|
1
|
3 |
|
import Token, { TokenType } from './Token' |
|
2
|
|
|
|
|
3
|
3 |
|
const OPERATORS = { |
|
4
|
|
|
'+': 'add', |
|
5
|
|
|
'-': 'sub', |
|
6
|
|
|
'*': 'mul', |
|
7
|
|
|
'/': 'div', |
|
8
|
|
|
'**': 'pow', |
|
9
|
|
|
'^': 'pow', |
|
10
|
|
|
'=': 'set', |
|
11
|
|
|
'==': 'equals', |
|
12
|
|
|
'<': 'isLessThan', |
|
13
|
|
|
'>': 'isGreaterThan', |
|
14
|
|
|
} |
|
15
|
|
|
|
|
16
|
|
|
/** |
|
17
|
|
|
* @class Lexer |
|
18
|
|
|
* @name Lexer |
|
19
|
|
|
*/ |
|
20
|
3 |
|
export class Lexer { |
|
21
|
|
|
s: string |
|
22
|
|
|
|
|
23
|
|
|
/** |
|
24
|
|
|
* Initialize a lexer. |
|
25
|
|
|
*/ |
|
26
|
|
|
constructor(input: string) { |
|
27
|
16 |
|
this.s = input |
|
28
|
|
|
} |
|
29
|
|
|
|
|
30
|
|
|
/** |
|
31
|
|
|
* Yields tokens as they are lexed. |
|
32
|
|
|
* |
|
33
|
|
|
*/ |
|
34
|
|
|
*lex(): Generator<Token> { |
|
35
|
14 |
|
let onSpace = false |
|
36
|
14 |
|
let onInteger = false |
|
37
|
14 |
|
let onWord = false |
|
38
|
14 |
|
let onOperator = false |
|
39
|
14 |
|
let buffer = '' |
|
40
|
|
|
|
|
41
|
|
|
// for each character |
|
42
|
14 |
|
for (let i = 0; i < this.s.length; i++) { |
|
43
|
59 |
|
const char = this.s[i] ?? '?' |
|
44
|
|
|
|
|
45
|
|
|
// if a space, combine with subsequent spaces into a separator |
|
46
|
59 |
|
if (/^\s$/.exec(char)) { |
|
47
|
12 |
|
if (onInteger) { |
|
48
|
4 |
|
yield new Token(TokenType.literal, buffer) |
|
49
|
4 |
|
onInteger = false |
|
50
|
4 |
|
buffer = '' |
|
51
|
8 |
|
} else if (onWord) { |
|
52
|
2 |
|
yield new Token(TokenType.identifier, buffer) |
|
53
|
2 |
|
onWord = false |
|
54
|
2 |
|
buffer = '' |
|
55
|
|
|
} |
|
56
|
12 |
|
onSpace = true |
|
57
|
12 |
|
buffer += char |
|
58
|
12 |
|
continue |
|
59
|
47 |
|
} else if (onSpace) { |
|
60
|
7 |
|
yield new Token(TokenType.separator, '') |
|
61
|
7 |
|
onSpace = false |
|
62
|
7 |
|
buffer = '' |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
// if a digit, combine with subsequent digits into a literal |
|
66
|
47 |
|
if (/^\d$/.exec(char)) { |
|
67
|
14 |
|
if (!onWord) { |
|
68
|
11 |
|
onInteger = true |
|
69
|
|
|
} |
|
70
|
14 |
|
buffer += char |
|
71
|
14 |
|
continue |
|
72
|
33 |
|
} else if (onInteger) { |
|
73
|
1 |
|
yield new Token(TokenType.literal, buffer) |
|
74
|
1 |
|
onInteger = false |
|
75
|
1 |
|
buffer = '' |
|
76
|
|
|
} |
|
77
|
|
|
|
|
78
|
|
|
// if a letter, combine with subsequent letters/numbers into an identifier |
|
79
|
33 |
|
if (/^[a-z]$/.exec(char) || (onWord && /^\d$/.exec(char))) { |
|
80
|
26 |
|
onWord = true |
|
81
|
26 |
|
buffer += char |
|
82
|
26 |
|
continue |
|
83
|
7 |
|
} else if (onWord) { |
|
84
|
1 |
|
yield new Token(TokenType.identifier, buffer) |
|
85
|
1 |
|
onWord = false |
|
86
|
1 |
|
buffer = '' |
|
87
|
|
|
} |
|
88
|
|
|
|
|
89
|
|
|
// if another char, combine with subsequent char and match an operator |
|
90
|
7 |
|
buffer += char |
|
91
|
7 |
|
if (buffer in OPERATORS) { |
|
92
|
3 |
|
yield new Token(TokenType.operator, buffer) |
|
93
|
3 |
|
onOperator = false |
|
94
|
3 |
|
buffer = '' |
|
95
|
3 |
|
continue |
|
96
|
|
|
} |
|
97
|
4 |
|
if (onOperator) { |
|
98
|
2 |
|
throw `Invalid operator "${buffer}"` |
|
99
|
|
|
} |
|
100
|
2 |
|
onOperator = true |
|
101
|
|
|
} |
|
102
|
|
|
|
|
103
|
12 |
|
if (onSpace) { |
|
104
|
3 |
|
yield new Token(TokenType.separator, '') |
|
105
|
|
|
} |
|
106
|
|
|
|
|
107
|
12 |
|
if (onInteger) { |
|
108
|
3 |
|
yield new Token(TokenType.literal, buffer) |
|
109
|
|
|
} |
|
110
|
|
|
|
|
111
|
12 |
|
if (onWord) { |
|
112
|
6 |
|
yield new Token(TokenType.identifier, buffer) |
|
113
|
|
|
} |
|
114
|
|
|
} |
|
115
|
|
|
} |
|
116
|
|
|
|
|
117
|
|
|
export default Lexer |
|
118
|
|
|
|
acerix /
cnum
