basic lexing

6 files changed, 80 insertions, 8 deletions

diff --git a/build.sh b/build.sh
index 93b7a6e..7986eeb 100755
--- a/build.sh
+++ b/build.sh
@@ -1,2 +1,2 @@
 #!/bin/sh
-gcc -std=c99 -Wall -Wextra -o golem *.c
+gcc -g -O0 -std=c99 -Wall -Wextra -o golem *.c
diff --git a/lexer.c b/lexer.c
index 47bd816..d1cba7f 100644
--- a/lexer.c
+++ b/lexer.c
@@ -1,12 +1,70 @@
 #include "lexer.h"
 #include "slice.h"
 #include "token.h"
+#include <ctype.h>
+#include <stdio.h>
 #include <stdlib.h>
 
+static unsigned int
+simple_token(char c)
+{
+	switch (c) {
+	case '(': return TOK_OPEN_PAREN;
+	case ')': return TOK_CLOS_PAREN;
+	case '{': return TOK_OPEN_CURL;
+	case '}': return TOK_CLOS_CURL;
+	default: return TOK_NONE;
+	}
+}
+
+static Slice
+slice_number(Slice slice)
+{
+	for (int i = slice.begin; i < slice.end; i++)
+		if (!isdigit(slice.str[i]))
+			return slice_sub(slice, slice.begin, i);
+	return slice;
+}
+
 Token *
 lexer(Slice slice)
 {
 	Token *toks = NULL;
-	token_append(&toks, token_create(slice, TOK_STRING));
+	int i = slice.begin;
+	while (i < slice.end) {
+		// Skip whitespaces.
+		if (isspace(slice.str[i])) {
+			i += 1;
+			continue;
+		}
+
+		// Ignore comments.
+		if (slice.str[i] == '/' && slice.str[i + 1] == '/') {
+			while (i < slice.end && slice.str[i] != '\n')
+				i += 1;
+			continue;
+		}
+
+		// Single character tokens.
+		if (simple_token(slice.str[i]) != TOK_NONE) {
+			token_append(&toks,
+			             token_create(slice_sub(slice, i, i + 1),
+			             simple_token(slice.str[i])));
+			i += 1;
+			continue;
+		}
+
+		// Number token.
+		if (isdigit(slice.str[i])) {
+			Slice number;
+			number = slice_number(slice_sub(slice, i, slice.end));
+			token_append(&toks, token_create(number, TOK_NUMBER));
+			i = number.end;
+			continue;
+		}
+
+		printf("skipping '%c'\n", slice.str[i]);
+		i += 1;
+	}
 	return toks;
 }
diff --git a/main.c b/main.c
index f3fff23..c724bd9 100644
--- a/main.c
+++ b/main.c
@@ -21,11 +21,12 @@ main(int argc, char **argv)
 
 	Slice slice = slice_from_str(drain(file));
 	fclose(file);
-	slice_print(slice);
-	if (slice.str != NULL)
-		free(slice.str);
 
 	Token *tokens = lexer(slice);
+	token_print(tokens);
 	token_destroy(tokens);
+
+	if (slice.str != NULL)
+		free(slice.str);
 	return 0;
 }
diff --git a/slice.c b/slice.c
index dba8086..8640d0f 100644
--- a/slice.c
+++ b/slice.c
@@ -6,9 +6,8 @@ Slice
 slice_print(Slice slice)
 {
 	putchar('"');
-	for (int i = slice.begin; i < slice.end; i++) {
+	for (int i = slice.begin; i < slice.end; i++)
 		putchar(slice.str[i]);
-	}
 	putchar('"');
 	putchar('\n');
 	return slice;
@@ -25,5 +24,5 @@ slice_from_str(char *str)
 Slice
 slice_sub(Slice src, int begin, int end)
 {
-	return (Slice){ src.str, src.begin + begin, src.begin + end };
+	return (Slice){ src.str, begin, end };
 }
diff --git a/token.c b/token.c
index f025004..22facec 100644
--- a/token.c
+++ b/token.c
@@ -1,4 +1,5 @@
 #include "token.h"
+#include "slice.h"
 #include <stdlib.h>
 #include <stdio.h>
 
@@ -40,3 +41,15 @@ token_destroy(Token *token)
 		token = next;
 	}
 }
+
+Token *
+token_print(Token *token)
+{
+	Token *tok = token;
+	while (token != NULL) {
+		printf("type: %2u  slice: ", token->type);
+		slice_print(token->slice);
+		token = token->next;
+	}
+	return tok;
+}
diff --git a/token.h b/token.h
index d3f63e5..a46593d 100644
--- a/token.h
+++ b/token.h
@@ -23,3 +23,4 @@ struct Token {
 Token *token_create(Slice slice, unsigned int type);
 Token *token_append(Token **list, Token *elem);
 void token_destroy(Token *token);
+Token *token_print(Token *token);