[Article Contest, first draft] D Slices

[Steven Schveighoffer]

On Thu, 19 May 2011 07:42:57 -0400, Lars T. Kyllingstad  
<public at kyllingen.nospamnet> wrote:

> On Wed, 18 May 2011 14:03:05 -0400, Steven Schveighoffer wrote:
>
>> Having seen quite a few incorrect descriptions of how D slices work
>> (particularly regarding appending), I wrote an article that tries to
>> describe how D slices work, and why they behave the way they do.
>>
>> Being one of the only places where I have web space, it's on my
>> dcollections site, I probably will move it to the D wiki eventually, but
>> I'm much more familiar with the Trac wiki syntax.
>>
>> Please, if you have any comments or recommendations, let me know.  I
>> certainly am no author, so I probably screwed a few things up :)
>>
>> http://www.dsource.org/projects/dcollections/wiki/ArrayArticle
>
> This is an excellent article!  It answers many questions about the new
> appending system which I've been meaning to ask you for a long time.

Thanks!

>
> One question, still:  It seems inconsistent that the value of capacity()
> normally includes the existing slice elements, yet suddenly returns 0
> when the array cannot be appended to.  Why doesn't it return  
> slice.length?

Well, I think someone has brought that up before, and I thought it was a  
good idea, but I found a reason recently for leaving it the way it is.

If an array has completely filled a memory block, and you have a slice  
that ends at the end of that memory block, the capacity will be equal to  
the slice length.  This might be a condition you want to know, vs a slice  
that does not end at the end of the array (which currently returns  
capacity of 0).

The example I found was a recent request for a way to safely remove an  
element from a slice.  I recommended to use assumeSafeAppend after  
removing the element.  The question was then asked, what if the slice is  
part of a larger array?  Calling assumeSafeAppend effectively invalidates  
all elements after the slice.  So you want to know positively that a slice  
ends at the end of the array, and only call assumeSafeAppend if it's safe  
to do so.

Maybe there is a better way to do this, but it turns out to be a useful  
inconsistency.  Keep in mind that calling capacity is not a simple quick  
lookup, so you don't want to have to call it multiple times.

Slices of non-heap arrays also return a capacity of 0.

-Steve