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
|
package main
import (
"bytes"
"fmt"
"io"
"io/ioutil"
"os"
"path/filepath"
"testing"
"github.com/n-peugnet/dna-backup/logger"
"github.com/n-peugnet/dna-backup/testutils"
"github.com/n-peugnet/dna-backup/utils"
)
func chunkCompare(t *testing.T, dataDir string, repo *Repo, testFiles []string, chunkCount int) {
reader, writer := io.Pipe()
chunks := make(chan []byte)
files := listFiles(dataDir)
go concatFiles(&files, writer)
go repo.chunkStream(reader, chunks)
offset := 0
buff := make([]byte, repo.chunkSize*chunkCount)
for _, f := range testFiles {
content, err := os.ReadFile(filepath.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 * repo.chunkSize
end := (i + 1) * repo.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%repo.chunkSize-10:])
t.Log("Actual:", content)
}
i++
}
if i != chunkCount {
t.Errorf("Incorrect number of chunks: %d, should be: %d", i, chunkCount)
}
}
func (r *Repo) chunkStream(stream io.Reader, chunks chan<- []byte) {
var buff []byte
var prev, read = r.chunkSize, 0
var err error
for err != io.EOF {
if prev == r.chunkSize {
buff = make([]byte, r.chunkSize)
prev, err = stream.Read(buff)
} else {
read, err = stream.Read(buff[prev:])
prev += read
}
if err != nil && err != io.EOF {
logger.Error(err)
}
if prev == r.chunkSize {
chunks <- buff
}
}
if prev != r.chunkSize {
chunks <- buff[:prev]
}
close(chunks)
}
func storeChunks(dest string, chunks <-chan []byte) {
i := 0
for c := range chunks {
path := filepath.Join(dest, fmt.Sprintf(chunkIdFmt, i))
err := os.WriteFile(path, c, 0664)
if err != nil {
logger.Error(err)
}
i++
}
}
func TestReadFiles1(t *testing.T) {
tmpDir := t.TempDir()
repo := NewRepo(tmpDir)
chunkCount := 590/repo.chunkSize + 1
dataDir := filepath.Join("testdata", "logs", "1")
files := []string{"logTest.log"}
chunkCompare(t, dataDir, repo, files, chunkCount)
}
func TestReadFiles2(t *testing.T) {
tmpDir := t.TempDir()
repo := NewRepo(tmpDir)
chunkCount := 22899/repo.chunkSize + 1
dataDir := filepath.Join("testdata", "logs", "2")
files := []string{"csvParserTest.log", "slipdb.log"}
chunkCompare(t, dataDir, repo, files, chunkCount)
}
func TestReadFiles3(t *testing.T) {
tmpDir := t.TempDir()
repo := NewRepo(tmpDir)
chunkCount := 119398/repo.chunkSize + 1
dataDir := filepath.Join("testdata", "logs")
files := []string{
filepath.Join("1", "logTest.log"),
filepath.Join("2", "csvParserTest.log"),
filepath.Join("2", "slipdb.log"),
filepath.Join("3", "indexingTreeTest.log"),
}
chunkCompare(t, dataDir, repo, files, chunkCount)
}
func TestNoSuchFile(t *testing.T) {
tmpDir := t.TempDir()
os.Symlink("./notexisting", filepath.Join(tmpDir, "linknotexisting"))
var buff bytes.Buffer
files := listFiles(tmpDir)
testutils.AssertLen(t, 1, files, "Files")
concatFiles(&files, utils.NopCloser(&buff))
testutils.AssertLen(t, 0, files, "Files")
testutils.AssertLen(t, 0, buff.Bytes(), "Buffer")
}
func TestLoadChunks(t *testing.T) {
resultDir := t.TempDir()
dataDir := filepath.Join("testdata", "logs")
repo := NewRepo(resultDir)
repo.chunkReadWrapper = utils.NopReadWrapper
repo.chunkWriteWrapper = utils.NopWriteWrapper
resultVersion := filepath.Join(resultDir, "00000")
resultChunks := filepath.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 IdentifiedChunk, 16)
files := listFiles(dataDir)
go concatFiles(&files, writer1)
go concatFiles(&files, writer2)
go repo.chunkStream(reader1, chunks1)
go repo.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 prepareChunks(dataDir string, repo *Repo, streamFunc func(*[]File, io.WriteCloser)) {
resultVersion := filepath.Join(repo.path, "00000")
resultChunks := filepath.Join(resultVersion, chunksName)
os.MkdirAll(resultChunks, 0775)
reader := getDataStream(dataDir, streamFunc)
chunks := make(chan []byte, 16)
go repo.chunkStream(reader, chunks)
storeChunks(resultChunks, chunks)
}
func getDataStream(dataDir string, streamFunc func(*[]File, io.WriteCloser)) io.Reader {
reader, writer := io.Pipe()
files := listFiles(dataDir)
go streamFunc(&files, writer)
return reader
}
func TestBsdiff(t *testing.T) {
resultDir := t.TempDir()
repo := NewRepo(resultDir)
dataDir := filepath.Join("testdata", "logs")
addedFile1 := filepath.Join(dataDir, "2", "slogTest.log")
addedFile2 := filepath.Join(dataDir, "3", "slogTest.log")
// Store initial chunks
prepareChunks(dataDir, repo, concatFiles)
// Modify data
ioutil.WriteFile(addedFile1, []byte("hello"), 0664)
defer os.Remove(addedFile1)
ioutil.WriteFile(addedFile2, make([]byte, 4000), 0664)
defer os.Remove(addedFile2)
// configure repo
repo.chunkReadWrapper = utils.NopReadWrapper
repo.chunkWriteWrapper = utils.NopWriteWrapper
// Load previously stored chunks
oldChunks := make(chan IdentifiedChunk, 16)
versions := repo.loadVersions()
go repo.loadChunks(versions, oldChunks)
repo.hashChunks(oldChunks)
// Read new data
newVersion := len(versions)
newPath := filepath.Join(repo.path, fmt.Sprintf(versionFmt, newVersion))
os.MkdirAll(newPath, 0775)
reader := getDataStream(dataDir, concatFiles)
recipe := repo.matchStream(reader, newVersion)
newChunks := extractDeltaChunks(recipe)
testutils.AssertLen(t, 2, newChunks, "New delta chunks:")
for _, c := range newChunks {
logger.Info("Patch size:", len(c.Patch))
if len(c.Patch) >= repo.chunkSize/10 {
t.Errorf("Bsdiff of chunk is too large: %d", len(c.Patch))
}
}
}
func TestCommit(t *testing.T) {
dest := t.TempDir()
source := filepath.Join("testdata", "logs")
expected := filepath.Join("testdata", "repo_8k")
repo := NewRepo(dest)
repo.chunkReadWrapper = utils.NopReadWrapper
repo.chunkWriteWrapper = utils.NopWriteWrapper
repo.Commit(source)
assertSameTree(t, assertCompatibleRepoFile, expected, dest, "Commit")
}
func TestCommitZlib(t *testing.T) {
dest := t.TempDir()
source := filepath.Join("testdata", "logs")
expected := filepath.Join("testdata", "repo_8k_zlib")
repo := NewRepo(dest)
repo.chunkReadWrapper = utils.ZlibReader
repo.chunkWriteWrapper = utils.ZlibWriter
repo.Commit(source)
assertSameTree(t, assertCompatibleRepoFile, expected, dest, "Commit")
}
func TestRestore(t *testing.T) {
dest := t.TempDir()
source := filepath.Join("testdata", "repo_8k")
expected := filepath.Join("testdata", "logs")
repo := NewRepo(source)
repo.chunkReadWrapper = utils.NopReadWrapper
repo.chunkWriteWrapper = utils.NopWriteWrapper
repo.Restore(dest)
assertSameTree(t, testutils.AssertSameFile, expected, dest, "Restore")
}
func TestRestoreZlib(t *testing.T) {
dest := t.TempDir()
source := filepath.Join("testdata", "repo_8k_zlib")
expected := filepath.Join("testdata", "logs")
repo := NewRepo(source)
repo.chunkReadWrapper = utils.ZlibReader
repo.chunkWriteWrapper = utils.ZlibWriter
repo.Restore(dest)
assertSameTree(t, testutils.AssertSameFile, expected, dest, "Restore")
}
func TestHashes(t *testing.T) {
dest := t.TempDir()
source := filepath.Join("testdata", "repo_8k")
chunks := make(chan IdentifiedChunk, 16)
storeQueue := make(chan chunkData, 16)
storeEnd := make(chan bool)
repo1 := NewRepo(source)
repo1.chunkReadWrapper = utils.NopReadWrapper
repo1.chunkWriteWrapper = utils.NopWriteWrapper
go repo1.loadChunks([]string{filepath.Join(source, "00000")}, chunks)
for c := range chunks {
fp, sk := repo1.hashChunk(c.GetId(), c.Reader())
content, err := io.ReadAll(c.Reader())
if err != nil {
t.Error(err)
}
storeQueue <- chunkData{
hashes: chunkHashes{fp, sk},
content: content,
id: c.GetId(),
}
}
repo2 := NewRepo(dest)
repo2.chunkReadWrapper = utils.NopReadWrapper
repo2.chunkWriteWrapper = utils.NopWriteWrapper
os.MkdirAll(filepath.Join(dest, "00000", chunksName), 0775)
go repo2.storageWorker(0, storeQueue, storeEnd)
close(storeQueue)
<-storeEnd
testutils.AssertLen(t, 0, repo2.fingerprints, "Fingerprints")
testutils.AssertLen(t, 0, repo2.sketches, "Sketches")
repo2.loadHashes([]string{filepath.Join(dest, "00000")})
testutils.AssertSame(t, repo1.fingerprints, repo2.fingerprints, "Fingerprint maps")
testutils.AssertSame(t, repo1.sketches, repo2.sketches, "Sketches maps")
}
func assertSameTree(t *testing.T, apply func(t *testing.T, expected string, actual string, prefix string), expected string, actual string, prefix string) {
actualFiles := listFiles(actual)
expectedFiles := listFiles(expected)
efCount := len(expectedFiles)
if efCount <= 0 {
t.Fatalf("No expected files: %d", efCount)
}
afCount := len(actualFiles)
if efCount != afCount {
t.Fatalf("Incorrect number of files: %d, should be %d", afCount, efCount)
}
for i, ef := range expectedFiles {
af := actualFiles[i]
efRelPath := ef.Path[len(expected):]
afRelPath := af.Path[len(actual):]
if efRelPath != afRelPath {
t.Fatalf("File path '%s' does not match '%s'", afRelPath, efRelPath)
}
apply(t, ef.Path, af.Path, prefix)
}
}
func assertCompatibleRepoFile(t *testing.T, expected string, actual string, prefix string) {
if filepath.Base(expected) == filesName {
// TODO: Check Filelist file
// eFiles := loadFileList(expected)
// aFiles := loadFileList(actual)
// testutils.AssertLen(t, len(eFiles), aFiles, prefix)
// for i, eFile := range eFiles {
// eFile.Path = filepath.FromSlash(eFile.Path)
// if eFile != aFiles[i] {
// t.Fatal(prefix, "file entry do not match:", aFiles[i], ", expected:", eFile)
// }
// }
} else if filepath.Base(expected) == recipeName {
// TODO: Check Recipe files
// eRecipe := loadRecipe(expected)
// aRecipe := loadRecipe(actual)
// testutils.AssertSame(t, eRecipe, aRecipe, prefix+"recipe")
} else if filepath.Base(expected) == hashesName {
// Hashes file is checked in TestHashes
} else {
// Chunk content file
testutils.AssertSameFile(t, expected, actual, prefix)
}
}
func assertChunkContent(t *testing.T, expected []byte, c Chunk, prefix string) {
buf, err := io.ReadAll(c.Reader())
if err != nil {
t.Fatal(err)
}
testutils.AssertSame(t, expected, buf, prefix+" Chunk content")
}
|