nooc

commit e54844c13e003ffc1e0b25fc806da61b0bea843a
parent 5df1c5b29ed3579abf5de5c9f07d2fcd9ed30f8f
Author: Nihal Jere <nihal@nihaljere.xyz>
Date:   Thu,  9 Dec 2021 14:25:16 -0600

lex: split into separate file

Diffstat:
M
Makefile
 | 
4
++--
A
lex.c
 | 
126
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
A
lex.h
 | 
1
+
M
main.c
 | 
123
+------------------------------------------------------------------------------

4 files changed, 130 insertions(+), 124 deletions(-)
diff --git a/Makefile b/Makefile
@@ -1,8 +1,8 @@
 .c.o:
 	$(CC) -c $< -o $@
 
-nooc: main.o array.o util.o x64.o elf.o
-	$(CC) main.o array.o x64.o util.o elf.o -o nooc
+nooc: main.o array.o util.o x64.o elf.o lex.o
+	$(CC) main.o array.o x64.o util.o elf.o lex.o -o nooc
 
 clean:
 	rm -f *.o nooc
diff --git a/lex.c b/lex.c
@@ -0,0 +1,126 @@
+#include <ctype.h>
+#include <stdint.h>
+#include <stdlib.h>
+
+#include "nooc.h"
+#include "util.h"
+#include "lex.h"
+
+#define ADVANCE(n) \
+			start.data += (n) ; \
+			start.len -= (n) ; \
+			col += (n) ;
+
+struct token *
+lex(struct slice start)
+{
+	size_t line = 1;
+	size_t col = 1;
+	struct token *head = calloc(1, sizeof(struct token));
+	if (!head)
+		return NULL;
+
+	struct token *cur = head;
+	while (start.len) {
+		if (isblank(*start.data)) {
+			ADVANCE(1);
+			continue;
+		}
+
+		if (*start.data == '\n') {
+			ADVANCE(1);
+			line += 1;
+			col = 1;
+			continue;
+		}
+
+		cur->line = line;
+		cur->col = col;
+
+		if (slice_cmplit(&start, "if") == 0) {
+			cur->type = TOK_IF;
+			ADVANCE(2);
+		} else if (slice_cmplit(&start, "let") == 0) {
+			cur->type = TOK_LET;
+			ADVANCE(3);
+		} else if (slice_cmplit(&start, "else") == 0) {
+			cur->type = TOK_ELSE;
+			ADVANCE(4);
+		} else if (slice_cmplit(&start, "loop") == 0) {
+			cur->type = TOK_LOOP;
+			ADVANCE(4);
+		} else if (slice_cmplit(&start, "return") == 0) {
+			cur->type = TOK_RETURN;
+			ADVANCE(6);
+		} else if (*start.data == '>') {
+			cur->type = TOK_GREATER;
+			ADVANCE(1);
+		} else if (*start.data == ',') {
+			cur->type = TOK_COMMA;
+			ADVANCE(1);
+		} else if (*start.data == '(') {
+			cur->type = TOK_LPAREN;
+			ADVANCE(1);
+		} else if (*start.data == ')') {
+			cur->type = TOK_RPAREN;
+			ADVANCE(1);
+		} else if (*start.data == '{') {
+			cur->type = TOK_LCURLY;
+			ADVANCE(1);
+		} else if (*start.data == '}') {
+			cur->type = TOK_RCURLY;
+			ADVANCE(1);
+		} else if (isdigit(*start.data)) {
+			cur->slice.data = start.data;
+			cur->slice.len = 1;
+			ADVANCE(1);
+			cur->type = TOK_NUM;
+			while (isdigit(*start.data)) {
+				ADVANCE(1);
+				cur->slice.len++;
+			}
+		} else if (*start.data == '"') {
+			ADVANCE(1);
+			cur->slice.data = start.data;
+			cur->type = TOK_STRING;
+			while (*start.data != '"') {
+				ADVANCE(1);
+				cur->slice.len++;
+			}
+
+			ADVANCE(1);
+		} else if (*start.data == '+') {
+			cur->type = TOK_PLUS;
+			ADVANCE(1);
+		} else if (*start.data == '-') {
+			cur->type = TOK_MINUS;
+			ADVANCE(1);
+		} else if (*start.data == '=') {
+			cur->type = TOK_EQUAL;
+			ADVANCE(1);
+		} else if (isalpha(*start.data)) {
+			cur->type = TOK_NAME;
+			cur->slice.data = start.data;
+			cur->slice.len = 1;
+			ADVANCE(1);
+			while (isalnum(*start.data)) {
+				ADVANCE(1);
+				cur->slice.len++;
+			}
+		} else {
+			error("invalid token", line, col);
+		}
+
+		cur->next = calloc(1, sizeof(struct token));
+
+		if (!cur->next)
+			die("lex: failed to allocate next token");
+
+		cur = cur->next;
+	}
+
+	cur->line = line;
+	cur->col = col;
+
+	return head;
+}
diff --git a/lex.h b/lex.h
@@ -0,0 +1 @@
+struct token *lex(struct slice start);
diff --git a/main.c b/main.c
@@ -16,6 +16,7 @@
 #include "nooc.h"
 #include "util.h"
 #include "elf.h"
+#include "lex.h"
 
 struct decls decls;
 struct assgns assgns;
@@ -23,128 +24,6 @@ struct exprs exprs;
 
 char *infile;
 
-#define ADVANCE(n) \
-			start.data += (n) ; \
-			start.len -= (n) ; \
-			col += (n) ;
-
-struct token *
-lex(struct slice start)
-{
-	size_t line = 1;
-	size_t col = 1;
-	struct token *head = calloc(1, sizeof(struct token));
-	if (!head)
-		return NULL;
-
-	struct token *cur = head;
-	while (start.len) {
-		if (isblank(*start.data)) {
-			ADVANCE(1);
-			continue;
-		}
-
-		if (*start.data == '\n') {
-			ADVANCE(1);
-			line += 1;
-			col = 1;
-			continue;
-		}
-
-		cur->line = line;
-		cur->col = col;
-
-		if (slice_cmplit(&start, "if") == 0) {
-			cur->type = TOK_IF;
-			ADVANCE(2);
-		} else if (slice_cmplit(&start, "let") == 0) {
-			cur->type = TOK_LET;
-			ADVANCE(3);
-		} else if (slice_cmplit(&start, "else") == 0) {
-			cur->type = TOK_ELSE;
-			ADVANCE(4);
-		} else if (slice_cmplit(&start, "loop") == 0) {
-			cur->type = TOK_LOOP;
-			ADVANCE(4);
-		} else if (slice_cmplit(&start, "return") == 0) {
-			cur->type = TOK_RETURN;
-			ADVANCE(6);
-		} else if (*start.data == '>') {
-			cur->type = TOK_GREATER;
-			ADVANCE(1);
-		} else if (*start.data == ',') {
-			cur->type = TOK_COMMA;
-			ADVANCE(1);
-		} else if (*start.data == '(') {
-			cur->type = TOK_LPAREN;
-			ADVANCE(1);
-		} else if (*start.data == ')') {
-			cur->type = TOK_RPAREN;
-			ADVANCE(1);
-		} else if (*start.data == '{') {
-			cur->type = TOK_LCURLY;
-			ADVANCE(1);
-		} else if (*start.data == '}') {
-			cur->type = TOK_RCURLY;
-			ADVANCE(1);
-		} else if (isdigit(*start.data)) {
-			cur->slice.data = start.data;
-			cur->slice.len = 1;
-			ADVANCE(1);
-			cur->type = TOK_NUM;
-			while (isdigit(*start.data)) {
-				ADVANCE(1);
-				cur->slice.len++;
-			}
-		} else if (*start.data == '"') {
-			ADVANCE(1);
-			cur->slice.data = start.data;
-			cur->type = TOK_STRING;
-			while (*start.data != '"') {
-				ADVANCE(1);
-				cur->slice.len++;
-			}
-
-			ADVANCE(1);
-		} else if (*start.data == '+') {
-			cur->type = TOK_PLUS;
-			ADVANCE(1);
-		} else if (*start.data == '-') {
-			cur->type = TOK_MINUS;
-			ADVANCE(1);
-		} else if (*start.data == '=') {
-			cur->type = TOK_EQUAL;
-			ADVANCE(1);
-		} else if (isalpha(*start.data)) {
-			cur->type = TOK_NAME;
-			cur->slice.data = start.data;
-			cur->slice.len = 1;
-			ADVANCE(1);
-			while (isalnum(*start.data)) {
-				ADVANCE(1);
-				cur->slice.len++;
-			}
-		} else {
-			error("invalid token", line, col);
-		}
-
-		cur->next = calloc(1, sizeof(struct token));
-
-		// FIXME: handle this properly
-		if (!cur->next)
-			return NULL;
-
-		cur = cur->next;
-	}
-
-	cur->line = line;
-	cur->col = col;
-
-	return head;
-}
-
-#undef ADVANCE
-
 struct data data_seg;
 
 void