Question

here is the code

        var customers = db.ExecuteQuery<Customer>(@"SELECT CustomerID, CompanyName, ContactName, ContactTitle, 
   Address, City, Region, PostalCode, Country, Phone, Fax
   FROM   dbo.Customers
   WHERE  City = {0}", "London");

foreach (Customer c in customers)
   Console.WriteLine(c.ContactName);

code execute sql and retun a customer record. my question how result can be stored in customer class automatically….which i do not understand. if u see this line of code
db.ExecuteQuery<Customer> from here we can understand that customer result will be return and customer data will be stored in customer class. how automatically data can be stored & assign to right property in customer class because a customer class CustomerID property name could be CustID….then what will happen.

the line db.ExecuteQuery<Customer> is very confusing for me and i just do not understand a new customer instance will be created with return customer data….so plzz discuss in detail.

Answer 1

ExecuteQuery just runs arbitrary SQL (after applying string.Format-esque parameterization) and then materializes the IDataReader results as a sequence of objects.

Note that the mappings between mapped types and the db-columns/type-members is well defined by the data-context (usually via attributes on the members, but not always), but IIRC ExecuteQuery does not apply these mapping (I’m happy to be corrected here). You can inspect this via db.Mapping.GetMetaType(typeof(Customer)).

The fundamentals here are no different to if you executed

var cust = db.Customers.ToList();

except the db.Customers version can apply more mappings etc.

The materialization itself (i.e. creating the object and setting members) is an example of reflection and meta-programming; most ORMs and micro-ORMs will work pretty much the same there: inspect the type (Customer above) and inspect the fields in the reader – and then build some code on-the-fly (usually cached) that will create the new objects and set the members.