1) Yes. Value types have an inherently lower cost. They are cheaper to create/destroy, cheaper to maintain, and (most of the time) cheaper to use.
2) A Hashtable isn't really suited to sorting or filtering. It's actually better if the keys in a hashtable are as unsorted as possible, because this allows optimal searching of its values. The quickest way to get your subset would be from a list pre-sorted on the value of the 'foo' field, then you'd only have to enumerate the matching objects. If you need a list that remains sorted all the time, try SortedList, although it can slow things done quite a bit. Array.Sort() will sort any array based on an IComparer, or on the objects' own ICompaable implementation. Jim > -----Original Message----- > From: Greg Gates [mailto:[EMAIL PROTECTED]] > Sent: 23 May 2002 16:24 > To: [EMAIL PROTECTED] > Subject: [DOTNET] Hashtables and structs > > > Hello everyone: > > I have two basic questions regarding hashtables and structs. > > 1) Is there any point in using a struct if the struct > contains a string, > given that a string is a reference type? > > 2) What is the most efficent way to sort and filter a > hashtable in .NET. > For example, if I have a hashtable that contains the > following key/values: > > A,object1 > B,object2 > C,object3 > > object1 has a foo field with a value of 4 > object2 has a foo field with a value of 1 > object3 has a foo field with a value of 2 > > What is the quickest way to retrieve an ordered subset of the > items where > the foo field is less than or equal to 3? > > thanks, Greg > > You can read messages from the DOTNET archive, unsubscribe > from DOTNET, or > subscribe to other DevelopMentor lists at http://discuss.develop.com. > You can read messages from the DOTNET archive, unsubscribe from DOTNET, or subscribe to other DevelopMentor lists at http://discuss.develop.com.
