Merge pull request #10 from s-kybound/main

Upgrade Scm-slang parser

.eslintignore

@@ -0,0 +1,2 @@
+node_modules
+dist

.eslintrc.yml

@@ -0,0 +1,8 @@
+env:
+  browser: true
+  es2021: true
+extends: standard-with-typescript
+parserOptions:
+  ecmaVersion: latest
+  sourceType: module
+rules: {}

README.md

@@ -4,29 +4,25 @@
 
 `scm-slang` is an **experimental** implementation of the [Scheme](https://www.scheme.org/) programming language designed for compatibiity with the online learning environment of [Source Academy](https://sourceacademy.org/). It aims to be faithful to the original *Structure and Interpretation of Computer Programs* (SICP) book whilst maintaining compatibility with modules designed for use with [`js-slang`](https://github.com/source-academy/js-slang), a subset of JavaScript intended for use with the SICP JS book in Source Academy.
 
-`scm-slang` provides a special Scheme parser that is able to parse Scheme code.
+`scm-slang` provides a special Scheme parser that is able to parse Scheme code. It also supports standard Scheme data structures such as Lists, Vectors and the Numeric tower.
 
 ## How does it work?
 
 `scm-slang` parses a subset of Scheme (minimally, enough to fulfil SICP chapters 1-4) and generates an `estree`-compatible AST. This way, `scm-slang` allows code written in SCM Source Languages to use modules written for JS Source Languages.
 
 ## Comparison with Revised⁷ Report on the Algorithmic Language Scheme
 
-`scm-slang` is designed following the [R7RS language specification](https://small.r7rs.org/) of Scheme. However, there are several key deviations that differentiate `scm-slang` from a complete implementation of R7RS Scheme:
+`scm-slang` ultimately follows the [R7RS language specification](https://small.r7rs.org/) of Scheme. However, there are several key deviations that differentiate `scm-slang` from a complete implementation of R7RS Scheme:
 
-- Continuations: Continuations do not currently have first-class status in Source. Hence, procedures such as `call/cc`, which operate on continuations, are not yet implemented. Subject to change, with the implementation of the Explict Control Evaluator in js-slang.
+- Continuations: The parser itself does not support continuations, but in conjunction with the Source Academy Explicit Control Evaluator, continuations are supported within the Source Academy ecosystem.
 
-- Macros: Not implemented. Subject to change with the future refactoring of scm-slang to use an intermediate AST before final estree translation.
+- Macros: Not implemented at the moment. Set for completion in Autumn-Fall 2024.
 
-- Defines of same variable in same scope: Not allowed at the moment. Subject to change with the future refactoring of scm-slang to use an intermediate AST before final estree translation.
-
-- Types: `scm-slang` does not support complex numbers or characters.
+- Types: `scm-slang` does not support characters or bytevectors at the moment.
 
 - Parentheses: `scm-slang` supports the use of square brackets (i.e. []) interchangably with parentheses in order to enhance the visual representation of Scheme code, similar to [Racket](https://racket-lang.org/) or [Guile Scheme](https://www.gnu.org/software/guile/). [See relevant discussion here](http://community.schemewiki.org/?scheme-faq-language)
 
-- Named let is not supported. Subject to change with the future refactoring of scm-slang to use an intermediate AST before final estree translation.
-
-- Variadic functions: `scm-slang` does not currently support variadic functions with usage of the `.` operator or the `case-lambda` syntax. 
+- Named let is not supported.
 
 - Import/Export: `scm-slang` follows a specialised import/export system that deviates from any standard Scheme implementation. It follows more closely to JavaScript syntax so as to maintain compatibility with current Source Academy modules.
 
@@ -44,7 +40,7 @@
 
 # Requirements
 
-- `node`: known working version: v16.19.0
+- `node`: known working version: v20.11.0
 
 # Usage
 

babel.config.js

@@ -0,0 +1 @@
+module.exports = { presets: ["@babel/preset-env"] };

jest.config.js

@@ -0,0 +1,7 @@
+module.exports = {
+  preset: "ts-jest",
+  transform: {
+    "^.+\\.(ts|tsx)?$": "ts-jest",
+    "^.+\\.(js|jsx)$": "babel-jest",
+  },
+};

package.json

@@ -1,40 +1,58 @@
 {
-    "name": "scm-slang",
-    "version": "0.5.0",
-    "license": "Apache-2.0",
-    "description": "Scheme-based implementations of Source, written in Typescript",
-    "keywords": [
-        "Scheme",
-        "interpreter",
-        "compiler",
-        "Source",
-        "SICP"
-    ],
-    "author": {
-        "name": "Source Academy",
-        "url": "https://github.com/source-academy/"
-    },
-    "repository": {
-        "type": "git",
-        "url": "git+https://github.com/source-academy/scm-slang.git"
-    },
-    "main": "index.js",
-    "bugs": {
-        "url": "https://github.com/source-academy/scm-slang/issues"
-    },
-    "homepage": "https://github.com/source-academy/scm-slang#readme",
-    "private": true,
-    "dependencies": {
-        "@types/estree": "^1.0.0",
-        "acorn": "^8.8.2",
-        "acorn-walk": "^8.2.0",
-        "js-base64": "^3.7.5"
-    },
-    "scripts": {},
-    "meta": {},
-    "devDependencies": {
-        "@types/node": "^18.14.2",
-        "astring": "^1.8.4",
-        "source-map": "^0.7.4"
-    }
+  "name": "scm-slang",
+  "version": "1.0.3",
+  "license": "Apache-2.0",
+  "description": "Scheme-based implementations of Source, written in Typescript",
+  "keywords": [
+    "Scheme",
+    "interpreter",
+    "compiler",
+    "Source",
+    "SICP"
+  ],
+  "author": {
+    "name": "Source Academy",
+    "url": "https://github.com/source-academy/"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/source-academy/scm-slang.git"
+  },
+  "main": "index.js",
+  "bugs": {
+    "url": "https://github.com/source-academy/scm-slang/issues"
+  },
+  "homepage": "https://github.com/source-academy/scm-slang#readme",
+  "private": true,
+  "dependencies": {
+    "@types/estree": "^1.0.0",
+    "acorn": "^8.8.2",
+    "acorn-walk": "^8.2.0",
+    "js-base64": "^3.7.5"
+  },
+  "scripts": {
+    "build-libs": "npx ts-node ./src/compile-libs.ts",
+    "test": "jest",
+    "lint": "eslint --ignore-path .eslintignore --ext .ts",
+    "format": "prettier --ignore-path .gitignore --write \"**/*.+(js|ts|json)\""
+  },
+  "meta": {},
+  "devDependencies": {
+    "@babel/preset-env": "^7.23.9",
+    "@types/jest": "^29.5.11",
+    "@types/node": "^18.14.2",
+    "@typescript-eslint/eslint-plugin": "^6.4.0",
+    "babel-jest": "^29.7.0",
+    "escodegen": "^2.1.0",
+    "eslint": "^8.0.1",
+    "eslint-config-standard-with-typescript": "^43.0.1",
+    "eslint-plugin-import": "^2.25.2",
+    "eslint-plugin-n": "^15.0.0 || ^16.0.0 ",
+    "eslint-plugin-promise": "^6.0.0",
+    "jest": "^29.7.0",
+    "prettier": "^3.2.4",
+    "source-map": "^0.7.4",
+    "ts-jest": "^29.1.2",
+    "typescript": "*"
+  }
 }

src/compile-libs.ts

@@ -0,0 +1,45 @@
+import path from "path";
+import fs from "fs";
+import { schemeParse } from "./transpiler";
+import { estreeEncode } from "./utils/encoder-visitor";
+const escodegen = require("escodegen");
+
+function transpile(inputFilePath: string, outputFilePath: string) {
+  fs.readFile(inputFilePath, "utf8", (err, data) => {
+    if (err) {
+      console.error(`Error reading file: ${err}`);
+      return;
+    }
+
+    // we transpile the file
+    const transpiledAST = schemeParse(data);
+    const encodedAST = estreeEncode(transpiledAST);
+    const transpiledProgram = escodegen.generate(encodedAST);
+
+    fs.writeFile(outputFilePath, transpiledProgram, (err) => {
+      if (err) {
+        console.error(`Error writing file: ${err}`);
+        return;
+      }
+      console.log(`${inputFilePath} has been transpiled to ${outputFilePath}`);
+    });
+  });
+}
+
+// get file paths from command line arguments
+const inputFilePath: string = process.argv[2];
+const outputFilePath: string = process.argv[3]
+  ? process.argv[3]
+  : inputFilePath.replace(".scm", ".js");
+
+// validate file paths
+if (!inputFilePath) {
+  console.error("Please provide an input file path and an output file path");
+}
+
+if (!(path.extname(inputFilePath) === ".scm")) {
+  console.error("Please provide a .scm file for compilation!");
+}
+
+// if everything is fine, we transpile the file
+transpile(inputFilePath, outputFilePath);

src/index.ts

@@ -1,11 +1,9 @@
-import { Tokenizer } from "./tokenizer";
-import { Parser } from "./parser";
-import { Program } from "estree";
 import { encode as b64Encode, decode as b64Decode } from "js-base64";
 
-export * from "./prelude-visitor";
-export * as TokenizerError from "./tokenizer-error";
-export * as ParserError from "./parser-error";
+export * from "./utils/encoder-visitor";
+export { LexerError } from "./transpiler";
+export { ParserError } from "./transpiler";
+export { schemeParse } from "./transpiler";
 
 const JS_KEYWORDS: string[] = [
   "break",
@@ -66,13 +64,13 @@ export function encode(identifier: string): string {
       "$scheme_" +
       b64Encode(identifier).replace(
         /([^a-zA-Z0-9_])/g,
-        (match: string) => `\$${match.charCodeAt(0)}\$`
+        (match: string) => `\$${match.charCodeAt(0)}\$`,
       )
     );
   } else {
     return identifier.replace(
       /([^a-zA-Z0-9_])/g,
-      (match: string) => `\$${match.charCodeAt(0)}\$`
+      (match: string) => `\$${match.charCodeAt(0)}\$`,
     );
   }
 }
@@ -89,18 +87,12 @@ export function decode(identifier: string): string {
       identifier
         .slice(8)
         .replace(/\$([0-9]+)\$/g, (_, code: string) =>
-          String.fromCharCode(parseInt(code))
-        )
+          String.fromCharCode(parseInt(code)),
+        ),
     );
   } else {
     return identifier.replace(/\$([0-9]+)\$/g, (_, code: string) =>
-      String.fromCharCode(parseInt(code))
+      String.fromCharCode(parseInt(code)),
     );
   }
 }
-
-export function schemeParse(source: string, chapter?: number): Program {
-  const tokenizer = new Tokenizer(source);
-  const parser = new Parser(source, tokenizer.scanTokens(), chapter);
-  return parser.parse();
-}