/// <summary>
/// Average: O(1) * number_of_permutation
/// Returns suggested spellings that match value within a given true Damereau-Levenshtein distance.
/// </summary>
/// <param name="delete_distance">Default: -1. -1 returns everything.</param>
public List <Result> Lookup(string value, int delete_distance = -1, Verbosity verbosity = Verbosity.Top)
{
if (delete_distance < 0)
{
delete_distance = this.DeleteDistance;
}
else if (delete_distance > this.DeleteDistance)
{
throw new ArgumentOutOfRangeException($"{nameof(delete_distance)} ({delete_distance}) cannot be > {nameof(this.DeleteDistance)} ({this.DeleteDistance}).", nameof(delete_distance));
}
var permutation_hash = new HashSet <string>();
var visited_results = new HashSet <string>();
var results = new List <Result>();
int best_distance = int.MaxValue; // applies only when verbosity==Closest
CreateDeletePermutations(value, delete_distance, permutation_hash);
foreach (var permutation in permutation_hash)
{
if (!m_dict.TryGetValue(permutation, out var node))
{
continue;
}
foreach (var originalString in node)
{
if (!visited_results.Add(originalString))
{
continue;
}
int cost = FuzzyStringMatch.LevenshteinDistance(value, originalString, 1, 1, 1, 1, delete_distance + 1);
if (cost <= delete_distance)
{
// if exact match, break loop
if (cost == 0 && verbosity == Verbosity.Top)
{
results.Clear();
results.Add(new Result(originalString, cost));
return(results);
}
// keep only results within the best delete distance found so far
if (verbosity == Verbosity.Closest)
{
if (cost > best_distance)
{
continue;
}
else if (cost < best_distance)
{
results.Clear();
best_distance = cost;
}
}
results.Add(new Result(originalString, cost));
}
}
}
results.Sort();
return(results);
}
static void Main(string[] args)
{
var dist22 = new SpellingSuggestor(3);
dist22.AddRange("manly,abcde,hamburger,apple".Split(','));
var fufuufuff = dist22.Lookup("applr").ToList();
var art2 = new AdaptiveRadixTree <string, string>();
art2.Add("dfgdfg", "213123");
var dist22333 = FuzzyStringMatch.CombinedWordEditDistance("H7N0K3", "H7N0K3");
var art = new AdaptiveRadixTree <string, int>();
var dict = new Dictionary <string, int>();
int aa = 0;
var items = AdaptiveRadixTreeTest.GenerateTestKeys(100000).ToArray();
foreach (var item in items)
{
art.Add(in item, aa);
dict.Add(item, aa);
aa++;
}
System.Diagnostics.Debug.Write(art.CalculateMetrics());
//// System.Diagnostics.Debug.Write(art.DebugDump(true));
//
var now = DateTime.UtcNow;
int readResults = 0;
// todo: read/write not using stream
for (int i = 0; i < items.Length; i++)
{
var item = items[i];
//readResults += art.PartialMatch(item.Substring(0, item.Length - 1) + ".", '.').Count();
//readResults += art.PartialMatch("." + item.Substring(1), '.').Count();
//readResults += art.RegExpMatch("[A-B-D]" + item.Substring(1)).Count();
//readResults += art.StartsWithKeys(item.Substring(0, item.Length - 1)).Count();
//readResults += art.RegExpNearNeighbors("[A-B-D]" + item.Substring(1), 0).Count();
// remove()
// fix path filter enumerator to call the filter method in-order
// try with just one entry
//var sdfsdf = art.RegExpNearNeighbors("BBBI", 2).ToList();
if (i == 99999)
{
"".ToLower();
}
if (!art.Remove(item)) // f**k at i=163 with offbranch code
{
"".ToLower();
}
//if(i > 145 && art.DebugDump(true).Contains("EXCEPTION"))
// "".ToString();
}
//var dict4 = new Dictionary<int, string>(100000);
//int coll = 0;
//for(int i = 0; i < 100000; i++) {
// //FuzzyStringMatch.GetStableHashCode(items[i % 100000]);
// //items[i % 100000].GetHashCode();
// if((i %100000) == 0)
// dict4.Clear();
// try {
// dict4.Add(FuzzyStringMatch.GetStableHashCode(items[i % 100000]), items[i % 100000]);
// } catch {
// coll++;
// }
//}
var diff = DateTime.UtcNow - now;
System.Console.WriteLine(diff.ToString());
//System.Console.WriteLine(readResults);
System.Console.ReadLine();
}
private static double CalculateLicensePlateSimilarity(string normalized_input, string normalized_match)
{
//return FuzzyStringMatch.CombinedWordSimilarity(normalized_input, normalized_match);
return(FuzzyStringMatch.CombinedWordEditDistance(normalized_input, normalized_match));
}
public int GetHashCode(string obj)
{
return(FuzzyStringMatch.GetStableHashCode(obj));
}
private static string NormalizeLicensePlate(string value)
{
//value = value.Replace(" ", string.Empty);
// note: UPPER case is relevant because license plate are printed in upper cases, and visual similarity is taken into account
return(FuzzyStringMatch.RemoveDiacritics(value.Trim()).ToUpperInvariant());
}
