mini英汉字典V3.0

      前面两个版本的mini自带都是用Trie树实现的，对于load一个200k左右的字典文件到内存中，通过任务管理器查看占用内存，发现v1.0和v1.2占用的内存大小分别在4.5m、3.8m左右，而本版本也是把所有字典文件load到内存中，这个版本所占内存大约1.2m。Trie树占用内存大的原因是申请的节点太多，对于存储一个英文单词，比如abcd，就要申请5个节点，尽管v1.0把节点的类型分为分支和叶节点，v1.2钟又对分支节点进行了细分，但是节点数目没有减少，又由于字典文件很简单，记录的内容比节点的内容还要小，所以占用的内存空间多了很多。

      本版本没有使用索引，将来的版本中将会使用索引技术，这样就不用把所有的字典装载到内存中，而是只装载必要的索引信息，会大大节省内存空间，理论上要远远小于字典文件，才是比较理想的实现。

 

下面贴下代码。

 

//MyMain.cpp

#define _CRTDBG_MAP_ALLOC #include <stdlib.h> #include <crtdbg.h> #include "AllHeads.h" int main(void) { //_CrtSetDbgFlag ( _CRTDBG_ALLOC_MEM_DF | _CRTDBG_LEAK_CHECK_DF ); HashDictionary *dic = new HashDictionary(); //new a instance of Hash Dictionary //_CrtDumpMemoryLeaks(); if( dic->LoadDictionary(NULL) != SUCCESS) // load Dictionary to memory { return -1; _CrtDumpMemoryLeaks(); } dic->run(); // accept input word to query, and return the results delete dic; //new char[10]; //_CrtDumpMemoryLeaks(); return 0; _CrtDumpMemoryLeaks(); //memory leak check }

//MyHashTable.h

#ifndef __MINGSPY_MYHASHTABLE_HH__ #define __MINGSPY_MYHASHTABLE_HH__ //#include "AllHeads.h" #include <string> #include <iostream> #include <iostream> #include <string> #include <memory> #include <list> #include <algorithm> #include <fstream> namespace mingspy{ using namespace std; const size_t DEFHASHSIZE = 1000; // default hash table size const double DEFHASHFACTOR = 0.5; // default hash table factor /* A hash function class, return the keys hash value */ class DefHashFunc { public: int operator ()( int key, int n){ return key%n;} int operator ()( const char * key , int n) { return ELFHash(key, n); } int operator ()( const string & key , int n) { return ELFHash(key.c_str(), n); } private: int ELFHash( const char * key, int n) { unsigned long h = 0; while( *key ) { h = ( h << 4 ) + *key++; unsigned long g = h & 0xF0000000L; if( g ) h^= g >> 24; h &= ~g; } return h % n; } }; /* open hash method - Base HashTable */ template<class Key,class Item, class HashF = DefHashFunc > class _HashTable { public: //MyHashTable(); _HashTable(size_t n = DEFHASHSIZE, double factor = DEFHASHFACTOR) :_size(n),_factor(factor),_hashFunc(new HashF()),currSize(0) { } virtual ~_HashTable() { cout << " Base Hash Table deconstructed!"<<endl; if ( _countof(szBuffer) ) delete _countof(szBuffer); } virtual bool put(const Key &k, const Item & item) =0; virtual bool get(const Key &k, Item & item) =0; virtual bool remove( const Key &k , Item & pItem) =0; virtual Item & operator []( const Key &k ) =0; size_t size() const { return _size;} bool ReSize(size_t nSize) { if ( nSize <= _size ) //not allow a smaller size return false; reHash(nSize); } virtual bool reHash( size_t nSize)=0; protected: size_t _size; // total size double _factor; // the hash factor to determin realloc HashF * _hashFunc; // size_t currSize; // current size }; // A record node to save the key and value template<class Key,class Item> struct Record{ Key key; Item item; ~Record(){ //cout<<"Record "<<key<<":"<<item<<" deleted"<<endl; } Record(const Key k, const Item itm):key(k),item(itm){} Record(const Key k):key(k),item(){} bool operator == (const Record & r) { return this->key == r.key; } }; /*目前只实现简单的Hash功能，对于动态分配暂时不考虑，将来研究SGI STL时再实现 */ template<class Key,class Item, class HashF = DefHashFunc > class HashTable: public _HashTable<Key, Item,HashF> { public: //MyHashTable(); HashTable(size_t n = DEFHASHSIZE, double factor = DEFHASHFACTOR ) :_HashTable<Key,Item,HashF>(n,factor) { _arr = new list<Record<Key,Item>> [n]; } ~HashTable() { delete [] _arr; } bool put(const Key &k, const Item & item) { int h = (*_hashFunc)(k,size()); list<Record<Key,Item>>::iterator it = find(_arr[h].begin(),_arr[h].end(),k); if( it != _arr[h].end()) return false; _arr[h].push_front(Record<Key,Item>(k,item)); return true; } bool get(const Key &k, Item &pitem) { int h = (*_hashFunc)(k,size()); list<Record<Key,Item>>::iterator it = find(_arr[h].begin(),_arr[h].end(),k); if( it == _arr[h].end()) return false; pitem = it->item; return true; } bool remove( const Key &k , Item & pItem) { int h = (*_hashFunc)(k,size()); list<Record<Key,Item>>::iterator it = find(_arr[h].begin(),_arr[h].end(),k); if( it == _arr[h].end()) return false; pItem = it->item; (_arr[h]).erase(it); return true; } Item & operator []( const Key &k ) { int h = (*_hashFunc)(k,size()); list<Record<Key,Item>>::iterator it = find(_arr[h].begin(),_arr[h].end(),k); if( it == _arr[h].end()) { it = _arr[h].insert(it, Record<Key,Item>(k)); } return it->item; } bool reHash( size_t nSize){return false;}; //暂时不实现 private: list<Record<Key,Item>> * _arr ; }; }; #endif

//Dictrionary.h

#ifndef __MINGSPY_DICTINARY_HH__ #define __MINGSPY_DICTINARY_HH__ //#include "AllHeads.h" #include "MyHashTable.h" using namespace mingspy; class _Dictionary { public: virtual int find(const char* ) = 0; virtual int LoadDictionary(const char * DicPatch) = 0; virtual int run() = 0; virtual ~_Dictionary(){} }; class HashDictionary { public: HashDictionary(){} ~HashDictionary(){} int find(const char *); int LoadDictionary(const char * DicPatch); int run() ; private: HashTable<string,string> ht; }; #endif

//AllHeads.h

#ifndef __MINGSPY_ALLHEADS_HH__ #define __MINGSPY_ALLHEADS_HH__ #include "Dictionary.h" #include "MyHashTable.h" #include <iostream> using namespace std; enum { FAILURE = 0,SUCCESS = 1 }; using namespace std; #endif

//Dictionary.cpp

#define _CRTDBG_MAP_ALLOC #include <stdlib.h> #include <crtdbg.h> #include "AllHeads.h" ostream & operator == (ostream & out, string & s) { out<<s.c_str(); return out; } int HashDictionary::LoadDictionary(const char * DicPatch) { const char * filePath =NULL; if( ! DicPatch ) filePath = "./ForeignTrade.dic"; else filePath = DicPatch; ifstream inf(filePath,ios::in); if ( inf.fail()) { cout<< " open "<< inf <<" failed" <<endl; return false; } char buf[1024]; char * p, *q; /*读取文件内容并装入内存（挂到字典树上） */ while( !inf.eof()) { memset(buf, 0, sizeof(buf)); inf.getline(buf, 1024); if( buf[0] == '@') continue; p = strchr(buf, '/t'); if( !p ) { continue; } *p++ = '/0'; q=buf; //while(*q && isalpha(*q)) //{ // q++; //} //if(*q) continue; ht.put(buf,p); //挂到树上 } inf.close(); return SUCCESS; } int HashDictionary::find(const char * key) { if( !key) return NULL; //找到存储单词的节点 string s; bool ret=ht.get(key, s); if ( !ret) { cout<<"["<<key<<"]"<<" Not find!"<<endl; return FAILURE; } else { cout<<s<<endl; } return SUCCESS; } int HashDictionary::run() { _CrtSetDbgFlag ( _CRTDBG_ALLOC_MEM_DF | _CRTDBG_LEAK_CHECK_DF ); char key[100]; memset(key, 0, sizeof(key)); cout<<"Input the word u want to find :"; cin>>key; bool ret; while( key[0] != '#' ) { ret = find(key); memset(key, 0, sizeof(key)); cout<<"Input the word u want to find :"; cin>>key; } return SUCCESS; }