IEnumerable.Select() when attribute is known only at runtime - c#

Say I have a data class like this and a list of its objects:
public class DataSet
{
public int A { get; set; }
public string B { get; set; }
public double C { get; set; }
}
var data = new List<DataSet>
{
new DataSet() { A = 1, B = "One", C = 1.1 },
new DataSet() { A = 2, B = "Two", C = 2.2 },
new DataSet() { A = 3, B = "Three", C = 3.3 }
};
I would like to do a Select() on the list, based on different properties. For example, if I need a list of property A, I could do this easily:
var listA = data.Select(x => x.A).ToList();
All good so far.
But in my program, I need to do the above, only, I wouldn't know whether I need a list of A or B or C until runtime. This 'knowledge' of what to select is stored in a list of strings, and I need to iterate it and extract only the appropriate lists. Something like this:
// GetKeys() will return the keys that I need to extract.
// So at one time keyList could have "A" and "B", another time "B" and "C" etc.
List<string> keyList = GetKeys();
foreach (var key in keyList)
{
// What do I do here?
data.Select(x =>???).ToList();
}
Is this possible at all? I'm fine with even a non-LINQ solution, if it achieves my goal.
EDIT:
Clarifying the requirement.
The end result I want is a separate list based on each 'key' mentioned above. So, something like
List<List<object>>
The count in outer list would be the count of keyList.
The inner list would have as many items as in DataSet.

This would probably not be the most efficient solution, but you could use Reflection for a fully dynamic solution:
private static List<List<object>> SelectDynamicData<T>(IEnumerable<T> data, List<string> properties)
{
// get the properties only once per call
// this isn't fast
var wantedProperties = typeof(T)
.GetProperties()
.Where(x => properties.Contains(x.Name))
.ToArray();
var result = new Dictionary<string, List<object>>();
foreach (var item in data)
{
foreach (var wantedProperty in wantedProperties)
{
if (!result.ContainsKey(wantedProperty.Name))
{
result.Add(wantedProperty.Name, new List<object>());
}
result[wantedProperty.Name].Add(wantedProperty.GetValue(item));
}
}
return result.Select(x => x.Value).ToList();
}
And, of course, you'd need to do a double foreach or a LINQ query to print that. For example:
var data = new List<DataSet>
{
new DataSet() { A = 1, B = "One", C = 1.1 },
new DataSet() { A = 2, B = "Two", C = 2.2 },
new DataSet() { A = 3, B = "Three", C = 3.3 }
};
var selectedData = SelectDynamicData(data, new List<string> { "A", "C" });
foreach (var list in selectedData)
{
foreach (object item in list)
{
Console.Write(item + ", ");
}
Console.WriteLine();
}

Using Creating Expression Trees by Using the API you can build an expression tree to represent the linq query you were hard coding in order to make it more dynamic.
Expression<Func<TModel, object>> GetPropertyExpression<TModel>(string propertyName) {
// Manually build the expression tree for
// the lambda expression v => v.PropertyName.
// (TModel v) =>
var parameter = Expression.Parameter(typeof(TModel), "v");
// (TModel v) => v.PropertyName
var property = Expression.Property(parameter, propertyName);
// (TModel v) => (object) v.PropertyName
var cast = Expression.Convert(property, typeof(object));
var expression = Expression.Lambda<Func<TModel, object>>(cast, parameter);
return expression;
}
Review the comments to understand the building of the expression tree.
This now can be used with the data to extract the desired result.
Following similar to what was provided in another answer it would be simplified to
List<List<object>> SelectDynamicData<T>(IEnumerable<T> data, List<string> properties) {
return properties
.Select(_ => data.Select(GetPropertyExpression<T>(_).Compile()).ToList())
.ToList();
}
Both methods are displayed in the following example
[TestMethod]
public void TestMethod1() {
var data = new List<DataSet>
{
new DataSet() { A = 1, B = "One", C = 1.1 },
new DataSet() { A = 2, B = "Two", C = 2.2 },
new DataSet() { A = 3, B = "Three", C = 3.3 }
};
var propertyKnownAtRuntime = "A";
var expression = GetPropertyExpression<DataSet>(propertyKnownAtRuntime);
var listA = data.Select(expression.Compile()).ToList();
//Produces
// { 1, 2, 3}
var listAC = SelectDynamicData(data, new List<string> { "A", "C" });
//Produces
//{
// { 1, 2, 3},
// { 1.1, 2.2, 3.3 }
//}
}

You can use reflection, for example
string key = "A";
var query = data.Select(x =>
{
var prop = x.GetType().GetProperty(key); //NOTE: if key does not exist this will return null
return prop.GetValue(x);
});
foreach (var value in query)
{
Console.WriteLine(value); //will print 1, 2, 3
}

Related

Using Linq Select to create a list that contains more items than the original list

I have a list of items. For example (although the list could be any length):
var inputList = new List<Input1>()
{
new Input1() { Test = "a" },
new Input1() { Test = "b" }
};
What I want to do is create a new list of:
a1, a2, b8, b9
That is the value of Test (i.e. a) with a suffix based on the value of Test.
In that order. Obviously, this is a minimum workable example, not the actual problem. So I'd like to use something like the .Select to split the data - something like this:
var outputList = inputList.Select(x =>
{
if (x.Test == "a")
{
return new Input1() { Test = "a1" };
//return new Input1() { Test = "a2" };
}
else if (x.Test == "b")
{
return new Input1() { Test = "b8" };
//return new Input1() { Test = "b9" };
}
else
{
return x;
}
});
Input1 for completeness:
class Input1
{
public string Test { get; set; }
}
That is, to return a list that contains items that were not in the original list.
I realise I can use a foreach, but I'm interested if there's a better / more concise way.
Suppose you have a method that transforms your single input into multiple inputs:
public static Input1[] Transform(Input1 x)
{
if (x.Test == "a") return new[] {new Input1("a1"), new Input1("a2")};
if (x.Test == "b") return new[] {new Input1("b8"), new Input1("b9")};
return new[] {x};
}
(This is just from your toy example - I guess you actually need a transformation that is more meaningful.)
Then you can just use SelectMany to get your desired result in the correct order:
inputList
.SelectMany(Transform);
If you're using C# 8.0 or above, you may use switch expression as follows:
var outputList =
inputList.SelectMany(x => x.Test switch
{
"a" => new[] { new Input1() { Test = "a1" }, new Input1() { Test = "a2" } },
"b" => new[] { new Input1() { Test = "b8" }, new Input1() { Test = "b9" } },
_ => new[] { x }
})
.ToList();

Lambda expression to loop through two concurrent dictionaries

I am trying to loop through two concurrent dictionaries like the code below, however I want to use a lambda expression instead
foreach (var s in sb_eventdata)
{
foreach (var f in final_data)
{
if (s.Value.Car.Equals(f.Value.Car))
{
Console.Writeline("Found!");
}
}
}
var values = sb_eventdata.Where(k => k.Value.Hometeam.Contains( ???? );
I'm really not sure what to pass into contains, I assume another lambda expression but what?
The closest linq expression to your loops would be:
var sb_eventdata = new Dictionary<string, string>{ {"a", "a"}, {"b", "b"}};
var final_data = new Dictionary<string, string>{{"a", "a"}, {"b", "b"}, {"c","c"}};
var result =
// first loop
sb_eventdata.Select(s =>
// second loop
final_data.Where(f => s.Value.Equals(f.Value)))
// flatten results (returns results from the first dictionary)
.SelectMany(x => x);
You can use a linq Intersect function to find like items in a list.
Then display all like items.
var foo = sb_eventdata.Select(o => o.Value.Car).Intersect(final_data.Select(o => o.Value.Car));
foreach (var item in foo)
{
Console.Writeline("Found!");
}
I think your friend is the Join() method.
In "LinqPad style":
void Main()
{
var a = new[] {
new Car("Opel",200),
new Car("Volkswagen",300),
new Car("Audi", 500)
};
var b = new[] {
new Car("Peugeot", 180),
new Car("Seat", 300),
new Car("Volvo", 480)
};
var c = a.Join(b, ak => ak.Value, bk => bk.Value, (ak,bk) => new {A=ak.Name,B=bk.Name,ak.Value});
c.Dump();
}
// Define other methods and classes here
class Car {
public string Name;
public int Value;
public Car (string name, int value) {
Name = name;
Value = value;
}
}
If you just want to know if both dictionary share at least one value, you can use Any:
if(sb_eventdata.Any(s =>
final_data.Any(f => s.Value.Car.Equals(f.Value.Car))))
Console.WriteLine("Found!");
or with Contains:
if(sb_eventdata.Any(s => final_data.ContainsValue(s.Value)))
Console.WriteLine("Found!");
and if you want to count how many of sb_eventdata are in final_data:
sb_eventdata.Where(s => final_data.ContainsValue(s.Value)).Count();

LINQ to Entities .Contains for a tuple (2 foreign keys)

Actually, a colleague of mine asked me this question and I haven't been able to come up with an answer. Here goes.
Given an entity with 2 foreign keys, say
public class MyTable
{
public int Key1 { get; set; }
public int Key2 { get; set; }
}
and 2 lists
public ICollection List1 => new List<int> { 1, 2, 3 };
public ICollection List2 => new List<int> { 4, 5, 6 };
he needs to query for all records where Key1 matches the value from List1 and Key2 matches the value from List2, e.g.
Key1 == 1 && Key2 == 4
that is, he wants to check for any given tuple from List1 and List2, (1, 4), (2, 5) and (3, 6).
Is there a straightforward way in EF to do this?
You can make a for loop to capture some local variable (in each loop) in the Where and use Concat (or Union - maybe with worse performance) to sum up all the result like this:
IQueryable<MyTable> q = null;
//suppose 2 lists have the same count
for(var i = 0; i < list1.Count; i++){
var k1 = list1[i];
var k2 = list2[i];
var s = context.myTables.Where(e => e.Key1 == k1 && e.Key2 == k2);
q = q == null ? s : q.Concat(s);
}
//materialize the result
var result = q.ToList();
NOTE: we can use Concat here because each sub-result should be unique (based on searching the keys). It surely has better performance than Union (ensuring uniqueness while we already know the sub-results are all unique beforehand - so it's unnecessary).
If you have a list of int (just integral numeric), you can also pair the keys into underscore separated string and use Contains normally like this:
var stringKeys = list1.Select((e,i) => e + "_" + list2[i]).ToList();
var result = context.myTables.Where(e => stringKeys.Contains(e.Key1 + "_" + e.Key2))
.ToList();
Building an Expression tree is also another approach but it's more complicated while I'm not sure if it has better performance.
Try this:
static IQueryable<TSource> WhereIn(this Table<TSource> table, List<object[]> list) where TSource : class
{
var query = table.AsQueryable();
foreach (object[] item in list)
{
Expression<Func<TSource, bool>> expr = WhereInExpression(item);
query = query.Where(expr);
}
return query;
}
static Expression<Func<TSource, bool>> WhereInExpression<TSource>(object[] item)
{
ParameterExpression parameterItem = Expression.Parameter(typeof(TSource), "expr");
BinaryExpression filter1 = Expression.Equal(LambdaExpression.PropertyOrField(parameterItem, "Key1"),
Expression.Constant(item[0]));
BinaryExpression filter2 = Expression.Equal(LambdaExpression.PropertyOrField(parameterItem, "Key2"),
Expression.Constant(item[1]));
BinaryExpression filter = LambdaExpression.And(filter1, filter2);
var expr = Expression.Lambda<Func<TSource, bool>>(filter, new ParameterExpression[] { parameterItem });
expr.Compile();
return expr;
}
Usage:
List<object[]> list = new List<object[]>() { new object[] { 1, 100 }, new object[] { 1, 101 }, new object[] { 2, 100 } };
var result = db.MyTable.WhereIn<MyTable>(list);
Confirmed to work with Entity framework
var tuples = List1.Cast<int>().Zip(List2.Cast<int>(), (l, r) => new { l, r });
var results = Orders.Where(o =>
tuples.Contains(new { l = (int)o.KeyOne, r = (int)o.KeyTwo })
);
Or simpler, if you define your lists as ICollection<int> or IList<int> (etc...):
var tuples = List1.Zip(List2, (l, r) => new { l, r });
var results = Orders.Where(o =>
tuples.Contains(new { l = (int)o.KeyOne, r = (int)o.KeyTwo })
);
Fiddle here : https://dotnetfiddle.net/YyyZBY
var List_1 = new List<int> { 1, 2, 3 };
var List_2 = new List<int> { 4, 5, 6 };
var TargetList = new List<MyTable>();
var index1=0;
var List1 = List_1.Select(x=>new { ind=index1++,value=x });
var index2=0;
var List2 = List_2.Select(x=>new { ind=index2++,value=x });
var values = from l1 in List1
join l2 in List2 on l1.ind equals l2.ind
select new {value1=l1.value,value2=l2.value };
var result = TargetList.Where(x=>values.Any(y=>y.value1==x.Key1&&y.value2==x.Key2));

unequal size lists to merge

I have searched without success to a similar situation as follows.
I have two lists, list A and list B.
List A is composed of 10 objects created from ClassA which contains only strings.
List B is composed of 100 objects created from ClassB which only contains decimals.
List A is the header information.
List B is the data information.
The relationship between the two lists is:
Row 1 of list A corresponds to rows 1-10 of list B.
Row 2 of list A corresponds to rows 11-20 of list B.
Row 3 of list A corresponds to rows 21-30 of list B.
etc.........
How can I combine these two lists so that when I display them on the console the user will see a header row followed immediately by the corresponding 10 data rows.
I apologize if this has been answered before.
Ok, that should work. Let me know in case I got anything wrong.
List<ClassA> listA = GetListA()// ...
List<ClassB> listB = GetListA()// ...
if(listB.Count % listA.Count != 0)
throw new Exception("Unable to match listA to listB");
var datasPerHeader = listB.Count / listA.Count;
for(int i = 0; i < listA.Count;i++)
{
ClassA header = listA[i];
IEnumerable<ListB> datas = listB.Skip(datasPerHeader*i).Take(datasPerHeader);
Console.WriteLine(header.ToString());
foreach(var data in datas)
{
Console.WriteLine("\t{0}", data.ToString());
}
}
Here is some code that should fulfill your request - I am going to find a link for the partition extension as I can't find it in my code anymore:
void Main()
{
List<string> strings = Enumerable.Range(1,10).Select(x=>x.ToString()).ToList();
List<decimal> decimals = Enumerable.Range(1,100).Select(x=>(Decimal)x).ToList();
var detailsRows = decimals.Partition(10)
.Select((details, row) => new {HeaderRow = row, DetailsRows = details});
var headerRows = strings.Select((header, row) => new {HeaderRow = row, Header = header});
var final = headerRows.Join(detailsRows, x=>x.HeaderRow, x=>x.HeaderRow, (header, details) => new {Header = header.Header, Details = details.DetailsRows});
}
public static class Extensions
{
public static IEnumerable<List<T>> Partition<T>(this IEnumerable<T> source, Int32 size)
{
for (int i = 0; i < Math.Ceiling(source.Count() / (Double)size); i++)
yield return new List<T>(source.Skip(size * i).Take(size));
}
}
That Partition method is the one that does the grunt work...
And here is the link to the article - LINK
EDIT 2
Here is better code for the Main() method... Rushed to answer and forgot brain:
void Main()
{
List<string> strings = Enumerable.Range(1,10).Select(x=>x.ToString()).ToList();
List<decimal> decimals = Enumerable.Range(1,100).Select(x=>(Decimal)x).ToList();
var detailsRows = decimals.Partition(10);
var headerRows = strings; //just renamed for clarity from other code
var final = headerRows.Zip(detailsRows, (header, details) => new {Header = header, Details = details});
}
This should be pretty straight forward unless I'm missing something.
var grouped = ListA.Select((value, index) =>
new {
ListAItem = value,
ListBItems = ListB.Skip(index * 10).Take(10)
})
.ToList();
Returns back an anonymous type you can loop through.
foreach (var group in grouped)
{
Console.WriteLine("List A: {0}", group.Name);
foreach (var listBItem in group.ListBItems)
{
Console.WriteLine("List B: {0}", listBItem.Name);
{
}
The easiest way may be something like this:
var listA = new List<string>() { "A", "B", "C", ... }
var listB = new List<decimal>() { 1m, 2m, 3m, ... }
double ratio = ((double)listA.Count) / listB.Count;
var results =
from i in Enumerable.Range(0, listB.Count)
select new { A = listA[(int)Math.Truncate(i * ratio)], B = listB[i] };
Or in fluent syntax:
double ratio = ((double)listA.Count) / listB.Count;
var results = Enumerable.Range(0, listB.Count)
.Select(i => new { A = listA[(int)Math.Truncate(i * ratio)], B = listB[i] });
Of course if you know you will always have 10 items in listB for each item in listA, you can simplify this to:
var results =
from i in Enumerable.Range(0, listB.Count)
select new { A = listA[i / 10], B = listB[i] };
Or in fluent syntax:
var results = Enumerable.Range(0, listB.Count)
.Select(i => new { A = listA[i / 10], B = listB[i] });
This will return a result set like
{ { "A", 1 },
{ "A", 2 },
{ "A", 3 }
..,
{ "A", 10 },
{ "B", 11 },
{ "B", 12 },
{ "B", 13 },
...
{ "B", 20 },
{ "C", 21 },
...
{ "J", 100 }
}

Combining lists in linq

in linq, is it possible to combine many lists (of the same type), such that two lists,
list 1 = {a,b,c} and list 2 = {x,y,z}
turns into {[1,a] , [1,b] , [1,c] , [2,x] , [2,y] , [2,z] }
where [] represents a pair containing a "list identifier"
The problem is from having decks of arbitrary cards, where each deck is a list in a collection of lists.
I'm trying to create a query such that I can select only cards in a certain deck, or cards similar to 2 or more decks.
This is probably a duplicate question, but I don't know how to search for the question further then I already have.
List<List<int>> lists;
var combined = lists.Select((l, idx) => new { List = l, Idx = idx })
.SelectMany(p => p.List.Select(i => Tuple.Create(p.Idx + 1, i)));
var list1 = new List<string>() {a,b,c};
var list2 = new List<string>() {x,y,z};
var combined = list1.Select(x => new { id = 1, v = x }).Concat(list2.Select(x => new { id = 2, v = x }));
Normally I'd suggest Enumerable.Zip for combining multiple lists, however you seem to actually want to concatenate multiple lists with a list counter.
public IEnumerable<Tuple<int,T>> Combine<T>(params IEnumerable<T>[] lists) {
return lists.Select((x,i) => x.Select(y => Tuple.Create(i+1,y))).SelectMany (l =>l);
}
UPDATE
Completely missed that SelectMany has the index option so the above code can be written as
public IEnumerable<Tuple<int,T>> Combine<T>(params IEnumerable<T>[] lists) {
return lists.SelectMany((x,i) => x.Select(y => Tuple.Create(i+1,y)));
}
Then you can do
var list1 = new List<string> { "a", "b", "c" };
var list2 = new List<string> { "x", "y", "z" };
var combined = Combine(list1,list2);
Combined will be enumerable of tuples, with Item1 being the list index identifier (starting at 1) and Item2 being the value.
This method will handle multiple lists so you could just as easily call it with:
var list3 = new List<string> { "f", "g" };
var combined = Combine(list1,list2,list3);
You can merge the lists like:
var first = new List<string> {"a","b","c"};
var second = new List<string> {"x","y","z"};
var merged = first.Select(item => new { ListIndex = 1, Value = item}).ToList();
merged.AddRange(second.Select(item => new { ListIndex = 2, Value = item});
//or use concat
var merged = first.Select(item => new { ListIndex = 1, Value = item});
.Concat(second.Select(item => new { ListIndex = 2, Value = item});
Alternatively if you have the sources in something like:
List<List<string>> lists = new List<List<string>>
{
new List<string> {"a","b","c"},
new List<string> {"x","y","z"}
};
you can do:
var merged = lists.SelectMany((item, index) =>
item.Select(s => new { ListIndex = index, Value = s}));
Note that this will produce a 0-based list, so if you really need a 1-base list, just do ListIndex = index +1.
Also, if you will use this a lot, I would create it as an specific entity, something like
struct ListIdentValue
{
public int ListIndex {get; private set;}
public string Value {get; private set;}
public ListIdentValue(int listIndex, string value) {...}
}
Try using Concat
new[] {'a','b','c'}
.Select(v=>new Tuple<int,char>(1, v))
.Concat(
new[] {'x','y','z'}.Select(v=>new Tuple<int,char>(2, v))
)
string[] a = { "a", "b", "c" };
string[] b = { "x", "z", "y" };
var t =
(
from ai in a
select new { listNo = 1, Item = ai }
).Union
(
from bi in b
select new { listNo = 2, Item = bi }
);
or
var t =
(
from ai in a
select new object[] { 1, ai }
).Union
(
from bi in b
select new object[] { 2, bi }
);

Categories

Resources