/// <summary>
/// Compute Levenshtein distance using provided weights for substitution.
/// </summary>
/// <param name="s1">The first string to compare.</param>
/// <param name="s2">The second string to compare.</param>
/// <returns>The computed weighted Levenshtein distance.</returns>
/// <exception cref="ArgumentNullException">If s1 or s2 is null.</exception>
public double Distance(string s1, string s2)
{
if (s1 == null)
{
throw new ArgumentNullException(nameof(s1));
}
if (s2 == null)
{
throw new ArgumentNullException(nameof(s2));
}
if (s1.Equals(s2))
{
return(0);
}
if (s1.Length == 0)
{
return(s2.Length);
}
if (s2.Length == 0)
{
return(s1.Length);
}
// create two work vectors of integer distances
double[] v0 = new double[s2.Length + 1];
double[] v1 = new double[s2.Length + 1];
double[] vtemp;
// initialize v0 (the previous row of distances)
// this row is A[0][i]: edit distance for an empty s
// the distance is just the number of characters to delete from t
for (int i = 0; i < v0.Length; i++)
{
v0[i] = i;
}
for (int i = 0; i < s1.Length; i++)
{
// calculate v1 (current row distances) from the previous row v0
// first element of v1 is A[i+1][0]
// edit distance is delete (i+1) chars from s to match empty t
v1[0] = i + 1;
// use formula to fill in the rest of the row
for (int j = 0; j < s2.Length; j++)
{
double cost = 0;
double insertioncost = 0;
double deletioncost = 0;
if (s1[i] != s2[j])
{
cost = _characterSubstitution.Cost(s1[i], s2[j]);
insertioncost = _characterSubstitution.InsertionCost(s2[j]);
deletioncost = _characterSubstitution.DeletionCost(s1[i]);
}
v1[j + 1] = Math.Min(
v1[j] + insertioncost, // Cost of insertion
Math.Min(
v0[j + 1] + deletioncost, // Cost of remove
v0[j] + cost)); // Cost of substitution
}
// copy v1 (current row) to v0 (previous row) for next iteration
//System.arraycopy(v1, 0, v0, 0, v0.length);
// Flip references to current and previous row
vtemp = v0;
v0 = v1;
v1 = vtemp;
}
return(v0[s2.Length]);
}
/// <summary>
/// Compute Levenshtein distance using provided weights for substitution.
/// </summary>
/// <param name="s1">The first string to compare.</param>
/// <param name="s2">The second string to compare.</param>
/// <param name="limit">The maximum result to compute before stopping. This
/// means that the calculation can terminate early if you
/// only care about strings with a certain similarity.
/// Set this to Double.MaxValue if you want to run the
/// calculation to completion in every case.</param>
/// <returns>The computed weighted Levenshtein distance.</returns>
/// <exception cref="ArgumentNullException">If s1 or s2 is null.</exception>
public double Distance(string s1, string s2, double limit)
{
if (s1 == null)
{
throw new ArgumentNullException(nameof(s1));
}
if (s2 == null)
{
throw new ArgumentNullException(nameof(s2));
}
if (s1.Equals(s2))
{
return(0);
}
if (s1.Length == 0)
{
return(s2.Length);
}
if (s2.Length == 0)
{
return(s1.Length);
}
// create two work vectors of floating point (i.e. weighted) distances
double[] v0 = new double[s2.Length + 1];
double[] v1 = new double[s2.Length + 1];
double[] vtemp;
// initialize v0 (the previous row of distances)
// this row is A[0][i]: edit distance for an empty s1
// the distance is the cost of inserting each character of s2
v0[0] = 0;
for (int i = 1; i < v0.Length; i++)
{
v0[i] = v0[i - 1] + InsertionCost(s2[i - 1]);
}
for (int i = 0; i < s1.Length; i++)
{
char s1i = s1[i];
double deletionCost = DeletionCost(s1i);
// calculate v1 (current row distances) from the previous row v0
// first element of v1 is A[i+1][0]
// Edit distance is the cost of deleting characters from s1
// to match empty t.
v1[0] = v0[0] + deletionCost;
double minv1 = v1[0];
// use formula to fill in the rest of the row
for (int j = 0; j < s2.Length; j++)
{
char s2j = s2[j];
double cost = 0;
if (s1i != s2j)
{
cost = _characterSubstitution.Cost(s1i, s2j);
}
double insertionCost = InsertionCost(s2j);
v1[j + 1] = Math.Min(
v1[j] + insertionCost, // Cost of insertion
Math.Min(
v0[j + 1] + deletionCost, // Cost of deletion
v0[j] + cost)); // Cost of substitution
minv1 = Math.Min(minv1, v1[j + 1]);
}
if (minv1 >= limit)
{
return(limit);
}
// copy v1 (current row) to v0 (previous row) for next iteration
// System.arraycopy(v1, 0, v0, 0, v0.length);
// Flip references to current and previous row
vtemp = v0;
v0 = v1;
v1 = vtemp;
}
return(v0[s2.Length]);
}