编译原理之正则表达式转NFA

2023-12-19 20:48:34

本文转载自http://chriszz.sinaapp.com/?p=257

输入一个正则表达式，输出一个NFA。

我的做法：输入一个字符串表示正则，输出则是把输出到一个.dot文件中并将dot文件编译成pdf，fedora需要sudo yum install dot，然后evince XXX.pdf就可以查看生成的NFA了。

具体算法是按照龙书上的Tompson算法来的。

废话不多说，放码过来：

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
/*
Author：ChrisZZ(zchrissirhcz@gmail.com)
Time:2013-12-25 14:13:09
输入：正则表达式
输出：自动机
算法步骤：
1.把正则表达式转化为后缀表达式
2.把后缀表达式转化为NFA
3.用dot语言把NFA输出到PDF
参考：
1.Regular Expression Matching Can Be Simple And Fast
http://swtch.com/~rsc/regexp/regexp1.html
2.龙书 chap3.7.4 从正则表达式构造NFA
3.YCC学长的project中dot语言的使用
其他说明：
1.需要安装dot，并添加到系统path中
2.在windows下运行时，控制台因为编码不支持可能导致中文提示无法显示
*/
#include <iostream>
#include <string>
#include <stdio.h>
#include <stack>
#include <string.h>
#include <stdexcept>
#include <stdlib.h>
 
using namespace std;
 
const int Match = 256;
const int Split = 257;//表示epsilon分支
 
struct Paren{//括号结构体
    int natom;
    int nalt;
};
 
string re2post(string re){
    Paren paren;//括号
    stack<struct Paren>parenStk;
    string postExpr="";
    int i, len=re.length();
    int nalt=0, natom=0;
    const string invalidRegExp = "非法的正则表达式";
    for(i=0; i<len; i++){
        if(isspace(re[i])) continue;
        if(isalpha(re[i])){
            if(natom>1){
                natom--;
                postExpr = postExpr + ‘.‘;
            }
            natom++;
            postExpr = postExpr + re[i];
        }
        else if(re[i]==‘(‘){
            if(natom>1){
                postExpr = postExpr + ‘.‘;
            }
            paren.natom = natom;
            paren.nalt = nalt;
            parenStk.push(paren);
            nalt = 0;
            natom = 0;
        }
        else if(re[i]==‘)‘){
            if(natom==0 || parenStk.empty())
                throw runtime_error(invalidRegExp+":括号不匹配");
            while(--natom>0){//比如((a|b)(c|d))模式，当上一次匹配完倒数第二个右括号后，natom为2，需要添加‘.‘
                postExpr = postExpr + ‘.‘;
            }
            while(nalt-->0){
                postExpr = postExpr + ‘|‘;
            }
            paren=parenStk.top();
            parenStk.pop();
            natom = paren.natom;
            nalt = paren.nalt;
            natom++;
        }
        else if(re[i]==‘*‘){
            if(natom==0)
                throw runtime_error(invalidRegExp+":提前出现‘*‘");
            postExpr = postExpr + re[i];
        }
        else if(re[i]==‘|‘){
            if(natom==0) throw runtime_error(invalidRegExp+":提前出现‘|‘");
            while(--natom>0){
                postExpr = postExpr + ‘.‘;
            }
            nalt++;
        }
        else
            throw runtime_error(invalidRegExp);
    }
    if(!parenStk.empty())
        throw runtime_error(invalidRegExp+":括号不匹配");
    while(--natom>0){
        postExpr = postExpr + ‘.‘;
    }
    while(nalt-->0){
        postExpr = postExpr + ‘|‘;
    }
    return postExpr;
}
 
class NFA;
 
/*
* c<256表示edge权重为c；
* c=256表示终结状态，匹配成功
* c=257表示分支（split）
*/
class State{
    friend class NFA;
    friend void nfa2graph(State* head, const string& re);
public:
    State(int c=256, State* out=NULL, State* out1=NULL){
        this->c = c;
        this->out = out;
        this->out1 = out1;
        this->id = 0;
    }
    void setId(int id){
        this->id = id;
    }
 
private:
    int c;
    int id;//状态的编号
    State* out;//从本状态出去的状态集合的头指针
    State* out1;//两个分支的情况
};
 
class NFA{
public:
    NFA(){
        head = NULL;
        tail = NULL;
    }
    NFA(const int& c){
        tail = new State(Match, NULL, NULL);
        head = new State(c, tail, NULL);
    }
    void doCat(NFA& nfa){
        tail->out = nfa.head;
        tail->c = Split;
        tail = nfa.tail;
        nfa.head = NULL;
        nfa.tail = NULL;
    }
    void doUnion(NFA& nfa){
        State* newHead = new State(Split, head, nfa.head);
        State* newTail = new State(Match, NULL, NULL);
        tail->c = Split;
        tail->out = newTail;
        nfa.tail->c = Split;
        nfa.tail->out = newTail;
        tail = newTail;
        head = newHead;
        nfa.head = NULL;
        nfa.tail = NULL;
    }
    void doStar(){
        State* newTail = new State(Match, NULL, NULL);
        State* newHead = new State(Split, head, newTail);
        tail->c = Split;
        tail->out = newTail;
        tail->out1 = head;
        tail = newTail;
        head = newHead;
    }
 
    void nfa2graph(const string& re){
        char myfile[100];
        printf("请输入一个文件名，用来保存生成的NFA-graph(不必提供后缀):\n");
        scanf("%s", myfile);
        printf("已将DOT文件存储在\"%s.dot\",\n", myfile);
        printf("PDF文件则存储在\"%s.dot.pdf\".\n", myfile);
        int i;
        while(myfile[i]!=‘\0‘)
            i++;
        myfile[i] = ‘.‘;
        myfile[i+1] = ‘d‘;
        myfile[i+2] = ‘o‘;
        myfile[i+3] = ‘t‘;
        myfile[i+4] = ‘\0‘;
 
        FILE *file = fopen(myfile, "w");
 
        fputs("digraph {\n", file);
        fputs("\t\"", file);
        int len=re.length();
        for(i=0; i<len; i++){
            fprintf(file, "%c", re[i]);
        }
 
        fputs("\" [shape = plaintext]\n", file);
        fputs("\trankdir = LR\n", file);
        fputs("\t\"\" [shape = point]\n", file);
        fputs("\t\"\" -> 1 [label = Start]\n\n", file);
 
        int id = 1;
 
        char circle[2000];
        memset(circle, 0, sizeof(circle));
        State* p;
        stack<State*> staStk;
 
        head->setId(id++);
        staStk.push(head);
 
        while(!staStk.empty()){
            p = staStk.top();
            staStk.pop();
            char flag = 1;
            cout << "p->c=" << p->c << endl;
            if(p->c < Match){
                cout << "p->out->id=" << p->out->id << endl;
                if(p->out->id==0){
                    p->out->id = id++;
                    cout << "id=" << id << endl;                }
                else
                    flag = 0;
                fprintf(file, "\t%d -> %d [label = \"%c\"]\n", p->id, (p->out)->id, p->c);
                State *what = p->out;
                if(flag) //push(*what);
                    staStk.push(what);
            } else if(p->c == Match){
                circle[p->id] = 1;
            } else{     //对应Split的情形
                if(p->out->id==0)
                    p->out->id = id++;
                else
                    flag = 0;
                fprintf(file, "\t%d -> %d [label = <ε>]\n", p->id, p->out->id);
                State *what = p->out;
                if(flag) staStk.push(what);
 
                if(p->out1!=NULL){
                    flag = 1;
 
                    if(p->out1->id==0)
                        p->out1->id = id++;
                    else
                        flag = 0;
                    fprintf(file, "\t%d -> %d [label = <ε>]\n", p->id, p->out1->id);
                    what = p->out1;
                    if(flag) staStk.push(what);
                }
            }
        }
 
        for(i=1; i<id; i++){
            fprintf(file, "\t%d [shape = circle", i);
            if(circle[i])
                fputs(", peripheries = 2", file);
            fprintf(file, "]\n");
        }
 
        fputs("}", file);
        fclose(file);
 
        char cmd[108];
        sprintf(cmd, "dot %s -O -Tpdf", myfile);
        if(system(cmd)==0){
            printf("成功生成pdf图像！\n");
            //printf("Linux用户可以使用evince file.pdf &命令打开~\n");
        }
        else
            printf("悲剧！生成pdf图像时出现错误..\n");
    }
private:
    State* head;
    State* tail;
};
 
NFA post2nfa(const string& postExpr){
    stack<NFA> nfaStk;
    NFA e1, e2, e;
    int i, len=postExpr.length();
    for(i=0; i<len; i++){
        switch(postExpr[i]){
        case ‘.‘:
            e2 = nfaStk.top();
            nfaStk.pop();
            e1 = nfaStk.top();
            nfaStk.pop();
            e1.doCat(e2);
            nfaStk.push(e1);
            break;
        case ‘|‘:
            e2 = nfaStk.top();
            nfaStk.pop();
            e1 = nfaStk.top();
            nfaStk.pop();
            e1.doUnion(e2);
            nfaStk.push(e1);
            break;
        case ‘*‘:
            e = nfaStk.top();
            nfaStk.pop();
            e.doStar();
            nfaStk.push(e);
            break;
        default://
            NFA alpha(postExpr[i]);
            nfaStk.push(alpha);
        }
    }
    e = nfaStk.top();
    nfaStk.pop();
    if(!nfaStk.empty())
        throw runtime_error("未知错误");
    return e;
}
 
int main(){
    string re;
    while(true){
        cout << "请输入一个正则表达式:\n";
        cin >> re;
        string postExpr = re2post(re);
        cout << "postExpr is : " << postExpr << endl;
        NFA nfa = post2nfa(postExpr);
        nfa.nfa2graph(re);
        cout << "继续吗?(y/n)\n" << endl;
        char c;
        cin >> c;
        while(c!=‘y‘ && c!=‘n‘){
            cout << "请输入‘y‘或‘n‘.\n";
            c=getchar();
        }
        if(c==‘n‘)
            break;
    }
    cout << "Bye~\n";
    return 0;
}

编译原理之正则表达式转NFA

上一篇：插入排序


下一篇：[OpenCV]获取摄像头视频