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
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
//! defines and implements `Lexer` (Tokenizer).
use std::cmp::Ordering;

use erg_common::cache::CacheSet;
use erg_common::config::ErgConfig;
use erg_common::config::Input;
use erg_common::traits::{Locational, Runnable, Stream};
use erg_common::{debug_power_assert, fn_name_full, normalize_newline, switch_lang};

use crate::error::{LexError, LexErrors, LexResult, LexerRunnerError, LexerRunnerErrors};
use crate::token::{Token, TokenCategory, TokenKind, TokenStream};
use TokenKind::*;

/// Lexerは使い捨てなので、Runnerを用意
pub struct LexerRunner {
    cfg: ErgConfig,
}

impl Runnable for LexerRunner {
    type Err = LexerRunnerError;
    type Errs = LexerRunnerErrors;
    const NAME: &'static str = "Erg lexer";

    #[inline]
    fn new(cfg: ErgConfig) -> Self {
        Self { cfg }
    }

    #[inline]
    fn input(&self) -> &Input {
        &self.cfg.input
    }

    #[inline]
    fn finish(&mut self) {}

    #[inline]
    fn clear(&mut self) {}

    fn exec(&mut self) -> Result<(), Self::Errs> {
        let lexer = Lexer::from_str(self.input().read());
        let ts = lexer
            .lex()
            .map_err(|errs| LexerRunnerErrors::convert(self.input(), errs))?;
        println!("{ts}");
        Ok(())
    }

    fn eval(&mut self, src: String) -> Result<String, LexerRunnerErrors> {
        let lexer = Lexer::from_str(src);
        if cfg!(feature = "debug") {
            let ts = lexer
                .lex()
                .map_err(|errs| LexerRunnerErrors::convert(self.input(), errs))?;
            println!("{ts}");
            Ok(ts.to_string())
        } else {
            Ok(lexer
                .lex()
                .map_err(|errs| LexerRunnerErrors::convert(self.input(), errs))?
                .to_string())
        }
    }
}

/// Lexes a source code and iterates tokens.
///
/// This can be used as an iterator or to generate a `TokenStream`.
#[derive(Debug)]
pub struct Lexer /*<'a>*/ {
    str_cache: CacheSet<str>,
    chars: Vec<char>,
    indent_stack: Vec<usize>,
    /// indicates the position in the entire source code
    cursor: usize,
    /// to determine the type of operators, etc.
    prev_token: Token,
    /// 0-origin, but Token.lineno will 1-origin
    lineno_token_starts: usize,
    /// 0-origin, indicates the column number in which the token appears
    col_token_starts: usize,
}

impl Lexer /*<'a>*/ {
    pub fn new(input: Input) -> Self {
        let normed = normalize_newline(&input.read());
        Lexer {
            str_cache: CacheSet::new(),
            chars: normed.chars().collect::<Vec<char>>(),
            indent_stack: vec![],
            cursor: 0,
            prev_token: Token::new(TokenKind::BOF, "", 0, 0),
            lineno_token_starts: 0,
            col_token_starts: 0,
        }
    }

    #[allow(clippy::should_implement_trait)]
    pub fn from_str(src: String) -> Self {
        let escaped = normalize_newline(&src);
        Lexer {
            str_cache: CacheSet::new(),
            chars: escaped.chars().collect::<Vec<char>>(),
            indent_stack: vec![],
            cursor: 0,
            prev_token: Token::new(TokenKind::BOF, "", 0, 0),
            lineno_token_starts: 0,
            col_token_starts: 0,
        }
    }

    pub fn lex(self) -> Result<TokenStream, LexErrors> {
        let mut result = TokenStream::empty();
        let mut errs = LexErrors::empty();
        for i in self {
            match i {
                Ok(token) => result.push(token),
                Err(err) => {
                    errs.push(err);
                }
            }
        }
        if errs.is_empty() {
            Ok(result)
        } else {
            Err(errs)
        }
    }

    fn emit_token(&mut self, kind: TokenKind, cont: &str) -> Token {
        let cont = self.str_cache.get(cont);
        // cannot use String::len() for multi-byte characters
        let cont_len = cont.chars().count();
        let token = Token::new(
            kind,
            cont,
            self.lineno_token_starts + 1,
            self.col_token_starts,
        );
        self.prev_token = token.clone();
        self.col_token_starts += cont_len;
        token
    }

    #[inline]
    fn accept(&mut self, kind: TokenKind, cont: &str) -> Option<LexResult<Token>> {
        Some(Ok(self.emit_token(kind, cont)))
    }

    fn deny_feature(&mut self, cont: &str, feat_name: &str) -> Option<LexResult<Token>> {
        let token = self.emit_token(Illegal, cont);
        Some(Err(LexError::feature_error(0, token.loc(), feat_name)))
    }

    const fn is_valid_symbol_ch(c: char) -> bool {
        match c {
            '0'..='9' => true,
            // control characters
            '\0' | '\u{0009}'..='\u{001F}' => false,
            // white spaces
            ' ' | '\u{00A0}' => false,
            '\u{007F}' | '\u{0085}' | '\u{05C1}' | '\u{05C2}' => false,
            '\u{0701}'..='\u{070d}' => false,
            '\u{07B2}'..='\u{07BF}' => false,
            '\u{1680}' | '\u{180E}' => false,
            '\u{2000}'..='\u{200F}' => false,
            '\u{2028}'..='\u{202F}' => false,
            '\u{205F}'..='\u{206F}' => false,
            '\u{3000}' | '\u{3164}' | '\u{FEFF}' => false,
            // operator characters + special markers
            '<' | '>' | '$' | '%' | '.' | ',' | ':' | ';' | '+' | '-' | '*' | '/' | '=' | '#'
            | '&' | '|' | '^' | '~' | '@' | '!' | '?' | '\\' => false,
            // enclosures
            '[' | ']' | '(' | ')' | '{' | '}' | '\"' | '\'' | '`' => false,
            _ => true,
        }
    }

    /// Detect `c` is a bidirectional overriding character.
    /// [CVE-2021-42574: homoglyph atack](https://blog.rust-lang.org/2021/11/01/cve-2021-42574.html) countermeasures.
    pub fn is_bidi(c: char) -> bool {
        matches!(c, '\u{200F}' | '\u{202B}' | '\u{202E}' | '\u{2067}')
    }

    #[inline]
    fn is_definable_operator(s: &str) -> bool {
        matches!(
            s,
            "+" | "-"
                | "*"
                | "/"
                | "//"
                | "**"
                | "%"
                | ".."
                | "..="
                | "~"
                | "&&"
                | "||"
                | "^^"
                | ">>"
                | "<<"
                | "=="
                | "!="
                | ">"
                | "<"
                | ">="
                | "<="
                | "dot"
                | "cross"
        )
    }

    // +, -, * etc. may be pre/bin
    // and, or, is, isnot, in, notin, as, dot, cross may be bin/function
    const fn is_bin_position(&self) -> Option<bool> {
        match self.prev_token.category() {
            // unary: `[ +`, `= +`, `+ +`, `, +`, `:: +`
            TokenCategory::LEnclosure
            | TokenCategory::BinOp
            | TokenCategory::UnaryOp
            | TokenCategory::Separator
            | TokenCategory::SpecialBinOp
            | TokenCategory::DefOp
            | TokenCategory::LambdaOp
            | TokenCategory::BOF => Some(false),
            // bin: `] +`, `1 +`, `true and[true]`
            TokenCategory::REnclosure | TokenCategory::Literal => Some(true),
            // bin: `fn +1`
            // NOTE: if semantic analysis shows `fn` is a function, should this be rewritten to be unary?
            TokenCategory::Symbol => Some(true),
            _ => None,
        }
    }

    fn is_zero(s: &str) -> bool {
        s.replace("-0", "").replace('0', "").is_empty()
    }

    /// emit_tokenで一気にcol_token_startsを移動させるのでここでは移動させない
    fn consume(&mut self) -> Option<char> {
        let now = self.cursor;
        self.cursor += 1;
        self.chars.get(now).copied()
    }

    fn peek_prev_ch(&self) -> Option<char> {
        if self.cursor == 0 {
            None
        } else {
            self.chars.get(self.cursor - 1).copied()
        }
    }

    #[inline]
    fn peek_cur_ch(&self) -> Option<char> {
        self.chars.get(self.cursor).copied()
    }

    #[inline]
    fn peek_next_ch(&self) -> Option<char> {
        self.chars.get(self.cursor + 1).copied()
    }

    fn lex_comment(&mut self) -> LexResult<()> {
        // debug_power_assert!(self.consume(), ==, Some('#'));
        let mut s = "".to_string();
        while self.peek_cur_ch().map(|cur| cur != '\n').unwrap_or(false) {
            if Self::is_bidi(self.peek_cur_ch().unwrap()) {
                let comment = self.emit_token(Illegal, &s);
                return Err(LexError::syntax_error(
                    0,
                    comment.loc(),
                    switch_lang!(
                        "japanese" => "不正なユニコード文字(双方向オーバーライド)がコメント中に使用されています",
                        "simplified_chinese" => "注释中使用了非法的unicode字符(双向覆盖)",
                        "traditional_chinese" => "註釋中使用了非法的unicode字符(雙向覆蓋)",
                        "english" => "invalid unicode character (bi-directional override) in comments",
                    ),
                    None,
                ));
            }
            s.push(self.consume().unwrap());
        }
        Ok(())
    }

    fn lex_space_indent_dedent(&mut self) -> Option<LexResult<Token>> {
        let is_toplevel = self.cursor > 0
            && !self.indent_stack.is_empty()
            && self.peek_prev_ch() == Some('\n')
            && self.peek_cur_ch() != Some(' ');
        if is_toplevel {
            let dedent = self.emit_token(Dedent, "");
            self.indent_stack.pop();
            self.col_token_starts = 0;
            return Some(Ok(dedent));
        }
        let mut spaces = "".to_string();
        while let Some(' ') = self.peek_cur_ch() {
            spaces.push(self.consume().unwrap());
        }
        // indent in the first line: error
        if !spaces.is_empty() && self.cursor == 0 {
            let space = self.emit_token(Illegal, &spaces);
            Some(Err(LexError::syntax_error(
                0,
                space.loc(),
                switch_lang!(
                    "japanese" => "インデントが不正です",
                    "simplified_chinese" => "无效缩进",
                    "traditional_chinese" => "無效縮進",
                    "english" => "invalid indent",
                ),
                None,
            )))
        } else if self.prev_token.is(Newline) {
            self.lex_indent_dedent(spaces)
        } else {
            self.col_token_starts += spaces.len();
            None
        }
    }

    /// The semantic correctness of the use of indent/dedent will be analyzed with `Parser`
    fn lex_indent_dedent(&mut self, spaces: String) -> Option<LexResult<Token>> {
        // same as the CPython's limit
        if spaces.len() > 100 {
            let token = self.emit_token(Indent, &spaces);
            return Some(Err(LexError::syntax_error(
                0,
                token.loc(),
                switch_lang!(
                    "japanese" => "インデントが深すぎます",
                    "simplified_chinese" => "缩进太深",
                    "traditional_chinese" => "縮進太深",
                    "english" => "indentation is too deep",
                ),
                Some(
                    switch_lang!(
                        "japanese" => "コードが複雑すぎます。処理を分割してください",
                        "simplified_chinese" => "代码过于复杂,请拆分过程",
                        "traditional_chinese" => "代碼過於復雜,請拆分過程",
                        "english" => "The code is too complicated. Please split the process",
                    )
                    .into(),
                ),
            )));
        }
        // ignore indents if the current line is a comment
        if let Some('#') = self.peek_cur_ch() {
            if let Err(e) = self.lex_comment() {
                return Some(Err(e));
            }
        }
        let mut is_valid_dedent = false;
        let calc_indent_and_validate = |sum: usize, x: &usize| {
            if sum + *x == spaces.len() {
                is_valid_dedent = true;
            }
            sum + *x
        };
        let sum_indent = self.indent_stack.iter().fold(0, calc_indent_and_validate);
        match sum_indent.cmp(&spaces.len()) {
            Ordering::Less => {
                let indent_len = spaces.len() - sum_indent;
                self.col_token_starts += sum_indent;
                let indent = self.emit_token(Indent, &" ".repeat(indent_len));
                self.indent_stack.push(indent_len);
                Some(Ok(indent))
            }
            Ordering::Greater => {
                if is_valid_dedent {
                    let dedent = self.emit_token(Dedent, "");
                    self.indent_stack.pop();
                    Some(Ok(dedent))
                } else {
                    let invalid_dedent = self.emit_token(Dedent, "");
                    Some(Err(LexError::syntax_error(
                        0,
                        invalid_dedent.loc(),
                        switch_lang!(
                            "japanese" => "インデントが不正です",
                            "simplified_chinese" => "无效缩进",
                            "traditional_chinese" => "無效縮進",
                            "english" => "invalid indent",
                        ),
                        None,
                    )))
                }
            }
            Ordering::Equal /* if indent_sum == space.len() */ => {
                self.col_token_starts += spaces.len();
                None
            }
        }
    }

    fn lex_exponent(&mut self, mantissa: String) -> LexResult<Token> {
        let mut num = mantissa;
        debug_power_assert!(self.peek_cur_ch(), ==, Some('e'));
        num.push(self.consume().unwrap()); // e
        num.push(self.consume().unwrap()); // + | -
        while let Some(cur) = self.peek_cur_ch() {
            if cur.is_ascii_digit() || cur == '_' {
                num.push(self.consume().unwrap());
            } else {
                break;
            }
        }
        Ok(self.emit_token(RatioLit, &num))
    }

    /// `_` will be removed at compiletime
    fn lex_num(&mut self, first_ch: char) -> LexResult<Token> {
        let mut num = first_ch.to_string();
        while let Some(ch) = self.peek_cur_ch() {
            match ch {
                // `.` may be a dot operator, don't consume
                '.' => {
                    return self.lex_num_dot(num);
                }
                n if n.is_ascii_digit() || n == '_' => {
                    num.push(self.consume().unwrap());
                }
                c if Self::is_valid_symbol_ch(c) => {
                    // exponent (e.g. 10e+3)
                    if c == 'e'
                        && (self.peek_next_ch() == Some('+') || self.peek_next_ch() == Some('-'))
                    {
                        return self.lex_exponent(num);
                    } else {
                        // IntLit * Symbol(e.g. 3x + 1)
                        let token = self.emit_token(Illegal, &(num + &c.to_string()));
                        return Err(LexError::feature_error(
                            line!() as usize,
                            token.loc(),
                            "*-less multiply",
                        ));
                    }
                }
                _ => {
                    break;
                }
            }
        }
        let kind = if num.starts_with('-') && !Self::is_zero(&num) {
            IntLit
        } else {
            NatLit
        };
        Ok(self.emit_token(kind, &num))
    }

    /// number '.' ~~
    /// Possibility: RatioLit or Int/NatLit call
    fn lex_num_dot(&mut self, mut num: String) -> LexResult<Token> {
        match self.peek_next_ch() {
            // RatioLit
            Some(n) if n.is_ascii_digit() => {
                num.push(self.consume().unwrap());
                self.lex_ratio(num)
            }
            // method call of IntLit
            // or range operator (e.g. 1..)
            Some(c) if Self::is_valid_symbol_ch(c) || c == '.' => {
                let kind = if num.starts_with('-') && !Self::is_zero(&num) {
                    IntLit
                } else {
                    NatLit
                };
                Ok(self.emit_token(kind, &num))
            }
            Some('_') => {
                self.consume();
                let token = self.emit_token(Illegal, &(num + "_"));
                Err(LexError::simple_syntax_error(0, token.loc()))
            }
            // RatioLit without zero (e.g. 3.)
            _ => {
                num.push(self.consume().unwrap());
                self.lex_ratio(num)
            }
        }
    }

    /// int_part_and_point must be like `12.`
    fn lex_ratio(&mut self, intpart_and_point: String) -> LexResult<Token> {
        let mut num = intpart_and_point;
        while let Some(cur) = self.peek_cur_ch() {
            if cur.is_ascii_digit() || cur == '_' {
                num.push(self.consume().unwrap());
            } else if cur == 'e' {
                return self.lex_exponent(num);
            } else {
                break;
            }
        }
        Ok(self.emit_token(RatioLit, &num))
    }

    fn lex_symbol(&mut self, first_ch: char) -> LexResult<Token> {
        let mut cont = first_ch.to_string();
        while let Some(c) = self.peek_cur_ch() {
            if Self::is_valid_symbol_ch(c) {
                cont.push(self.consume().unwrap());
            } else {
                break;
            }
        }
        if let Some('!') = self.peek_cur_ch() {
            cont.push(self.consume().unwrap());
        }
        if cont.is_empty() {
            let token = self.emit_token(Illegal, &self.peek_cur_ch().unwrap().to_string());
            return Err(LexError::compiler_bug(
                0,
                token.loc(),
                fn_name_full!(),
                line!(),
            ));
        }
        // dot: scalar product, cross: vector product
        // An alphabetical operator can also declare as a function, so checking is necessary
        // e.g. and(true, true, true) = true
        let kind = match &cont[..] {
            "and" => AndOp,
            "or" => OrOp,
            "in" => InOp,
            "notin" => NotInOp,
            "is" => IsOp,
            "isnot" => IsNotOp,
            "dot" => DotOp,
            "cross" => CrossOp,
            "ref" => RefOp,
            "ref!" => RefMutOp,
            // これらはリテラルというより定数だが便宜的にリテラルということにしておく
            "True" | "False" => BoolLit,
            "None" => NoneLit,
            "NotImplemented" => NoImplLit,
            "Ellipsis" => EllipsisLit,
            "Inf" => InfLit,
            "_" => UBar,
            _ => Symbol,
        };
        Ok(self.emit_token(kind, &cont))
    }

    fn lex_str(&mut self) -> LexResult<Token> {
        let mut s = "\"".to_string();
        while let Some(c) = self.peek_cur_ch() {
            if c == '\"' && !s.ends_with('\\') {
                s.push(self.consume().unwrap());
                let token = self.emit_token(StrLit, &s);
                return Ok(token);
            } else {
                let c = self.consume().unwrap();
                s.push(c);
                if Self::is_bidi(c) {
                    let token = self.emit_token(Illegal, &s);
                    return Err(LexError::syntax_error(
                        0,
                        token.loc(),
                        switch_lang!(
                            "japanese" => "不正なユニコード文字(双方向オーバーライド)が文字列中に使用されています",
                            "simplified_chinese" => "注释中使用了非法的unicode字符(双向覆盖)",
                            "traditional_chinese" => "註釋中使用了非法的unicode字符(雙向覆蓋)",
                            "english" => "invalid unicode character (bi-directional override) in string literal",
                        ),
                        None,
                    ));
                }
            }
        }
        let token = self.emit_token(Illegal, &s);
        Err(LexError::syntax_error(
            0,
            token.loc(),
            switch_lang!(
                "japanese" => "文字列が\"によって閉じられていません",
                "simplified_chinese" => "字符串没有被\"关闭",
                "traditional_chinese" => "字符串没有被\"关闭",
                "english" => "the string is not closed by \"",
            ),
            None,
        ))
    }
}

impl Iterator for Lexer /*<'a>*/ {
    type Item = LexResult<Token>;

    fn next(&mut self) -> Option<Self::Item> {
        if self.prev_token.is(TokenKind::EOF) {
            return None;
        }
        let indent_dedent = self.lex_space_indent_dedent();
        if indent_dedent.is_some() {
            return indent_dedent;
        }
        if let Some('#') = self.peek_cur_ch() {
            if let Err(e) = self.lex_comment() {
                return Some(Err(e));
            }
        }
        match self.consume() {
            Some('(') => self.accept(LParen, "("),
            Some(')') => self.accept(RParen, ")"),
            Some('[') => self.accept(LSqBr, "["),
            Some(']') => self.accept(RSqBr, "]"),
            Some('{') => self.accept(LBrace, "{"),
            Some('}') => self.accept(RBrace, "}"),
            Some('<') => match self.peek_cur_ch() {
                Some('.') => {
                    self.consume();
                    if let Some('.') = self.peek_cur_ch() {
                        self.consume();
                        match self.peek_cur_ch() {
                            Some('<') => {
                                self.consume();
                                self.accept(Open, "<..<")
                            }
                            _ => self.accept(LeftOpen, "<.."),
                        }
                    } else {
                        let token = self.emit_token(Illegal, "<.");
                        Some(Err(LexError::syntax_error(
                            0,
                            token.loc(),
                            switch_lang!(
                                "japanese" => "<.という演算子はありません",
                                "simplified_chinese" => "没有这样的运算符:<.",
                                "traditional_chinese" => "沒有這樣的運算符:<.",
                                "english" => "no such operator: <.",
                            ),
                            None,
                        )))
                    }
                }
                Some('=') => {
                    self.consume();
                    self.accept(LessEq, "<=")
                }
                Some('<') => {
                    self.consume();
                    self.accept(Shl, "<<")
                }
                _ => self.accept(Less, "<"),
            },
            Some('>') => match self.peek_cur_ch() {
                Some('=') => {
                    self.consume();
                    self.accept(GreEq, ">=")
                }
                Some('>') => {
                    self.consume();
                    self.accept(Shr, ">>")
                }
                _ => self.accept(Gre, ">"),
            },
            Some('.') => match self.peek_cur_ch() {
                Some('.') => {
                    self.consume();
                    match self.peek_cur_ch() {
                        Some('<') => {
                            self.consume();
                            self.accept(RightOpen, "..<")
                        }
                        Some('.') => {
                            self.consume();
                            self.accept(EllipsisLit, "...")
                        }
                        _ => self.accept(Closed, ".."),
                    }
                }
                // prev_token is Symbol => TupleAttribute
                // else: RatioLit (e.g. .0)
                Some(c) if c.is_ascii_digit() && !self.prev_token.is(Symbol) => {
                    Some(self.lex_ratio(".".into()))
                }
                _ => self.accept(Dot, "."),
            },
            Some(',') => self.accept(Comma, ","),
            Some(':') => match self.peek_cur_ch() {
                Some(':') => {
                    self.consume();
                    self.accept(DblColon, "::")
                }
                Some('=') => {
                    self.consume();
                    self.accept(Walrus, ":=")
                }
                Some('>') => {
                    self.consume();
                    self.accept(SupertypeOf, ":>")
                }
                _ => self.accept(Colon, ":"),
            },
            Some(';') => self.accept(Semi, ";"),
            Some('&') => {
                if let Some('&') = self.peek_cur_ch() {
                    self.consume();
                    self.accept(BitAnd, "&&")
                } else {
                    // let kind = if self.is_bin_position().unwrap() { Amper } else { PreAmp };
                    self.accept(Amper, "&")
                }
            }
            Some('|') => match self.peek_cur_ch() {
                Some('|') => {
                    self.consume();
                    self.accept(BitOr, "||")
                }
                _ => self.accept(VBar, "|"),
            },
            Some('^') => {
                if let Some('^') = self.peek_cur_ch() {
                    self.consume();
                    self.accept(BitXor, "^^")
                } else {
                    self.accept(Caret, "^")
                }
            }
            Some('~') => self.accept(PreBitNot, "~"),
            // TODO:
            Some('$') => self.deny_feature("$", "shared variables"),
            Some('@') => self.accept(AtSign, "@"),
            Some('=') => match self.peek_cur_ch() {
                Some('=') => {
                    self.consume();
                    self.accept(DblEq, "==")
                }
                Some('>') => {
                    self.consume();
                    self.accept(ProcArrow, "=>")
                }
                _ => self.accept(Equal, "="),
            },
            Some('!') => {
                if let Some('=') = self.peek_cur_ch() {
                    self.consume();
                    self.accept(NotEq, "!=")
                } else {
                    self.accept(Mutate, "!")
                }
            }
            Some('?') => self.accept(Try, "?"),
            Some('+') => {
                let kind = match self.is_bin_position() {
                    Some(true) => Plus,
                    Some(false) => PrePlus,
                    None => {
                        let token = self.emit_token(Illegal, "+");
                        return Some(Err(LexError::simple_syntax_error(0, token.loc())));
                    }
                };
                self.accept(kind, "+")
            }
            Some('-') => match self.peek_cur_ch() {
                Some('>') => {
                    self.consume();
                    self.accept(FuncArrow, "->")
                }
                _ => {
                    match self.is_bin_position() {
                        Some(true) => self.accept(Minus, "-"),
                        Some(false) => {
                            // IntLit (negative number)
                            if self
                                .peek_cur_ch()
                                .map(|t| t.is_ascii_digit())
                                .unwrap_or(false)
                            {
                                Some(self.lex_num('-'))
                            } else {
                                self.accept(PreMinus, "-")
                            }
                        }
                        None => {
                            let token = self.emit_token(Illegal, "-");
                            Some(Err(LexError::simple_syntax_error(0, token.loc())))
                        }
                    }
                }
            },
            Some('*') => match self.peek_cur_ch() {
                Some('*') => {
                    self.consume();
                    self.accept(Pow, "**")
                }
                _ => {
                    let kind = match self.is_bin_position() {
                        Some(true) => Star,
                        _ => {
                            let token = self.emit_token(Illegal, "*");
                            return Some(Err(LexError::simple_syntax_error(0, token.loc())));
                        }
                    };
                    self.accept(kind, "*")
                }
            },
            Some('/') => match self.peek_cur_ch() {
                Some('/') => {
                    self.consume();
                    self.accept(FloorDiv, "//")
                }
                _ => self.accept(Slash, "/"),
            },
            Some('%') => self.accept(Mod, "%"),
            // Newline
            // 改行記号はLexer新規生成時に全て\nにreplaceしてある
            Some('\n') => {
                let token = self.emit_token(Newline, "\n");
                self.lineno_token_starts += 1;
                self.col_token_starts = 0;
                Some(Ok(token))
            }
            Some('\t') => {
                let token = self.emit_token(Illegal, "\t");
                Some(Err(LexError::syntax_error(
                    0,
                    token.loc(),
                    switch_lang!(
                        "japanese" => "タブ文字は使用できません",
                        "simplified_chinese" => "不能将制表符用作空格",
                        "traditional_chinese" => "不能將製表符用作空格",
                        "english" => "cannot use a tab as a space",
                    ),
                    Some(
                        switch_lang!(
                            "japanese" => "スペース( )を使用してください",
                            "simplified_chinese" => "使用空格( )",
                            "traditional_chinese" => "使用空格( )",
                            "english" => "use spaces ( )",
                        )
                        .into(),
                    ),
                )))
            }
            // TODO:
            Some('\\') => self.deny_feature("\\", "ignoring line break"),
            // StrLit
            Some('\"') => Some(self.lex_str()),
            // TODO:
            Some('\'') => self.deny_feature("'", "raw identifier"),
            // Symbolized operators (シンボル化された演算子)
            // e.g. `-`(l, r) = l + (-r)
            Some('`') => {
                let mut op = "".to_string();
                while let Some(c) = self.consume() {
                    if c == '`' {
                        if Self::is_definable_operator(&op[..]) {
                            return self.accept(Symbol, &op);
                        } else {
                            let token = self.emit_token(Illegal, &op);
                            return Some(Err(LexError::syntax_error(
                                0,
                                token.loc(),
                                switch_lang!(
                                    "japanese" => format!("`{}`はユーザー定義できません", &token.content),
                                    "simplified_chinese" => format!("`{}`不能由用户定义", &token.content),
                                    "traditional_chinese" => format!("`{}`不能由用戶定義", &token.content),
                                    "english" => format!("`{}` cannot be defined by user", &token.content),
                                ),
                                None,
                            )));
                        }
                    }
                    op.push(c);
                }
                let token = self.emit_token(Illegal, &op);
                Some(Err(LexError::syntax_error(
                    0,
                    token.loc(),
                    switch_lang!(
                        "japanese" => format!("バッククォート(`)が閉じられていません"),
                        "simplified_chinese" => format!("反引号(`)未关闭"),
                        "traditional_chinese" => format!("反引號(`)未關閉"),
                        "english" => format!("back quotes (`) not closed"),
                    ),
                    None,
                )))
            }
            // IntLit or RatioLit
            Some(n) if n.is_ascii_digit() => Some(self.lex_num(n)),
            // Symbol (includes '_')
            Some(c) if Self::is_valid_symbol_ch(c) => Some(self.lex_symbol(c)),
            // Invalid character (e.g. space-like character)
            Some(invalid) => {
                let token = self.emit_token(Illegal, &invalid.to_string());
                Some(Err(LexError::syntax_error(
                    0,
                    token.loc(),
                    switch_lang!(
                        "japanese" => format!("この文字は使用できません: '{invalid}'"),
                        "simplified_chinese" => format!("无效字符:'{invalid}'"),
                        "traditional_chinese" => format!("無效字符:'{invalid}'"),
                        "english" => format!("invalid character: '{invalid}'"),
                    ),
                    None,
                )))
            }
            None => {
                if self.indent_stack.is_empty() {
                    self.accept(EOF, "")
                } else {
                    self.indent_stack.pop();
                    self.accept(Dedent, "")
                }
            }
        }
    }
}