-
Notifications
You must be signed in to change notification settings - Fork 5
/
lexer.go
277 lines (230 loc) · 4.79 KB
/
lexer.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
package main
import (
"fmt"
"strings"
"unicode/utf8"
)
// ItemType is type of lexer items
type ItemType int
func (typ ItemType) String() string {
strType := "UNDEFINED"
switch typ {
case IERR:
strType = "IERR"
case INUMBER:
strType = "INUMBER"
case ILPAR:
strType = "ILPAR"
case IRPAR:
strType = "IRPAR"
case IADD:
strType = "IADD"
case ISUB:
strType = "ISUB"
case IMUL:
strType = "IMUL"
case IDIV:
strType = "IDIV"
case EOF:
strType = "EOF"
}
return strType
}
// Item types from lexer items
const (
// EOF Item type denoting end of input
EOF ItemType = -1
// lexing error occured
IERR ItemType = iota
// positive integer
INUMBER
// left parenthesis
ILPAR
// right parenthesis
IRPAR
// plus + symbol
IADD
// minus - symbol
ISUB
// multiply * symbol
IMUL
// divide / symbol
IDIV
)
// Tokens understood by lexer
const (
numbers = "0123456789"
operators = "+-*/"
white = " \n\r\t"
lpar = "("
rpar = ")"
)
// Type for lexing state machine, function returns another function which represents state transition
type stateFn func(*Lexer) stateFn
// LexItem are items emitted by lexer
type LexItem struct {
Typ ItemType
Pos int
Val string
}
// Debug stringify lex item
func (li LexItem) String() string {
return fmt.Sprintf("Type: %s, Val: %q, Pos: %d", li.Typ, li.Val, li.Pos)
}
// Lexer class containing state of lexer
type Lexer struct {
// input text
text string
// start index (of input text) of currently lexing token
start int
// current position (of input text) of lexer
Pos int
// width of last read rune
width int
// output channel of lexer
items chan LexItem
}
func (l *Lexer) dumpState() {
fmt.Printf("%#v\n", l)
}
// Move to next ASCII or UTF-8 character/rune
func (l *Lexer) next() rune {
if l.Pos >= len(l.text) {
l.width = 0
return -1 // EOF
}
r, w := utf8.DecodeRuneInString(l.text[l.Pos:])
l.width = w
l.Pos += w
return r
}
// Go back one character/rune
func (l *Lexer) backup() {
l.Pos -= l.width
_, w := utf8.DecodeLastRuneInString(l.text[:l.Pos])
l.width = w
}
// Check what rune is next in input
func (l *Lexer) peek() rune {
r := l.next()
l.backup()
return r
}
// Consume next rune if its one of 'runes' parameter, otherwise no effect
func (l *Lexer) consume(runes string) bool {
if strings.ContainsRune(runes, l.next()) {
return true
}
l.backup()
return false
}
// Consume all following runes that are one of 'runes'
func (l *Lexer) consumeAll(runes string) {
for l.consume(runes) {
}
}
// Ignores all un-emitted characters (moves start to current pos)
func (l *Lexer) ignore() {
l.start = l.Pos
l.width = 0
}
// Emits lexer item to output channel
func (l *Lexer) emit(typ ItemType) {
l.items <- LexItem{
Typ: typ,
Pos: l.Pos,
Val: l.text[l.start:l.Pos],
}
l.start = l.Pos
}
// Helper func that emits error item IERR with message
func (l *Lexer) errorf(format string, args ...interface{}) stateFn {
l.items <- LexItem{
Pos: l.Pos,
Typ: IERR,
Val: fmt.Sprintf(format, args...),
}
return nil
}
// Starting state of state machine, peeks forward and decides what lexing function should be used
func lexFn(l *Lexer) stateFn {
r := l.peek()
switch {
case r == -1:
l.emit(EOF)
return nil
case strings.ContainsRune(white, r):
return lexWhite
case strings.ContainsRune(operators, r):
return lexOperator
case strings.ContainsRune(numbers, r):
return lexNumber
case strings.ContainsRune(lpar, r):
return lexLpar
case strings.ContainsRune(rpar, r):
return lexRpar
default:
return l.errorf("Invalid symbol: %q", r)
}
}
// Lexes operators
func lexOperator(l *Lexer) stateFn {
op := l.next()
switch op {
case '+':
l.emit(IADD)
case '-':
l.emit(ISUB)
case '*':
l.emit(IMUL)
case '/':
l.emit(IDIV)
default:
return l.errorf("lexOperator: inValid operator: %q", op)
}
return lexFn
}
// Lexes left parenthesis
func lexLpar(l *Lexer) stateFn {
l.consume(lpar)
l.emit(ILPAR)
return lexFn
}
// Lexes right parenthesis
func lexRpar(l *Lexer) stateFn {
l.consume(rpar)
l.emit(IRPAR)
return lexFn
}
// lexes numbers
func lexNumber(l *Lexer) stateFn {
l.consumeAll(numbers)
l.emit(INUMBER)
return lexFn
}
// Lexes whitespaces and thrashes them (no emitting)
func lexWhite(l *Lexer) stateFn {
l.consumeAll(white)
l.ignore()
return lexFn
}
// Items method gets channel of lex items
func (l *Lexer) Items() chan LexItem {
return l.items
}
// Run method will is the core part of this lexer
// It fires off the state machine, starting with lexFn and takes return function as next state
// once some function returns nil (e.g. EOF), it stops lexing and returns
func (l *Lexer) Run() {
defer close(l.items)
for fun := lexFn; fun != nil; {
fun = fun(l)
}
}
// Lex is constructor for lexer
func Lex(text string) *Lexer {
return &Lexer{
items: make(chan LexItem),
text: text,
}
}