Unique list of items using LINQ
I've been using LINQ for a while now, but seem to be stuck on something with regards to Unique items, I have the folling list:
List<Stock> stock = new List<Stock>();
This has the following Properties: string ID , string Type, string Description, example:
public class Stock
{
public string ID { get; set; }
public string Type { get; set; }
public string Description { get; set; }
}
I want to have a LINQ query that will group the items in Stock by their Type and return this as a new List of Stock (it has to be the same type as the original Stock List).
Example Data:
ID Type Description
----------------------------------------------
1 Kitchen Appliance Dishwasher
2 Living Room Television
3 Kitchen Appliance Washing Machine
4 开发者_如何学JAVA Kitchen Appliance Fridge
...
My Linq query wants to be able to return all the Kitchen Appliances for Example.
So I would pass this as a "type" into the query and it would return the items 1, 3 and 4 from this example list.This list returned must also be of type: List<Stock>
.
Essentially I want a list of the unique items by type, kind of like an SQL Distinct query, how do I achieve this in LINQ?
Alternative solutions are fine but must be Silverlight / C# client code only.Just another clarification is that I also may not provide the parameter "Kitchen Appliance" and may just want the unique ones, for example It would return Kitchen Appliance and Living Room once each only to kind of like a category no matter how many of that Type there are.
Flexible approach
Use GroupBy
and ToDictionary
to create a dictionary of List<Stock>
values keyed on the Type
property:
var appliancesByType = stock
.GroupBy(item => item.Type)
.ToDictionary(grp => grp.Key, grp => grp.ToList());
Then you can access the types themselves as well as a list for any given type quite easily:
// List of unique type names only
List<string> stockTypes = appliancesByType.Keys.ToList();
// Or: list of one stock item per type
List<Stock> exampleStocks = appliancesByType
.Select(kvp => kvp.Value[0])
.ToList();
// List of stock items for a given type
List<Stock> kitchenAppliances = appliancesByType["Kitchen Appliance"];
This approach really takes care of all your needs, as I see it. But for some other options, see below.
Alternate (quick & dirty) approach
You can always just use Where
to get the items of the type you want, then ToList
to put these items in a new List<Stock>
:
List<Stock> kitchenAppliances = stock
.Where(item => item.Type == "Kitchen Appliance")
.ToList();
In response to this last part:
Just another clarification is that I also may not provide the parameter "Kitchen Appliance" and may just want the unique ones, for example It would return Kitchen Appliance and Living Room once each only to kind of like a category no matter how many of that Type there are.
Here, you seem to be wanting something completely different: basically the behavior provided by Distinct
. For this functionality, you could essentially go with Soonts's answer (optionally, returning an IEnumerable<tKey>
instead of IEnumerable<tSource>
), or you could just leverage Distinct
in combination with Select
to avoid the need to implement an IEqualityComparer<Stock>
(see below).
Update
In response to your clarification, here's my recommendation: two methods, one for each purpose (Single Responsibility Principle):
// This will return a list of all Stock objects having the specified Type
static List<Stock> GetItemsForType(string type)
{
return stock
.Where(item => item.Type == type)
.ToList();
}
// This will return a list of the names of all Type values (no duplicates)
static List<string> GetStockTypes()
{
return stock
.Select(item => item.Type)
.Distinct()
.ToList();
}
Sounds like it's a simple Where clause needed.
List<Stock> kitchen= stock.Where(s=>s.Type=="Kitchen Appliance")
.OrderBy(s=>s.Description).ToList();
If you wanted strictly the Types
contained in the source list:
string[] typesFound = stock.Select(s=>s.Type).Distinct().ToArray();
static class EnumerableEx
{
// Selectively skip some elements from the input sequence based on their key uniqueness.
// If several elements share the same key value, skip all but the 1-st one.
public static IEnumerable<tSource> uniqueBy<tSource, tKey>( this IEnumerable<tSource> src, Func<tSource, tKey> keySelecta )
{
HashSet<tKey> res = new HashSet<tKey>();
foreach( tSource e in src )
{
tKey k = keySelecta( e );
if( res.Contains( k ) )
continue;
res.Add( k );
yield return e;
}
}
}
// Then later in the code
List<Stock> res = src.uniqueBy( elt => elt.Type ).ToList()
var kitchenAppliances = stocks.Where(stock => stock.Type == "Kitchen Appliance");
I probably don't understand the question, but this sounds to me like a pretty simple Linq-query:
List<Stock> stock = new List<Stock>();
... populate your list as per your example
List<Stock> kitchenAppliances =
(from obj in stock
where obj.Type == "Kitchen Appliance"
select obj).ToList();
or if you prefer the extension-method syntax:
List<Stock> kitchenAppliances =
stock.Where(obj => obj.Type == "Kitchen Appliance").ToList();
What I don't understand is your usage of distinct and unique in this context. The objects are already unique, and it seems to me that you want a basic query "give me all kitchen appliances".
You could also do something along these lines, in this case I am taking lucene results then pulling off umbraco nodes to an anonymous type
var datasource = (from n in SearchResults
select new
{
PageLink = uQuery.GetNode(n.Id).NiceUrl,
uQuery.GetNode(n.Id).Name,
Date = uQuery.GetNode(n.Id).GetProperty("startDate"),
IntroText = string.IsNullOrEmpty(uQuery.GetNode(n.Id).GetProperty("introText").Value) ? string.Empty : uQuery.GetNode(n.Id).GetProperty("introText").Value,
Image = ImageCheck(uQuery.GetNode(n.Id).GetProperty("smallImage").Value)
}).Distinct().ToList();
精彩评论