scan identifiers

lexer/identifier_lexer.test.ts

@@ -0,0 +1,48 @@
+import { expect, test, describe } from "bun:test";
+import { scan_identifier } from "./identifier_lexer";
+
+
+describe("Identifier Lexer", () => {
+    test("should return an identifier token", () => {
+        const code = "a";
+        const token = scan_identifier(code, 0);
+
+        expect(token).toEqual([{ v: "a", token_type: "identifier" }, 1]);
+    });
+
+    test("should scan an underscore", () => {
+        const code = "_";
+        const token = scan_identifier(code, 0);
+
+        expect(token).toEqual([{ v: "_", token_type: "identifier" }, 1]);
+    });
+
+    test("should scan an identifier with an underscore", () => {
+        const code = "a_";
+        const token = scan_identifier(code, 0);
+
+        expect(token).toEqual([{ v: "a_", token_type: "identifier" }, 2]);
+    });
+
+    test("should scan an identifier that starts with an underscore", () => {
+        const code = "_a";
+        const token = scan_identifier(code, 0);
+
+        expect(token).toEqual([{ v: "_a", token_type: "identifier" }, 2]);
+    });
+
+    test("should scan an identifier with numbers and uppercase letters", () => {
+        const code = "aA1";
+        const token = scan_identifier(code, 0);
+
+        expect(token).toEqual([{ v: "aA1", token_type: "identifier" }, 3]);
+    });
+
+    test("should scan a keyword", () => {
+        const code = "val";
+        const token = scan_identifier(code, 0);
+
+        expect(token).toEqual([{ v: "val", token_type: "keyword" }, 3]);
+    });
+});
+

lexer/identifier_lexer.ts

@@ -0,0 +1,35 @@
+import type { Token } from "./lexer.ts";
+import { is_identifier_char } from "./utils.ts";
+
+/**
+ * Scans an identifier, at the given position in the input string.
+ * This function assumes that the character at the given position is a letter.
+ */
+export function scan_identifier(input: string, starting_position: number): [Token, number] {
+    let value = input[starting_position];
+    let pos = starting_position + 1;
+
+    while (pos < input.length) {
+        const c = input[pos];
+
+        if (is_identifier_char(c)) {
+            pos += 1;
+            value += c;
+        }
+        else {
+            break;
+        }
+    }
+
+    return [{ v: value, token_type: check_keyword(value) }, pos];
+}
+
+function check_keyword(value: string): string {
+    const keywords = ["case", "static", "const", "enum", "loop", "use", "break", "catch", "continue", "do", "else", "finally", "for", "fun", "if", "in", "fn", "nil", "return", "throw", "try", "while", "type", "match", "with", "of", "abstract", "class", "interface", "private", "pub", "map", "override", "open", "init", "val", "var", "mut", "clone"];
+
+    if (keywords.includes(value)) {
+        return "keyword";
+    }
+    return "identifier";
+}
+

lexer/lexer.ts

@@ -1,5 +1,6 @@
-import { lex_number } from "./number_lexer.ts";
-import { is_digit } from "./utils.ts";
+import { scan_identifier } from "./identifier_lexer.ts";
+import { scan_number } from "./number_lexer.ts";
+import { is_digit, is_lowercase } from "./utils.ts";
 
 export type Token = {
     v: string,
@@ -37,6 +38,7 @@ export function lex(code: string): Array<Token> {
         let next_token: Token | null = null;
         let next_position: number | null = null;
 
+        // try to scan a number
         if (is_digit(c)) {
             // if the current default token is not empty, push it to the tokens array
             if (current_default_token !== "") {
@@ -45,7 +47,20 @@ export function lex(code: string): Array<Token> {
             }
 
             // lex a number
-            const [token, next] = lex_number(code, current_pos);
+            const [token, next] = scan_number(code, current_pos);
+            current_pos = next;
+            tokens.push(token);
+            continue;
+        }
+        // try to scan an identifier/keyword
+        else if (is_lowercase(c) || c === "_") {
+            // if the current default token is not empty, push it to the tokens array
+            if (current_default_token !== "") {
+                tokens.push({ v: current_default_token, token_type: "" });
+                current_default_token = "";
+            }
+
+            const [token, next] = scan_identifier(code, current_pos);
             current_pos = next;
             tokens.push(token);
             continue;

lexer/number_lexer.test.ts

@@ -1,17 +1,17 @@
 import { expect, test, describe } from "bun:test";
-import { lex_number } from "./number_lexer";
+import { scan_number } from "./number_lexer";
 
 describe("Number Lexer", () => {
     test("should return a whole number token", () => {
         const code = "1";
-        const token = lex_number(code, 0);
+        const token = scan_number(code, 0);
 
         expect(token).toEqual([{ v: "1", token_type: "number" }, 1]);
     });
 
     test("should return a whole number token pt 2", () => {
         const code = "12345";
-        const token = lex_number(code, 0);
+        const token = scan_number(code, 0);
 
         expect(token).toEqual([{ v: "12345", token_type: "number" }, 5]);
     });

lexer/number_lexer.ts

@@ -10,7 +10,7 @@ import { is_digit } from "./utils.ts";
  * @param pos the position to start scanning from
  * @returns 
  */
-export function lex_number(input: string, pos: number): [Token, number] {
+export function scan_number(input: string, pos: number): [Token, number] {
     const [token_value, next] = scan_decimal(input, pos);
 
     return [{ v: token_value, token_type: "number" }, next];

lexer/utils.ts

@@ -1,3 +1,15 @@
 export function is_digit(c: string): boolean {
     return c >= '0' && c <= '9';
-}
+}
+
+export function is_lowercase(c: string): boolean {
+    return c >= 'a' && c <= 'z';
+}
+
+export function is_uppercase(c: string): boolean {
+    return c >= 'A' && c <= 'Z';
+}
+
+export function is_identifier_char(c: string): boolean {
+    return is_lowercase(c) || is_uppercase(c) || is_digit(c) || c === '_';
+}

md/learn/index.md

@@ -76,7 +76,7 @@ val has_key = haystack.contains("needle")
 ]
 
 // THP
-Obj {
+.{
     names: #("Toni", "Stark"), // Tuple
     age: 33,
     numbers: [32, 64, 128]