Advertisement
2002VB String Manipulation #20516

_A Smart String Comparison

This takes 2 strings and returns the percent alike that they are. (i.e. "test string number 1" is 86.48% similar to "teststring numb 2") This function is very useful! You can use it in databases to match data that may have errors in it. Examples being people's names, company names, addresses, or anything else where you may encounter misspellings or inconsistencies in the data. Your feedback and/or votes are greatly appreciated! -- NEW - updated to use byte arrays instead of strings, 50-300% performance improvement!

AI

AI Summary: This codebase represents a historical implementation of the logic described in the metadata. Our preservation engine analyzes the structure to provide context for modern developers.

Source Code
original-source
Private b1() As Byte
Private b2() As Byte
Public Function Simil(String1 As String, String2 As String) As Double
  Dim l1 As Long
  Dim l2 As Long
  Dim l As Long
  Dim r As Double
  If UCase(String1) = UCase(String2) Then
    r = 1
  Else
    l1 = Len(String1)
    l2 = Len(String2)
    If l1 = 0 Or l2 = 0 Then
      r = 0
    Else
      ReDim b1(1 To l1): ReDim b2(1 To l2)
      For l = 1 To l1
        b1(l) = Asc(UCase(Mid(String1, l, 1)))
      Next
      For l = 1 To l2
        b2(l) = Asc(UCase(Mid(String2, l, 1)))
      Next
      r = SubSim(1, l1, 1, l2) / (l1 + l2) * 2
    End If
  End If
  Simil = r
  Erase b1
  Erase b2
End Function
Private Function SubSim(st1 As Long, end1 As Long, st2 As Long, end2 As Long) As Long
  Dim c1 As Long
  Dim c2 As Long
  Dim ns1 As Long
  Dim ns2 As Long
  Dim i As Long
  Dim max As Long
  If st1 > end1 Or st2 > end2 Or st1 <= 0 Or st2 <= 0 Then Exit Function
  For c1 = st1 To end1
    For c2 = st2 To end2
      i = 0
      Do Until b1(c1 + i) <> b2(c2 + i)
        i = i + 1
        If i > max Then
          ns1 = c1
          ns2 = c2
          max = i
        End If
        If c1 + i > end1 Or c2 + i > end2 Then Exit Do
      Loop
    Next
  Next
  max = max + SubSim(ns1 + max, end1, ns2 + max, end2)
  max = max + SubSim(st1, ns1 - 1, st2, ns2 - 1)
  SubSim = max
End Function
Original Comments (3)
Recovered from Wayback Machine