怎么用c语言编一个词法分析器

怎么用c语言编一个词法分析器

简而言之就是先画一个状态图，然后根据图来编码就行
一个简单的xml的词法分析器供参考

#include
#include
#include

typedef struct
{
char *p;
int len;
}
xml_Text;

typedef enum {
xml_tt_U,
xml_tt_H,
xml_tt_E,
xml_tt_B,
xml_tt_BE,
xml_tt_T
}
xml_TokenType;

typedef struct
{
xml_Text text;
xml_TokenType type;
}
xml_Token;

int xml_initText(xml_Text *pText, char *s)
{
pText->p = s;
pText->len = strlen(s);
return 0;
}

int xml_initToken(xml_Token *pToken, xml_Text *pText)
{
pToken->text.p = pText->p;
pToken->text.len = 0;
pToken->type = xml_tt_U;
return 0;
}

int xml_print(xml_Text *pText)
{
int i;
for (i = 0; i < pText->len; i++)
{
putchar(pText->p[i]);
}
return 0;
}

int xml_println(xml_Text *pText)
{
xml_print(pText);
putchar('\n');
return 0;
}

int xml_getToken(xml_Text *pText, xml_Token *pToken)
{
char *start = pToken->text.p + pToken->text.len;
char *p = start;
char *end = pText->p + pText->len;
int state = 0;

pToken->text.p = p;
pToken->type = xml_tt_U;

for (; p < end; p++)
{
switch(state)
{
case 0:
switch(*p)
{
case '<':
state = 1;
break;
default:
state = 7;
break;
}
break;
case 1:
switch(*p)
{
case '?':
state = 2;
break;
case '/':
state = 4;
break;
default:
state = 5;
break;
}
break;
case 2:
switch(*p)
{
case '?':
state = 3;
break;
default:
state = 2;
break;
}
break;
case 3:
switch(*p)
{
case '>':
pToken->text.len = p - start + 1;
pToken->type = xml_tt_H;
return 1;
default:
state = -1;
break;
}
break;
case 4:
switch(*p)
{
case '>':
pToken->text.len = p - start + 1;
pToken->type = xml_tt_E;
return 1;
default:
state = 4;
break;
}
break;
case 5:
switch(*p)
{
case '>':
pToken->text.len = p - start + 1;
pToken->type = xml_tt_B;
return 1;
case '/':
state = 6;
break;
default:
state = 5;
break;
}
break;
case 6:
switch(*p)
{
case '>':
pToken->text.len = p - start + 1;
pToken->type = xml_tt_BE;
return 1;
default:
state = -1;
break;
}
break;
case 7:
switch(*p)
{
case '<':
p--;
pToken->text.len = p - start + 1;
pToken->type = xml_tt_T;
return 1;
default:
state = 7;
break;
}
break;
default:
pToken->text.len = p - start + 1;
pToken->type = xml_tt_T;
return 1;
}
}
return 0;
}

int main()
{
int ret = 0;

xml_Text xml;
xml_initText(&xml, " ss hoho");

xml_Token token;
xml_initToken(&token, &xml);

ret = xml_getToken(&xml, &token);
printf("ret=%d;text=",ret);
xml_print(&token.text);
printf(";type=%d;\n\n", token.type);

ret = xml_getToken(&xml, &token);
printf("ret=%d;text=",ret);
xml_print(&token.text);
printf(";type=%d;\n\n", token.type);

ret = xml_getToken(&xml, &token);
printf("ret=%d;text=",ret);
xml_print(&token.text);
printf(";type=%d;\n\n", token.type);

ret = xml_getToken(&xml, &token);
printf("ret=%d;text=",ret);
xml_print(&token.text);
printf(";type=%d;\n\n", token.type);

ret = xml_getToken(&xml, &token);
printf("ret=%d;text=",ret);
xml_print(&token.text);
printf(";type=%d;\n\n", token.type);

ret = xml_getToken(&xml, &token);
printf("ret=%d;text=",ret);
xml_print(&token.text);
printf(";type=%d;\n\n", token.type);

ret = xml_getToken(&xml, &token);
printf("ret=%d;text=",ret);
xml_print(&token.text);
printf(";type=%d;\n\n", token.type);

return 0;
}