grindlemire/go-tui code browser

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
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317

package tuigen

import (
	"unicode/utf8"
)

// Lexer tokenizes .gsx source files.
type Lexer struct {
	filename string
	source   string
	pos      int  // current position in source
	readPos  int  // next position to read
	ch       rune // current character
	line     int  // current line (1-based)
	column   int  // current column (1-based)

	// Track the start position of current token
	tokenLine     int
	tokenColumn   int
	tokenStartPos int // byte offset where current token starts

	// Comments collected since last ConsumeComments() call
	pendingComments []*Comment

	// Track the end line of the last comment that was collected
	// (persists across ConsumeComments calls to detect blank lines between comment batches)
	lastCommentEndLine int

	errors *ErrorList
}

// NewLexer creates a new Lexer for the given source.
func NewLexer(filename, source string) *Lexer {
	l := &Lexer{
		filename: filename,
		source:   source,
		line:     1,
		column:   0,
		errors:   NewErrorList(),
	}
	l.readChar()
	return l
}

// Errors returns any errors encountered during lexing.
func (l *Lexer) Errors() *ErrorList {
	return l.errors
}

// readChar advances to the next character in the source.
func (l *Lexer) readChar() {
	// Track if previous char was a newline for line counting
	prevWasNewline := l.ch == '\n'

	if l.readPos >= len(l.source) {
		l.ch = 0 // EOF
		l.pos = l.readPos
		if prevWasNewline {
			l.line++
			l.column = 1
		} else {
			l.column++
		}
		return
	}

	r, size := utf8.DecodeRuneInString(l.source[l.readPos:])
	l.ch = r
	l.pos = l.readPos
	l.readPos += size

	if prevWasNewline {
		l.line++
		l.column = 1
	} else {
		l.column++
	}
}

// peekChar returns the next character without advancing.
func (l *Lexer) peekChar() rune {
	if l.readPos >= len(l.source) {
		return 0
	}
	r, _ := utf8.DecodeRuneInString(l.source[l.readPos:])
	return r
}

// startToken marks the beginning of a new token.
func (l *Lexer) startToken() {
	l.tokenLine = l.line
	l.tokenColumn = l.column
	l.tokenStartPos = l.pos
}

// makeToken creates a token with the current start position.
func (l *Lexer) makeToken(typ TokenType, literal string) Token {
	// Reset lastCommentEndLine for non-newline tokens
	// This ensures we don't detect false "blank lines" between comment groups
	// that are actually separated by code
	if typ != TokenNewline && typ != TokenEOF {
		l.lastCommentEndLine = 0
	}
	return Token{
		Type:     typ,
		Literal:  literal,
		Line:     l.tokenLine,
		Column:   l.tokenColumn,
		StartPos: l.tokenStartPos,
	}
}

// position returns the current Position for error reporting.
func (l *Lexer) position() Position {
	return Position{
		File:   l.filename,
		Line:   l.tokenLine,
		Column: l.tokenColumn,
	}
}

// LexerState captures the lexer's position for save/restore.
type LexerState struct {
	pos, readPos int
	ch           rune
	line, column int

	tokenLine, tokenColumn int
	tokenStartPos          int
	pendingComments        []*Comment
	lastCommentEndLine     int
}

// SaveState captures the current lexer state for speculative parsing.
func (l *Lexer) SaveState() LexerState {
	return LexerState{
		pos:                l.pos,
		readPos:            l.readPos,
		ch:                 l.ch,
		line:               l.line,
		column:             l.column,
		tokenLine:          l.tokenLine,
		tokenColumn:        l.tokenColumn,
		tokenStartPos:      l.tokenStartPos,
		pendingComments:    l.pendingComments,
		lastCommentEndLine: l.lastCommentEndLine,
	}
}

// RestoreState restores the lexer to a previously saved state.
func (l *Lexer) RestoreState(s LexerState) {
	l.pos = s.pos
	l.readPos = s.readPos
	l.ch = s.ch
	l.line = s.line
	l.column = s.column
	l.tokenLine = s.tokenLine
	l.tokenColumn = s.tokenColumn
	l.tokenStartPos = s.tokenStartPos
	l.pendingComments = s.pendingComments
	l.lastCommentEndLine = s.lastCommentEndLine
}

// Source returns the raw source string. Used by the parser for lookahead scanning.
func (l *Lexer) Source() string {
	return l.source
}

// Next returns the next token from the source.
func (l *Lexer) Next() Token {
	// Skip whitespace and collect any comments
	l.skipWhitespaceAndCollectComments()

	l.startToken()

	switch l.ch {
	case 0:
		return l.makeToken(TokenEOF, "")

	case '\n':
		l.readChar()
		return l.makeToken(TokenNewline, "\n")

	case '(':
		l.readChar()
		return l.makeToken(TokenLParen, "(")

	case ')':
		l.readChar()
		return l.makeToken(TokenRParen, ")")

	case '{':
		l.readChar()
		return l.makeToken(TokenLBrace, "{")

	case '}':
		l.readChar()
		return l.makeToken(TokenRBrace, "}")

	case '[':
		l.readChar()
		return l.makeToken(TokenLBracket, "[")

	case ']':
		l.readChar()
		return l.makeToken(TokenRBracket, "]")

	case '<':
		if l.peekChar() == '/' {
			l.readChar() // consume <
			l.readChar() // consume /
			return l.makeToken(TokenLAngleSlash, "</")
		}
		l.readChar()
		return l.makeToken(TokenLAngle, "<")

	case '>':
		l.readChar()
		return l.makeToken(TokenRAngle, ">")

	case '/':
		if l.peekChar() == '>' {
			l.readChar() // consume /
			l.readChar() // consume >
			return l.makeToken(TokenSlashAngle, "/>")
		}
		l.readChar()
		return l.makeToken(TokenSlash, "/")

	case '=':
		l.readChar()
		return l.makeToken(TokenEquals, "=")

	case ',':
		l.readChar()
		return l.makeToken(TokenComma, ",")

	case '.':
		// Could be . or ... or a number like .5
		if isDigit(l.peekChar()) {
			return l.readNumber()
		}
		l.readChar()
		return l.makeToken(TokenDot, ".")

	case ':':
		if l.peekChar() == '=' {
			l.readChar() // consume :
			l.readChar() // consume =
			return l.makeToken(TokenColonEquals, ":=")
		}
		l.readChar()
		return l.makeToken(TokenColon, ":")

	case ';':
		l.readChar()
		return l.makeToken(TokenSemicolon, ";")

	case '&':
		l.readChar()
		return l.makeToken(TokenAmpersand, "&")

	case '|':
		l.readChar()
		return l.makeToken(TokenPipe, "|")

	case '*':
		l.readChar()
		return l.makeToken(TokenStar, "*")

	case '+':
		l.readChar()
		return l.makeToken(TokenPlus, "+")

	case '-':
		l.readChar()
		return l.makeToken(TokenMinus, "-")

	case '!':
		l.readChar()
		return l.makeToken(TokenBang, "!")

	case '_':
		// Could be _ or an identifier starting with _
		if isLetter(l.peekChar()) || isDigit(l.peekChar()) {
			return l.readIdentifier()
		}
		l.readChar()
		return l.makeToken(TokenUnderscore, "_")

	case '@':
		return l.readAtKeyword()

	case '"':
		return l.readString()

	case '\'':
		return l.readRune()

	case '`':
		return l.readRawString()

	default:
		if isLetter(l.ch) {
			return l.readIdentifier()
		}
		if isDigit(l.ch) {
			return l.readNumber()
		}

		// Unknown character - treat as symbol for text content
		// (e.g., Unicode arrows, emoji, etc.)
		ch := l.ch
		l.readChar()
		return l.makeToken(TokenSymbol, string(ch))
	}
}