summaryrefslogtreecommitdiff
path: root/src/pkg/go/scanner/scanner.go
blob: 1e259d5ed2a4f1007cc8cbae778b7d8757e5fa2e (plain)
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
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

// Package scanner implements a scanner for Go source text.
// It takes a []byte as source which can then be tokenized
// through repeated calls to the Scan method.
//
package scanner

import (
	"bytes"
	"fmt"
	"go/token"
	"path/filepath"
	"strconv"
	"unicode"
	"unicode/utf8"
)

// An ErrorHandler may be provided to Scanner.Init. If a syntax error is
// encountered and a handler was installed, the handler is called with a
// position and an error message. The position points to the beginning of
// the offending token.
//
type ErrorHandler func(pos token.Position, msg string)

// A Scanner holds the scanner's internal state while processing
// a given text.  It can be allocated as part of another data
// structure but must be initialized via Init before use.
//
type Scanner struct {
	// immutable state
	file *token.File  // source file handle
	dir  string       // directory portion of file.Name()
	src  []byte       // source
	err  ErrorHandler // error reporting; or nil
	mode Mode         // scanning mode

	// scanning state
	ch         rune // current character
	offset     int  // character offset
	rdOffset   int  // reading offset (position after current character)
	lineOffset int  // current line offset
	insertSemi bool // insert a semicolon before next newline

	// public state - ok to modify
	ErrorCount int // number of errors encountered
}

const bom = 0xFEFF // byte order mark, only permitted as very first character

// Read the next Unicode char into s.ch.
// s.ch < 0 means end-of-file.
//
func (s *Scanner) next() {
	if s.rdOffset < len(s.src) {
		s.offset = s.rdOffset
		if s.ch == '\n' {
			s.lineOffset = s.offset
			s.file.AddLine(s.offset)
		}
		r, w := rune(s.src[s.rdOffset]), 1
		switch {
		case r == 0:
			s.error(s.offset, "illegal character NUL")
		case r >= 0x80:
			// not ASCII
			r, w = utf8.DecodeRune(s.src[s.rdOffset:])
			if r == utf8.RuneError && w == 1 {
				s.error(s.offset, "illegal UTF-8 encoding")
			} else if r == bom && s.offset > 0 {
				s.error(s.offset, "illegal byte order mark")
			}
		}
		s.rdOffset += w
		s.ch = r
	} else {
		s.offset = len(s.src)
		if s.ch == '\n' {
			s.lineOffset = s.offset
			s.file.AddLine(s.offset)
		}
		s.ch = -1 // eof
	}
}

// A mode value is a set of flags (or 0).
// They control scanner behavior.
//
type Mode uint

const (
	ScanComments    Mode = 1 << iota // return comments as COMMENT tokens
	dontInsertSemis                  // do not automatically insert semicolons - for testing only
)

// Init prepares the scanner s to tokenize the text src by setting the
// scanner at the beginning of src. The scanner uses the file set file
// for position information and it adds line information for each line.
// It is ok to re-use the same file when re-scanning the same file as
// line information which is already present is ignored. Init causes a
// panic if the file size does not match the src size.
//
// Calls to Scan will invoke the error handler err if they encounter a
// syntax error and err is not nil. Also, for each error encountered,
// the Scanner field ErrorCount is incremented by one. The mode parameter
// determines how comments are handled.
//
// Note that Init may call err if there is an error in the first character
// of the file.
//
func (s *Scanner) Init(file *token.File, src []byte, err ErrorHandler, mode Mode) {
	// Explicitly initialize all fields since a scanner may be reused.
	if file.Size() != len(src) {
		panic(fmt.Sprintf("file size (%d) does not match src len (%d)", file.Size(), len(src)))
	}
	s.file = file
	s.dir, _ = filepath.Split(file.Name())
	s.src = src
	s.err = err
	s.mode = mode

	s.ch = ' '
	s.offset = 0
	s.rdOffset = 0
	s.lineOffset = 0
	s.insertSemi = false
	s.ErrorCount = 0

	s.next()
	if s.ch == bom {
		s.next() // ignore BOM at file beginning
	}
}

func (s *Scanner) error(offs int, msg string) {
	if s.err != nil {
		s.err(s.file.Position(s.file.Pos(offs)), msg)
	}
	s.ErrorCount++
}

var prefix = []byte("//line ")

func (s *Scanner) interpretLineComment(text []byte) {
	if bytes.HasPrefix(text, prefix) {
		// get filename and line number, if any
		if i := bytes.LastIndex(text, []byte{':'}); i > 0 {
			if line, err := strconv.Atoi(string(text[i+1:])); err == nil && line > 0 {
				// valid //line filename:line comment;
				filename := filepath.Clean(string(text[len(prefix):i]))
				if !filepath.IsAbs(filename) {
					// make filename relative to current directory
					filename = filepath.Join(s.dir, filename)
				}
				// update scanner position
				s.file.AddLineInfo(s.lineOffset+len(text)+1, filename, line) // +len(text)+1 since comment applies to next line
			}
		}
	}
}

func (s *Scanner) scanComment() string {
	// initial '/' already consumed; s.ch == '/' || s.ch == '*'
	offs := s.offset - 1 // position of initial '/'
	hasCR := false

	if s.ch == '/' {
		//-style comment
		s.next()
		for s.ch != '\n' && s.ch >= 0 {
			if s.ch == '\r' {
				hasCR = true
			}
			s.next()
		}
		if offs == s.lineOffset {
			// comment starts at the beginning of the current line
			s.interpretLineComment(s.src[offs:s.offset])
		}
		goto exit
	}

	/*-style comment */
	s.next()
	for s.ch >= 0 {
		ch := s.ch
		if ch == '\r' {
			hasCR = true
		}
		s.next()
		if ch == '*' && s.ch == '/' {
			s.next()
			goto exit
		}
	}

	s.error(offs, "comment not terminated")

exit:
	lit := s.src[offs:s.offset]
	if hasCR {
		lit = stripCR(lit)
	}

	return string(lit)
}

func (s *Scanner) findLineEnd() bool {
	// initial '/' already consumed

	defer func(offs int) {
		// reset scanner state to where it was upon calling findLineEnd
		s.ch = '/'
		s.offset = offs
		s.rdOffset = offs + 1
		s.next() // consume initial '/' again
	}(s.offset - 1)

	// read ahead until a newline, EOF, or non-comment token is found
	for s.ch == '/' || s.ch == '*' {
		if s.ch == '/' {
			//-style comment always contains a newline
			return true
		}
		/*-style comment: look for newline */
		s.next()
		for s.ch >= 0 {
			ch := s.ch
			if ch == '\n' {
				return true
			}
			s.next()
			if ch == '*' && s.ch == '/' {
				s.next()
				break
			}
		}
		s.skipWhitespace() // s.insertSemi is set
		if s.ch < 0 || s.ch == '\n' {
			return true
		}
		if s.ch != '/' {
			// non-comment token
			return false
		}
		s.next() // consume '/'
	}

	return false
}

func isLetter(ch rune) bool {
	return 'a' <= ch && ch <= 'z' || 'A' <= ch && ch <= 'Z' || ch == '_' || ch >= 0x80 && unicode.IsLetter(ch)
}

func isDigit(ch rune) bool {
	return '0' <= ch && ch <= '9' || ch >= 0x80 && unicode.IsDigit(ch)
}

func (s *Scanner) scanIdentifier() string {
	offs := s.offset
	for isLetter(s.ch) || isDigit(s.ch) {
		s.next()
	}
	return string(s.src[offs:s.offset])
}

func digitVal(ch rune) int {
	switch {
	case '0' <= ch && ch <= '9':
		return int(ch - '0')
	case 'a' <= ch && ch <= 'f':
		return int(ch - 'a' + 10)
	case 'A' <= ch && ch <= 'F':
		return int(ch - 'A' + 10)
	}
	return 16 // larger than any legal digit val
}

func (s *Scanner) scanMantissa(base int) {
	for digitVal(s.ch) < base {
		s.next()
	}
}

func (s *Scanner) scanNumber(seenDecimalPoint bool) (token.Token, string) {
	// digitVal(s.ch) < 10
	offs := s.offset
	tok := token.INT

	if seenDecimalPoint {
		offs--
		tok = token.FLOAT
		s.scanMantissa(10)
		goto exponent
	}

	if s.ch == '0' {
		// int or float
		offs := s.offset
		s.next()
		if s.ch == 'x' || s.ch == 'X' {
			// hexadecimal int
			s.next()
			s.scanMantissa(16)
			if s.offset-offs <= 2 {
				// only scanned "0x" or "0X"
				s.error(offs, "illegal hexadecimal number")
			}
		} else {
			// octal int or float
			seenDecimalDigit := false
			s.scanMantissa(8)
			if s.ch == '8' || s.ch == '9' {
				// illegal octal int or float
				seenDecimalDigit = true
				s.scanMantissa(10)
			}
			if s.ch == '.' || s.ch == 'e' || s.ch == 'E' || s.ch == 'i' {
				goto fraction
			}
			// octal int
			if seenDecimalDigit {
				s.error(offs, "illegal octal number")
			}
		}
		goto exit
	}

	// decimal int or float
	s.scanMantissa(10)

fraction:
	if s.ch == '.' {
		tok = token.FLOAT
		s.next()
		s.scanMantissa(10)
	}

exponent:
	if s.ch == 'e' || s.ch == 'E' {
		tok = token.FLOAT
		s.next()
		if s.ch == '-' || s.ch == '+' {
			s.next()
		}
		s.scanMantissa(10)
	}

	if s.ch == 'i' {
		tok = token.IMAG
		s.next()
	}

exit:
	return tok, string(s.src[offs:s.offset])
}

func (s *Scanner) scanEscape(quote rune) {
	offs := s.offset

	var i, base, max uint32
	switch s.ch {
	case 'a', 'b', 'f', 'n', 'r', 't', 'v', '\\', quote:
		s.next()
		return
	case '0', '1', '2', '3', '4', '5', '6', '7':
		i, base, max = 3, 8, 255
	case 'x':
		s.next()
		i, base, max = 2, 16, 255
	case 'u':
		s.next()
		i, base, max = 4, 16, unicode.MaxRune
	case 'U':
		s.next()
		i, base, max = 8, 16, unicode.MaxRune
	default:
		s.next() // always make progress
		s.error(offs, "unknown escape sequence")
		return
	}

	var x uint32
	for ; i > 0 && s.ch != quote && s.ch >= 0; i-- {
		d := uint32(digitVal(s.ch))
		if d >= base {
			s.error(s.offset, "illegal character in escape sequence")
			break
		}
		x = x*base + d
		s.next()
	}
	// in case of an error, consume remaining chars
	for ; i > 0 && s.ch != quote && s.ch >= 0; i-- {
		s.next()
	}
	if x > max || 0xD800 <= x && x < 0xE000 {
		s.error(offs, "escape sequence is invalid Unicode code point")
	}
}

func (s *Scanner) scanChar() string {
	// '\'' opening already consumed
	offs := s.offset - 1

	n := 0
	for s.ch != '\'' {
		ch := s.ch
		n++
		s.next()
		if ch == '\n' || ch < 0 {
			s.error(offs, "character literal not terminated")
			n = 1
			break
		}
		if ch == '\\' {
			s.scanEscape('\'')
		}
	}

	s.next()

	if n != 1 {
		s.error(offs, "illegal character literal")
	}

	return string(s.src[offs:s.offset])
}

func (s *Scanner) scanString() string {
	// '"' opening already consumed
	offs := s.offset - 1

	for s.ch != '"' {
		ch := s.ch
		s.next()
		if ch == '\n' || ch < 0 {
			s.error(offs, "string not terminated")
			break
		}
		if ch == '\\' {
			s.scanEscape('"')
		}
	}

	s.next()

	return string(s.src[offs:s.offset])
}

func stripCR(b []byte) []byte {
	c := make([]byte, len(b))
	i := 0
	for _, ch := range b {
		if ch != '\r' {
			c[i] = ch
			i++
		}
	}
	return c[:i]
}

func (s *Scanner) scanRawString() string {
	// '`' opening already consumed
	offs := s.offset - 1

	hasCR := false
	for s.ch != '`' {
		ch := s.ch
		s.next()
		if ch == '\r' {
			hasCR = true
		}
		if ch < 0 {
			s.error(offs, "string not terminated")
			break
		}
	}

	s.next()

	lit := s.src[offs:s.offset]
	if hasCR {
		lit = stripCR(lit)
	}

	return string(lit)
}

func (s *Scanner) skipWhitespace() {
	for s.ch == ' ' || s.ch == '\t' || s.ch == '\n' && !s.insertSemi || s.ch == '\r' {
		s.next()
	}
}

// Helper functions for scanning multi-byte tokens such as >> += >>= .
// Different routines recognize different length tok_i based on matches
// of ch_i. If a token ends in '=', the result is tok1 or tok3
// respectively. Otherwise, the result is tok0 if there was no other
// matching character, or tok2 if the matching character was ch2.

func (s *Scanner) switch2(tok0, tok1 token.Token) token.Token {
	if s.ch == '=' {
		s.next()
		return tok1
	}
	return tok0
}

func (s *Scanner) switch3(tok0, tok1 token.Token, ch2 rune, tok2 token.Token) token.Token {
	if s.ch == '=' {
		s.next()
		return tok1
	}
	if s.ch == ch2 {
		s.next()
		return tok2
	}
	return tok0
}

func (s *Scanner) switch4(tok0, tok1 token.Token, ch2 rune, tok2, tok3 token.Token) token.Token {
	if s.ch == '=' {
		s.next()
		return tok1
	}
	if s.ch == ch2 {
		s.next()
		if s.ch == '=' {
			s.next()
			return tok3
		}
		return tok2
	}
	return tok0
}

// Scan scans the next token and returns the token position, the token,
// and its literal string if applicable. The source end is indicated by
// token.EOF.
//
// If the returned token is a literal (token.IDENT, token.INT, token.FLOAT,
// token.IMAG, token.CHAR, token.STRING) or token.COMMENT, the literal string
// has the corresponding value.
//
// If the returned token is a keyword, the literal string is the keyword.
//
// If the returned token is token.SEMICOLON, the corresponding
// literal string is ";" if the semicolon was present in the source,
// and "\n" if the semicolon was inserted because of a newline or
// at EOF.
//
// If the returned token is token.ILLEGAL, the literal string is the
// offending character.
//
// In all other cases, Scan returns an empty literal string.
//
// For more tolerant parsing, Scan will return a valid token if
// possible even if a syntax error was encountered. Thus, even
// if the resulting token sequence contains no illegal tokens,
// a client may not assume that no error occurred. Instead it
// must check the scanner's ErrorCount or the number of calls
// of the error handler, if there was one installed.
//
// Scan adds line information to the file added to the file
// set with Init. Token positions are relative to that file
// and thus relative to the file set.
//
func (s *Scanner) Scan() (pos token.Pos, tok token.Token, lit string) {
scanAgain:
	s.skipWhitespace()

	// current token start
	pos = s.file.Pos(s.offset)

	// determine token value
	insertSemi := false
	switch ch := s.ch; {
	case isLetter(ch):
		lit = s.scanIdentifier()
		if len(lit) > 1 {
			// keywords are longer than one letter - avoid lookup otherwise
			tok = token.Lookup(lit)
			switch tok {
			case token.IDENT, token.BREAK, token.CONTINUE, token.FALLTHROUGH, token.RETURN:
				insertSemi = true
			}
		} else {
			insertSemi = true
			tok = token.IDENT
		}
	case '0' <= ch && ch <= '9':
		insertSemi = true
		tok, lit = s.scanNumber(false)
	default:
		s.next() // always make progress
		switch ch {
		case -1:
			if s.insertSemi {
				s.insertSemi = false // EOF consumed
				return pos, token.SEMICOLON, "\n"
			}
			tok = token.EOF
		case '\n':
			// we only reach here if s.insertSemi was
			// set in the first place and exited early
			// from s.skipWhitespace()
			s.insertSemi = false // newline consumed
			return pos, token.SEMICOLON, "\n"
		case '"':
			insertSemi = true
			tok = token.STRING
			lit = s.scanString()
		case '\'':
			insertSemi = true
			tok = token.CHAR
			lit = s.scanChar()
		case '`':
			insertSemi = true
			tok = token.STRING
			lit = s.scanRawString()
		case ':':
			tok = s.switch2(token.COLON, token.DEFINE)
		case '.':
			if '0' <= s.ch && s.ch <= '9' {
				insertSemi = true
				tok, lit = s.scanNumber(true)
			} else if s.ch == '.' {
				s.next()
				if s.ch == '.' {
					s.next()
					tok = token.ELLIPSIS
				}
			} else {
				tok = token.PERIOD
			}
		case ',':
			tok = token.COMMA
		case ';':
			tok = token.SEMICOLON
			lit = ";"
		case '(':
			tok = token.LPAREN
		case ')':
			insertSemi = true
			tok = token.RPAREN
		case '[':
			tok = token.LBRACK
		case ']':
			insertSemi = true
			tok = token.RBRACK
		case '{':
			tok = token.LBRACE
		case '}':
			insertSemi = true
			tok = token.RBRACE
		case '+':
			tok = s.switch3(token.ADD, token.ADD_ASSIGN, '+', token.INC)
			if tok == token.INC {
				insertSemi = true
			}
		case '-':
			tok = s.switch3(token.SUB, token.SUB_ASSIGN, '-', token.DEC)
			if tok == token.DEC {
				insertSemi = true
			}
		case '*':
			tok = s.switch2(token.MUL, token.MUL_ASSIGN)
		case '/':
			if s.ch == '/' || s.ch == '*' {
				// comment
				if s.insertSemi && s.findLineEnd() {
					// reset position to the beginning of the comment
					s.ch = '/'
					s.offset = s.file.Offset(pos)
					s.rdOffset = s.offset + 1
					s.insertSemi = false // newline consumed
					return pos, token.SEMICOLON, "\n"
				}
				lit = s.scanComment()
				if s.mode&ScanComments == 0 {
					// skip comment
					s.insertSemi = false // newline consumed
					goto scanAgain
				}
				tok = token.COMMENT
			} else {
				tok = s.switch2(token.QUO, token.QUO_ASSIGN)
			}
		case '%':
			tok = s.switch2(token.REM, token.REM_ASSIGN)
		case '^':
			tok = s.switch2(token.XOR, token.XOR_ASSIGN)
		case '<':
			if s.ch == '-' {
				s.next()
				tok = token.ARROW
			} else {
				tok = s.switch4(token.LSS, token.LEQ, '<', token.SHL, token.SHL_ASSIGN)
			}
		case '>':
			tok = s.switch4(token.GTR, token.GEQ, '>', token.SHR, token.SHR_ASSIGN)
		case '=':
			tok = s.switch2(token.ASSIGN, token.EQL)
		case '!':
			tok = s.switch2(token.NOT, token.NEQ)
		case '&':
			if s.ch == '^' {
				s.next()
				tok = s.switch2(token.AND_NOT, token.AND_NOT_ASSIGN)
			} else {
				tok = s.switch3(token.AND, token.AND_ASSIGN, '&', token.LAND)
			}
		case '|':
			tok = s.switch3(token.OR, token.OR_ASSIGN, '|', token.LOR)
		default:
			// next reports unexpected BOMs - don't repeat
			if ch != bom {
				s.error(s.file.Offset(pos), fmt.Sprintf("illegal character %#U", ch))
			}
			insertSemi = s.insertSemi // preserve insertSemi info
			tok = token.ILLEGAL
			lit = string(ch)
		}
	}
	if s.mode&dontInsertSemis == 0 {
		s.insertSemi = insertSemi
	}

	return
}