Compare commits
No commits in common. "bba969922fb7dc8a80ef52285cafbc625f964335" and "d1db7a0bd976416ea5f38744f0075fa2d47ba56e" have entirely different histories.
bba969922f
...
d1db7a0bd9
3 changed files with 4 additions and 148 deletions
|
@ -9,7 +9,7 @@ fn run(allocator: *Allocator, data: []u8) !void {
|
|||
var stdout_file = try std.io.getStdOut();
|
||||
const stdout = &stdout_file.outStream().stream;
|
||||
|
||||
var scanner = try Scanner.init(allocator, data);
|
||||
var scanner = Scanner.init(allocator, data);
|
||||
var tokens = try scanner.scanTokens();
|
||||
var it = tokens.iterator();
|
||||
|
||||
|
@ -21,9 +21,6 @@ fn run(allocator: *Allocator, data: []u8) !void {
|
|||
.Slice => |value| {
|
||||
try value.printToken(stdout);
|
||||
},
|
||||
.Number => |value| {
|
||||
try value.printToken(stdout);
|
||||
},
|
||||
}
|
||||
|
||||
hadError = false;
|
||||
|
|
143
src/scanner.zig
143
src/scanner.zig
|
@ -4,88 +4,19 @@ const token = @import("token.zig");
|
|||
const main = @import("main.zig");
|
||||
|
||||
const TokenList = std.ArrayList(token.Token);
|
||||
const TokenType = token.TokenType;
|
||||
|
||||
fn isDigit(char: u8) bool {
|
||||
return char >= '0' and char <= '9';
|
||||
}
|
||||
|
||||
fn isAlpha(c: u8) bool {
|
||||
return (c >= 'a' and c <= 'z') or
|
||||
(c >= 'A' and c <= 'Z') or
|
||||
c == '_';
|
||||
}
|
||||
|
||||
fn isAlphaNumeric(char: u8) bool {
|
||||
return isAlpha(char) or isDigit(char);
|
||||
}
|
||||
|
||||
// hashmaps don't work on HashMaps for some reason. anyways.
|
||||
pub const KeywordMap = std.AutoHashMap([]const u8, u6);
|
||||
|
||||
fn initKeywordMap(allocator: *std.mem.Allocator) !KeywordMap {
|
||||
var map = KeywordMap.init(allocator);
|
||||
|
||||
const keywords = [][]const u8{
|
||||
"and"[0..],
|
||||
"class"[0..],
|
||||
"else"[0..],
|
||||
"false"[0..],
|
||||
"for"[0..],
|
||||
"fun"[0..],
|
||||
"if"[0..],
|
||||
"nil"[0..],
|
||||
"or"[0..],
|
||||
"print"[0..],
|
||||
"return"[0..],
|
||||
"super"[0..],
|
||||
"this"[0..],
|
||||
"true"[0..],
|
||||
"var"[0..],
|
||||
"while"[0..],
|
||||
};
|
||||
|
||||
const tags = []TokenType{
|
||||
TokenType.AND,
|
||||
TokenType.CLASS,
|
||||
TokenType.ELSE,
|
||||
TokenType.FALSE,
|
||||
TokenType.FOR,
|
||||
TokenType.FUN,
|
||||
TokenType.IF,
|
||||
TokenType.NIL,
|
||||
TokenType.OR,
|
||||
TokenType.PRINT,
|
||||
TokenType.RETURN,
|
||||
TokenType.SUPER,
|
||||
TokenType.THIS,
|
||||
TokenType.TRUE,
|
||||
TokenType.VAR,
|
||||
TokenType.WHILE,
|
||||
};
|
||||
|
||||
for (keywords) |keyword, idx| {
|
||||
var tag = @enumToInt(tags[idx]);
|
||||
_ = try map.put(keyword, tag);
|
||||
}
|
||||
|
||||
return map;
|
||||
}
|
||||
|
||||
pub const Scanner = struct {
|
||||
source: []u8,
|
||||
tokens: TokenList,
|
||||
keywords: KeywordMap,
|
||||
|
||||
start: usize = 0,
|
||||
current: usize = 0,
|
||||
line: usize = 1,
|
||||
|
||||
pub fn init(allocator: *std.mem.Allocator, data: []u8) !Scanner {
|
||||
pub fn init(allocator: *std.mem.Allocator, data: []u8) Scanner {
|
||||
return Scanner{
|
||||
.source = data,
|
||||
.tokens = TokenList.init(allocator),
|
||||
.keywords = try initKeywordMap(allocator),
|
||||
};
|
||||
}
|
||||
|
||||
|
@ -124,18 +55,6 @@ pub const Scanner = struct {
|
|||
});
|
||||
}
|
||||
|
||||
/// Keep in mind Lox only has a single number type and that is a float one.
|
||||
fn addNumberToken(self: *Scanner, ttype: token.TokenType, num: f32) !void {
|
||||
try self.addToken(token.Token{
|
||||
.Number = token.NumberToken.init(
|
||||
ttype,
|
||||
self.currentLexeme(),
|
||||
self.line,
|
||||
num,
|
||||
),
|
||||
});
|
||||
}
|
||||
|
||||
fn addToken(
|
||||
self: *Scanner,
|
||||
tok: token.Token,
|
||||
|
@ -195,58 +114,6 @@ pub const Scanner = struct {
|
|||
);
|
||||
}
|
||||
|
||||
fn peekNext(self: *Scanner) u8 {
|
||||
if (self.current + 1 >= self.source.len) return 0;
|
||||
return self.source[self.current + 1];
|
||||
}
|
||||
|
||||
/// Consume a number
|
||||
fn doNumber(self: *Scanner) !void {
|
||||
while (isDigit(self.peek())) {
|
||||
_ = self.advance();
|
||||
}
|
||||
|
||||
// check if its a number like 12.34, where the '.' character
|
||||
// exists and the one next to it is a digit.
|
||||
if (self.peek() == '.' and isDigit(self.peekNext())) {
|
||||
_ = self.advance();
|
||||
|
||||
while (isDigit(self.peek())) {
|
||||
_ = self.advance();
|
||||
}
|
||||
}
|
||||
|
||||
// after going through all of the number, we can just use fmt.parseFloat
|
||||
|
||||
var num = try std.fmt.parseFloat(
|
||||
f32,
|
||||
self.source[self.start..self.current],
|
||||
);
|
||||
|
||||
try self.addNumberToken(.NUMBER, num);
|
||||
}
|
||||
|
||||
fn doIdentifier(self: *Scanner) !void {
|
||||
while (isAlphaNumeric(self.peek())) {
|
||||
_ = self.advance();
|
||||
}
|
||||
|
||||
// after reading the identifier, we check
|
||||
// if it is any of our keywords, if it is, then we add
|
||||
// the specificed keyword type. if not, just .IDENTIFIER
|
||||
var text = self.source[self.start..self.current];
|
||||
var type_opt = self.keywords.get(text);
|
||||
var toktype: TokenType = undefined;
|
||||
|
||||
if (type_opt) |kv| {
|
||||
toktype = @intToEnum(TokenType, kv.value);
|
||||
} else {
|
||||
toktype = TokenType.IDENTIFIER;
|
||||
}
|
||||
|
||||
try self.addSimpleToken(toktype);
|
||||
}
|
||||
|
||||
/// Scan through our tokens and add them to the Scanner's token list.
|
||||
fn scanToken(self: *Scanner) !void {
|
||||
var c = self.advance();
|
||||
|
@ -287,13 +154,7 @@ pub const Scanner = struct {
|
|||
'"' => try self.doString(),
|
||||
|
||||
else => {
|
||||
if (isDigit(c)) {
|
||||
try self.doNumber();
|
||||
} else if (isAlpha(c)) {
|
||||
try self.doIdentifier();
|
||||
} else {
|
||||
try main.doError(self.line, "Unexpected character");
|
||||
}
|
||||
try main.doError(self.line, "Unexpected character");
|
||||
},
|
||||
}
|
||||
}
|
||||
|
|
|
@ -1,6 +1,6 @@
|
|||
const std = @import("std");
|
||||
|
||||
pub const TokenType = enum(u6) {
|
||||
pub const TokenType = enum {
|
||||
// Single-character tokens.
|
||||
LEFT_PAREN,
|
||||
RIGHT_PAREN,
|
||||
|
@ -98,10 +98,8 @@ pub fn TokenFactory(
|
|||
|
||||
pub const SimpleToken = TokenFactory(void);
|
||||
pub const SliceToken = TokenFactory([]u8);
|
||||
pub const NumberToken = TokenFactory(f32);
|
||||
|
||||
pub const Token = union(enum) {
|
||||
Simple: SimpleToken,
|
||||
Slice: SliceToken,
|
||||
Number: NumberToken,
|
||||
};
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue