X-Git-Url: https://www.kengrimes.com/gitweb/?p=henge%2Fwebcc.git;a=blobdiff_plain;f=src%2Fapc%2Flexer_lex.rl;fp=src%2Fapc%2Flexer_lex.rl;h=cb068d6e237e2bd23cf884c9ad4d659660c30455;hp=0000000000000000000000000000000000000000;hb=dc6e33a26d6488e388817d4dd255dcdff22b4a2a;hpb=a4acd734f2064a6876849652e7f896ecade6497a

diff --git a/src/apc/lexer_lex.rl b/src/apc/lexer_lex.rl
new file mode 100644
index 0000000..cb068d6
--- /dev/null
+++ b/src/apc/lexer_lex.rl
@@ -0,0 +1,123 @@
+/* Ragel State Machine for tokenizing text */
+#include <stdio.h>
+#include <string.h>
+#include <apc/parser.tab.h>
+
+extern void lexer_pushtok(int, YYSTYPE);
+
+int lexer_lex(const char*);
+int ipow(int, int);
+int ttov(const char* str, int);
+uint64_t ttor(const char* str, int);
+char* ttos(const char* str, int);
+
+
+#define MAX_TOK_LEN 64
+#define MAX_TOKENS 16
+#define MAX_STR_SIZE (MAX_TOK_LEN * MAX_TOKENS)
+
+
+%%{
+  machine token_matcher;
+
+  # set up yylval and tok_t to be pushed to stack
+  action set_ref {
+                   tok_t = REF;                      \
+                   yylval.ref = ttor(ts, p-ts);      \
+                   lexer_pushtok(tok_t, yylval);     \
+                   ts = p;   }
+
+  action set_val { tok_t = NUM;                      \
+                   yylval.val = ttov(ts, p-ts);      \
+                   lexer_pushtok(tok_t, yylval);     \
+                   ts = p;   }
+
+  action set_name { tok_t = NAME;                    \
+                    yylval.str = ttos(ts, p-ts);     \
+                    lexer_pushtok(tok_t, yylval);    \
+                    ts = p;   }
+
+  # instantiate machines for each possible token
+  ref = '0x' xdigit+ %set_ref;
+  val = digit+ %set_val;
+  name = alpha+ %set_name;
+  tok = ref | val | name;
+
+  main := (tok . '_')* . tok;
+}%%
+
+
+%%write data;
+
+/*  0xxdigit+  => tok_t REF, yylval.ref = uint64_t
+    [0-9]+  => tok_t NUM, yylval.val = int
+    [a-zA-Z]+ => tok_t NAME, yylval.str = char*       */
+
+/* Scan filename and push the its tokens
+   onto the stack */
+int lexer_lex (const char* str)
+{
+  const char *p, *pe, *ts, *eof;
+  int  cs, tok_t ; //tok_t == token type
+
+  p = ts = str;
+  pe = p + strlen(str) + 1;
+  %%write init;
+  %%write exec;
+
+  lexer_pushtok(tok_t, yylval);
+
+  printf (str);
+  return 1;
+}
+
+int ipow(int base, int exp)
+{
+  int result = 1;
+  while (exp)
+    {
+      if (exp & 1)
+        result = result * base;
+      exp = exp >> 1;
+      base *= base;
+    }
+
+  return result;
+}
+
+/*  Token to Value */
+int ttov(const char* str, int len)
+{
+  int i, val = 0;
+
+  for (i = 0; i < len; i++)
+    {
+      val += ((str[len - (i + 1)] - '0') * ipow(10,i));
+    }
+
+  return val;
+}
+
+uint64_t ttor(const char* str, int len)
+{
+  int i;
+  uint64_t num = 0;
+
+  for (i = 0; i < len; i++)
+    {
+      num += ((str[len - (i + 1)] - '0') * ipow(10,i));
+    }
+
+  return num;
+}
+
+char* ttos(const char* str, int len)
+{
+  int i;
+  char token_buf[MAX_TOK_LEN];
+
+  memmove(token_buf, str, len);
+  token_buf[len+1] = '\0';
+
+  return strdup(token_buf);
+}