Wednesday 25 November 2015

Locking and unlocking specific bytes of a file in C#

The following code shows a program that can lock and unlock specific bytes of file in C# using FileStream::Lock and FileStream::Unlock in System.IO. To test out that locking can protect designated bytes of a file, start up multiple instances of the program and try writing to the file after it has been locked. This will protect the file to be modified by other processes, which is handy if you want to ensure data integrity of files your system or application uses and expect to have a given contents and format. FileStream class allows us to work on files on a byte level and do detailed I/O operations on files in .NET.


using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Text;
using System.Threading.Tasks;

namespace TestOutFileLocking
{

    class FStreamLock
    {

        static void Main(string[] args)
        {
            var uniEncoding = new UnicodeEncoding();
            string lastRecordText = "The last processed record number was: ";
            int textLength = uniEncoding.GetByteCount(lastRecordText);
            int recordNumber = 13;
            int byteCount = uniEncoding.GetByteCount(recordNumber.ToString());
            string tempString;

            using (var fileStream = new FileStream("Test#@@#.dat", FileMode.OpenOrCreate, FileAccess.ReadWrite, FileShare.ReadWrite))
            {
                //Write the original file data. 
                if (fileStream.Length == 0)
                {
                    tempString = lastRecordText + recordNumber.ToString(); 
                    fileStream.Write(uniEncoding.GetBytes(tempString), 0, uniEncoding.GetByteCount(tempString));
                }

                //Allow the user to choose the operation. 
                char consoleInput = 'R'; 
                byte[] readText = new byte[fileStream.Length];
                while (consoleInput != 'X')
                {
                    Console.Write("\nEnter 'R' to read, 'W' to write, 'L' to " + "lock, 'U' to unlock, anything else to exit: ");

                    if ((tempString = Console.ReadLine()).Length == 0)
                    {
                        break;
                    }
                    consoleInput = char.ToUpper(tempString[0]);

                    switch (consoleInput)
                    {
                        //Read data from the file and 
                        //write it to the console 
                        case 'R':
                            try
                            {
                                fileStream.Seek(0, SeekOrigin.Begin);
                                fileStream.Read(readText, 0, (int) fileStream.Length);
                                tempString = new string(uniEncoding.GetChars(readText, 0, readText.Length));
                                Console.WriteLine(tempString);
                                recordNumber = int.Parse(tempString.Substring(tempString.IndexOf(':') + 2)); 

                            }
                            //Catch the IOException generated if the 
                            //specified part of the file is locked. 
                            catch (IOException e)
                            {
                                Console.WriteLine("{0}: The read " +
                                 "operation could not be performed " +
                                 "because the specified part of the " +
                                 "file is locked.",
                                 e.GetType().Name);
                                
                            }

                            break;

                            //Update the file. 
                        case 'W':

                            try
                            {
                                fileStream.Seek(textLength, SeekOrigin.Begin);
                                fileStream.Read(readText, textLength - 1, byteCount);
                                tempString = new string(uniEncoding.GetChars(readText, textLength - 1, byteCount));
                                recordNumber = int.Parse(tempString) + 1;
                                fileStream.Seek(textLength, SeekOrigin.Begin);
                                fileStream.Write(uniEncoding.GetBytes(recordNumber.ToString()), 0, byteCount);
                                fileStream.Flush(); 
                                Console.WriteLine("Record has een updated.");
                            }
                            catch (IOException e)
                            {
                                Console.WriteLine(
                                    "{0}: The write operation could not " +
                                    "be performed because the specified " +
                                    "part of the file is locked.",
                                    e.GetType().Name);
                            } //try-catch 
                            break;

                            //Lock the specified part of the file. 
                        case 'L':

                            try
                            {
                                fileStream.Lock(textLength - 1, byteCount);
                                Console.WriteLine("The specified part " + "of file has been locked.");
                            }
                            catch (IOException err)
                            {
                                Console.WriteLine("{0}: The specified part of file is" + " already locked.", err.GetType().Name);
                            } //try-catch 
                            break; 

                        case 'U':
                            try
                            {
                                fileStream.Unlock(textLength - 1, byteCount);
                                Console.WriteLine("The specified part of file has been unlocked.");

                            }
                            catch (IOException err)
                            {
                                Console.WriteLine("{0}: The specified part of file is " + "not locked by the current prcess.", err.GetType().Name);
                            }
                            break; 

                        default:
                            consoleInput = 'X';
                            break; 

                    } //switch 

                }

            } //using 



        }

    }

}


Thursday 12 November 2015

Reading a text file in VB6 and putting the contents into an array

VB6 or Visual Basic 6.0 is not used in many new projects, as it is replaced by .NET. But sometimes you are assigned at work to read and further develop or migrate Legacy code. Let's look at some simple VB6 code. We first read the contents of a file and put each line into a string array. Then we add the items to a listbox control.

Private Sub Command3_Click()

 Dim someFileHandle As Integer
 Dim fileName As String
 Dim someStrings() As String
 
 someFileHandle = FreeFile

 fileName = App.Path + "\fox.txt"
 
 ReDim someStrings(1000) As String

 Open fileName For Input As #someFileHandle
 
 Dim lineNo As Integer
 
 Do Until EOF(someFileHandle)
  Input #someFileHandle, someStrings(lineNo)
  lineNo = lineNo + 1
 Loop
 
 ReDim Preserve someStrings(lineNo - 1) As String
  
 List1.Clear
  
 For x = 0 To lineNo - 1
  List1.AddItem someStrings(x)
 Next x

End Sub


First we get a file handle to the file we want to open. We declare an integer and use the FreeFile method to get a filehandle. We then use the Open function to open a file and assign the file handle. Note the use of the pound sign (#) here. We also declare a large string array, which is one dimensional. We use the ReDim Preserve Method to resize the array to save some memory space and preserve the content. We use the Input Method to put each line into an array element before this is done. Note the use of EOF here. We finally loop through the array and add each array item to a listbox control. So now you should have some basic overview how you can read a file in VB6 into an array structure and loop through its content. How neat is that!