aboutsummaryrefslogtreecommitdiff
path: root/vendor/github.com/go-logfmt/logfmt/decode.go
blob: 04e0efff5fc74a645472ffa9403f4362022310e3 (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
package logfmt

import (
	"bufio"
	"bytes"
	"fmt"
	"io"
	"unicode/utf8"
)

// A Decoder reads and decodes logfmt records from an input stream.
type Decoder struct {
	pos     int
	key     []byte
	value   []byte
	lineNum int
	s       *bufio.Scanner
	err     error
}

// NewDecoder returns a new decoder that reads from r.
//
// The decoder introduces its own buffering and may read data from r beyond
// the logfmt records requested.
func NewDecoder(r io.Reader) *Decoder {
	dec := &Decoder{
		s: bufio.NewScanner(r),
	}
	return dec
}

// ScanRecord advances the Decoder to the next record, which can then be
// parsed with the ScanKeyval method. It returns false when decoding stops,
// either by reaching the end of the input or an error. After ScanRecord
// returns false, the Err method will return any error that occurred during
// decoding, except that if it was io.EOF, Err will return nil.
func (dec *Decoder) ScanRecord() bool {
	if dec.err != nil {
		return false
	}
	if !dec.s.Scan() {
		dec.err = dec.s.Err()
		return false
	}
	dec.lineNum++
	dec.pos = 0
	return true
}

// ScanKeyval advances the Decoder to the next key/value pair of the current
// record, which can then be retrieved with the Key and Value methods. It
// returns false when decoding stops, either by reaching the end of the
// current record or an error.
func (dec *Decoder) ScanKeyval() bool {
	dec.key, dec.value = nil, nil
	if dec.err != nil {
		return false
	}

	line := dec.s.Bytes()

	// garbage
	for p, c := range line[dec.pos:] {
		if c > ' ' {
			dec.pos += p
			goto key
		}
	}
	dec.pos = len(line)
	return false

key:
	const invalidKeyError = "invalid key"

	start, multibyte := dec.pos, false
	for p, c := range line[dec.pos:] {
		switch {
		case c == '=':
			dec.pos += p
			if dec.pos > start {
				dec.key = line[start:dec.pos]
				if multibyte && bytes.IndexRune(dec.key, utf8.RuneError) != -1 {
					dec.syntaxError(invalidKeyError)
					return false
				}
			}
			if dec.key == nil {
				dec.unexpectedByte(c)
				return false
			}
			goto equal
		case c == '"':
			dec.pos += p
			dec.unexpectedByte(c)
			return false
		case c <= ' ':
			dec.pos += p
			if dec.pos > start {
				dec.key = line[start:dec.pos]
				if multibyte && bytes.IndexRune(dec.key, utf8.RuneError) != -1 {
					dec.syntaxError(invalidKeyError)
					return false
				}
			}
			return true
		case c >= utf8.RuneSelf:
			multibyte = true
		}
	}
	dec.pos = len(line)
	if dec.pos > start {
		dec.key = line[start:dec.pos]
		if multibyte && bytes.IndexRune(dec.key, utf8.RuneError) != -1 {
			dec.syntaxError(invalidKeyError)
			return false
		}
	}
	return true

equal:
	dec.pos++
	if dec.pos >= len(line) {
		return true
	}
	switch c := line[dec.pos]; {
	case c <= ' ':
		return true
	case c == '"':
		goto qvalue
	}

	// value
	start = dec.pos
	for p, c := range line[dec.pos:] {
		switch {
		case c == '=' || c == '"':
			dec.pos += p
			dec.unexpectedByte(c)
			return false
		case c <= ' ':
			dec.pos += p
			if dec.pos > start {
				dec.value = line[start:dec.pos]
			}
			return true
		}
	}
	dec.pos = len(line)
	if dec.pos > start {
		dec.value = line[start:dec.pos]
	}
	return true

qvalue:
	const (
		untermQuote  = "unterminated quoted value"
		invalidQuote = "invalid quoted value"
	)

	hasEsc, esc := false, false
	start = dec.pos
	for p, c := range line[dec.pos+1:] {
		switch {
		case esc:
			esc = false
		case c == '\\':
			hasEsc, esc = true, true
		case c == '"':
			dec.pos += p + 2
			if hasEsc {
				v, ok := unquoteBytes(line[start:dec.pos])
				if !ok {
					dec.syntaxError(invalidQuote)
					return false
				}
				dec.value = v
			} else {
				start++
				end := dec.pos - 1
				if end > start {
					dec.value = line[start:end]
				}
			}
			return true
		}
	}
	dec.pos = len(line)
	dec.syntaxError(untermQuote)
	return false
}

// Key returns the most recent key found by a call to ScanKeyval. The returned
// slice may point to internal buffers and is only valid until the next call
// to ScanRecord.  It does no allocation.
func (dec *Decoder) Key() []byte {
	return dec.key
}

// Value returns the most recent value found by a call to ScanKeyval. The
// returned slice may point to internal buffers and is only valid until the
// next call to ScanRecord.  It does no allocation when the value has no
// escape sequences.
func (dec *Decoder) Value() []byte {
	return dec.value
}

// Err returns the first non-EOF error that was encountered by the Scanner.
func (dec *Decoder) Err() error {
	return dec.err
}

func (dec *Decoder) syntaxError(msg string) {
	dec.err = &SyntaxError{
		Msg:  msg,
		Line: dec.lineNum,
		Pos:  dec.pos + 1,
	}
}

func (dec *Decoder) unexpectedByte(c byte) {
	dec.err = &SyntaxError{
		Msg:  fmt.Sprintf("unexpected %q", c),
		Line: dec.lineNum,
		Pos:  dec.pos + 1,
	}
}

// A SyntaxError represents a syntax error in the logfmt input stream.
type SyntaxError struct {
	Msg  string
	Line int
	Pos  int
}

func (e *SyntaxError) Error() string {
	return fmt.Sprintf("logfmt syntax error at pos %d on line %d: %s", e.Pos, e.Line, e.Msg)
}
nihil fit ex nihilo