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
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
// this_file: crates/core/src/streaming/buffered/lexer.rs
//! Buffered lexer for streaming JSON parsing
//!
//! This lexer can handle incremental input and properly tokenize across buffer boundaries.
use crate::ast::Token;
use crate::error::{Error, Result, Span};
use crate::lexer::{LexerConfig, LexerMode};
/// State of the lexer for resumable parsing
#[derive(Debug, Clone)]
pub enum LexerState {
/// Normal state, expecting any token
Normal,
/// Inside a string, with quote character
InString { quote: char, escape_next: bool },
/// Inside a single-line comment
InSingleLineComment,
/// Inside a multi-line comment
InMultiLineComment { star_seen: bool },
/// Building a number
InNumber,
/// Building an identifier (for keywords and unquoted keys)
InIdentifier,
}
/// Buffered lexer that can handle incremental input
pub struct BufferedLexer {
/// Current state for resumable parsing
state: LexerState,
/// Buffer for the current token being built
token_buffer: String,
/// Position in the overall input stream
position: usize,
/// Start position of current token
token_start: usize,
/// Configuration
config: LexerConfig,
}
impl BufferedLexer {
/// Create a new buffered lexer
pub fn new(config: LexerConfig) -> Self {
Self {
state: LexerState::Normal,
token_buffer: String::new(),
position: 0,
token_start: 0,
config,
}
}
/// Feed a chunk of input to the lexer
/// Returns tokens that were completed and whether more input is needed
pub fn feed(&mut self, input: &str) -> Result<(Vec<(Token, Span)>, bool)> {
let mut tokens = Vec::new();
let mut chars = input.chars().peekable();
let mut pending = None;
while let Some(ch) = pending.take().or_else(|| chars.next()) {
match &self.state {
LexerState::Normal => {
match ch {
// Whitespace
' ' | '\t' | '\r' => {
// Skip whitespace
}
'\n' => {
tokens.push((Token::Newline, self.make_span(1)));
}
// Structural characters
'{' => tokens.push((Token::LeftBrace, self.make_span(1))),
'}' => tokens.push((Token::RightBrace, self.make_span(1))),
'[' => tokens.push((Token::LeftBracket, self.make_span(1))),
']' => tokens.push((Token::RightBracket, self.make_span(1))),
',' => tokens.push((Token::Comma, self.make_span(1))),
':' => tokens.push((Token::Colon, self.make_span(1))),
// Strings
'"' | '\'' => {
self.state = LexerState::InString { quote: ch, escape_next: false };
self.token_start = self.position;
self.token_buffer.clear();
}
// Comments
'/' if self.config.mode == LexerMode::Forgiving => {
if let Some(&next_ch) = chars.peek() {
match next_ch {
'/' => {
chars.next(); // consume second /
self.position += 1;
self.state = LexerState::InSingleLineComment;
}
'*' => {
chars.next(); // consume *
self.position += 1;
self.state = LexerState::InMultiLineComment { star_seen: false };
}
_ => {
// Not a comment, treat as error
return Err(Error::UnexpectedChar('/', self.position));
}
}
} else {
// Need more input to determine if it's a comment
self.token_buffer.push(ch);
self.token_start = self.position;
self.position += 1;
return Ok((tokens, true));
}
}
// Numbers
'-' | '0'..='9' => {
self.state = LexerState::InNumber;
self.token_start = self.position;
self.token_buffer.clear();
self.token_buffer.push(ch);
}
// Identifiers (null, true, false, unquoted keys)
'a'..='z' | 'A'..='Z' | '_' => {
self.state = LexerState::InIdentifier;
self.token_start = self.position;
self.token_buffer.clear();
self.token_buffer.push(ch);
}
_ => {
return Err(Error::UnexpectedChar(ch, self.position));
}
}
}
LexerState::InString { quote, escape_next } => {
if *escape_next {
// Just add the escaped character
self.token_buffer.push(ch);
self.state = LexerState::InString { quote: *quote, escape_next: false };
} else if ch == '\\' {
self.token_buffer.push(ch);
self.state = LexerState::InString { quote: *quote, escape_next: true };
} else if ch == *quote {
// String complete - increment position first to include the closing quote
self.position += 1;
tokens.push((Token::String, self.make_span_from_start()));
self.state = LexerState::Normal;
self.token_buffer.clear();
continue; // Skip the normal position increment
} else {
self.token_buffer.push(ch);
}
}
LexerState::InSingleLineComment => {
if ch == '\n' {
self.state = LexerState::Normal;
tokens.push((Token::Newline, self.make_span(1)));
}
// Otherwise, ignore characters in comment
}
LexerState::InMultiLineComment { star_seen } => {
if *star_seen && ch == '/' {
// End of multi-line comment
self.state = LexerState::Normal;
} else {
self.state = LexerState::InMultiLineComment { star_seen: ch == '*' };
}
}
LexerState::InNumber => {
if ch.is_numeric() || ch == '.' || ch == 'e' || ch == 'E' || ch == '+' || ch == '-' {
self.token_buffer.push(ch);
} else {
// Number complete
tokens.push((Token::Number, self.make_span_from_start()));
self.state = LexerState::Normal;
self.token_buffer.clear();
// Reprocess this character
pending = Some(ch);
continue;
}
}
LexerState::InIdentifier => {
if ch.is_alphanumeric() || ch == '_' {
self.token_buffer.push(ch);
} else {
// Identifier complete
let token = match self.token_buffer.as_str() {
"null" => Token::Null,
"true" => Token::True,
"false" => Token::False,
_ => {
// Unquoted key or error
if self.config.mode == LexerMode::Forgiving {
Token::UnquotedString
} else {
return Err(Error::UnexpectedChar(
self.token_buffer.chars().next().unwrap_or(' '),
self.token_start
));
}
}
};
tokens.push((token, self.make_span_from_start()));
self.state = LexerState::Normal;
self.token_buffer.clear();
// Reprocess this character
pending = Some(ch);
continue;
}
}
}
self.position += 1;
}
// Check if we need more input
let needs_more = !matches!(self.state, LexerState::Normal);
Ok((tokens, needs_more))
}
/// Flush any remaining tokens (call at end of input)
pub fn flush(&mut self) -> Result<Vec<(Token, Span)>> {
match &self.state {
LexerState::Normal => Ok(vec![]),
LexerState::InString { .. } => Err(Error::UnterminatedString(self.token_start)),
LexerState::InSingleLineComment => {
self.state = LexerState::Normal;
Ok(vec![])
}
LexerState::InMultiLineComment { .. } => {
Err(Error::UnexpectedEof(self.position))
}
LexerState::InNumber => {
let token = vec![(Token::Number, self.make_span_from_start())];
self.state = LexerState::Normal;
self.token_buffer.clear();
Ok(token)
}
LexerState::InIdentifier => {
let token = match self.token_buffer.as_str() {
"null" => Token::Null,
"true" => Token::True,
"false" => Token::False,
_ => {
if self.config.mode == LexerMode::Forgiving {
Token::UnquotedString
} else {
return Err(Error::UnexpectedChar(
self.token_buffer.chars().next().unwrap_or(' '),
self.token_start
));
}
}
};
let tokens = vec![(token, self.make_span_from_start())];
self.state = LexerState::Normal;
self.token_buffer.clear();
Ok(tokens)
}
}
}
/// Get current position
pub fn position(&self) -> usize {
self.position
}
/// Create a span for a single character token
fn make_span(&self, len: usize) -> Span {
Span {
start: self.position,
end: self.position + len,
}
}
/// Create a span from token start to current position
fn make_span_from_start(&self) -> Span {
Span {
start: self.token_start,
end: self.position,
}
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn feed_reprocesses_number_and_identifier_delimiters_in_order() {
let mut lexer = BufferedLexer::new(LexerConfig::default());
let (tokens, needs_more) = lexer.feed("[12,true]").unwrap();
assert_eq!(
tokens,
vec![
(Token::LeftBracket, Span::new(0, 1)),
(Token::Number, Span::new(1, 3)),
(Token::Comma, Span::new(3, 4)),
(Token::True, Span::new(4, 8)),
(Token::RightBracket, Span::new(8, 9)),
]
);
assert!(!needs_more);
assert_eq!(lexer.position(), 9);
}
#[test]
fn feed_preserves_delimiters_across_chunks_and_flushes_eof_identifier() {
let mut lexer = BufferedLexer::new(LexerConfig::default());
let (first_tokens, first_needs_more) = lexer.feed("[12").unwrap();
let (second_tokens, second_needs_more) = lexer.feed(",true").unwrap();
let eof_tokens = lexer.flush().unwrap();
assert_eq!(first_tokens, vec![(Token::LeftBracket, Span::new(0, 1))]);
assert!(first_needs_more);
assert_eq!(
second_tokens,
vec![
(Token::Number, Span::new(1, 3)),
(Token::Comma, Span::new(3, 4)),
]
);
assert!(second_needs_more);
assert_eq!(eof_tokens, vec![(Token::True, Span::new(4, 8))]);
assert_eq!(lexer.position(), 8);
}
}