Opening File in exclusive or shared mode

Sometimes you need to read file that can be used by other processes as well and in that case you if you just want to read the content of the file and let the other process to write in it, you actually have to open a file in shared mode. Code below can allow you to open a file in shared mode.

"FileShare.ReadWrite" enum allow you to specify the level of lock you want on the file. Here it will allow other processes to read and write to the file/

using (Stream s = new FileStream(fullFilePath,
                                 FileMode.Open,
                                 FileAccess.Read,
                                 FileShare.ReadWrite))
{
  // here use the stream s
}

In case if you don't want other processes to interfere or read your file you can specify "FileShare.None" which will acquire exclusive access to the file.


using (Stream s = new FileStream(fullFilePath,
                                 FileMode.Open,
                                 FileAccess.Read,
                                 FileShare.None))
{
  // here use the stream s
}

 

Parallel Processing with Defined Degree of Parallelism using Task Parallelism

Sometimes you really want to achieve very simple task but when it comes to program it, you end up thinking about dead locks, thread sync, max threads, processing time etc. I have faced a similar situation where i want to process some items mutually exclusive with each other but i can't process more than 4 items at a time. so i end up using Task Parallelism which makes it really easy and understandable. Other way of doing this is by using ThreadPool and ManualRest locks that is you can say a retro fit for this problem.

To illustrate the problem, let assume you want to parse the html content of a list of urls on a site but you can't process more than 4 urls at a time to avoid overloading your web server and bandwidth. Here is how it can be easily achieved using Task Parallelism:


          //fetch all list of urls to scan.
           List<string> urlsToScan = GetUrlsToScan();
           
            //apply parallel foreach on every url but with Degree of Parallelism to 4 threads at a time.
            Parallel.ForEach(urlsToScan,
                new ParallelOptions()
                {
                    MaxDegreeOfParallelism = 4
                }, (s) =>
                {
                    Console.WriteLine("{0} - Processing url: {1}", DateTime.Now, s);
                    ScrapUrl(s);
                });

Hope this will help a lot to solve many such problems very easily. Happy Coding!!