mirror of
https://github.com/Dushistov/sdcv.git
synced 2025-12-16 09:51:56 +00:00
distance: c++11 plus remove not used code
This commit is contained in:
@@ -34,13 +34,11 @@ The Levenshtein distance algorithm has been used in:
|
|||||||
*/
|
*/
|
||||||
|
|
||||||
|
|
||||||
#include <stdlib.h>
|
#include <cstdlib>
|
||||||
#include <string.h>
|
#include <cstring>
|
||||||
//#include <stdio.h>
|
|
||||||
|
|
||||||
#include "distance.h"
|
#include "distance.h"
|
||||||
|
|
||||||
#define OPTIMIZE_ED
|
|
||||||
/*
|
/*
|
||||||
Cover transposition, in addition to deletion,
|
Cover transposition, in addition to deletion,
|
||||||
insertion and substitution. This step is taken from:
|
insertion and substitution. This step is taken from:
|
||||||
@@ -54,19 +52,17 @@ Enhanced Dynamic Programming ASM Algorithm"
|
|||||||
/*Implementation of Levenshtein distance*/
|
/*Implementation of Levenshtein distance*/
|
||||||
/****************************************/
|
/****************************************/
|
||||||
|
|
||||||
EditDistance::EditDistance()
|
/*Gets the minimum of three values */
|
||||||
|
static inline int minimum(const int a, const int b, const int c)
|
||||||
{
|
{
|
||||||
currentelements = 2500; // It's enough for most conditions :-)
|
int min = a;
|
||||||
d = (int*)malloc(sizeof(int)*currentelements);
|
if ( b < min )
|
||||||
|
min = b;
|
||||||
|
if ( c < min )
|
||||||
|
min = c;
|
||||||
|
return min;
|
||||||
}
|
}
|
||||||
|
|
||||||
EditDistance::~EditDistance()
|
|
||||||
{
|
|
||||||
// printf("size:%d\n",currentelements);
|
|
||||||
if (d) free(d);
|
|
||||||
}
|
|
||||||
|
|
||||||
#ifdef OPTIMIZE_ED
|
|
||||||
int EditDistance::CalEditDistance(const gunichar *s,const gunichar *t,const int limit)
|
int EditDistance::CalEditDistance(const gunichar *s,const gunichar *t,const int limit)
|
||||||
/*Compute levenshtein distance between s and t, this is using QUICK algorithm*/
|
/*Compute levenshtein distance between s and t, this is using QUICK algorithm*/
|
||||||
{
|
{
|
||||||
@@ -92,7 +88,7 @@ int EditDistance::CalEditDistance(const gunichar *s,const gunichar *t,const int
|
|||||||
{
|
{
|
||||||
n--;m--;
|
n--;m--;
|
||||||
}
|
}
|
||||||
if ( m==0 || n==0 || d==(int*)0 )
|
if ( m==0 || n==0 || d==nullptr )
|
||||||
return (m+n);
|
return (m+n);
|
||||||
if ( m < n )
|
if ( m < n )
|
||||||
{
|
{
|
||||||
@@ -112,8 +108,8 @@ int EditDistance::CalEditDistance(const gunichar *s,const gunichar *t,const int
|
|||||||
if ( m*n > currentelements )
|
if ( m*n > currentelements )
|
||||||
{
|
{
|
||||||
currentelements = m*n*2; // double the request
|
currentelements = m*n*2; // double the request
|
||||||
d = (int*)realloc(d,sizeof(int)*currentelements);
|
d = static_cast<int*>(realloc(d, sizeof(int) * currentelements));
|
||||||
if ( (int*)0 == d )
|
if ( nullptr == d )
|
||||||
return (m+n);
|
return (m+n);
|
||||||
}
|
}
|
||||||
// step 2, init matrix
|
// step 2, init matrix
|
||||||
@@ -156,48 +152,3 @@ int EditDistance::CalEditDistance(const gunichar *s,const gunichar *t,const int
|
|||||||
// d(n-1,m-1)
|
// d(n-1,m-1)
|
||||||
return d[n*m-1];
|
return d[n*m-1];
|
||||||
}
|
}
|
||||||
#else
|
|
||||||
int EditDistance::CalEditDistance(const char *s,const char *t,const int limit)
|
|
||||||
{
|
|
||||||
//Step 1
|
|
||||||
int k,i,j,n,m,cost;
|
|
||||||
n=strlen(s);
|
|
||||||
m=strlen(t);
|
|
||||||
if( n!=0 && m!=0 && d!=(int*)0 )
|
|
||||||
{
|
|
||||||
m++;n++;
|
|
||||||
if ( m*n > currentelements )
|
|
||||||
{
|
|
||||||
currentelements = m*n*2;
|
|
||||||
d = (int*)realloc(d,sizeof(int)*currentelements);
|
|
||||||
if ( (int*)0 == d )
|
|
||||||
return (m+n);
|
|
||||||
}
|
|
||||||
//Step 2
|
|
||||||
for(k=0;k<n;k++)
|
|
||||||
d[k]=k;
|
|
||||||
for(k=0;k<m;k++)
|
|
||||||
d[k*n]=k;
|
|
||||||
//Step 3 and 4
|
|
||||||
for(i=1;i<n;i++)
|
|
||||||
for(j=1;j<m;j++)
|
|
||||||
{
|
|
||||||
//Step 5
|
|
||||||
if(s[i-1]==t[j-1])
|
|
||||||
cost=0;
|
|
||||||
else
|
|
||||||
cost=1;
|
|
||||||
//Step 6
|
|
||||||
d[j*n+i]=minimum(d[(j-1)*n+i]+1,d[j*n+i-1]+1,d[(j-1)*n+i-1]+cost);
|
|
||||||
#ifdef COVER_TRANSPOSITION
|
|
||||||
if ( i>=2 && j>=2 && (d[j*n+i]-d[(j-2)*n+i-2]==2)
|
|
||||||
&& (s[i-2]==t[j-1]) && (s[i-1]==t[j-2]) )
|
|
||||||
d[j*n+i]--;
|
|
||||||
#endif
|
|
||||||
}
|
|
||||||
return d[n*m-1];
|
|
||||||
}
|
|
||||||
else
|
|
||||||
return (n+m);
|
|
||||||
}
|
|
||||||
#endif
|
|
||||||
|
|||||||
@@ -1,26 +1,23 @@
|
|||||||
#ifndef DISTANCE_H
|
#pragma once
|
||||||
#define DISTANCE_H
|
|
||||||
|
|
||||||
|
#include <cstdlib>
|
||||||
#include <glib.h>
|
#include <glib.h>
|
||||||
|
|
||||||
class EditDistance {
|
class EditDistance {
|
||||||
|
public:
|
||||||
|
EditDistance() {
|
||||||
|
currentelements = 2500; // It's enough for most conditions :-)
|
||||||
|
d = static_cast<int *>(malloc(sizeof(int)*currentelements));
|
||||||
|
}
|
||||||
|
~EditDistance() {
|
||||||
|
if (d != nullptr)
|
||||||
|
free(d);
|
||||||
|
}
|
||||||
|
EditDistance(const EditDistance&) = delete;
|
||||||
|
EditDistance& operator=(const EditDistance&) = delete;
|
||||||
|
int CalEditDistance( const gunichar *s, const gunichar *t, const int limit );
|
||||||
private:
|
private:
|
||||||
int *d;
|
int *d;
|
||||||
int currentelements;
|
int currentelements;
|
||||||
/*Gets the minimum of three values */
|
|
||||||
inline int minimum( const int a, const int b, const int c )
|
|
||||||
{
|
|
||||||
int min = a;
|
|
||||||
if ( b < min )
|
|
||||||
min = b;
|
|
||||||
if ( c < min )
|
|
||||||
min = c;
|
|
||||||
return min;
|
|
||||||
};
|
|
||||||
public:
|
|
||||||
EditDistance( );
|
|
||||||
~EditDistance( );
|
|
||||||
int CalEditDistance( const gunichar *s, const gunichar *t, const int limit );
|
|
||||||
};
|
};
|
||||||
|
|
||||||
#endif
|
|
||||||
|
|||||||
Reference in New Issue
Block a user