lexer

.clang-format

@@ -0,0 +1,14 @@
+BasedOnStyle: WebKit
+IndentWidth: 4
+ColumnLimit: 80
+IndentCaseLabels: true
+BreakBeforeBraces: Custom
+BraceWrapping:
+    AfterFunction: true
+    SplitEmptyFunction: false
+AlignAfterOpenBracket: BlockIndent
+AlignOperands: AlignAfterOperator
+BreakBeforeBinaryOperators: true
+BinPackArguments: false
+BinPackParameters: false
+

.gitignore

@@ -0,0 +1,4 @@
+matemateak
+*.o
+compile_flags.txt
+

Makefile

@@ -0,0 +1,23 @@
+
+CC = gcc
+
+CFLAGS = -std=c17 -Wall -Wextra -Wpedantic -Wconversion
+LFLAGS = -lm
+
+C_FILES = $(shell find src/ -name *.c)
+HEADER_FILES = $(shell find src/ -name *.h)
+
+OBJECT_FILES = $(patsubst %.c, %.o, $(C_FILES))
+
+matemateak: $(OBJECT_FILES)
+	$(CC) -o $@ $(LFLAGS) $^
+
+%.o: %.c $(HEADER_FILES)
+	$(CC) -c -o $@ $(CFLAGS) $<
+
+compile_flags.txt:
+	echo -xc $(C_FLAGS) | sed 's/\s\+/\n/g' > compile_flags.txt
+
+clean:
+	$(RM) $(OBJECT_FILES) matemateak
+

src/lexer.c

@@ -0,0 +1,123 @@
+#include "lexer.h"
+#include <stdbool.h>
+
+bool lexer_done(const Lexer* lexer) { return lexer->index >= lexer->length; }
+
+char lexer_current(const Lexer* lexer) { return lexer->text[lexer->index]; }
+
+void lexer_step(Lexer* lexer)
+{
+    lexer->index += 1;
+    if (!lexer_done(lexer)) {
+        if (lexer_current(lexer) == '\n') {
+            lexer->line += 1;
+            lexer->col = 1;
+        } else {
+            lexer->col += 1;
+        }
+    }
+}
+
+Position lexer_pos(const Lexer* lexer)
+{
+    return (Position) {
+        .index = lexer->index,
+        .line = lexer->line,
+        .col = lexer->col,
+    };
+}
+
+Token lexer_token(const Lexer* lexer, TokenType type, Position start)
+{
+    return (Token) {
+        .type = type,
+        .index = start.index,
+        .length = lexer->index - start.index,
+        .line = start.line,
+        .col = start.col,
+    };
+}
+
+Token lexer_step_and_token(Lexer* lexer, TokenType type, Position start)
+{
+    lexer_step(lexer);
+    return lexer_token(lexer, type, start);
+}
+
+bool lexer_is_whitespace(char value)
+{
+    return value == ' ' || value == '\t' || value == '\r' || value == '\n';
+}
+
+Token lexer_skip_whitespace(Lexer* lexer)
+{
+    lexer_step(lexer);
+    while (!lexer_done(lexer) && lexer_is_whitespace(lexer_current(lexer)))
+        lexer_step(lexer);
+    return lexer_next(lexer);
+}
+
+bool lexer_is_int(char value) { return value >= '0' && value <= '9'; }
+
+bool lexer_is_int_start(char value) { return lexer_is_int(value); }
+
+Token lexer_int_token(Lexer* lexer)
+{
+    Position start = lexer_pos(lexer);
+    lexer_step(lexer);
+    while (!lexer_done(lexer) && lexer_is_int(lexer_current(lexer)))
+        lexer_step(lexer);
+    return lexer_token(lexer, TokenTypeInt, start);
+}
+
+void lexer(Lexer* lexer, const char* text, size_t length)
+{
+    *lexer = (Lexer) {
+        .text = text,
+        .length = length,
+        .index = 0,
+        .line = 1,
+        .col = 1,
+    };
+}
+
+Token lexer_level3(Lexer* lexer)
+{
+    if (lexer_is_int_start(lexer_current(lexer)))
+        return lexer_int_token(lexer);
+    else
+        return lexer_step_and_token(
+            lexer, TokenTypeInvalidChar, lexer_pos(lexer)
+        );
+}
+
+Token lexer_level2(Lexer* lexer)
+{
+    Position start = lexer_pos(lexer);
+    switch (lexer_current(lexer)) {
+        case '+':
+            return lexer_step_and_token(lexer, TokenTypePlus, start);
+        case '-':
+            return lexer_step_and_token(lexer, TokenTypeMinus, start);
+        case '*':
+            return lexer_step_and_token(lexer, TokenTypeAsterisk, start);
+        case '/':
+            return lexer_step_and_token(lexer, TokenTypeSlash, start);
+        case '(':
+            return lexer_step_and_token(lexer, TokenTypeLParen, start);
+        case ')':
+            return lexer_step_and_token(lexer, TokenTypeRParen, start);
+        default:
+            return lexer_level3(lexer);
+    }
+}
+
+Token lexer_next(Lexer* lexer)
+{
+    if (lexer_done(lexer))
+        return lexer_token(lexer, TokenTypeEof, lexer_pos(lexer));
+    else if (lexer_is_whitespace(lexer_current(lexer)))
+        return lexer_skip_whitespace(lexer);
+    else
+        return lexer_level2(lexer);
+}

src/lexer.h

@@ -0,0 +1,38 @@
+#ifndef LEXER_H
+#define LEXER_H
+
+#include <stddef.h>
+
+typedef struct {
+    size_t index;
+    int line, col;
+} Position;
+
+typedef enum {
+    TokenTypeEof,
+    TokenTypeInvalidChar,
+    TokenTypeInt,
+    TokenTypePlus,
+    TokenTypeMinus,
+    TokenTypeAsterisk,
+    TokenTypeSlash,
+    TokenTypeLParen,
+    TokenTypeRParen,
+} TokenType;
+
+typedef struct {
+    TokenType type;
+    size_t index, length;
+    int line, col;
+} Token;
+
+typedef struct {
+    const char* text;
+    size_t index, length;
+    int line, col;
+} Lexer;
+
+void lexer(Lexer* lexer, const char* text, size_t length);
+Token lexer_next(Lexer* lexer);
+
+#endif

src/main.c

@@ -0,0 +1,4 @@
+#include <stdio.h>
+#include <stdlib.h>
+
+int main(void) { printf("hello world\n"); }