From: Robert Klemme <bob.news@...>
Date: 2005-08-04T02:11:05+09:00
Subject: Re: Cascading <=> comparisons

Garance A Drosehn <drosihn@gmail.com> wrote:
> On 7/16/05, Robert Klemme <bob.news@gmx.net> wrote:
>> Garance A Drosehn <drosihn@gmail.com> wrote:
>>
>> Noone explicitely commented on the use of hash.keys.sort:
>> this is quite inefficient.  Instead doing the sort on the hash
>> directly is much more efficient:
>
> I can see how this would be faster wrt execution, but if I use
> hashobj.sort then I end up with an array of arrays which have
> both the key-names and their values.  Memory-wise, isn't that
> going to be significantly larger than an array which only has
> the key-names?  When I wrote this, I mainly thought about the
> memory usage, and I just assumed that the performance would
> be "about-the-same" either way...
>
> I'll check over my actual script wrt this idea.  It's probably true
> that I would rather optimize the speed than the memory usage,
> in which case I might take advantage of this.  I'll certainly use
> it in other scripts I write!  Thanks.
>
> (note that in the actual script, the values are not simple integers,
> but objects which in turn have a lot of instance-variables...)

That doesn't make a difference as they are not copied.  Mem overhead is just 
the temp arrays.

Kind regards

    robert