>
>
>
V3171. Potentially negative value is us…


V3171. Potentially negative value is used as the size of an array.

The analyzer detected that a potentially negative value of a variable or expression might be used as the size of an array that is created.

Consider an example:

void ProcessBytes(byte[] bytes)
{
  int length = BitConverter.ToUInt16(bytes, 0);
  int[] newArr = new int[length - 2];
  ....
}

The value returned by the 'ToUInt16' method and assigned to the 'length' variable may be zero. This will happen if the first two bytes in the 'bytes' array are zero. This way, when you create the 'newArr' array, its length will be a negative value. This will result in the 'OverflowException' type exception.

Fixed version of the 'ProcessBytes' method with an additional check might look like this:

void ProcessBytes(byte[] bytes)
{
  int length = BitConverter.ToUInt16(bytes, 0);

  if (length < 2)
    return;

  int[] newArr = new int[length - 2];
  ....
}

Here is another code example. When you call the 'SubArray' method with a certain values of input arguments, a negative value can be used as a length of the array:

public byte[] GetSubArray(byte[] bytes)
{
  return bytes.SubArray(4, 2);
}

public static T[] SubArray<T>(this T[] arr, int index, int length)
{
  if (length < 0)
    throw new Exception($"Incorrect length value: {length}.");

  if (index < 0)
    throw new Exception($"Incorrect index value: {index}.");

  if (arr.Length < index + length)
    length = arr.Length - index;

  var subArr = new T[length];
  Array.Copy(arr, index, subArr, 0, length);
  return subArr;
}

Here is the problem with the 'SubArray' method. It does not take into account cases when the length of the 'arr' array may be less than the value in the 'index' variable. In this case, the 'length' variable will get a negative value. Assume that the length of the 'arr' array is 3, the value of the 'index' variable is 4. The 'length' variable will get the value of -1 during the method's execution. There will be an attempt to create an array with a negative size.

The fixed version of the 'SubArray' method may look like this:

public static T[] SubArray<T>(this T[] arr, int index, int length)
{
  if (length < 0)
    throw new Exception($"Incorrect length value: {length}.");

  if (index < 0 || arr.Length <= index)
    throw new Exception($"Incorrect index value: {index}.");

  if (arr.Length < index + length)
    length = arr.Length - index;

  var subArr = new T[length];
  Array.Copy(arr, index, subArr, 0, length);
  return subArr;
}

This diagnostic is classified as: