. Show 4 replies
IUID interface is what you need to implement to provide a unique id for you objects. Once you've done that, as you said, you'll, need to play run certain tests to see how it fares in performance.
You might want to publish your finding for the forum!
IUID is interesting but I doubt its related to the question, as I understand question is if flex uses hash code to implement collections which is lot more efficient than liniar search,(assuming every object has unique hashcode) I think its more of a ActionScript (ecmascript) question than flex. I doubt there is bucket search support in flex collections but would like to hear from adobe team as well I am also interested, +keeping it alive ..
Yes, my original question is unanswered. It probably can only be answered by someone who knows the implementation under the hood - probably a member of the Adobe Flex team.
I am not sure if it is necessarily an Ecmascript/Actionscript questions since I can conceive of an implementation of ListCollectionView that could examine the object's contents, at least partially, and use the information to do better than O(n) operations. On that note, I was thinking perhaps implementing IUID might help the Flex library do better than O(n), but it seems a bit far-fetched.
In any case, it seems worth the time of somebody from the Flex / ActionScript team to offer some insight into this important issue (since it relates to scale and fundamental aspects of Flex)
Thanks in advance & thanks for the responses so far.
I want to chime in here too, as the answer has relevance for the work I am doing. Anyone know the answer? Thanks.