Does anyone know if there is a good equivalent to Java's Set
collection in C#? I know that you can somewhat mimic a set using a Dictionary
or a HashTable
by populating but ignoring the values, but that's not a very elegant way.
If you're using .NET 3.5, you can use HashSet<T>
. It's true that .NET doesn't cater for sets as well as Java does though.
The Wintellect PowerCollections may help too.
Try HashSet:
The HashSet(Of T) class provides high-performance set operations. A set is a collection that contains no duplicate elements, and whose elements are in no particular order... The capacity of a HashSet(Of T) object is the number of elements that the object can hold. A HashSet(Of T) object's capacity automatically increases as elements are added to the object. The HashSet(Of T) class is based on the model of mathematical sets and provides high-performance set operations similar to accessing the keys of the Dictionary(Of TKey, TValue) or Hashtable collections. In simple terms, the HashSet(Of T) class can be thought of as a Dictionary(Of TKey, TValue) collection without values. A HashSet(Of T) collection is not sorted and cannot contain duplicate elements...
If you're using .NET 4.0 or later:
In the case where you need sorting then use SortedSet<T>
. Otherwise if you don't, then use HashSet<T>
since it's O(1)
for search and manipulate operations. Whereas SortedSet<T>
is O(log n)
for search and manipulate operations.
I use Iesi.Collections http://www.codeproject.com/KB/recipes/sets.aspx
It's used in lot of OSS projects, I first came across it in NHibernate
I use a wrapper around a Dictionary<T, object>
, storing nulls in the values. This gives O(1) add, lookup and remove on the keys, and to all intents and purposes acts like a set.
Have a look at PowerCollections over at CodePlex. Apart from Set and OrderedSet it has a few other usefull collection types such as Deque, MultiDictionary, Bag, OrderedBag, OrderedDictionary and OrderedMultiDictionary.
For more collections, there is also the C5 Generic Collection Library.
I know this is an old thread, but I was running into the same problem and found HashSet to be very unreliable because given the same seed, GetHashCode() returned different codes. So, I thought, why not just use a List and hide the add method like this
public class UniqueList<T> : List<T>
{
public new void Add(T obj)
{
if(!Contains(obj))
{
base.Add(obj);
}
}
}
Because List uses the Equals method solely to determine equality, you can define the Equals method on your T type to make sure you get the desired results.
List.Contains
is of O(n)
complexity which means that your Add
method now becomes O(n)
complexity as well. Assuming the inner collection doesn't need to be resized, Add
for both List
and HashMap
should be of O(1)
complexity. TLDR: This will work, but it's hacky and less efficient.
Success story sharing