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
|
package main
import (
"bytes"
"io"
"io/ioutil"
"log"
"os"
"path"
"reflect"
"testing"
"github.com/gabstv/go-bsdiff/pkg/bsdiff"
)
func chunkCompare(t *testing.T, dataDir string, testFiles []string, chunkCount int) {
reader, writer := io.Pipe()
chunks := make(chan []byte)
files := listFiles(dataDir)
go concatFiles(files, writer)
go chunkStream(reader, chunks)
offset := 0
buff := make([]byte, chunkSize*chunkCount)
for _, f := range testFiles {
content, err := os.ReadFile(path.Join(dataDir, f))
if err != nil {
t.Error("Error reading test data file")
}
for i := range content {
buff[offset+i] = content[i]
}
offset += len(content)
}
i := 0
for c := range chunks {
start := i * chunkSize
end := (i + 1) * chunkSize
if end > offset {
end = offset
}
content := buff[start:end]
if bytes.Compare(c, content) != 0 {
t.Errorf("Chunk %d does not match file content", i)
// for i, b := range c {
// fmt.Printf("E: %d, A: %d\n", b, content[i])
// }
t.Log("Expected: ", c[:10], "...", c[end%chunkSize-10:])
t.Log("Actual:", content)
}
i++
}
if i != chunkCount {
t.Errorf("Incorrect number of chunks: %d, should be: %d", i, chunkCount)
}
}
func TestReadFiles1(t *testing.T) {
chunkCount := 590/chunkSize + 1
dataDir := path.Join("test", "data", "logs", "1")
files := []string{"logTest.log"}
chunkCompare(t, dataDir, files, chunkCount)
}
func TestReadFiles2(t *testing.T) {
chunkCount := 22899/chunkSize + 1
dataDir := path.Join("test", "data", "logs", "2")
files := []string{"csvParserTest.log", "slipdb.log"}
chunkCompare(t, dataDir, files, chunkCount)
}
func TestReadFiles3(t *testing.T) {
chunkCount := 119398/chunkSize + 1
dataDir := path.Join("test", "data", "logs")
files := []string{
path.Join("1", "logTest.log"),
path.Join("2", "csvParserTest.log"),
path.Join("2", "slipdb.log"),
path.Join("3", "indexingTreeTest.log"),
}
chunkCompare(t, dataDir, files, chunkCount)
}
func TestLoadChunks(t *testing.T) {
resultDir := t.TempDir()
dataDir := path.Join("test", "data", "logs")
repo := NewRepo(resultDir)
resultVersion := path.Join(resultDir, "00000")
resultChunks := path.Join(resultVersion, chunksName)
os.MkdirAll(resultChunks, 0775)
reader1, writer1 := io.Pipe()
reader2, writer2 := io.Pipe()
chunks1 := make(chan []byte, 16)
chunks2 := make(chan []byte, 16)
chunks3 := make(chan StoredChunk, 16)
files := listFiles(dataDir)
go concatFiles(files, writer1)
go concatFiles(files, writer2)
go chunkStream(reader1, chunks1)
go chunkStream(reader2, chunks2)
storeChunks(resultChunks, chunks1)
versions := []string{resultVersion}
go repo.loadChunks(versions, chunks3)
i := 0
for c2 := range chunks2 {
c3 := <-chunks3
buff, err := io.ReadAll(c3.Reader())
if err != nil {
t.Errorf("Error reading from chunk %d: %s\n", c3, err)
}
if bytes.Compare(c2, buff) != 0 {
t.Errorf("Chunk %d does not match file content", i)
t.Log("Expected: ", c2[:10], "...")
t.Log("Actual:", buff)
}
i++
}
}
func TestExtractNewChunks(t *testing.T) {
chunks := []Chunk{
&TempChunk{value: []byte{'a'}},
&LoadedChunk{id: &ChunkId{0, 0}},
&TempChunk{value: []byte{'b'}},
&TempChunk{value: []byte{'c'}},
&LoadedChunk{id: &ChunkId{0, 1}},
}
newChunks := extractNewChunks(chunks)
if len(newChunks) != 2 {
t.Error("New chunks should contain 2 slices")
t.Log("Actual: ", newChunks)
}
if len(newChunks[1]) != 2 {
t.Error("New chunks second slice should contain 2 chunks")
t.Log("Actual: ", newChunks[0])
}
if !reflect.DeepEqual(newChunks[1][0], chunks[2]) {
t.Error("New chunks do not match")
t.Log("Expected: ", chunks[2])
t.Log("Actual: ", newChunks[1][0])
}
}
func TestStoreLoadFiles(t *testing.T) {
resultDir := t.TempDir()
dataDir := path.Join("test", "data", "logs")
resultFiles := path.Join(resultDir, filesName)
files1 := listFiles(dataDir)
storeFileList(resultFiles, files1)
files2 := loadFileList(resultFiles)
if len(files1) != 4 {
t.Errorf("Incorrect number of files: %d, should be %d\n", len(files1), 4)
}
for i, f := range files1 {
if f != files2[i] {
t.Errorf("Loaded file data %d does not match stored one", i)
t.Log("Expected: ", f)
t.Log("Actual: ", files2[i])
}
}
}
func TestBsdiff(t *testing.T) {
resultDir := t.TempDir()
dataDir := path.Join("test", "data", "logs")
addedFile := path.Join(dataDir, "2", "slogTest.log")
resultVersion := path.Join(resultDir, "00000")
resultChunks := path.Join(resultVersion, chunksName)
os.MkdirAll(resultChunks, 0775)
reader, writer := io.Pipe()
chunks := make(chan []byte, 16)
files := listFiles(dataDir)
go concatFiles(files, writer)
go chunkStream(reader, chunks)
storeChunks(resultChunks, chunks)
input := []byte("hello")
ioutil.WriteFile(addedFile, input, 0664)
defer os.Remove(addedFile)
reader, writer = io.Pipe()
oldChunks := make(chan StoredChunk, 16)
files = listFiles(dataDir)
repo := NewRepo(resultDir)
versions := repo.loadVersions()
go repo.loadChunks(versions, oldChunks)
go concatFiles(files, writer)
fingerprints, sketches := hashChunks(oldChunks)
recipe := repo.matchStream(reader, fingerprints)
newChunks := extractNewChunks(recipe)
log.Println("Checking new chunks:", len(newChunks[0]))
for _, chunks := range newChunks {
for _, c := range chunks {
id, exists := findSimilarChunk(c, sketches)
log.Println(id, exists)
if exists {
patch := new(bytes.Buffer)
stored := id.Reader(repo.path)
new := c.Reader()
bsdiff.Reader(stored, new, patch)
log.Println("Patch size:", patch.Len())
if patch.Len() >= chunkSize/10 {
t.Errorf("Bsdiff of chunk is too large: %d", patch.Len())
}
}
}
}
}
|
