Sorry about that — feel free to email me at the associated address if it’s not clear.

private static Expression GetEqualsExpr(ParameterExpression param,
String property,
TColumn value)
{
Expression prop = Expression.Property(param, property);
Expression val = Expression.Constant(value);
return Expression.Equal(prop, val);
}

public static Expression<Func> GetWhereAndLambda(Dictionary filters)

I think that you can make it even more powerful using Type parameters for the Type of object you are filtering… So, e.g., the GetEquals method becomes something like:

private Expression GetEqualsExpr(ParameterExpression param,
String property,
TColumn value)
{
Expression prop = Expression.Property(param, property);
Expression val = Expression.Constant(value);
return Expression.Equal(prop, val);
}

signature for GetWhereLambda becomes something like:

public static Expression<Func> GetWhereAndLambda

Thank you!

I tried your code and figured out that to finally use the expression in the Where-Method I had to call the Compile-Method.

The final line should therefore look like this:

IEnumerable filtered = students.Where(myLambda.Compile());

Hopefully this helps other people facing the same trouble.

Tuhank’s

My user is going to hand me an ArrayList (let’s call it input) with tags/categories, along with an integer for the ID

so I get values of input[0]=”foo”, input[1]=”bar”, input[2]=”baz”

I want to construct a query that does:

query = (from c in db.theTable
where c.ID == inputID
&& c.tags ==

//This is where I run into trouble – I want to loop through the //arraylist with a foreach (or for or whatever) and construct a series //of OR conditions to put into the query here) – I need to be able to //have an arbitrary number of conditions

select c);

The code I added to your code basically lets you pass in any field and value and it will sort out the type matching at run time – so we can pass the following and it be ok:

[Field name (type) -- value]
FirstName (varchar(50)) — “Amye”
LastName (varchar(50)) — “Couves”
Age (Int32) — “30″ (note here that this would normally cause the code to fail but the ChangeType function changes “30″ to 30 at runtime.

As I said – i think some try {} catch {} is needed around the convert as it would throw an exception on converting “bob” to an integer (the user might put the wrong data in a search field by mistake) but this is easy to add and can then be passed back to the front end as an error.

Thanks again!