finn-lang/Program.cs

using System;
using System.IO;
using System.Collections;
using System.Collections.Generic;

namespace Finn;

class Program
{
	static bool hadError = false;

	static void Main(string[] args)
	{
		if (args.Length > 1)
		{
			Console.WriteLine("Usage: finn [script]");
			Environment.Exit(64);
		}
		else if (args.Length == 1)
		{
			runFile(args[0]);
		}
		else
		{
			runPrompt();
		}
	}

	static void runFile(string path)
	{
		var src = File.ReadAllText(path);
		run(src);
	}

	static void run(string src)
	{
		var scanner = new Scanner(src);
		List<Token> tokens = scanner.scanTokens();

		foreach (Token token in tokens)
		{
			Console.WriteLine(token);
		}
	}

	static void runPrompt()
	{
		var input = new BufferedStream(Console.OpenStandardInput());
		var reader = new StreamReader(input);

		while (true)
		{
			Console.Write("> ");
			var line = reader.ReadLine();
			if (line == null) break;
			run(line);
		}
	}

	public static void error(int line, String message)
	{
		report(line, "", message);
	}

	static void report(int line, String where, String message)
	{
		Console.WriteLine($"[line {line}] Error{where}: {message}");
		hadError = true;
	}
}

enum TokenType
{
	LParen, RParen,
	LBrace, RBrace,
	LBracket, RBracket,
	Less, Greater,
	Pipe,
	Colon,
	Tick,
	Backtick,
	Comma,
	Period,
	Equal,
	Plus, Minus, Asterisk, Slash,
	PlusPlus,
	DoubleEqual,
	Bang, BangEqual,
	LessEqual, GreaterEqual,
	SingleArrow,
	DoubleArrow,
	PipeRight,
	PipeLeft,
	ComposeRight,
	ComposeLeft,
	When,
	Is,
	Let,
	In,
	With,
	Fn,
	Type,
	Alias,
	Recurse,
	Blank,
	Identifier, Number, String,
	EOF
}

class Token
{
	public readonly TokenType type;
	public readonly String lexeme;
	public readonly Object? literal;
	public readonly int line;

	public Token(TokenType type, String lexeme, Object? literal, int line)
	{
		this.type = type;
		this.lexeme = lexeme;
		this.literal = literal;
		this.line = line;
	}

	public override string ToString()
	{
		return $"{type} {lexeme} {literal}";
	}
}

class Scanner
{
	private readonly String source;
	private readonly List<Token> tokens = new List<Token>();
	private int start = 0;
	private int current = 0;
	private int line = 1;

	private static readonly Dictionary<String, TokenType> keywords = new Dictionary<string, TokenType>()
	{
		{"when", TokenType.When},
		{"is", TokenType.Is},
		{"let", TokenType.Let},
		{"in", TokenType.In},
		{"with", TokenType.With},
		{"fn", TokenType.Fn},
		{"type", TokenType.Type},
		{"alias", TokenType.Alias},
		{"recurse", TokenType.Recurse},
		{"_", TokenType.Blank},
	};

	public Scanner(String source)
	{
		this.source = source;
	}

	private bool isAtEnd()
	{
		return current >= source.Length;
	}

	private char advance()
	{
		return source[current++];
	}

	private void addToken(TokenType type)
	{
		addToken(type, null);
	}

	private void addToken(TokenType type, Object? literal)
	{
		String text = source.Substring(start, current - start);
		tokens.Add(new Token(type, text, literal, line));
	}

	private bool match(char expected)
	{
		if (isAtEnd()) return false;
		if (source[current] != expected) return false;

		current++;
		return true;
	}

	private bool match(Predicate<char> pred)
	{
		if (isAtEnd()) return false;
		if (!pred(source[current])) return false;

		current++;
		return true;
	}

	private char? peek()
	{
		if (isAtEnd()) return null;
		return source[current];
	}

	private void stringLiteral()
	{
		while (peek() != '"' && !isAtEnd())
		{
			if (peek() == '\n') line++;
			advance();
		}

		if (isAtEnd())
		{
			Program.error(line, "Unterminated string.");
			return;
		}

		// The closing ".
		advance();

		// Trim the surrounding quotes.
		String value = source.Substring(start + 1, current - start - 2);
		addToken(TokenType.String, value);
	}

	private bool isDigitOrSeparator(char c)
	{
		return Char.IsAsciiDigit(c) || c == '_';
	}

	private char? peekNext()
	{
		if (current + 1 >= source.Length) return null;
		return source[current + 1];
	}
	private void numberLiteral()
	{
		while (match(isDigitOrSeparator)) ;

		// Look for a fractional part.
		if (peek() == '.' && isDigitOrSeparator(peekNext() ?? '\0'))
		{
			match('.');
			while (match(isDigitOrSeparator)) ;
		}
		double value = Double.Parse(source.Substring(start, current - start));
		addToken(TokenType.Number, value);
	}

	private bool isIdentifierStartChar(char c)
	{
		return Char.IsAsciiLetter(c) || c == '_';
	}

	private bool isIdentifierChar(char c)
	{
		return isIdentifierStartChar(c) || Char.IsAsciiDigit(c);
	}

	private void identifier()
	{
		while (match(isIdentifierChar)) ;
		String text = source.Substring(start, current - start);
		TokenType type;
		var isKeyword = keywords.TryGetValue(text, out type);
		addToken(isKeyword ? type : TokenType.Identifier);
	}

	private void scanToken()
	{
		char c = advance();
		switch (c)
		{
			case '(': addToken(TokenType.LParen); break;
			case ')': addToken(TokenType.RParen); break;
			case '{': addToken(TokenType.LBrace); break;
			case '}': addToken(TokenType.RBrace); break;
			case '[': addToken(TokenType.LBracket); break;
			case ']': addToken(TokenType.RBracket); break;
			case ':': addToken(TokenType.Colon); break;
			case '\'': addToken(TokenType.Tick); break;
			case '`': addToken(TokenType.Backtick); break;
			case ',': addToken(TokenType.Comma); break;
			case '.': addToken(TokenType.Period); break;
			case '*': addToken(TokenType.Asterisk); break;
			case '/': addToken(TokenType.Slash); break;
			case '-':
				addToken(match('>') ? TokenType.SingleArrow : TokenType.Minus);
				break;
			case '!':
				addToken(match('=') ? TokenType.BangEqual : TokenType.Bang);
				break;
			case '+':
				addToken(match('+') ? TokenType.PlusPlus : TokenType.Plus);
				break;
			case '=':
				addToken(match('=') ? TokenType.DoubleEqual :
					match('>') ? TokenType.DoubleArrow :
					TokenType.Equal);
				break;
			case '<':
				addToken(match('=') ? TokenType.LessEqual :
					match('<') ? TokenType.ComposeLeft :
					match('|') ? TokenType.PipeLeft :
					TokenType.Less);
				break;
			case '>':
				addToken(match('=') ? TokenType.GreaterEqual :
					match('>') ? TokenType.ComposeRight : TokenType.Greater);
				break;
			case '|':
				addToken(match('>') ? TokenType.PipeRight : TokenType.Pipe);
				break;
			case '"': stringLiteral(); break;
			case '#':
				while (peek() != '\n' && !isAtEnd()) advance();
				break;
			case ' ':
			case '\r':
			case '\t':
				break;
			case '\n':
				line++;
				break;
			default:
				if (Char.IsAsciiDigit(c))
				{
					numberLiteral();
				}
				else if (isIdentifierStartChar(c))
				{
					identifier();
				}
				else
				{
					Program.error(line, "Unexpected character.");
				}
				break;
		}
	}
	public List<Token> scanTokens()
	{
		while (!isAtEnd())
		{
			start = current;
			scanToken();
		}

		tokens.Add(new Token(TokenType.EOF, "", null, line));
		return tokens;
	}
}