aboutsummaryrefslogtreecommitdiff
path: root/vendor/google.golang.org/protobuf/internal/encoding/text/decode.go
blob: eb10ea10261ae54f2b2b47a2e64f5d1dae2946cc (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
// Copyright 2018 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 text

import (
	"bytes"
	"fmt"
	"io"
	"regexp"
	"strconv"
	"unicode/utf8"

	"google.golang.org/protobuf/internal/errors"
)

// Decoder is a token-based textproto decoder.
type Decoder struct {
	// lastCall is last method called, either readCall or peekCall.
	// Initial value is readCall.
	lastCall call

	// lastToken contains the last read token.
	lastToken Token

	// lastErr contains the last read error.
	lastErr error

	// openStack is a stack containing the byte characters for MessageOpen and
	// ListOpen kinds. The top of stack represents the message or the list that
	// the current token is nested in. An empty stack means the current token is
	// at the top level message. The characters '{' and '<' both represent the
	// MessageOpen kind.
	openStack []byte

	// orig is used in reporting line and column.
	orig []byte
	// in contains the unconsumed input.
	in []byte
}

// NewDecoder returns a Decoder to read the given []byte.
func NewDecoder(b []byte) *Decoder {
	return &Decoder{orig: b, in: b}
}

// ErrUnexpectedEOF means that EOF was encountered in the middle of the input.
var ErrUnexpectedEOF = errors.New("%v", io.ErrUnexpectedEOF)

// call specifies which Decoder method was invoked.
type call uint8

const (
	readCall call = iota
	peekCall
)

// Peek looks ahead and returns the next token and error without advancing a read.
func (d *Decoder) Peek() (Token, error) {
	defer func() { d.lastCall = peekCall }()
	if d.lastCall == readCall {
		d.lastToken, d.lastErr = d.Read()
	}
	return d.lastToken, d.lastErr
}

// Read returns the next token.
// It will return an error if there is no valid token.
func (d *Decoder) Read() (Token, error) {
	defer func() { d.lastCall = readCall }()
	if d.lastCall == peekCall {
		return d.lastToken, d.lastErr
	}

	tok, err := d.parseNext(d.lastToken.kind)
	if err != nil {
		return Token{}, err
	}

	switch tok.kind {
	case comma, semicolon:
		tok, err = d.parseNext(tok.kind)
		if err != nil {
			return Token{}, err
		}
	}
	d.lastToken = tok
	return tok, nil
}

const (
	mismatchedFmt = "mismatched close character %q"
	unexpectedFmt = "unexpected character %q"
)

// parseNext parses the next Token based on given last kind.
func (d *Decoder) parseNext(lastKind Kind) (Token, error) {
	// Trim leading spaces.
	d.consume(0)
	isEOF := false
	if len(d.in) == 0 {
		isEOF = true
	}

	switch lastKind {
	case EOF:
		return d.consumeToken(EOF, 0, 0), nil

	case bof:
		// Start of top level message. Next token can be EOF or Name.
		if isEOF {
			return d.consumeToken(EOF, 0, 0), nil
		}
		return d.parseFieldName()

	case Name:
		// Next token can be MessageOpen, ListOpen or Scalar.
		if isEOF {
			return Token{}, ErrUnexpectedEOF
		}
		switch ch := d.in[0]; ch {
		case '{', '<':
			d.pushOpenStack(ch)
			return d.consumeToken(MessageOpen, 1, 0), nil
		case '[':
			d.pushOpenStack(ch)
			return d.consumeToken(ListOpen, 1, 0), nil
		default:
			return d.parseScalar()
		}

	case Scalar:
		openKind, closeCh := d.currentOpenKind()
		switch openKind {
		case bof:
			// Top level message.
			// 	Next token can be EOF, comma, semicolon or Name.
			if isEOF {
				return d.consumeToken(EOF, 0, 0), nil
			}
			switch d.in[0] {
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			case ';':
				return d.consumeToken(semicolon, 1, 0), nil
			default:
				return d.parseFieldName()
			}

		case MessageOpen:
			// Next token can be MessageClose, comma, semicolon or Name.
			if isEOF {
				return Token{}, ErrUnexpectedEOF
			}
			switch ch := d.in[0]; ch {
			case closeCh:
				d.popOpenStack()
				return d.consumeToken(MessageClose, 1, 0), nil
			case otherCloseChar[closeCh]:
				return Token{}, d.newSyntaxError(mismatchedFmt, ch)
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			case ';':
				return d.consumeToken(semicolon, 1, 0), nil
			default:
				return d.parseFieldName()
			}

		case ListOpen:
			// Next token can be ListClose or comma.
			if isEOF {
				return Token{}, ErrUnexpectedEOF
			}
			switch ch := d.in[0]; ch {
			case ']':
				d.popOpenStack()
				return d.consumeToken(ListClose, 1, 0), nil
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			default:
				return Token{}, d.newSyntaxError(unexpectedFmt, ch)
			}
		}

	case MessageOpen:
		// Next token can be MessageClose or Name.
		if isEOF {
			return Token{}, ErrUnexpectedEOF
		}
		_, closeCh := d.currentOpenKind()
		switch ch := d.in[0]; ch {
		case closeCh:
			d.popOpenStack()
			return d.consumeToken(MessageClose, 1, 0), nil
		case otherCloseChar[closeCh]:
			return Token{}, d.newSyntaxError(mismatchedFmt, ch)
		default:
			return d.parseFieldName()
		}

	case MessageClose:
		openKind, closeCh := d.currentOpenKind()
		switch openKind {
		case bof:
			// Top level message.
			// Next token can be EOF, comma, semicolon or Name.
			if isEOF {
				return d.consumeToken(EOF, 0, 0), nil
			}
			switch ch := d.in[0]; ch {
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			case ';':
				return d.consumeToken(semicolon, 1, 0), nil
			default:
				return d.parseFieldName()
			}

		case MessageOpen:
			// Next token can be MessageClose, comma, semicolon or Name.
			if isEOF {
				return Token{}, ErrUnexpectedEOF
			}
			switch ch := d.in[0]; ch {
			case closeCh:
				d.popOpenStack()
				return d.consumeToken(MessageClose, 1, 0), nil
			case otherCloseChar[closeCh]:
				return Token{}, d.newSyntaxError(mismatchedFmt, ch)
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			case ';':
				return d.consumeToken(semicolon, 1, 0), nil
			default:
				return d.parseFieldName()
			}

		case ListOpen:
			// Next token can be ListClose or comma
			if isEOF {
				return Token{}, ErrUnexpectedEOF
			}
			switch ch := d.in[0]; ch {
			case closeCh:
				d.popOpenStack()
				return d.consumeToken(ListClose, 1, 0), nil
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			default:
				return Token{}, d.newSyntaxError(unexpectedFmt, ch)
			}
		}

	case ListOpen:
		// Next token can be ListClose, MessageStart or Scalar.
		if isEOF {
			return Token{}, ErrUnexpectedEOF
		}
		switch ch := d.in[0]; ch {
		case ']':
			d.popOpenStack()
			return d.consumeToken(ListClose, 1, 0), nil
		case '{', '<':
			d.pushOpenStack(ch)
			return d.consumeToken(MessageOpen, 1, 0), nil
		default:
			return d.parseScalar()
		}

	case ListClose:
		openKind, closeCh := d.currentOpenKind()
		switch openKind {
		case bof:
			// Top level message.
			// Next token can be EOF, comma, semicolon or Name.
			if isEOF {
				return d.consumeToken(EOF, 0, 0), nil
			}
			switch ch := d.in[0]; ch {
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			case ';':
				return d.consumeToken(semicolon, 1, 0), nil
			default:
				return d.parseFieldName()
			}

		case MessageOpen:
			// Next token can be MessageClose, comma, semicolon or Name.
			if isEOF {
				return Token{}, ErrUnexpectedEOF
			}
			switch ch := d.in[0]; ch {
			case closeCh:
				d.popOpenStack()
				return d.consumeToken(MessageClose, 1, 0), nil
			case otherCloseChar[closeCh]:
				return Token{}, d.newSyntaxError(mismatchedFmt, ch)
			case ',':
				return d.consumeToken(comma, 1, 0), nil
			case ';':
				return d.consumeToken(semicolon, 1, 0), nil
			default:
				return d.parseFieldName()
			}

		default:
			// It is not possible to have this case. Let it panic below.
		}

	case comma, semicolon:
		openKind, closeCh := d.currentOpenKind()
		switch openKind {
		case bof:
			// Top level message. Next token can be EOF or Name.
			if isEOF {
				return d.consumeToken(EOF, 0, 0), nil
			}
			return d.parseFieldName()

		case MessageOpen:
			// Next token can be MessageClose or Name.
			if isEOF {
				return Token{}, ErrUnexpectedEOF
			}
			switch ch := d.in[0]; ch {
			case closeCh:
				d.popOpenStack()
				return d.consumeToken(MessageClose, 1, 0), nil
			case otherCloseChar[closeCh]:
				return Token{}, d.newSyntaxError(mismatchedFmt, ch)
			default:
				return d.parseFieldName()
			}

		case ListOpen:
			if lastKind == semicolon {
				// It is not be possible to have this case as logic here
				// should not have produced a semicolon Token when inside a
				// list. Let it panic below.
				break
			}
			// Next token can be MessageOpen or Scalar.
			if isEOF {
				return Token{}, ErrUnexpectedEOF
			}
			switch ch := d.in[0]; ch {
			case '{', '<':
				d.pushOpenStack(ch)
				return d.consumeToken(MessageOpen, 1, 0), nil
			default:
				return d.parseScalar()
			}
		}
	}

	line, column := d.Position(len(d.orig) - len(d.in))
	panic(fmt.Sprintf("Decoder.parseNext: bug at handling line %d:%d with lastKind=%v", line, column, lastKind))
}

var otherCloseChar = map[byte]byte{
	'}': '>',
	'>': '}',
}

// currentOpenKind indicates whether current position is inside a message, list
// or top-level message by returning MessageOpen, ListOpen or bof respectively.
// If the returned kind is either a MessageOpen or ListOpen, it also returns the
// corresponding closing character.
func (d *Decoder) currentOpenKind() (Kind, byte) {
	if len(d.openStack) == 0 {
		return bof, 0
	}
	openCh := d.openStack[len(d.openStack)-1]
	switch openCh {
	case '{':
		return MessageOpen, '}'
	case '<':
		return MessageOpen, '>'
	case '[':
		return ListOpen, ']'
	}
	panic(fmt.Sprintf("Decoder: openStack contains invalid byte %s", string(openCh)))
}

func (d *Decoder) pushOpenStack(ch byte) {
	d.openStack = append(d.openStack, ch)
}

func (d *Decoder) popOpenStack() {
	d.openStack = d.openStack[:len(d.openStack)-1]
}

// parseFieldName parses field name and separator.
func (d *Decoder) parseFieldName() (tok Token, err error) {
	defer func() {
		if err == nil && d.tryConsumeChar(':') {
			tok.attrs |= hasSeparator
		}
	}()

	// Extension or Any type URL.
	if d.in[0] == '[' {
		return d.parseTypeName()
	}

	// Identifier.
	if size := parseIdent(d.in, false); size > 0 {
		return d.consumeToken(Name, size, uint8(IdentName)), nil
	}

	// Field number. Identify if input is a valid number that is not negative
	// and is decimal integer within 32-bit range.
	if num := parseNumber(d.in); num.size > 0 {
		if !num.neg && num.kind == numDec {
			if _, err := strconv.ParseInt(string(d.in[:num.size]), 10, 32); err == nil {
				return d.consumeToken(Name, num.size, uint8(FieldNumber)), nil
			}
		}
		return Token{}, d.newSyntaxError("invalid field number: %s", d.in[:num.size])
	}

	return Token{}, d.newSyntaxError("invalid field name: %s", errRegexp.Find(d.in))
}

// parseTypeName parses Any type URL or extension field name. The name is
// enclosed in [ and ] characters. The C++ parser does not handle many legal URL
// strings. This implementation is more liberal and allows for the pattern
// ^[-_a-zA-Z0-9]+([./][-_a-zA-Z0-9]+)*`). Whitespaces and comments are allowed
// in between [ ], '.', '/' and the sub names.
func (d *Decoder) parseTypeName() (Token, error) {
	startPos := len(d.orig) - len(d.in)
	// Use alias s to advance first in order to use d.in for error handling.
	// Caller already checks for [ as first character.
	s := consume(d.in[1:], 0)
	if len(s) == 0 {
		return Token{}, ErrUnexpectedEOF
	}

	var name []byte
	for len(s) > 0 && isTypeNameChar(s[0]) {
		name = append(name, s[0])
		s = s[1:]
	}
	s = consume(s, 0)

	var closed bool
	for len(s) > 0 && !closed {
		switch {
		case s[0] == ']':
			s = s[1:]
			closed = true

		case s[0] == '/', s[0] == '.':
			if len(name) > 0 && (name[len(name)-1] == '/' || name[len(name)-1] == '.') {
				return Token{}, d.newSyntaxError("invalid type URL/extension field name: %s",
					d.orig[startPos:len(d.orig)-len(s)+1])
			}
			name = append(name, s[0])
			s = s[1:]
			s = consume(s, 0)
			for len(s) > 0 && isTypeNameChar(s[0]) {
				name = append(name, s[0])
				s = s[1:]
			}
			s = consume(s, 0)

		default:
			return Token{}, d.newSyntaxError(
				"invalid type URL/extension field name: %s", d.orig[startPos:len(d.orig)-len(s)+1])
		}
	}

	if !closed {
		return Token{}, ErrUnexpectedEOF
	}

	// First character cannot be '.'. Last character cannot be '.' or '/'.
	size := len(name)
	if size == 0 || name[0] == '.' || name[size-1] == '.' || name[size-1] == '/' {
		return Token{}, d.newSyntaxError("invalid type URL/extension field name: %s",
			d.orig[startPos:len(d.orig)-len(s)])
	}

	d.in = s
	endPos := len(d.orig) - len(d.in)
	d.consume(0)

	return Token{
		kind:  Name,
		attrs: uint8(TypeName),
		pos:   startPos,
		raw:   d.orig[startPos:endPos],
		str:   string(name),
	}, nil
}

func isTypeNameChar(b byte) bool {
	return (b == '-' || b == '_' ||
		('0' <= b && b <= '9') ||
		('a' <= b && b <= 'z') ||
		('A' <= b && b <= 'Z'))
}

func isWhiteSpace(b byte) bool {
	switch b {
	case ' ', '\n', '\r', '\t':
		return true
	default:
		return false
	}
}

// parseIdent parses an unquoted proto identifier and returns size.
// If allowNeg is true, it allows '-' to be the first character in the
// identifier. This is used when parsing literal values like -infinity, etc.
// Regular expression matches an identifier: `^[_a-zA-Z][_a-zA-Z0-9]*`
func parseIdent(input []byte, allowNeg bool) int {
	var size int

	s := input
	if len(s) == 0 {
		return 0
	}

	if allowNeg && s[0] == '-' {
		s = s[1:]
		size++
		if len(s) == 0 {
			return 0
		}
	}

	switch {
	case s[0] == '_',
		'a' <= s[0] && s[0] <= 'z',
		'A' <= s[0] && s[0] <= 'Z':
		s = s[1:]
		size++
	default:
		return 0
	}

	for len(s) > 0 && (s[0] == '_' ||
		'a' <= s[0] && s[0] <= 'z' ||
		'A' <= s[0] && s[0] <= 'Z' ||
		'0' <= s[0] && s[0] <= '9') {
		s = s[1:]
		size++
	}

	if len(s) > 0 && !isDelim(s[0]) {
		return 0
	}

	return size
}

// parseScalar parses for a string, literal or number value.
func (d *Decoder) parseScalar() (Token, error) {
	if d.in[0] == '"' || d.in[0] == '\'' {
		return d.parseStringValue()
	}

	if tok, ok := d.parseLiteralValue(); ok {
		return tok, nil
	}

	if tok, ok := d.parseNumberValue(); ok {
		return tok, nil
	}

	return Token{}, d.newSyntaxError("invalid scalar value: %s", errRegexp.Find(d.in))
}

// parseLiteralValue parses a literal value. A literal value is used for
// bools, special floats and enums. This function simply identifies that the
// field value is a literal.
func (d *Decoder) parseLiteralValue() (Token, bool) {
	size := parseIdent(d.in, true)
	if size == 0 {
		return Token{}, false
	}
	return d.consumeToken(Scalar, size, literalValue), true
}

// consumeToken constructs a Token for given Kind from d.in and consumes given
// size-length from it.
func (d *Decoder) consumeToken(kind Kind, size int, attrs uint8) Token {
	// Important to compute raw and pos before consuming.
	tok := Token{
		kind:  kind,
		attrs: attrs,
		pos:   len(d.orig) - len(d.in),
		raw:   d.in[:size],
	}
	d.consume(size)
	return tok
}

// newSyntaxError returns a syntax error with line and column information for
// current position.
func (d *Decoder) newSyntaxError(f string, x ...interface{}) error {
	e := errors.New(f, x...)
	line, column := d.Position(len(d.orig) - len(d.in))
	return errors.New("syntax error (line %d:%d): %v", line, column, e)
}

// Position returns line and column number of given index of the original input.
// It will panic if index is out of range.
func (d *Decoder) Position(idx int) (line int, column int) {
	b := d.orig[:idx]
	line = bytes.Count(b, []byte("\n")) + 1
	if i := bytes.LastIndexByte(b, '\n'); i >= 0 {
		b = b[i+1:]
	}
	column = utf8.RuneCount(b) + 1 // ignore multi-rune characters
	return line, column
}

func (d *Decoder) tryConsumeChar(c byte) bool {
	if len(d.in) > 0 && d.in[0] == c {
		d.consume(1)
		return true
	}
	return false
}

// consume consumes n bytes of input and any subsequent whitespace or comments.
func (d *Decoder) consume(n int) {
	d.in = consume(d.in, n)
	return
}

// consume consumes n bytes of input and any subsequent whitespace or comments.
func consume(b []byte, n int) []byte {
	b = b[n:]
	for len(b) > 0 {
		switch b[0] {
		case ' ', '\n', '\r', '\t':
			b = b[1:]
		case '#':
			if i := bytes.IndexByte(b, '\n'); i >= 0 {
				b = b[i+len("\n"):]
			} else {
				b = nil
			}
		default:
			return b
		}
	}
	return b
}

// Any sequence that looks like a non-delimiter (for error reporting).
var errRegexp = regexp.MustCompile(`^([-+._a-zA-Z0-9\/]+|.)`)

// isDelim returns true if given byte is a delimiter character.
func isDelim(c byte) bool {
	return !(c == '-' || c == '+' || c == '.' || c == '_' ||
		('a' <= c && c <= 'z') ||
		('A' <= c && c <= 'Z') ||
		('0' <= c && c <= '9'))
}