/
merge.go
380 lines (347 loc) · 9.79 KB
/
merge.go
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
package fileset
import (
"io"
"io/ioutil"
"github.com/pachyderm/pachyderm/src/server/pkg/storage/chunk"
"github.com/pachyderm/pachyderm/src/server/pkg/storage/fileset/index"
"github.com/pachyderm/pachyderm/src/server/pkg/storage/fileset/tar"
)
// MergeReader merges a file's content that shows up across
// multiple fileset streams.
// A file's content is ordered based on the lexicographical order of
// the tagged content, so the output file content is produced by
// performing a merge of the tagged content.
type MergeReader struct {
pq *priorityQueue
}
func newMergeReader(rs []*Reader) *MergeReader {
var fileStreams []stream
for _, r := range rs {
fileStreams = append(fileStreams, &fileStream{r: r})
}
return &MergeReader{pq: newPriorityQueue(fileStreams)}
}
// Iterate iterates over the file merge readers in the merged fileset.
func (mr *MergeReader) Iterate(f func(*FileMergeReader) error) error {
return mr.pq.iterate(func(ss []stream, _ ...string) error {
// Convert generic streams to file streams.
var fss []*fileStream
for _, s := range ss {
fss = append(fss, s.(*fileStream))
}
// Create file merge reader and execute callback.
var frs []*FileReader
for _, fs := range fss {
fr, err := fs.r.Next()
if err != nil {
return err
}
frs = append(frs, fr)
}
return f(newFileMergeReader(frs))
})
}
// WriteTo writes the merged fileset to the passed in fileset writer.
func (mr *MergeReader) WriteTo(w *Writer) error {
return mr.pq.iterate(func(ss []stream, next ...string) error {
// Convert generic streams to file streams.
var fss []*fileStream
for _, s := range ss {
fss = append(fss, s.(*fileStream))
}
// Copy single file stream.
if len(fss) == 1 {
return fss[0].r.Iterate(func(fr *FileReader) error {
return w.CopyFile(fr)
}, next...)
}
// Create file merge reader and call its WriteTo function.
var frs []*FileReader
for _, fs := range fss {
fr, err := fs.r.Next()
if err != nil {
return err
}
frs = append(frs, fr)
}
return newFileMergeReader(frs).WriteTo(w)
})
}
// Get writes the merged fileset.
func (mr *MergeReader) Get(w io.Writer) error {
// Write a tar entry for each file merge reader.
if err := mr.Iterate(func(fmr *FileMergeReader) error {
return fmr.Get(w)
}); err != nil {
return err
}
// Close a tar writer to create tar EOF padding.
return tar.NewWriter(w).Close()
}
type fileStream struct {
r *Reader
idx *index.Index
}
func (fs *fileStream) next() error {
var err error
fs.idx, err = fs.r.Peek()
return err
}
func (fs *fileStream) key() string {
return fs.idx.Path
}
// FileMergeReader is an abstraction for reading a merged file.
type FileMergeReader struct {
frs []*FileReader
hdr *tar.Header
tsmr *TagSetMergeReader
}
func newFileMergeReader(frs []*FileReader) *FileMergeReader {
// (bryce) need to handle delete operations.
return &FileMergeReader{
frs: frs,
tsmr: newTagSetMergeReader(frs),
}
}
// Index returns the index for the merged file.
func (fmr *FileMergeReader) Index() *index.Index {
// (bryce) need to merge and compute new hashes when client
// wants the stable hash for the content. Only returning
// returning index with path and size for now.
idx := &index.Index{}
idx.Path = fmr.frs[0].Index().Path
for _, fr := range fmr.frs {
idx.SizeBytes += fr.Index().SizeBytes
}
return idx
}
// Header returns the tar header for the merged file.
func (fmr *FileMergeReader) Header() (*tar.Header, error) {
if fmr.hdr == nil {
// Compute the size of the headers being merged.
var size int64
for _, fr := range fmr.frs {
hdr, err := fr.Header()
if err != nil {
return nil, err
}
size += hdr.Size
}
// Use the header from the highest priority file
// reader and update the size.
// (bryce) might want to copy the header?
var err error
fmr.hdr, err = fmr.frs[len(fmr.frs)-1].Header()
if err != nil {
return nil, err
}
fmr.hdr.Size = size
}
return fmr.hdr, nil
}
// WriteTo writes the merged file to the passed in fileset writer.
func (fmr *FileMergeReader) WriteTo(w *Writer) error {
hdr, err := fmr.Header()
if err != nil {
return err
}
// Write merged header.
if err := w.WriteHeader(hdr); err != nil {
return err
}
// Write merged content.
return fmr.tsmr.WriteTo(w)
}
// Get writes the merged file.
// (bryce) it might be cleaner to check if w is of type *Writer then use WriteTo rather than Get.
func (fmr *FileMergeReader) Get(w io.Writer) error {
hdr, err := fmr.Header()
if err != nil {
return err
}
// Write merged header.
tw := tar.NewWriter(w)
if err := tw.WriteHeader(hdr); err != nil {
return err
}
// Write merged content.
if err := fmr.tsmr.Get(tw); err != nil {
return err
}
return tw.Flush()
}
// TagSetMergeReader returns the tagset merge reader for the file.
// This is how you would get just the data in the file (excludes the tar
// header and padding).
func (fmr *FileMergeReader) TagSetMergeReader() (*TagSetMergeReader, error) {
if _, err := fmr.Header(); err != nil {
return nil, err
}
return fmr.tsmr, nil
}
// TagSetMergeReader is an abstraction for reading the merged tagged data in a merged file.
type TagSetMergeReader struct {
frs []*FileReader
pq *priorityQueue
}
func newTagSetMergeReader(frs []*FileReader) *TagSetMergeReader {
var tagStreams []stream
for _, fr := range frs {
tagStreams = append(tagStreams, &tagStream{fr: fr})
}
return &TagSetMergeReader{
frs: frs,
pq: newPriorityQueue(tagStreams),
}
}
// Iterate iterates over the tag merge readers in the merged tagset.
func (tsmr *TagSetMergeReader) Iterate(f func(*TagMergeReader) error) error {
return tsmr.pq.iterate(func(ss []stream, _ ...string) error {
// Convert generic stream to tag stream.
var tss []*tagStream
for _, s := range ss {
tss = append(tss, s.(*tagStream))
}
// Check if we are at the padding tag, which indicates we are done.
tag, err := tss[0].fr.PeekTag()
if err != nil {
return err
}
if tag.Id == paddingTag {
for _, ts := range tss {
if err := ts.fr.NextTagReader().Get(ioutil.Discard); err != nil {
return err
}
}
return nil
}
// Create tag merge reader and execute callback.
var trs []*chunk.TagReader
for _, ts := range tss {
trs = append(trs, ts.fr.NextTagReader())
}
return f(newTagMergeReader(trs))
})
}
// WriteTo writes the merged tagset to the passed in fileset writer.
func (tsmr *TagSetMergeReader) WriteTo(w *Writer) error {
return tsmr.pq.iterate(func(ss []stream, next ...string) error {
// Convert generic stream to tag stream.
var tss []*tagStream
for _, s := range ss {
tss = append(tss, s.(*tagStream))
}
// Check if we are at the padding tag, which indicates we are done.
tag, err := tss[0].fr.PeekTag()
if err != nil {
return err
}
if tag.Id == paddingTag {
for _, ts := range tss {
if err := ts.fr.NextTagReader().Get(ioutil.Discard); err != nil {
return err
}
}
return nil
}
// Copy single tag stream.
if len(ss) == 1 {
if len(next) == 0 {
next = []string{paddingTag}
}
return tss[0].fr.Iterate(func(dr *chunk.DataReader) error {
return w.CopyTags(dr.BoundReader(next...))
}, next...)
}
// Create tag merge reader and call its WriteTo function.
var trs []*chunk.TagReader
for _, ts := range tss {
trs = append(trs, ts.fr.NextTagReader())
}
return newTagMergeReader(trs).WriteTo(w)
})
}
// Get writes the merged tagset.
func (tsmr *TagSetMergeReader) Get(w io.Writer) error {
return tsmr.Iterate(func(tmr *TagMergeReader) error {
return tmr.Get(w)
})
}
type tagStream struct {
fr *FileReader
tag *chunk.Tag
}
func (ts *tagStream) next() error {
var err error
ts.tag, err = ts.fr.PeekTag()
return err
}
func (ts *tagStream) key() string {
return ts.tag.Id
}
// TagMergeReader is an abstraction for reading a merged tag.
// This abstraction is necessary because a tag in a file can appear
// across multiple filesets.
type TagMergeReader struct {
trs []*chunk.TagReader
}
func newTagMergeReader(trs []*chunk.TagReader) *TagMergeReader {
return &TagMergeReader{trs: trs}
}
// Iterate iterates over the data readers for the tagged data being merged.
func (tmr *TagMergeReader) Iterate(f func(*chunk.DataReader) error) error {
for _, tr := range tmr.trs {
if err := tr.Iterate(f); err != nil {
return err
}
}
return nil
}
// WriteTo writes the merged tagged data to the passed in fileset writer.
func (tmr *TagMergeReader) WriteTo(w *Writer) error {
return tmr.Iterate(func(dr *chunk.DataReader) error {
return w.CopyTags(dr)
})
}
// Get writes the merged tagged data.
func (tmr *TagMergeReader) Get(w io.Writer) error {
return tmr.Iterate(func(dr *chunk.DataReader) error {
return dr.Iterate(func(_ *chunk.Tag, r io.Reader) error {
_, err := io.Copy(w, r)
return err
})
})
}
// ShardFunc is a callback that returns a PathRange for each shard.
type ShardFunc func(*index.PathRange) error
// shard creates shards (path ranges) from the fileset streams being merged.
// A shard is created when the size of the content for a path range is greater than
// the passed in shard threshold.
// For each shard, the callback is called with the path range for the shard.
func shard(mr *MergeReader, shardThreshold int64, f ShardFunc) error {
var size int64
pathRange := &index.PathRange{}
if err := mr.Iterate(func(fmr *FileMergeReader) error {
if pathRange.Lower == "" {
pathRange.Lower = fmr.Index().Path
}
size += fmr.Index().SizeBytes
// A shard is created when we have encountered more than shardThreshold content bytes.
if size >= shardThreshold {
pathRange.Upper = fmr.Index().Path
if err := f(pathRange); err != nil {
return err
}
size = 0
pathRange = &index.PathRange{}
}
return nil
}); err != nil {
return err
}
// Edge case where the last shard is created at the last file.
if pathRange.Lower == "" {
return nil
}
return f(pathRange)
}