词法分析程序（Lexical Analyzer）要求：
- 从左至右扫描构成源程序的字符流
- 识别出有词法意义的单词（Lexemes）
- 返回单词记录（单词类别，单词本身）
- 滤掉空格
- 跳过注释
- 发现词法错误
程序结构：
输入：字符流（什么输入方式，什么数据结构保存）
处理：
–遍历（什么遍历方式）
–词法规则
输出：单词流（什么输出形式）
–二元组
单词类别：
1.标识符（10）
2.无符号数（11）
3.保留字（一词一码）
4.运算符（一词一码）
5.界符（一词一码）
|
单词符号 |
种别码 |
单词符号 |
种别码 |
|
begin |
1 |
: |
17 |
|
if |
2 |
:= |
18 |
|
then |
3 |
< |
20 |
|
while |
4 |
<= |
21 |
|
do |
5 |
<> |
22 |
|
end |
6 |
> |
23 |
|
l(l|d)* |
10 |
>= |
24 |
|
dd* |
11 |
= |
25 |
|
+ |
13 |
; |
26 |
|
- |
14 |
( |
27 |
|
* |
15 |
) |
28 |
|
/ |
16 |
# |
0 |
答案
1.词法分析程序源码
注：必须使用支持C99标准的编译器，否则无法编译成功，在这里我使用的是gcc 9.1编译器
1 #include <stdio.h>
2 #include <stdlib.h>
3 #include <string.h>
4 #include <regex.h>
5
6 void analyzer(char* code_string, int len);
7 int table(char* res);
8 int8_t is_char_or_digit(char res);
9 char* file_to_string(const char* path, int *len);
10 int re_match(const char* pattern, const char* res);
11
12 int main(int argc, char *args[])
13 {
14 int content_len;
15 char* content = file_to_string(args[1], &content_len);
16
17 analyzer(content, content_len);
18
19 free(content);
20 return 0;
21 }
22
23 void analyzer(char* code_string, int len)
24 {
25 // Create a buffer to store a single character.
26 char* character = (char*)calloc(1, sizeof(char));
27
28 for (int i = 0; i < len; ++i)
29 {
30 // Value the buffer with current character.
31 *character = code_string[i];
32
33 // Skip space and return.
34 if (*character == ' ' || *character == '\n' || *character == '\r' || *character == '\t')
35 continue;
36 // Match reserved-words or symbols.
37 else if ( is_char_or_digit(*character) )
38 {
39 // Get the symbol's or the reserved-word's length
40 int end_of_symbol, matched_len = 0;
41 for (end_of_symbol = i; end_of_symbol < len; ++end_of_symbol)
42 {
43 if ( !is_char_or_digit(code_string[end_of_symbol]) )
44 {
45 matched_len = end_of_symbol - i;
46 break;
47 }
48 }
49
50 // Create a buffer to store the symbol.
51 char* matched = (char*)calloc(matched_len, sizeof(char));
52 memcpy(matched, code_string+i, matched_len * sizeof(char));
53
54 i = end_of_symbol - 1; // Shift the cursor.
55
56 printf( "(%s, %d)\n", matched, table(matched) );
57 free(matched);
58 }
59 else if ( *character == ':' && code_string[i+1] == '=')
60 {
61 printf( "(%s, %d)\n", ":=", table(":=") );
62 i++; // Shift the cursor.
63 }
64 else if ( *character == '<' && code_string[i+1] == '>')
65 {
66 printf( "(%s, %d)\n", "<>", table("<>") );
67 i++; // Shift the cursor.
68 }
69 else if ( *character == '<' && code_string[i+1] == '=')
70 {
71 printf( "(%s, %d)\n", "<=", table("<=") );
72 i++; // Shift the cursor.
73 }
74 else if ( *character == '>' && code_string[i+1] == '=')
75 {
76 printf( "(%s, %d)\n", ">=", table(">=") );
77 i++; // Shift the cursor.
78 }
79 else if ( *character == '+' && code_string[i+1] == '=')
80 {
81 printf( "(%s, %d)\n", "+=", table("+=") );
82 i++; // Shift the cursor.
83 }
84 else
85 printf( "(%c, %d)\n", *character, table(character) );
86 }
87
88 free(character); // Free the buffer
89 }
90
91 int table(char* res)
92 {
93 const char* ld_pattern = "^[0-9a-zA-Z]*$";
94 const char* dd_pattern = "[0-9]";
95
96 if ( strcmp(res, "begin") == 0 ) return 1;
97 else if ( strcmp(res, "if") == 0 ) return 2;
98 else if ( strcmp(res, "then") == 0 )return 3;
99 else if ( strcmp(res, "while") == 0 )return 4;
100 else if ( strcmp(res, "do") == 0 ) return 5;
101 else if ( strcmp(res, "end") == 0 ) return 6;
102 else if ( strcmp(res, "+") == 0 ) return 13;
103 else if ( strcmp(res, "-") == 0 ) return 14;
104 else if ( strcmp(res, "*") == 0 ) return 15;
105 else if ( strcmp(res, "/") == 0 ) return 16;
106 else if ( strcmp(res, ":") == 0 ) return 17;
107 else if ( strcmp(res, ":=") == 0 ) return 18;
108 else if ( strcmp(res, "<") == 0 ) return 20;
109 else if ( strcmp(res, "<>") == 0 ) return 21;
110 else if ( strcmp(res, "<=") == 0 ) return 22;
111 else if ( strcmp(res, ">") == 0 ) return 23;
112 else if ( strcmp(res, ">=") == 0 ) return 24;
113 else if ( strcmp(res, "+=") == 0 ) return 25;
114 else if ( strcmp(res, "=") == 0 ) return 26;
115 else if ( strcmp(res, ";") == 0 ) return 27;
116 else if ( strcmp(res, "(") == 0 ) return 28;
117 else if ( strcmp(res, ")") == 0 ) return 29;
118 else if ( strcmp(res, "{") == 0 ) return 30;
119 else if ( strcmp(res, "}") == 0 ) return 31;
120 else if ( strcmp(res, "#") == 0 ) return 0;
121 else if ( re_match(dd_pattern, res) == 0 ) return 11;
122 else if ( re_match(ld_pattern, res) == 0 ) return 10;
123 else return -1;
124 }
125
126 int8_t is_char_or_digit(char res)
127 {
128 if ( (res >= '0' && res <= '9') || (res >= 'a' && res <= 'z') || (res >= 'A' && res <= 'Z') )
129 return 1;
130 else
131 return 0;
132 }
133
134 char* file_to_string(const char* path, int *len)
135 {
136 FILE *fp;
137 if ( !(fp = fopen(path, "rb")) )
138 return NULL;
139
140 fseek(fp, 0, SEEK_END);
141 *len = ftell(fp);
142
143 fseek(fp , 0 , SEEK_SET);
144 char* content = (char*)malloc(sizeof(char) * (*len));
145
146 fread(content, sizeof(char), *len, fp);
147
148 fclose(fp);
149
150 return content;
151 }
152
153 int re_match(const char* pattern, const char* res)
154 {
155 regmatch_t pmatch[1];
156 size_t nmatch = 1;
157 regex_t reg;
158
159 regcomp(®, pattern, REG_EXTENDED);
160 int status = regexec(®, res, nmatch, pmatch, 0);
161 regfree(®);
162
163 return status;
164 }
2.运行结果
准备好测试用的伪代码文件test.txt如下：
运行可执行文件时，将test.txt的文件路径作为运行参数，程序会按照指定路径读取，结果如下：
