中文分词系统之-导入文本文件类词库建立哈希表索引-JAVA源码

/* /*分词系统的词典导入的接口类,主要实现文本文件中词典的导入,及创建哈希表。且实现初步的哈希表关键字查询 *输入参数:Iterator other/int max *输出数据: *author 张顺明 *email  mingzhingjien@hotmail.com **/ package chinese_participle; import java.io.*; import java.util.Vector; // Referenced classes of package beyondchtacs: //                 FreDicHashProcessing public class InPutTextFile {          HashProcessing hashproce;          Vector vector = new Vector();          static int fdMax = 0x21a31;               public InPutTextFile(String filename)          {              //fdHp = new HashProcessing(fdMax);                       File readfile = new File(filename);              try{                    InputStreamReader readinput = new InputStreamReader(new FileInputStream(readfile));                  BufferedReader reader = new BufferedReader(readinput);                  String s;                    while((s = reader.readLine()) != null)                    {                       vector.addElement(s);                    }                    hashproce=new HashProcessing(vector);                    reader.close();                    readinput.close();                  }              catch(Exception ey)                  {                    System.out.println(ey);                  }                             return;             }             ///////////////////////////////////////////////////检查某词是否在词典里面             public boolean isInVector(String x)             {               for (int i=0;i<vector.size();i++)              {                if( x.equals( vector.elementAt( i ).toString() ) );                return true;              }              return true;             }             //////////////////////////////////////////////////检查某词是否在新建的哈希表里面             public boolean isInHash(String x)             {                 return hashproce.isActive( hashproce.hashStr,hashproce.findAddress( x ) ) ;                         } } /* /*分词系统的词典导入的哈希表,是基于二次探测构建的 *输入参数:Iterator other/int max *输出数据: *author 张顺明 *email  mingzhingjien@hotmail.com **/ package chinese_participle; import java.util.*; public class HashProcessing {             private int max;          public HashEntry hashStr[];          private int currentSize=0;          private int occupied=0;          private int modCount=0;               public HashProcessing()          {                      hashStr = new HashEntry[max];              clear();          }               public HashProcessing(Collection other)          {                      hashStr = new HashEntry[other.size()*2];              clear();                       Iterator itr = other.iterator();                while( itr.hasNext() )                 add( itr.next() );          }          //////////////////////////////////////////////二次探测哈希表的isActive方法          public       boolean isActive(HashEntry[] arr,int pos)          {           return arr[pos]!=null&&arr[pos].isActive;          }          //////////////////////////////////////////////二次探测哈希表的remove方法和clear方法          public boolean remove( Object x)          {           int address=findAddress( x );           if( !isActive( hashStr,address ) )             return false;           hashStr[address].isActive=false;           currentSize--;           modCount++;           if(currentSize<hashStr.length/8)            rehash();           return true;          }          public void clear()          {           currentSize=occupied=0;           modCount++;           for(int i=0;i<hashStr.length;i++)           hashStr[i]=null;          }          //////////////////////////////////////////////查找对应地址函数          public int findAddress( Object x )          {           int colliCounter = 0;              int address =( x==null )? 0:Math.abs( x.hashCode( )% hashStr.length );                       while( hashStr[address]!=null )              {               if(x==null)               {                if(hashStr[address].element==null)                break;               }               else if (x.equals(hashStr[address].element))               break;               address+=2*++colliCounter-1;               if(address>=hashStr.length)                  address-=hashStr.length;                         }              return address;          }          ////////////////////////////////////////////////二次探测哈希表的add方法          public boolean add(Object x)          {           int address=findAddress(x);           if( isActive( hashStr,address ) )              return false;                   hashStr[address]=new HashEntry(x,true);            currentSize++;            occupied++;            modCount++;                   if( occupied>hashStr.length/2 )               rehash();            return true;          }               ////////////////////////////////////////////////实现再哈希          public void rehash()          {           HashEntry[] oldArry=hashStr;           //Create a new,empty table           hashStr=new HashEntry[nextPrime( 4*currentSize ) ];           currentSize=0;           occupied=0;           //Copy table over           for(int i=0;i<oldArry.length;i++)             if( isActive( oldArry,i ) )                add( oldArry[i].element );          } /*        ////////////////////////////////////////////////          public static void createHash(String keyStr)          {              int colliCounter = 0;              int address =(keyStr==null)? 0:hashMod(keyStr);              do              {                  if(hashStr[address] == null)                  {                      hashStr[address] = keyStr;                      break;                  }                  colliCounter++;                  address = collisionOffset(address, colliCounter);              } while(true);          } */          ///////////////////////////////////////////////二次探测哈希表的查找函数          public       Object hashSearch(Object x)          {               int address=findAddress( x );            if( hashStr[address]==null )               return null;            return hashStr[address].element;          }          public boolean contains(Object x)          {            return isActive( hashStr,findAddress( x ) );          }          //////////////////////////////////////////////寻找一个合适的质数         public boolean isPrime(int n)         {           int i=n%2;           if(i==0)            return false;           for(;i>1;i--)           {             if(n/i==0) break;                return true;           }             return true;         }         public int nextPrime(int n)         {           if( n%2==0)              n++;           for( ;!isPrime( n);n+=2 )           ;           return n;         } }