Tagged: linq

JSON Serialization and Circular Reference Error !

I have a table that references a child table with reference to parent (!), I am using LINQ to SQL, for a paricular scenario, when I want to serialize my result using Json(obj) in my controller code, I got an error saying.. “A circular reference was detected while serializing an object of type..”

This issue is very well explained by Rick Strahl in his blog (link here)

If we are using WCF services, setting SerializationMode of the data context to “Unidirectional” would do the trick. In my case I was using MVC and it was a controller action invoke from javascript!! Also I did not want to try setting the Access Level to internal because I was not sure whether I get deep serialization if I do that! (did not try, because I was half convinced that It may not work for me)

Issue was Json() call in my controller to convert my object to Json and send back as JsonResult… So I ended up serializing it in my biz layer, biz layer method now returned a Json String, changed my controller to return ActionResult/ContentResult and used Content(jsonString) to send back a ContentResult.

It worked because Javascript library expected a Json Literal and it got that. All is good for now!

A LINQ Thought

LINQ stands for Language Integrated Query. I want to stress on “Integrated” because I think the strength of LINQ comes from the fact that it is a data query engine that is integrated with .Net languages. This very reason gives us many features that we as developers are comfortable
using in a language into our data manipulation operations.

Lets take a simple example,

var query
= from customer in CustomerTable.AsEnumerable()
where customer.Field("Name").ToLower().Contains("John")
select customer;

In the above example, we are using simple string functions that are part of C# string class to do our comparisons for our data fetch.
Lets take another example, where we will use a C# function to evaluate a sub-query/expression,

var query
= from customer in CustomerTable.AsEnumerable()
where
customer.Field("Name").ToLower().Contains("John")
select new Customer 
{
Name = customer.Field("Name"),
State = customer.Field("State"),
CustomerType = GetCustomerType(customer.Field("State"))
};

In the above example, GetCustomerType() function is used inline with a LINQ query to evaluate some business rule.
In short, with great flexibility like this, LINQ has made life of developers much easier!

Cheers!

Using LINQ to manipulate data in DataSet/DataTable

Have you tried a micro ORM for your data access? Click to read more


LINQ (Language Integrated Query) provides a very easy and fast way to manipulate data that is cached in a DataSet. In .Net applications, DataSet represents a disconnected cache of data. Your application typically needs to search, filter thru this data in order to display this data according to the need. DataView provides a very limited options
when it comes to creating a filtered view of the data. LINQ extends these capabilities to a greater extend.

A LINQ query opertaion consists of 3 actions (Ref:MSDN): obtain the data source, create the query and execute the query.

Any datasource that implements the IEnumerable(T) generic interface can be queried thru LINQ. So DataTable objects are good candidates to do any LINQ query opertions, we will see using the following examples, how some common tasks can be done using LINQ queries.

For our example, we will consider that our DataSet has one(1) table with the following schema,

dtCustomer (CustomerID,CustomerName,Address,City,PostalCode,State,Country,PhoneNumer)

A simple select:

IEnumerable query =
    from customer in dtCustomer.AsEnumerable()
    select customer;

Till this point, we have the LINQ query ready, but not executed it yet, query is executed when we actually use it.

foreach (DataRow dr in query)
{
    Console.WriteLine(dr.Field("CustomerName"));
}

At this point, our query is executed and it prints the names of the customer.

We can also, get the resulset as a DataView by simply doing,

DataView view = query.AsDataView();

Most times, when we are dealing with DataSet/DataTable, data we will be creating a DataView as result of our LINQ query. ToList(), ToArray() methods are also very useful when you want to get your resultset
as a generic list or Array (Think AJAX/JSON!).

Lambda Expressions can be used as parameters to LINQ queries.

IEnumerable customers =
    query.Where(c => c.Field("CustomerName").Containes("John"));

//All names that contain "John"
foreach (DataRow cust in customers)
{
    Console.WriteLine(cust.Field("CustomerName"));
}

Simple Where Clause:

EnumerableRowCollection query
              = from customer in dtCustomer.AsEnumerable()
                where customer.Field("State") == "NJ"
                select customer;
            DataView njview = query.AsDataView();

Pretty simple, njview represents all customers that live in NJ.
You can extend the example to add more criteria to your where clause…

EnumerableRowCollection query
              = from customer in dtCustomer.AsEnumerable()
                where customer.Field("State") == "NJ" && customer.Field("PostalCode") == "08807"
                select customer;
            DataView njview = query.AsDataView();

It is useful to note that when you write your where clause, you leverage the power of your C# (or VB.Net) language features to search and filter your resultset using LINQ.

So, A SQL where clause is

where customer.Field("State") == "NJ"
     where customer.Field("State") != "NJ"

A SQL Like clause is

where customer.Field("CustomerName").Containes("John")

Skip and Take allows to get the skip the first n rows or get the top n rows as a result of the query.\

EnumerableRowCollection query
              = (from customer in dtCustomer.AsEnumerable()
                where customer.Field("State") == "NJ"
                select customer).Skip(3);

EnumerableRowCollection query
              = (from customer in dtCusomter.AsEnumerable()
                where customer.Field("State") == "NJ"
                select customer).Take(3);

Simple ORDER BY clause:

EnumerableRowCollection query
              = from customer in dtCustomer.AsEnumerable()
                orderby customer.Field("CustomerName")
                select customer;

Above query, gets the result order by customer name (ascending is default). And if you want it by descending order,

EnumerableRowCollection query
              = from customer in dtCusomter.AsEnumerable()
                orderby customer.Field("CustomerName")  descending
                select customer;

Reverse ORDER:

EnumerableRowCollection query
              = (from customer in dtCustomer.AsEnumerable()
                orderby customer.Field("CustomerName")
                select customer.Reverse();

Simple JOIN Operator:

var customers = ds.Tables["Customer"].AsEnumerable();
var orders = ds.Tables["Order"].AsEnumerable();
var query =
    		from customer in customers
    		join order in orders
		on order.Field("CustomerID")
    		equals customer.Field("Customer")
	into custOrders
	select custOrders;

All the examples given above is just the tip of iceberg on what you can do with LINQ and Lambda expressions. There is tons of samples and articles available on Internet on this, if you are looking for a way to simplyfy and speed up your business processing logic in your middle tier, LINQ is something you need to adopt!

Cheers!