searchRReserve(char reserve[],char s)， searchReserve(char reserve[][20],char s[])函数用于避免出现重复，既每出现一个token类别都到原有的类别数组中搜索一下。；输出在Scanner（）函数中直接传输到文件。
2.3 流程图
图 3 token序列识别流程图
三、源程序代码：（加入注释）
#include <iostream>
#include<stdio.h>
#include<string.h>
#include<stdlib.h>
using namespace std;
//保留字表
static char reserveWord[32][20] =
{
"auto", "break", "case", "char", "const", "continue",
"default", "do", "double", "else", "enum", "extern",
"float", "for", "goto", "if", "int", "long",
"register", "return", "short", "signed", "sizeof", "static",
"struct", "switch", "typedef", "union", "unsigned", "void",
"volatile", "while"
};
//界运算符
static char operatorOrDelimiter[36][10]=
{
"+","-","*","http://www.likecs.com/","<","<=",">",">=","=","==",
"!=",";","(",")","^",",","\"","\'","#","&",
"&&","|","||","%","~","<<",">>","[","]","{",
"}","\\",".","\?",":","!"
};
static char IDentifierTbl[1000][20]= {""}; //标识符表i
static char SingleChar[1000]= {""}; //单个字符表ct
static char StringChar[1000][20]= {""}; //字符串表S
static char ConstantTbl[1000][20]= {""}; //常数表C
int countct=0;//单个字符表ct
int countc=0;//常数表C
int counts=0; //字符串表S
int countt=0;//标识符表i
char resourceProject[10000];//输入的源程序存放处，最大可以存放10000个字符。
//查找保留字，若成功查找，则返回种别码
//否则返回-1，代表查找不成功，即为标识符
int searchReserve(char reserveWord[ ][20], char s[]);
bool IsLetter(char letter);
bool IsDigit(char digit);
void filterResource(char r[],int pProject);
void Scanner(int &syn,char resourceProject[],char token[],int &pProject);
int searchReserve(char reserve[][20],char s[])
{
for(int i=0; reserve[i][0]!='\0'; i++)
{
if(strcmp(reserve[i],s)==0)
return i+1;
}
return -1;
}
int searchRReserve(char reserve[],char s)
{
for(int i=0; reserve[i]!='\0'; i++)
{
if(reserve[i]==s)
return i+1;
}
return -1;
}
bool IsLetter(char letter)
{
if(letter>='a'&&letter<='z'||letter>='A'&&letter<='Z'||letter=='_')
return true;
else
return false;
}
bool IsDigit(char digit)
{
if(digit>='0'&&digit<='9')
return true;
else
return false;
}
void filterResource(char r[],int pProject)
{
char tempString[10000];
int count=0;
for(int i=0; i<=pProject; i++)
{
if(r[i]=='http://www.likecs.com/'&&r[i+1]=='http://www.likecs.com/')
while(r[i]!='\n')
i++;
if(r[i]=='http://www.likecs.com/'&&r[i+1]=='*')
{
i+=2;
while(r[i]!='*'||r[i+1]!='http://www.likecs.com/')
{
i++;
if(r[i]=='$')
{
cout<<"注释出错"<<endl;
exit(0);
}
}
i+=2;
}
if(r[i]!='\n'&&r[i]!='\t'&&r[i]!='\v'&&r[i]!='\r')
{
tempString[count++]=r[i];
}
}
tempString[count]='\0';
strcpy(r,tempString);
}
void Scanner(int &syn,char resourceProject[],char token[],int &pProject)
{
FILE *fp1;
if((fp1=fopen("E:\\2017.txt","at"))==NULL)
{
cout<<"cam not open";
exit(0);
}
int i,count=0;//count用来做token指示器，收集有用字符
char ch;
ch=resourceProject[pProject];
while(ch==' ')
{
pProject++;
ch=resourceProject[pProject];
}
for(i=0; i<20; i++)
{
token[i]='\0';//收集前先清零
}
if(IsLetter(resourceProject[pProject]))
{
token[count++]=resourceProject[pProject];
pProject++;
while(IsLetter(resourceProject[pProject])||IsDigit(resourceProject[pProject]))
{
token[count++]=resourceProject[pProject];
pProject++;
}
token[count]='\0';
syn=searchReserve(reserveWord,token);
if(syn!=-1)
{
cout<<"{"<<"k"<<","<<syn<<","<<reserveWord[syn-1]<<"}"<<endl;
fprintf(fp1, "{k , %d ,%s }\n", syn,reserveWord[syn-1]);
}
if(syn==-1)
{
syn=searchReserve(IDentifierTbl,token);
if(syn==-1)
{
strcpy(IDentifierTbl[countt++],token);
syn=countt;
}
cout<<"{"<<"i"<<","<<syn<<","<<IDentifierTbl[syn-1]<<"}"<<endl;
fprintf(fp1, "{i , %d ,%s }\n", syn,IDentifierTbl[syn-1]);
}
}
else if(IsDigit(resourceProject[pProject]))
{
while(IsDigit(resourceProject[pProject]))
{
token[count++]=resourceProject[pProject];
pProject++;
}
token[count]='\0';
syn=searchReserve(ConstantTbl,token);
if(syn==-1)
{
strcpy(ConstantTbl[countc++],token);
syn=countc;
}
cout<<"{"<<"c"<<","<<syn<<","<<ConstantTbl[syn-1]<<"}"<<endl;
fprintf(fp1, "{c , %d ,%s }\n", syn,ConstantTbl[syn-1]);
}
else if(resourceProject[pProject]=='\'')
{
pProject++;
if(IsLetter(resourceProject[pProject]))
{
pProject++;
if(resourceProject[pProject]=='\'')
{
pProject--;
syn=searchRReserve(SingleChar,resourceProject[pProject]);
if(syn==-1)
{
SingleChar[countct++]=resourceProject[pProject];
syn=countct;
}
}
}
cout<<"{"<<"CT"<<","<<syn<<","<<resourceProject[syn-1]<<"}"<<endl;
fprintf(fp1, "{CT , %d ,%c }\n", syn,resourceProject[syn-1]);
pProject+=2;
}
else if(resourceProject[pProject]=='"')
{
pProject++;
while(resourceProject[pProject]!='"')
{
token[count++]=resourceProject[pProject];
pProject++;
}
token[count]='\0';
syn=searchReserve(StringChar,token);
if(syn==-1)
{
strcpy(StringChar[counts++],token);
syn=counts;
}
cout<<"{"<<"S"<<","<<syn<<","<<StringChar[syn-1]<<"}"<<endl;
fprintf(fp1, "{S , %d ,%s }\n", syn,StringChar[syn-1]);
pProject++;
}
else if (ch == '+' || ch == '-' || ch == '*' || ch == 'http://www.likecs.com/' || ch == ';' || ch == '(' || ch == ')' || ch == '^'
|| ch == ',' || ch == '\"' || ch == '\'' || ch == '~' || ch == '#' || ch == '%' || ch == '['
|| ch == ']' || ch == '{' || ch == '}' || ch == '\\' || ch == '.' || ch == '\?' || ch == ':')
{
//若为运算符或者界符，查表得到结果
token[0] = resourceProject[pProject];
token[1] = '\0';//形成单字符串
for (i = 0; i<36; i++)
{
//查运算符界符表
if (strcmp(token, operatorOrDelimiter[i]) == 0)
{
syn = i+1;//获得种别码，使用了一点技巧，使之呈线性映射
break;//查到即推出
}
}
cout<<"{"<<"p"<<","<<syn<<","<<operatorOrDelimiter[syn-1]<<"}"<<endl;
fprintf(fp1, "{p , %d ,%s }\n", syn,operatorOrDelimiter[syn-1]);
pProject++;//指针下移，为下一扫描做准备
}
else if(resourceProject[pProject]=='<')
{
//<,<=,<<
pProject++;
if(resourceProject[pProject]=='=') syn=6;
else if(resourceProject[pProject]=='<')
syn=26;
else
{
pProject--;
syn=5;
}
cout<<"{"<<"p"<<","<<syn<<","<<operatorOrDelimiter[syn-1]<<"}"<<endl;
fprintf(fp1, "{p , %d ,%s }\n", syn,operatorOrDelimiter[syn-1]);
pProject++;
}
else if(resourceProject[pProject]=='>')
{
//>,>=,>>
pProject++;
if(resourceProject[pProject]=='=') syn=8;
else if(resourceProject[pProject]=='>') syn=27;
else
{
pProject--;
syn=7;
}
cout<<"{"<<"p"<<","<<syn<<","<<operatorOrDelimiter[syn-1]<<"}"<<endl;
fprintf(fp1, "{p , %d ,%s }\n", syn,operatorOrDelimiter[syn-1]);
pProject++;
}
else if(resourceProject[pProject]=='=')
{
//=,==
pProject++;
if(resourceProject[pProject]=='=') syn=10;
else
{
pProject--;
syn=9;
}
cout<<"{"<<"p"<<","<<syn<<","<<operatorOrDelimiter[syn-1]<<"}"<<endl;
fprintf(fp1, "{p , %d ,%s }\n", syn,operatorOrDelimiter[syn-1]);
pProject++;
}
else if(resourceProject[pProject]=='!')
{
// !,!=
pProject++;
if(resourceProject[pProject]=='=') syn=11;
else
{
pProject--;
syn=36;
}
cout<<"{"<<"p"<<","<<syn<<","<<operatorOrDelimiter[syn-1]<<"}"<<endl;
fprintf(fp1, "{p , %d ,%s }\n", syn,operatorOrDelimiter[syn-1]);
pProject++;
}
else if(resourceProject[pProject]=='&')
{
//&,&&
pProject++;
if(resourceProject[pProject]=='&') syn=21;
else
{
pProject--;
syn=20;
}
cout<<"{"<<"p"<<","<<syn<<","<<operatorOrDelimiter[syn-1]<<"}"<<endl;
fprintf(fp1, "{p , %d ,%s }\n", syn,operatorOrDelimiter[syn-1]);
pProject++;
}
else if(resourceProject[pProject]=='|')
{
//|,||
pProject++;
if(resourceProject[pProject]=='|') syn=23;
else
{
pProject--;
syn=22;
}
cout<<"{"<<"p"<<","<<syn<<","<<operatorOrDelimiter[syn-1]<<"}"<<endl;
fprintf(fp1, "{p , %d ,%s }\n", syn,operatorOrDelimiter[syn-1]);
pProject++;
}
else if(resourceProject[pProject]=='$')
{
//$
syn=0;
}
else
{
cout<<"error: no exist "<<resourceProject[pProject]<<endl;
exit(0);
}
fclose(fp1);
}
int main()
{
char token[20]= {0}; //每次扫描的时候存储已经扫描的结果。