我最近正在学编译原理，我有c语言实现的词法分析程序，不知可不可以，识别的是TEST语言的单词。
创新互联长期为上千家客户提供的网站建设服务，团队从业经验10年，关注不同地域、不同群体，并针对不同对象提供差异化的产品和服务；打造开放共赢平台，与合作伙伴共同营造健康的互联网生态环境。为云岩企业提供专业的做网站、成都做网站，云岩网站改版等技术服务。拥有10余年丰富建站经验和众多成功案例,为您定制开发。
#includestdio.h
#includectype.h
#includestring.h
#define keywordSum 8
char * keyword[keywordSum] = {"do", "else", "for", "if", "int", "read", "while", "write"};
char singleword[50] = "+-*(){};,:";
char doubleword[10] = "=!|";
char Scanin[300], Scanout[300];
FILE * fin, * fout;
int binaryFind(int low, int high, char * c1, char ** c2) {
int mid;
if(low high) return -1;
mid = (low+high)/2;
if(strcmp(c1, c2[mid]) == 0) return mid;
else if(strcmp(c1, c2[mid]) 0) return binaryFind(mid+1, high, c1, c2);
else return binaryFind(low, mid-1, c1, c2);
}
int TESTscan() {
char ch, token[40];
int es = 0, j, n;
printf("请输入源文件名(包括路径):");
scanf("%s", Scanin);
printf("请输入词法分析输出文件名(包括路径):");
scanf("%s", Scanout);
if((fin=fopen(Scanin, "r")) == NULL) {
printf("\n打开词法分析输入文件出错!\n");
return 1;
}
if((fout=fopen(Scanout, "w")) == NULL) {
printf("\n创建词法分析输出文件出错!\n");
return 2;
}
// printf("%c", getc(fin));
ch = getc(fin);
while(ch != EOF) {
while(ch==' ' || ch=='\n' || ch=='\t') {
ch = getc(fin);
}
if(isalpha(ch)) { //标识符
token[0] = ch;
j = 1;
ch = getc(fin);
while(isalnum(ch)) { //判断当前字符是否是字母或数字
token[j++] = ch;
ch = getc(fin);
}
token[j] = '\0';
// printf("%s", token);
n = binaryFind(0, keywordSum-1, token, keyword);
if(n 0 ) {
fprintf(fout, "%s\t%s\n", "ID", token);
} else {
fprintf(fout, "%s\t%s\n", token, token);
}
} else if(isdigit(ch)) { //数字
token[0] = ch;
j = 1;
ch = getc(fin);
while(isdigit(ch)) {
token[j++] = ch;
ch = getc(fin);
}
token[j] = '\0';
fprintf(fout, "%s\t%s\n", "NUM", token);
} else if(strchr(singleword, ch) 0) { //singleword
token[0] = ch;
token[1] = '\0';
ch = getc(fin);
fprintf(fout, "%s\t%s\n", token, token);
} else if(strchr(doubleword, ch) 0) { //doubleword
token[0] = ch;
ch = getc(fin);
if(ch=='=' (token[0]==''||token[0]=='' || token[0] == '!')) {
token[1] = ch;
token[2] = '\0';
ch = getc(fin);
} else if((ch=='')||(ch=='|')||(ch=='=') ch==token[0]) {
token[1] = ch;
token[2] = '\0';
ch = getc(fin);
} else {
token[1] = '\0';
}
fprintf(fout, "%s\t%s\n", token, token);
} else if(ch == '/') { //注释
ch = getc(fin);
if(ch == '*') {
char ch1;
ch1 = getc(fin);
do {
ch = ch1;
ch1 = getc(fin);
} while((ch!='*'||ch1!='/') ch1!=EOF);
ch = getc(fin);
} else {
token[0] = '/';
token[1] = '\0';
fprintf(fout, "%s\t%s\n", token, token);
}
} else {
token[0] = ch;
token[1] = '\0';
ch = getc(fin);
es = 3;
fprintf(fout, "%s\t%s\n", "ERROR", token);
}
}
fclose(fin);
fclose(fout);
return es;
}
void main() {
int es = 0;
es = TESTscan();
if(es 0) {
printf("词法分析有错, 编译停止!\n");
} else {
printf("词法分析成功!\n");
}
}
首先看下我们要分析的代码段如下：
输出结果如下：
输出结果(a).PNG
输出结果(b).PNG
输出结果(c).PNG
括号里是一个二元式：（单词类别编码，单词位置编号）
代码如下：
?
1234567891011121314
package Yue.LexicalAnalyzer; import java.io.*; /* * 主程序 */public class Main { public static void main(String[] args) throws IOException { Lexer lexer = new Lexer(); lexer.printToken(); lexer.printSymbolsTable(); }}
?
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283
package Yue.LexicalAnalyzer; import java.io.*;import java.util.*; /* * 词法分析并输出 */public class Lexer { /*记录行号*/ public static int line = 1; /*存放最新读入的字符*/ char character = ' '; /*保留字*/ HashtableString, KeyWord keywords = new HashtableString, KeyWord(); /*token序列*/ private ArrayListToken tokens = new ArrayListToken(); /*符号表*/ private ArrayListSymbol symtable = new ArrayListSymbol(); /*读取文件变量*/ BufferedReader reader = null; /*保存当前是否读取到了文件的结尾*/ private Boolean isEnd = false; /* 是否读取到文件的结尾 */ public Boolean getReaderState() { return this.isEnd; } /*打印tokens序列*/ public void printToken() throws IOException { FileWriter writer = new FileWriter("E:\\lex.txt"); System.out.println("词法分析结果如下："); System.out.print("杜悦-2015220201031\r\n\n"); writer.write("杜悦-2015220201031\r\n\r\n"); while (getReaderState() == false) { Token tok = scan(); String str = "line " + tok.line + "\t(" + tok.tag + "," + tok.pos + ")\t\t" + tok.name + ": " + tok.toString() + "\r\n"; writer.write(str); System.out.print(str); } writer.flush(); } /*打印符号表*/ public void printSymbolsTable() throws IOException { FileWriter writer = new FileWriter("E:\\symtab1.txt"); System.out.print("\r\n\r\n符号表\r\n"); System.out.print("编号\t行号\t名称\r\n"); writer.write("符号表\r\n"); writer.write("编号 " + "\t行号 " + "\t名称 \r\n"); IteratorSymbol e = symtable.iterator(); while (e.hasNext()) { Symbol symbol = e.next(); String desc = symbol.pos + "\t" + symbol.line + "\t" + symbol.toString(); System.out.print(desc + "\r\n"); writer.write(desc + "\r\n"); } writer.flush(); } /*打印错误*/ public void printError(Token tok) throws IOException{ FileWriter writer = new FileWriter("E:\\error.txt"); System.out.print("\r\n\r\n错误词法如下：\r\n"); writer.write("错误词法如下：\r\n"); String str = "line " + tok.line + "\t(" + tok.tag + "," + tok.pos + ")\t\t" + tok.name + ": " + tok.toString() + "\r\n"; writer.write(str); } /*添加保留字*/ void reserve(KeyWord w) { keywords.put(w.lexme, w); } public Lexer() { /*初始化读取文件变量*/ try { reader = new BufferedReader(new FileReader("E:\\输入.txt")); } catch (IOException e) { System.out.print(e); } /*添加保留字*/ this.reserve(KeyWord.begin); this.reserve(KeyWord.end); this.reserve(KeyWord.integer); this.reserve(KeyWord.function); this.reserve(KeyWord.read); this.reserve(KeyWord.write); this.reserve(KeyWord.aIf); this.reserve(KeyWord.aThen); this.reserve(KeyWord.aElse); } /*按字符读*/ public void readch() throws IOException { character = (char) reader.read(); if ((int) character == 0xffff) { this.isEnd = true; } } /*判断是否匹配*/ public Boolean readch(char ch) throws IOException { readch(); if (this.character != ch) { return false; } this.character = ' '; return true; } /*数字的识别*/ public Boolean isDigit() throws IOException { if (Character.isDigit(character)) { int value = 0; while (Character.isDigit(character)) { value = 10 * value + Character.digit(character, 10); readch(); } Num n = new Num(value); n.line = line; tokens.add(n); return true; } else return false; } /*保留字、标识符的识别*/ public Boolean isLetter() throws IOException { if (Character.isLetter(character)) { StringBuffer sb = new StringBuffer(); /*首先得到整个的一个分割*/ while (Character.isLetterOrDigit(character)) { sb.append(character); readch(); } /*判断是保留字还是标识符*/ String s = sb.toString(); KeyWord w = keywords.get(s); /*如果是保留字的话，w不应该是空的*/ if (w != null) { w.line = line; tokens.add(w); } else { /*否则就是标识符，此处多出记录标识符编号的语句*/ Symbol sy = new Symbol(s); Symbol mark = sy; //用于标记已存在标识符 Boolean isRepeat = false; sy.line = line; for (Symbol i : symtable) { if (sy.toString().equals(i.toString())) { mark = i; isRepeat = true; } } if (!isRepeat) { sy.pos = symtable.size() + 1; symtable.add(sy); } else if (isRepeat) { sy.pos = mark.pos; } tokens.add(sy); } return true; } else return false; } /*符号的识别*/ public Boolean isSign() throws IOException { switch (character) { case '#': readch(); AllEnd.allEnd.line = line; tokens.add(AllEnd.allEnd); return true; case '\r': if (readch('\n')) { readch(); LineEnd.lineEnd.line = line; tokens.add(LineEnd.lineEnd); line++; return true; } case '(': readch(); Delimiter.lpar.line = line; tokens.add(Delimiter.lpar); return true; case ')': readch(); Delimiter.rpar.line = line; tokens.add(Delimiter.rpar); return true; case ';': readch(); Delimiter.sem.line = line; tokens.add(Delimiter.sem); return true; case '+': readch(); CalcWord.add.line = line; tokens.add(CalcWord.add); return true; case '-': readch(); CalcWord.sub.line = line; tokens.add(CalcWord.sub); return true; case '*': readch(); CalcWord.mul.line = line; tokens.add(CalcWord.mul); return true; case '/': readch(); CalcWord.div.line = line; tokens.add(CalcWord.div); return true; case ':': if (readch('=')) { readch(); CalcWord.assign.line = line; tokens.add(CalcWord.assign); return true; } break; case '': if (readch('=')) { readch(); CalcWord.ge.line = line; tokens.add(CalcWord.ge); return true; } break; case '': if (readch('=')) { readch(); CalcWord.le.line = line; tokens.add(CalcWord.le); return true; } break; case '!': if (readch('=')) { readch(); CalcWord.ne.line = line; tokens.add(CalcWord.ne); return true; } break; } return false; } /*下面开始分割关键字，标识符等信息*/ public Token scan() throws IOException { Token tok; while (character == ' ') readch(); if (isDigit() || isSign() || isLetter()) { tok = tokens.get(tokens.size() - 1); } else { tok = new Token(character); printError(tok); } return tok; }}
StringTokenizer 是出于兼容性的原因而被保留的遗留类（虽然在新代码中并不鼓励使用它）。建议所有寻求此功能的人使用 String 的 split 方法或 java.util.regex 包。
下面的示例阐明了如何使用 String.split 方法将字符串分解为基本标记：
String[] result = "this is a test".split("\\s");
for (int x=0; xresult.length; x++)
System.out.println(result[x]);
输出以下字符串：
this
is
a
test
首先你得定义 XML的语言结构， 比如 开始标签结束标签标签中的属性标签中的属性值标签夹住的内容 然后定义他们的关系，用表达式表达出来 最后写一个程序识别你定义的元素，然后根据关系验证元素之间的关系是否正常。 简而言之，先抽象每个元素，然后建立元素的先后顺序关系， 最后写程序识别元素，根据关系验证他们出现的顺序 好比，定义主语，谓语，宾语 然后定义主谓宾的出现顺序， 入 主语，谓语，然后宾语 最后写个程序，能识别，主谓宾，然后再验证他们的顺序，就这样的。 内容较多， 呵呵。 需要用到的技术： 正则表达式，用来识别一些关键性的元素 还有那个编译原理里面定义的什么来着的 如 关键词=字母|（字母，数字） 忘了，你可以翻书看看
我也做过这个作业
package source;
import java.util.LinkedList;
public class LexicalAnalysis
{
//私有变量声明
private LinkedListWord optr = new LinkedListWord();
private String exp;
//词法分析
public LinkedListWord lexical_analysis(String exp)
{
char ch = '\0'; //当前文件指针内容
int index = 0; //文件指针
StringBuffer strToken = new StringBuffer("");
//扫描处理字符串
while(true)
{
ch = exp.charAt(index);
index++;
//标识符(字母开头，数字或字符组成)
if(Character.isLetter(ch))
{
while(Character.isLetter(ch) || Character.isDigit(ch))
{
strToken.append(ch);
ch = exp.charAt(index);
index++;
}
index--;
String str = strToken.toString();
if(str.equals("if"))
optr.add(new Word(str, 13));
else if(str.equals("else"))
optr.add(new Word(str, 14));
else if(str.equals("then"))
optr.add(new Word(str, 15));
else
optr.add(new Word(str, 26));
}
//数字
else if(Character.isDigit(ch))
{
while(Character.isDigit(ch))
{
strToken.append(ch);
ch = exp.charAt(index);
index++;
}
index--;
optr.add(new Word(strToken.toString(), 26));
}
//加号或自加
else if(ch == '+')
{
ch = exp.charAt(index);
index++;
if(ch == '+')
optr.add(new Word("++", 21));
else if(ch == '=')
optr.add(new Word("+=", 16));
else
{
index--;
optr.add(new Word("+", 19));
}
}
//加号或自加
else if(ch == '-')
{
ch = exp.charAt(index);
index++;
if(ch == '-')
optr.add(new Word("--", 21));
else if(ch == '=')
optr.add(new Word("-=", 16));
else
{
index--;
optr.add(new Word("-", 19));
}
}
//乘法或乘幂
else if(ch == '*')
{
ch = exp.charAt(index);
index++;
if(ch == '*')
optr.add(new Word("**", 20));
else if(ch == '=')
optr.add(new Word("*=", 16));
else
{
index--;
optr.add(new Word("*", 20));
}
}
//除法或注释
else if(ch == '/')
{
ch = exp.charAt(index);
index++;
//多行注释
if(ch == '*')
{
while(true)
{
ch = exp.charAt(index);
index++;
if(ch == '*')
{
ch = exp.charAt(index);
index++;
if(ch == '/') break;
else if(ch == '\n')
{
exp = Input.newLine();
index = 0;
ch = exp.charAt(index);
index++;
}
else index--;
}
else if(ch == '#')
{
int tIndex = index - 1;
if(exp.length() tIndex+9)
{
String end = exp.substring(tIndex, tIndex+9);
if(end.equals("#?e_N_d?#")) break;
}
else
{
System.out.println("非法符号\'#\'后的语句忽略！");
exp = Input.newLine();
index = 0;
break;
}
}
else if(ch == '\n')
{
exp = Input.newLine();
index = 0;
}
}
}
//单行注释
else if(ch == '/')
break;
else if(ch == '=')
optr.add(new Word("/=", 16));
else
{
index--;
optr.add(new Word("/", 20));
}
}
//大于或大于等于或右移
else if(ch == '')
{
ch = exp.charAt(index);
index++;
if(ch == '=')
optr.add(new Word("=", 18));
else if(ch == '')
optr.add(new Word("", 20));
else
{
index--;
optr.add(new Word("", 18));
}
}
//小于或小于等于或左移
else if(ch == '')
{
ch = exp.charAt(index);
index++;
if(ch == '=')
optr.add(new Word("=", 18));
else if(ch == '')
optr.add(new Word("", 20));
else
{
index--;
optr.add(new Word("", 18));
}
}
//赋值或等于
else if(ch == '=')
{
ch = exp.charAt(index);
index++;
if(ch == '=')
optr.add(new Word("==", 18));
else
{
index--;
optr.add(new Word("=", 16));
}
}
//或运算按位或
else if(ch == '|')
{
ch = exp.charAt(index);
index++;
if(ch == '|')
optr.add(new Word("||", 17));
else
{
index--;
optr.add(new Word("|", 20));
}
}
//与运算
else if(ch == '')
{
ch = exp.charAt(index);
index++;
if(ch == '')
optr.add(new Word("", 17));
else
{
index--;
optr.add(new Word("", 20));
}
}
//非运算或不等于
else if(ch == '!')
{
ch = exp.charAt(index);
index++;
if(ch == '=')
optr.add(new Word("!=", 18));
else
{
index--;
optr.add(new Word("!", 21));
}
}
//按位亦或
else if(ch == '^')
optr.add(new Word("^", 20));
//取模运算
else if(ch == '%')
optr.add(new Word("%", 20));
//左括号
else if(ch == '(')
optr.add(new Word("(", 22));
//右括号
else if(ch == ')')
optr.add(new Word(")", 23));
//左大括号
else if(ch == '{')
optr.add(new Word("{", 24));
//右大括号
else if(ch == '}')
optr.add(new Word("}", 25));
//结束扫描标志为：#?e_N_d?#
else if(ch == '\n')
{
break;
}
else if(ch == '#')
{
int tIndex = index - 1;
if(exp.length() tIndex+9)
{
String end = exp.substring(tIndex, tIndex+9);
if(end.equals("#?e_N_d?#"))
{
optr.add(new Word("#", 27));
break;
}
}
else
{
System.out.println("非法符号\'#\'后的语句忽略！");
optr.add(new Word("#", 27));
break;
}
}
//清空扫描串
strToken.setLength(0);
}
return optr;
}
}
