Transform Block

Characteristics:

• Accepts and sends data

• Has an output queue

• Map function like Func(T, U)

Transform Block Characteristics:

• Accepts and sends data

• Has an output queue

• Map function like Func<T, U>

• Default value for processing is 1 (no parallel execution, but can be overridden)

Receive Function Block

• Accepts an item

• Will make the Receive call

• Return it to the program and move on

Parallelism without Logic changes - Enable parallelism at the constructor

Transform process number items is set to 1 by default. We can specify it by providing the options at the constructor.

var transformBlock = new TransformBlock<int, string>(n =>

{

TaskDelary(TimeSpan.FromSeconds(1));

return n.ToString();

new ExecutionDataflowBlockOptions(){ MaxDegreeOfParallelism = 2}

);

This is an example of how TPL Dataflow allows parallelism implementation without having to change the Execution and/or Business Logic.

Item1 Item2 -> f(x) -> Item2 Item1

Input Queue -> f(x) -> Output Queue

class Program
{
    static async Task Main(string[] args)
    {
        // Create a TransformBlock that doubles the input integer
        var doublerBlock = new TransformBlock<int, int>(input => input * 2);

        // Post some data to the block
        for (int i = 1; i <= 10; i++)
        {
            await doublerBlock.SendAsync(i);
        }

        // Complete the block, indicating that no more data will be posted
        doublerBlock.Complete();

        // Read the output data from the block
        while (await doublerBlock.OutputAvailableAsync())
        {
            var output = await doublerBlock.ReceiveAsync();
            Console.WriteLine(output);
        }
    }
}

In this example:

We create a TransformBlock<int, int> where the input type is int and the output type is also int. The block will double the input integer and produce it as output.
We post integers from 1 to 10 into the block asynchronously using SendAsync.
After all data has been posted, we complete the block using Complete(). This indicates that no more data will be sent to the block.
Finally, we read the output from the block using OutputAvailableAsync() to check if there's output available and ReceiveAsync() to receive the output.

No files yet, migration hasn't completed yet!