编译原理 第二章 一个简单的编译器.ppt

带有符号表功能的词法分析器 int lexan() { char lexbuf[100]; char c; while (1) { 从输入字符序列读取一个字符，保存到c; if (c == ‘ ‘ || c == ‘\t’) ; else if (c == ‘\n’) lineno++; else if (isdigit(c)) { tokenval = c及后续数字组成的数值; return NUM; } 带有符号表功能的词法分析器 else if (isalpha(c)) { 将c及后续字母、数字放入lexbuf; p = lookup(lexbuf); if (p = 0) p = insert(lexbuf, ID); tokenval = p; return 符号表项p的token值; } else { tokenval = NONE; return 字符c对应的整数; } } } Lcc的符号表 字符串的管理 //string.c static struct string { char *str; int len; struct string *link; } *buckets[1024]; char *stringn(const char *str, int len) { int i; unsigned int h; const char *end; struct string *p; assert(str); for (h = 0, i = len, end = str; i 0; i--) // Hash h = (h1) + scatter[*(unsigned char *)end++]; h = NELEMS(buckets)-1; 字符串管理 for (p = buckets[h]; p; p = p-link) //搜索 if (len == p-len) { const char *s1 = str; char *s2 = p-str; do { if (s1 == end) return p-str; } while (*s1++ == *s2++); } { static char *next, *strlimit; //创建 if (len + 1 = strlimit - next) { //现有空间不够 int n = len + 4*1024; next = allocate(n, PERM); //分配空间 strlimit = next + n; } 字符串管理 p-len = len; for (p-str = next; str end; ) *next++ = *str++; //复制字符串 *next++ = 0; p-link = buckets[h]; //放入Hash表 buckets[h] = p; return p-str; } } 符号表 //sym.c struct table { int level; Table previous; struct entry { struct symbol sym; struct entry *link; } *buckets[256]; Symbol all; }; 符号表 Symbol lookup(const char *name, Table tp) { struct entry *p; unsigned h = (unsigned long)name(HASHSIZE-1); assert(tp); do for (p = tp-buckets[h]; p; p = p-link) if (name == p-sym.name) return p-sym; while ((tp = tp-previous) != NULL); return NULL; } 符号表 Symbol install(const char *name, Table *tpp, int level, int arena) { Table tp = *tpp; struct entry *p; unsigned h = (unsigned long)name(HASHSIZE-1); assert(level == 0 || level = tp-level); if (level 0 tp-level l