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
|
package compressutil
import (
"bytes"
"compress/gzip"
"compress/lzw"
"fmt"
"io"
)
const (
// A byte value used as a canary prefix for the compressed information
// which is used to distinguish if a JSON input is compressed or not.
// The value of this constant should not be a first character of any
// valid JSON string.
// Byte value used as canary when using Gzip format
CompressionCanaryGzip byte = 'G'
// Byte value used as canary when using Lzw format
CompressionCanaryLzw byte = 'L'
CompressionTypeLzw = "lzw"
CompressionTypeGzip = "gzip"
)
// CompressionConfig is used to select a compression type to be performed by
// Compress and Decompress utilities.
// Supported types are:
// * CompressionTypeLzw
// * CompressionTypeGzip
//
// When using CompressionTypeGzip, the compression levels can also be chosen:
// * gzip.DefaultCompression
// * gzip.BestSpeed
// * gzip.BestCompression
type CompressionConfig struct {
// Type of the compression algorithm to be used
Type string
// When using Gzip format, the compression level to employ
GzipCompressionLevel int
}
// Compress places the canary byte in a buffer and uses the same buffer to fill
// in the compressed information of the given input. The configuration supports
// two type of compression: LZW and Gzip. When using Gzip compression format,
// if GzipCompressionLevel is not specified, the 'gzip.DefaultCompression' will
// be assumed.
func Compress(data []byte, config *CompressionConfig) ([]byte, error) {
var buf bytes.Buffer
var writer io.WriteCloser
var err error
if config == nil {
return nil, fmt.Errorf("config is nil")
}
// Write the canary into the buffer and create writer to compress the
// input data based on the configured type
switch config.Type {
case CompressionTypeLzw:
buf.Write([]byte{CompressionCanaryLzw})
writer = lzw.NewWriter(&buf, lzw.LSB, 8)
case CompressionTypeGzip:
buf.Write([]byte{CompressionCanaryGzip})
switch {
case config.GzipCompressionLevel == gzip.BestCompression,
config.GzipCompressionLevel == gzip.BestSpeed,
config.GzipCompressionLevel == gzip.DefaultCompression:
// These are valid compression levels
default:
// If compression level is set to NoCompression or to
// any invalid value, fallback to Defaultcompression
config.GzipCompressionLevel = gzip.DefaultCompression
}
writer, err = gzip.NewWriterLevel(&buf, config.GzipCompressionLevel)
default:
return nil, fmt.Errorf("unsupported compression type")
}
if err != nil {
return nil, fmt.Errorf("failed to create a compression writer; err: %v", err)
}
if writer == nil {
return nil, fmt.Errorf("failed to create a compression writer")
}
// Compress the input and place it in the same buffer containing the
// canary byte.
if _, err = writer.Write(data); err != nil {
return nil, fmt.Errorf("failed to compress input data; err: %v", err)
}
// Close the io.WriteCloser
if err = writer.Close(); err != nil {
return nil, err
}
// Return the compressed bytes with canary byte at the start
return buf.Bytes(), nil
}
// Decompress checks if the first byte in the input matches the canary byte.
// If the first byte is a canary byte, then the input past the canary byte
// will be decompressed using the method specified in the given configuration.
// If the first byte isn't a canary byte, then the utility returns a boolean
// value indicating that the input was not compressed.
func Decompress(data []byte) ([]byte, bool, error) {
var err error
var reader io.ReadCloser
if data == nil || len(data) == 0 {
return nil, false, fmt.Errorf("'data' being decompressed is empty")
}
switch {
case data[0] == CompressionCanaryGzip:
// If the first byte matches the canary byte, remove the canary
// byte and try to decompress the data that is after the canary.
if len(data) < 2 {
return nil, false, fmt.Errorf("invalid 'data' after the canary")
}
data = data[1:]
reader, err = gzip.NewReader(bytes.NewReader(data))
case data[0] == CompressionCanaryLzw:
// If the first byte matches the canary byte, remove the canary
// byte and try to decompress the data that is after the canary.
if len(data) < 2 {
return nil, false, fmt.Errorf("invalid 'data' after the canary")
}
data = data[1:]
reader = lzw.NewReader(bytes.NewReader(data), lzw.LSB, 8)
default:
// If the first byte doesn't match the canary byte, it means
// that the content was not compressed at all. Indicate the
// caller that the input was not compressed.
return nil, true, nil
}
if err != nil {
return nil, false, fmt.Errorf("failed to create a compression reader; err: %v", err)
}
if reader == nil {
return nil, false, fmt.Errorf("failed to create a compression reader")
}
// Close the io.ReadCloser
defer reader.Close()
// Read all the compressed data into a buffer
var buf bytes.Buffer
if _, err = io.Copy(&buf, reader); err != nil {
return nil, false, err
}
return buf.Bytes(), false, nil
}
|
