Introducing HashSet<T> [Kim Hamilton]

re: Introducing HashSet [Kim Hamilton]

Wesner Moise — Thu, 09 Nov 2006 22:04:35 GMT

FXCop rule prohibits naming a type after a language keyword. "set" is also used in C#, but also in VB6, though it is either gone or optional in VB.NET.

It would be nice to support operator overloading for sets.

re: Introducing HashSet [Kim Hamilton]

Alan Dean — Thu, 09 Nov 2006 22:04:35 GMT

Because "set" is a reserved contextual keyword

http://msdn2.microsoft.com/en-gb/library/x53a06bb.aspx

re: Introducing HashSet [Kim Hamilton]

Tom — Thu, 09 Nov 2006 22:05:23 GMT

I would strongly prefer the name Set instead of HashSet. As you state, HashSet refers too strongly to the implementation. Users of the class don't really care about the implementation, we just care about the functionality and performance characteristics.

re: Introducing HashSet [Kim Hamilton]

TheMuuj — Thu, 09 Nov 2006 23:34:58 GMT

Even though you no longer have to use Set when assigning references in VB, it is still a keyword. It is used in declaring property accessors:

Property Text() As String

Get

Return _text

End

Set(ByVal value As String)

_text = value

End Set

End Property

I'm not sure if Set is a true keyword or a contextual keyword, but either way there's potential for confusion.

re: Introducing HashSet [Kim Hamilton]

Timothy Fries — Fri, 10 Nov 2006 00:19:47 GMT

PowerCollections uses Set<T>... I can't say I've ever had a problem with keyword collision with it.

Introducing HashSet

DotNetKicks.com — Fri, 10 Nov 2006 00:57:49 GMT

You've been kicked (a good thing) - Trackback from DotNetKicks.com

System.Collections.Generic."Hash"Set?

Kael's WebLog — Fri, 10 Nov 2006 01:48:10 GMT

HashSet is a new generic collection that has been added to the System.Collections.Generic namespace.

re: Introducing HashSet [Kim Hamilton]

dono — Fri, 10 Nov 2006 07:30:16 GMT

> Any guesses as to why we didn’t

> go with the name Set?

I assume you mean Set<T>.

That makes the VB argument mostly irrelevant.

I am _really_ glad that you did not call it Set<T>. I eagerly await the day that ISet<T> is introduced along with other set types, TreeSet<T>, OrderSet<T> etc.

However, I think it is a mistake to introduce HashSet<T> now without an ISet<T>. A lot of code will need to be rewritten later (such as updating signatures) because of this choice.

I hope the type is unsealed. Lets say I want to design a TreeSet<T>. It really does not make sense for me to extend HashSet<T> to do this. Yet without an ISet<T>, there really aren't any other options. I could ignore HashSet<T> entirely and write everything from scratch. However, that does not fit well into the _framework_.

Please reconsider this issue.

re: Introducing HashSet [Kim Hamilton]

Doug — Fri, 10 Nov 2006 12:58:54 GMT

A Set type is great to have. I've gotten by with Dictionary<KeyType, object> so far (leaving the value as null), but this is even better.

Another type that I use often is Dictionary<KeyType, List<ValueType>> - a key maps to a set of zero or more values. I would love to see something like this in the BCL.

re: Introducing HashSet [Kim Hamilton]

Israel Aece — Fri, 10 Nov 2006 14:46:30 GMT

Hello Kim,

When and where this collection will be available?

System.Collections.Generic.HashSet

Israel Aéce — Fri, 10 Nov 2006 14:50:10 GMT

The BCL Team work in a new type collection called (temporary) HashSet that is a collection containing

Then a HashList ?

Sébastien Ros — Fri, 10 Nov 2006 15:18:00 GMT

I think this is a great idea, as it is really a missing feature currently. Thus why don't you go a step beyond by creating an HashList which would be o(1) (or did I miss it)

re: Introducing HashSet [Kim Hamilton]

Kim Hamilton — Fri, 10 Nov 2006 21:38:44 GMT

Hi Isreal,

It's in our latest Orcas CTP, which you can download here:

http://www.microsoft.com/downloads/details.aspx?FamilyId=C09B5A2D-EB6A-44B6-8BBD-3764A2FDA9CE&displaylang=en

It's in System.Core.dll, in the System.Collections.Generic namespace.

Thanks,

Kim

re: Then a HashList?

Kim Hamilton — Mon, 13 Nov 2006 10:03:19 GMT

Hi Sebastien,

I'm interpreting HashList as a collection in which you can perform both index-based lookups and hash-based lookups. In other words, a generalization of NameValueCollection (which accepts only Strings).

We haven't gotten a lot of requests for this type of collection, and as with NVC, it would have some additional overhead. So we'd still recommend picking one or the other (List or HashSet) if you don't actually require both types of lookups. See the related blog on NVC overhead here:

http://blogs.msdn.com/bclteam/archive/2006/09/05/741660.aspx

Thanks,

Kim

re: Introducing HashSet [Kim Hamilton]

Kim Hamilton — Mon, 13 Nov 2006 10:15:27 GMT

Hi Dono,

>> A lot of code will need to be rewritten later (such as updating signatures) because of this choice.

We had a lot of discussion on this point during the design of HashSet (which Justin summarized); we'll most likely follow up soon with some more details.

Thanks,

Kim

re: Introducing HashSet [Kim Hamilton]

Sébastien Ros — Sat, 18 Nov 2006 12:03:00 GMT

Kim,

By HashList I was referring to an IList<T> implementation which would be O(1) for Add, Remove and Contains. Isn't it one of the main matters of HashSet ?

re: Introducing HashSet [Kim Hamilton]

diegov — Wed, 22 Nov 2006 19:08:13 GMT

What I don't get is the reason you change the semantics of Add so lightly.

I would prefer that you were more explicit. My suggestion:

Create a public bool TryAdd() and then a void Add() that throws on duplicates as usual. Of course, you can implement Add() as a wrapper around TryAdd() if convinient.

re: Introducing HashSet [Kim Hamilton]

Kim Hamilton — Wed, 22 Nov 2006 22:03:27 GMT

Hi Diegov,

Earlier versions of our design used Add/TryAdd, but we removed it to be consistent with the set interpretation of add. Even though a set is maintained such that it doesn't contain duplicate items, attempting to add a duplicate isn't exceptional. We provided the Add that returns bool for cases where the user may want different control flow based on whether the item was already present in the set, but this shouldn't be achieved through exceptions (with their associated perf cost, etc).

So in fact, this isn't a change of semantics of Add. You can think of it this way -- if the items are equal according to the equality comparer, then it's an implementation choice whether the original item is replaced by the new duplicate item. After Add is finished, the item is present in the set.

I should also mention that we came to this decision only after much debate. Using Add/TryAdd was a common suggestion, but more than 2/3rds of people reviewing the API thought that having Add throw on duplicates would be surprising for a set.

Some readers may be wondering why IDictionary throws on duplicate adds. That's because it takes both a key and a value, and although the key may be the same, the values may differ.

Thanks,

Kim

re: Introducing HashSet [Kim Hamilton]

Kim Hamilton — Wed, 22 Nov 2006 22:13:01 GMT

Hi Sebastien,

>> Isn't it one of the main matters of HashSet ?

Not exactly; HashSet is unordered and doesn't contain duplicates, but an IList is ordered and may contain duplicates.

Thanks,

Kim

re: Introducing HashSet [Kim Hamilton]

diegov — Wed, 22 Nov 2006 23:30:17 GMT

Well, I guess the majority rules :) Anyway, the use of IEqualityComparer<T>.GetHashCode() gives the slight feeling that this is also a dictionary, only the key is function of the value. I guess then, it is not completely possible to abstract of the fact that HashSet is implemented as with a hash. Which is another reason not to call it Set :)

System.Collections.Generic.HashSet

Daniel Moth — Thu, 08 Feb 2007 02:50:37 GMT

System.Collections.Generic.HashSet

.Net Framework 3.5 - System.Core.dll

Mike Chaliy's Blog — Thu, 15 Feb 2007 17:41:48 GMT

Незабаром в нас буде нова версія .Net Framework 3.5, яка схоже, як і 3.0, буде спиратися на базовий .Net

(Hash)Set in the 3.5 BCL

James Manning's blog — Tue, 12 Jun 2007 16:47:22 GMT

Since it's coming up on 2 years since I complained that the BCL didn't have Set<T> I figured I

New .NET 3.5 core stuff

Patrick Smacchia [MVP C#] — Wed, 08 Aug 2007 22:22:05 GMT

While installing VisualStudio 2008 Beta2 I was surprised that the NET framework 2.0 installation got

Generic function to removeDuplicates from Generic List

Michael Freidgeim's Blog — Wed, 12 Sep 2007 06:42:49 GMT

Generic function to removeDuplicates from Generic List

What’s new in the .Net Framework 3.5 (Orcas)

Jack Gudenkauf (JackG) WebLog — Thu, 13 Sep 2007 21:32:04 GMT

Visual Studio (VS) 2008 and .Net 3.5, code named “Orcas”, will be released in the near future. This release

Nachlese: TechTalk - Visual Studio 2008 & .NET 3.5

Dariusz quatscht — Tue, 23 Oct 2007 02:30:51 GMT

Mein TechTalk ist nun zu Ende. Meine letzte Station heute in Berlin war Lustig und Amüsant, ich hoffe

System.Core.dll of .NET Framework 3.5

Pedram Rezaei's Ramblings — Wed, 07 Nov 2007 14:25:46 GMT

I recently wrote an MSDN Flash article on the hidden gems of System.Core.dll of .NET Framework 3.5 which

System.Core.dll of .NET Framework 3.5

Noticias externas — Wed, 07 Nov 2007 15:10:16 GMT

I recently wrote an MSDN Flash article on the hidden gems of System.Core.dll of .NET Framework 3.5 which

.NET Framework 3.5 Now Available! [Justin Van Patten]

BCL Team Blog — Tue, 20 Nov 2007 03:42:05 GMT

.NET Framework 3.5 and Visual Studio 2008 have officially shipped! Soma has the announcement on his blog

The New Generic Kid on the Block - HashSet<T>

Rik Robinson's Blog — Tue, 22 Apr 2008 22:03:43 GMT

In mathematics, a Set is typically thought of as a collection of distinct objects that is usually defined