Line data Source code
1 : // Copyright 2011 The LevelDB-Go and Pebble Authors. All rights reserved. Use
2 : // of this source code is governed by a BSD-style license that can be found in
3 : // the LICENSE file.
4 :
5 : package sstable
6 :
7 : import (
8 : "cmp"
9 : "context"
10 : "encoding/binary"
11 : "fmt"
12 : "io"
13 : "path/filepath"
14 : "runtime"
15 : "slices"
16 : "time"
17 :
18 : "github.com/cespare/xxhash/v2"
19 : "github.com/cockroachdb/crlib/crtime"
20 : "github.com/cockroachdb/crlib/fifo"
21 : "github.com/cockroachdb/errors"
22 : "github.com/cockroachdb/pebble/internal/base"
23 : "github.com/cockroachdb/pebble/internal/bytealloc"
24 : "github.com/cockroachdb/pebble/internal/cache"
25 : "github.com/cockroachdb/pebble/internal/crc"
26 : "github.com/cockroachdb/pebble/internal/invariants"
27 : "github.com/cockroachdb/pebble/internal/keyspan"
28 : "github.com/cockroachdb/pebble/internal/sstableinternal"
29 : "github.com/cockroachdb/pebble/objstorage"
30 : "github.com/cockroachdb/pebble/objstorage/objstorageprovider"
31 : "github.com/cockroachdb/pebble/objstorage/objstorageprovider/objiotracing"
32 : "github.com/cockroachdb/pebble/sstable/block"
33 : "github.com/cockroachdb/pebble/sstable/colblk"
34 : "github.com/cockroachdb/pebble/sstable/rowblk"
35 : "github.com/cockroachdb/pebble/sstable/valblk"
36 : "github.com/cockroachdb/pebble/vfs"
37 : )
38 :
39 : var errReaderClosed = errors.New("pebble/table: reader is closed")
40 :
41 : type loadBlockResult int8
42 :
43 : const (
44 : loadBlockOK loadBlockResult = iota
45 : // Could be due to error or because no block left to load.
46 : loadBlockFailed
47 : loadBlockIrrelevant
48 : )
49 :
50 : // Reader is a table reader.
51 : type Reader struct {
52 : readable objstorage.Readable
53 :
54 : // The following fields are copied from the ReadOptions.
55 : cacheOpts sstableinternal.CacheOptions
56 : keySchema *colblk.KeySchema
57 : loadBlockSema *fifo.Semaphore
58 : deniedUserProperties map[string]struct{}
59 : filterMetricsTracker *FilterMetricsTracker
60 : logger base.LoggerAndTracer
61 :
62 : Comparer *base.Comparer
63 : Compare Compare
64 : Equal Equal
65 : Split Split
66 :
67 : tableFilter *tableFilterReader
68 :
69 : err error
70 :
71 : indexBH block.Handle
72 : filterBH block.Handle
73 : rangeDelBH block.Handle
74 : rangeKeyBH block.Handle
75 : valueBIH valblk.IndexHandle
76 : propertiesBH block.Handle
77 : metaindexBH block.Handle
78 : footerBH block.Handle
79 :
80 : Properties Properties
81 : tableFormat TableFormat
82 : checksumType block.ChecksumType
83 :
84 : // metaBufferPool is a buffer pool used exclusively when opening a table and
85 : // loading its meta blocks. metaBufferPoolAlloc is used to batch-allocate
86 : // the BufferPool.pool slice as a part of the Reader allocation. It's
87 : // capacity 3 to accommodate the meta block (1), and both the compressed
88 : // properties block (1) and decompressed properties block (1)
89 : // simultaneously.
90 : metaBufferPool block.BufferPool
91 : metaBufferPoolAlloc [3]block.AllocedBuffer
92 : }
93 :
94 : var _ CommonReader = (*Reader)(nil)
95 :
96 : // Close the reader and the underlying objstorage.Readable.
97 1 : func (r *Reader) Close() error {
98 1 : r.cacheOpts.Cache.Unref()
99 1 :
100 1 : if r.readable != nil {
101 1 : r.err = firstError(r.err, r.readable.Close())
102 1 : r.readable = nil
103 1 : }
104 :
105 1 : if r.err != nil {
106 1 : return r.err
107 1 : }
108 : // Make any future calls to Get, NewIter or Close return an error.
109 1 : r.err = errReaderClosed
110 1 : return nil
111 : }
112 :
113 : // NewPointIter returns an iterator for the point keys in the table.
114 : //
115 : // If transform.HideObsoletePoints is set, the callee assumes that filterer
116 : // already includes obsoleteKeyBlockPropertyFilter. The caller can satisfy this
117 : // contract by first calling TryAddBlockPropertyFilterForHideObsoletePoints.
118 : func (r *Reader) NewPointIter(
119 : ctx context.Context,
120 : transforms IterTransforms,
121 : lower, upper []byte,
122 : filterer *BlockPropertiesFilterer,
123 : filterBlockSizeLimit FilterBlockSizeLimit,
124 : stats *base.InternalIteratorStats,
125 : statsAccum IterStatsAccumulator,
126 : rp ReaderProvider,
127 1 : ) (Iterator, error) {
128 1 : return r.newPointIter(
129 1 : ctx, transforms, lower, upper, filterer, filterBlockSizeLimit,
130 1 : stats, statsAccum, rp, nil)
131 1 : }
132 :
133 : // TryAddBlockPropertyFilterForHideObsoletePoints is expected to be called
134 : // before the call to NewPointIter, to get the value of hideObsoletePoints and
135 : // potentially add a block property filter.
136 : func (r *Reader) TryAddBlockPropertyFilterForHideObsoletePoints(
137 : snapshotForHideObsoletePoints base.SeqNum,
138 : fileLargestSeqNum base.SeqNum,
139 : pointKeyFilters []BlockPropertyFilter,
140 1 : ) (hideObsoletePoints bool, filters []BlockPropertyFilter) {
141 1 : hideObsoletePoints = r.tableFormat >= TableFormatPebblev4 &&
142 1 : snapshotForHideObsoletePoints > fileLargestSeqNum
143 1 : if hideObsoletePoints {
144 1 : pointKeyFilters = append(pointKeyFilters, obsoleteKeyBlockPropertyFilter{})
145 1 : }
146 1 : return hideObsoletePoints, pointKeyFilters
147 : }
148 :
149 : func (r *Reader) newPointIter(
150 : ctx context.Context,
151 : transforms IterTransforms,
152 : lower, upper []byte,
153 : filterer *BlockPropertiesFilterer,
154 : filterBlockSizeLimit FilterBlockSizeLimit,
155 : stats *base.InternalIteratorStats,
156 : statsAccum IterStatsAccumulator,
157 : rp ReaderProvider,
158 : vState *virtualState,
159 1 : ) (Iterator, error) {
160 1 : // NB: pebble.tableCache wraps the returned iterator with one which performs
161 1 : // reference counting on the Reader, preventing the Reader from being closed
162 1 : // until the final iterator closes.
163 1 : var res Iterator
164 1 : var err error
165 1 : if r.Properties.IndexType == twoLevelIndex {
166 1 : if r.tableFormat.BlockColumnar() {
167 1 : res, err = newColumnBlockTwoLevelIterator(
168 1 : ctx, r, vState, transforms, lower, upper, filterer, filterBlockSizeLimit,
169 1 : stats, statsAccum, rp, nil /* bufferPool */)
170 1 : } else {
171 1 : res, err = newRowBlockTwoLevelIterator(
172 1 : ctx, r, vState, transforms, lower, upper, filterer, filterBlockSizeLimit,
173 1 : stats, statsAccum, rp, nil /* bufferPool */)
174 1 : }
175 1 : } else {
176 1 : if r.tableFormat.BlockColumnar() {
177 1 : res, err = newColumnBlockSingleLevelIterator(
178 1 : ctx, r, vState, transforms, lower, upper, filterer, filterBlockSizeLimit,
179 1 : stats, statsAccum, rp, nil /* bufferPool */)
180 1 : } else {
181 1 : res, err = newRowBlockSingleLevelIterator(
182 1 : ctx, r, vState, transforms, lower, upper, filterer, filterBlockSizeLimit,
183 1 : stats, statsAccum, rp, nil /* bufferPool */)
184 1 : }
185 : }
186 1 : if err != nil {
187 1 : // Note: we don't want to return res here - it will be a nil
188 1 : // single/twoLevelIterator, not a nil Iterator.
189 1 : return nil, err
190 1 : }
191 1 : return res, nil
192 : }
193 :
194 : // NewIter returns an iterator for the point keys in the table. It is a
195 : // simplified version of NewPointIter and should only be used for tests and
196 : // tooling.
197 : //
198 : // NewIter must only be used when the Reader is guaranteed to outlive any
199 : // LazyValues returned from the iter.
200 1 : func (r *Reader) NewIter(transforms IterTransforms, lower, upper []byte) (Iterator, error) {
201 1 : // TODO(radu): we should probably not use bloom filters in this case, as there
202 1 : // likely isn't a cache set up.
203 1 : return r.NewPointIter(
204 1 : context.TODO(), transforms, lower, upper, nil, AlwaysUseFilterBlock,
205 1 : nil /* stats */, nil /* statsAccum */, MakeTrivialReaderProvider(r))
206 1 : }
207 :
208 : // NewCompactionIter returns an iterator similar to NewIter but it also increments
209 : // the number of bytes iterated. If an error occurs, NewCompactionIter cleans up
210 : // after itself and returns a nil iterator.
211 : func (r *Reader) NewCompactionIter(
212 : transforms IterTransforms,
213 : statsAccum IterStatsAccumulator,
214 : rp ReaderProvider,
215 : bufferPool *block.BufferPool,
216 1 : ) (Iterator, error) {
217 1 : return r.newCompactionIter(transforms, statsAccum, rp, nil, bufferPool)
218 1 : }
219 :
220 : func (r *Reader) newCompactionIter(
221 : transforms IterTransforms,
222 : statsAccum IterStatsAccumulator,
223 : rp ReaderProvider,
224 : vState *virtualState,
225 : bufferPool *block.BufferPool,
226 1 : ) (Iterator, error) {
227 1 : if vState != nil && vState.isSharedIngested {
228 1 : transforms.HideObsoletePoints = true
229 1 : }
230 1 : if r.Properties.IndexType == twoLevelIndex {
231 1 : if !r.tableFormat.BlockColumnar() {
232 1 : i, err := newRowBlockTwoLevelIterator(
233 1 : context.Background(),
234 1 : r, vState, transforms, nil /* lower */, nil /* upper */, nil,
235 1 : NeverUseFilterBlock, nil /* stats */, statsAccum, rp, bufferPool)
236 1 : if err != nil {
237 0 : return nil, err
238 0 : }
239 1 : i.SetupForCompaction()
240 1 : return i, nil
241 : }
242 1 : i, err := newColumnBlockTwoLevelIterator(
243 1 : context.Background(),
244 1 : r, vState, transforms, nil /* lower */, nil /* upper */, nil,
245 1 : NeverUseFilterBlock, nil /* stats */, statsAccum, rp, bufferPool)
246 1 : if err != nil {
247 0 : return nil, err
248 0 : }
249 1 : i.SetupForCompaction()
250 1 : return i, nil
251 : }
252 1 : if !r.tableFormat.BlockColumnar() {
253 1 : i, err := newRowBlockSingleLevelIterator(
254 1 : context.Background(), r, vState, transforms, nil /* lower */, nil, /* upper */
255 1 : nil, NeverUseFilterBlock, nil /* stats */, statsAccum, rp, bufferPool)
256 1 : if err != nil {
257 0 : return nil, err
258 0 : }
259 1 : i.SetupForCompaction()
260 1 : return i, nil
261 : }
262 1 : i, err := newColumnBlockSingleLevelIterator(
263 1 : context.Background(), r, vState, transforms, nil /* lower */, nil, /* upper */
264 1 : nil, NeverUseFilterBlock, nil /* stats */, statsAccum, rp, bufferPool)
265 1 : if err != nil {
266 0 : return nil, err
267 0 : }
268 1 : i.SetupForCompaction()
269 1 : return i, nil
270 : }
271 :
272 : // NewRawRangeDelIter returns an internal iterator for the contents of the
273 : // range-del block for the table. Returns nil if the table does not contain
274 : // any range deletions.
275 : func (r *Reader) NewRawRangeDelIter(
276 : ctx context.Context, transforms FragmentIterTransforms,
277 1 : ) (iter keyspan.FragmentIterator, err error) {
278 1 : if r.rangeDelBH.Length == 0 {
279 1 : return nil, nil
280 1 : }
281 : // TODO(radu): plumb stats here.
282 1 : h, err := r.readRangeDelBlock(ctx, noEnv, noReadHandle, r.rangeDelBH)
283 1 : if err != nil {
284 1 : return nil, err
285 1 : }
286 1 : if r.tableFormat.BlockColumnar() {
287 1 : iter = colblk.NewKeyspanIter(r.Compare, h, transforms)
288 1 : } else {
289 1 : iter, err = rowblk.NewFragmentIter(r.cacheOpts.FileNum, r.Compare, r.Comparer.CompareRangeSuffixes, r.Split, h, transforms)
290 1 : if err != nil {
291 0 : return nil, err
292 0 : }
293 : }
294 1 : return keyspan.MaybeAssert(iter, r.Compare), nil
295 : }
296 :
297 : // NewRawRangeKeyIter returns an internal iterator for the contents of the
298 : // range-key block for the table. Returns nil if the table does not contain any
299 : // range keys.
300 : func (r *Reader) NewRawRangeKeyIter(
301 : ctx context.Context, transforms FragmentIterTransforms,
302 1 : ) (iter keyspan.FragmentIterator, err error) {
303 1 : if r.rangeKeyBH.Length == 0 {
304 1 : return nil, nil
305 1 : }
306 : // TODO(radu): plumb stats here.
307 1 : h, err := r.readRangeKeyBlock(ctx, noEnv, noReadHandle, r.rangeKeyBH)
308 1 : if err != nil {
309 1 : return nil, err
310 1 : }
311 1 : if r.tableFormat.BlockColumnar() {
312 1 : iter = colblk.NewKeyspanIter(r.Compare, h, transforms)
313 1 : } else {
314 1 : iter, err = rowblk.NewFragmentIter(r.cacheOpts.FileNum, r.Compare, r.Comparer.CompareRangeSuffixes, r.Split, h, transforms)
315 1 : if err != nil {
316 0 : return nil, err
317 0 : }
318 : }
319 1 : return keyspan.MaybeAssert(iter, r.Compare), nil
320 : }
321 :
322 : // readBlockEnv contains arguments used when reading a block which apply to all
323 : // the block reads performed by a higher-level operation.
324 : type readBlockEnv struct {
325 : // stats and iterStats are slightly different. stats is a shared struct
326 : // supplied from the outside, and represents stats for the whole iterator
327 : // tree and can be reset from the outside (e.g. when the pebble.Iterator is
328 : // being reused). It is currently only provided when the iterator tree is
329 : // rooted at pebble.Iterator. iterStats contains an sstable iterator's
330 : // private stats that are reported to a CategoryStatsCollector when this
331 : // iterator is closed. In the important code paths, the CategoryStatsCollector
332 : // is managed by the tableCacheContainer.
333 : Stats *base.InternalIteratorStats
334 : IterStats *iterStatsAccumulator
335 :
336 : // BufferPool is not-nil if we read blocks into a buffer pool and not into the
337 : // cache. This is used during compactions.
338 : BufferPool *block.BufferPool
339 : }
340 :
341 : // BlockServedFromCache updates the stats when a block was found in the cache.
342 1 : func (env *readBlockEnv) BlockServedFromCache(blockLength uint64) {
343 1 : if env.Stats != nil {
344 1 : env.Stats.BlockBytes += blockLength
345 1 : env.Stats.BlockBytesInCache += blockLength
346 1 : }
347 1 : if env.IterStats != nil {
348 1 : env.IterStats.reportStats(blockLength, blockLength, 0)
349 1 : }
350 : }
351 :
352 : // BlockRead updates the stats when a block had to be read.
353 1 : func (env *readBlockEnv) BlockRead(blockLength uint64, readDuration time.Duration) {
354 1 : if env.Stats != nil {
355 1 : env.Stats.BlockBytes += blockLength
356 1 : env.Stats.BlockReadDuration += readDuration
357 1 : }
358 1 : if env.IterStats != nil {
359 1 : env.IterStats.reportStats(blockLength, 0, readDuration)
360 1 : }
361 : }
362 :
363 : // noEnv is the empty readBlockEnv which reports no stats and does not use a
364 : // buffer pool.
365 : var noEnv = readBlockEnv{}
366 :
367 : // noReadHandle is used when we don't want to pass a ReadHandle to one of the
368 : // read block methods.
369 : var noReadHandle objstorage.ReadHandle = nil
370 :
371 1 : var noInitBlockMetadataFn = func(*block.Metadata, []byte) error { return nil }
372 :
373 : // readMetaindexBlock reads the metaindex block.
374 : func (r *Reader) readMetaindexBlock(
375 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle,
376 1 : ) (block.BufferHandle, error) {
377 1 : ctx = objiotracing.WithBlockType(ctx, objiotracing.MetadataBlock)
378 1 : return r.readBlockInternal(ctx, env, readHandle, r.metaindexBH, noInitBlockMetadataFn)
379 1 : }
380 :
381 : // readTopLevelIndexBlock reads the top-level index block.
382 : func (r *Reader) readTopLevelIndexBlock(
383 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle,
384 1 : ) (block.BufferHandle, error) {
385 1 : return r.readIndexBlock(ctx, env, readHandle, r.indexBH)
386 1 : }
387 :
388 : // readIndexBlock reads a top-level or second-level index block.
389 : func (r *Reader) readIndexBlock(
390 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle, bh block.Handle,
391 1 : ) (block.BufferHandle, error) {
392 1 : ctx = objiotracing.WithBlockType(ctx, objiotracing.MetadataBlock)
393 1 : return r.readBlockInternal(ctx, env, readHandle, bh, r.initIndexBlockMetadata)
394 1 : }
395 :
396 : // initIndexBlockMetadata initializes the Metadata for a data block. This will
397 : // later be used (and reused) when reading from the block.
398 1 : func (r *Reader) initIndexBlockMetadata(metadata *block.Metadata, data []byte) error {
399 1 : if r.tableFormat.BlockColumnar() {
400 1 : return colblk.InitIndexBlockMetadata(metadata, data)
401 1 : }
402 1 : return nil
403 : }
404 :
405 : func (r *Reader) readDataBlock(
406 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle, bh block.Handle,
407 1 : ) (block.BufferHandle, error) {
408 1 : ctx = objiotracing.WithBlockType(ctx, objiotracing.DataBlock)
409 1 : return r.readBlockInternal(ctx, env, readHandle, bh, r.initDataBlockMetadata)
410 1 : }
411 :
412 : // initDataBlockMetadata initializes the Metadata for a data block. This will
413 : // later be used (and reused) when reading from the block.
414 1 : func (r *Reader) initDataBlockMetadata(metadata *block.Metadata, data []byte) error {
415 1 : if r.tableFormat.BlockColumnar() {
416 1 : return colblk.InitDataBlockMetadata(r.keySchema, metadata, data)
417 1 : }
418 1 : return nil
419 : }
420 :
421 : func (r *Reader) readFilterBlock(
422 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle, bh block.Handle,
423 1 : ) (block.BufferHandle, error) {
424 1 : ctx = objiotracing.WithBlockType(ctx, objiotracing.FilterBlock)
425 1 : return r.readBlockInternal(ctx, env, readHandle, bh, noInitBlockMetadataFn)
426 1 : }
427 :
428 : func (r *Reader) readRangeDelBlock(
429 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle, bh block.Handle,
430 1 : ) (block.BufferHandle, error) {
431 1 : ctx = objiotracing.WithBlockType(ctx, objiotracing.MetadataBlock)
432 1 : return r.readBlockInternal(ctx, env, readHandle, bh, r.initKeyspanBlockMetadata)
433 1 : }
434 :
435 : func (r *Reader) readRangeKeyBlock(
436 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle, bh block.Handle,
437 1 : ) (block.BufferHandle, error) {
438 1 : ctx = objiotracing.WithBlockType(ctx, objiotracing.MetadataBlock)
439 1 : return r.readBlockInternal(ctx, env, readHandle, bh, r.initKeyspanBlockMetadata)
440 1 : }
441 :
442 : // initKeyspanBlockMetadata initializes the Metadata for a rangedel or range key
443 : // block. This will later be used (and reused) when reading from the block.
444 1 : func (r *Reader) initKeyspanBlockMetadata(metadata *block.Metadata, data []byte) error {
445 1 : if r.tableFormat.BlockColumnar() {
446 1 : return colblk.InitKeyspanBlockMetadata(metadata, data)
447 1 : }
448 1 : return nil
449 : }
450 :
451 : func (r *Reader) readValueBlock(
452 : ctx context.Context, env readBlockEnv, readHandle objstorage.ReadHandle, bh block.Handle,
453 1 : ) (block.BufferHandle, error) {
454 1 : ctx = objiotracing.WithBlockType(ctx, objiotracing.ValueBlock)
455 1 : return r.readBlockInternal(ctx, env, readHandle, bh, noInitBlockMetadataFn)
456 1 : }
457 :
458 : func checkChecksum(
459 : checksumType block.ChecksumType, b []byte, bh block.Handle, fileNum base.DiskFileNum,
460 1 : ) error {
461 1 : expectedChecksum := binary.LittleEndian.Uint32(b[bh.Length+1:])
462 1 : var computedChecksum uint32
463 1 : switch checksumType {
464 1 : case block.ChecksumTypeCRC32c:
465 1 : computedChecksum = crc.New(b[:bh.Length+1]).Value()
466 1 : case block.ChecksumTypeXXHash64:
467 1 : computedChecksum = uint32(xxhash.Sum64(b[:bh.Length+1]))
468 0 : default:
469 0 : return errors.Errorf("unsupported checksum type: %d", checksumType)
470 : }
471 :
472 1 : if expectedChecksum != computedChecksum {
473 1 : return base.CorruptionErrorf(
474 1 : "pebble/table: invalid table %s (checksum mismatch at %d/%d)",
475 1 : fileNum, errors.Safe(bh.Offset), errors.Safe(bh.Length))
476 1 : }
477 1 : return nil
478 : }
479 :
480 : // DeterministicReadBlockDurationForTesting is for tests that want a
481 : // deterministic value of the time to read a block (that is not in the cache).
482 : // The return value is a function that must be called before the test exits.
483 1 : func DeterministicReadBlockDurationForTesting() func() {
484 1 : drbdForTesting := deterministicReadBlockDurationForTesting
485 1 : deterministicReadBlockDurationForTesting = true
486 1 : return func() {
487 1 : deterministicReadBlockDurationForTesting = drbdForTesting
488 1 : }
489 : }
490 :
491 : var deterministicReadBlockDurationForTesting = false
492 :
493 : // readBlockInternal should not be used directly; one of the read*Block methods
494 : // should be used instead.
495 : func (r *Reader) readBlockInternal(
496 : ctx context.Context,
497 : env readBlockEnv,
498 : readHandle objstorage.ReadHandle,
499 : bh block.Handle,
500 : initBlockMetadataFn func(*block.Metadata, []byte) error,
501 1 : ) (handle block.BufferHandle, _ error) {
502 1 : if h := r.cacheOpts.Cache.Get(r.cacheOpts.CacheID, r.cacheOpts.FileNum, bh.Offset); h.Valid() {
503 1 : // Cache hit.
504 1 : if readHandle != nil {
505 1 : readHandle.RecordCacheHit(ctx, int64(bh.Offset), int64(bh.Length+block.TrailerLen))
506 1 : }
507 1 : env.BlockServedFromCache(bh.Length)
508 1 : return block.CacheBufferHandle(h), nil
509 : }
510 :
511 : // Cache miss.
512 :
513 1 : if sema := r.loadBlockSema; sema != nil {
514 1 : if err := sema.Acquire(ctx, 1); err != nil {
515 0 : // An error here can only come from the context.
516 0 : return block.BufferHandle{}, err
517 0 : }
518 1 : defer sema.Release(1)
519 : }
520 :
521 1 : compressed := block.Alloc(int(bh.Length+block.TrailerLen), env.BufferPool)
522 1 : readStopwatch := makeStopwatch()
523 1 : var err error
524 1 : if readHandle != nil {
525 1 : err = readHandle.ReadAt(ctx, compressed.BlockData(), int64(bh.Offset))
526 1 : } else {
527 1 : err = r.readable.ReadAt(ctx, compressed.BlockData(), int64(bh.Offset))
528 1 : }
529 1 : readDuration := readStopwatch.stop()
530 1 : // Call IsTracingEnabled to avoid the allocations of boxing integers into an
531 1 : // interface{}, unless necessary.
532 1 : if readDuration >= slowReadTracingThreshold && r.logger.IsTracingEnabled(ctx) {
533 1 : _, file1, line1, _ := runtime.Caller(1)
534 1 : _, file2, line2, _ := runtime.Caller(2)
535 1 : r.logger.Eventf(ctx, "reading block of %d bytes took %s (fileNum=%s; %s/%s:%d -> %s/%s:%d)",
536 1 : int(bh.Length+block.TrailerLen), readDuration.String(),
537 1 : r.cacheOpts.FileNum,
538 1 : filepath.Base(filepath.Dir(file2)), filepath.Base(file2), line2,
539 1 : filepath.Base(filepath.Dir(file1)), filepath.Base(file1), line1)
540 1 : }
541 1 : if err != nil {
542 1 : compressed.Release()
543 1 : return block.BufferHandle{}, err
544 1 : }
545 1 : env.BlockRead(bh.Length, readDuration)
546 1 : if err := checkChecksum(r.checksumType, compressed.BlockData(), bh, r.cacheOpts.FileNum); err != nil {
547 1 : compressed.Release()
548 1 : return block.BufferHandle{}, err
549 1 : }
550 :
551 1 : typ := block.CompressionIndicator(compressed.BlockData()[bh.Length])
552 1 : compressed.Truncate(int(bh.Length))
553 1 :
554 1 : var decompressed block.Value
555 1 : if typ == block.NoCompressionIndicator {
556 1 : decompressed = compressed
557 1 : } else {
558 1 : // Decode the length of the decompressed value.
559 1 : decodedLen, prefixLen, err := block.DecompressedLen(typ, compressed.BlockData())
560 1 : if err != nil {
561 0 : compressed.Release()
562 0 : return block.BufferHandle{}, err
563 0 : }
564 :
565 1 : decompressed = block.Alloc(decodedLen, env.BufferPool)
566 1 : err = block.DecompressInto(typ, compressed.BlockData()[prefixLen:], decompressed.BlockData())
567 1 : compressed.Release()
568 1 : if err != nil {
569 0 : decompressed.Release()
570 0 : return block.BufferHandle{}, err
571 0 : }
572 : }
573 1 : if err := initBlockMetadataFn(decompressed.BlockMetadata(), decompressed.BlockData()); err != nil {
574 0 : decompressed.Release()
575 0 : return block.BufferHandle{}, err
576 0 : }
577 1 : h := decompressed.MakeHandle(r.cacheOpts.Cache, r.cacheOpts.CacheID, r.cacheOpts.FileNum, bh.Offset)
578 1 : return h, nil
579 : }
580 :
581 : func (r *Reader) readMetaindex(
582 : ctx context.Context, readHandle objstorage.ReadHandle, filters map[string]FilterPolicy,
583 1 : ) error {
584 1 : // We use a BufferPool when reading metaindex blocks in order to avoid
585 1 : // populating the block cache with these blocks. In heavy-write workloads,
586 1 : // especially with high compaction concurrency, new tables may be created
587 1 : // frequently. Populating the block cache with these metaindex blocks adds
588 1 : // additional contention on the block cache mutexes (see #1997).
589 1 : // Additionally, these blocks are exceedingly unlikely to be read again
590 1 : // while they're still in the block cache except in misconfigurations with
591 1 : // excessive sstables counts or a table cache that's far too small.
592 1 : r.metaBufferPool.InitPreallocated(r.metaBufferPoolAlloc[:0])
593 1 : // When we're finished, release the buffers we've allocated back to memory
594 1 : // allocator. We don't expect to use metaBufferPool again.
595 1 : defer r.metaBufferPool.Release()
596 1 : metaEnv := readBlockEnv{
597 1 : BufferPool: &r.metaBufferPool,
598 1 : }
599 1 :
600 1 : b, err := r.readMetaindexBlock(ctx, metaEnv, readHandle)
601 1 : if err != nil {
602 1 : return err
603 1 : }
604 1 : data := b.BlockData()
605 1 : defer b.Release()
606 1 :
607 1 : if uint64(len(data)) != r.metaindexBH.Length {
608 0 : return base.CorruptionErrorf("pebble/table: unexpected metaindex block size: %d vs %d",
609 0 : errors.Safe(len(data)), errors.Safe(r.metaindexBH.Length))
610 0 : }
611 :
612 1 : var meta map[string]block.Handle
613 1 : meta, r.valueBIH, err = decodeMetaindex(data)
614 1 : if err != nil {
615 0 : return err
616 0 : }
617 :
618 1 : if bh, ok := meta[metaPropertiesName]; ok {
619 1 : b, err = r.readBlockInternal(ctx, metaEnv, readHandle, bh, noInitBlockMetadataFn)
620 1 : if err != nil {
621 1 : return err
622 1 : }
623 1 : r.propertiesBH = bh
624 1 : err := r.Properties.load(b.BlockData(), r.deniedUserProperties)
625 1 : b.Release()
626 1 : if err != nil {
627 0 : return err
628 0 : }
629 : }
630 :
631 1 : if bh, ok := meta[metaRangeDelV2Name]; ok {
632 1 : r.rangeDelBH = bh
633 1 : } else if _, ok := meta[metaRangeDelV1Name]; ok {
634 0 : // This version of Pebble requires a format major version at least as
635 0 : // high as FormatFlushableIngest (see pebble.FormatMinSupported). In
636 0 : // this format major verison, we have a guarantee that we've compacted
637 0 : // away all RocksDB sstables. It should not be possible to encounter an
638 0 : // sstable with a v1 range deletion block but not a v2 range deletion
639 0 : // block.
640 0 : err := errors.Newf("pebble/table: unexpected range-del block type: %s", metaRangeDelV1Name)
641 0 : return errors.Mark(err, base.ErrCorruption)
642 0 : }
643 :
644 1 : if bh, ok := meta[metaRangeKeyName]; ok {
645 1 : r.rangeKeyBH = bh
646 1 : }
647 :
648 1 : for name, fp := range filters {
649 1 : if bh, ok := meta["fullfilter."+name]; ok {
650 1 : r.filterBH = bh
651 1 : r.tableFilter = newTableFilterReader(fp, r.filterMetricsTracker)
652 1 : break
653 : }
654 : }
655 1 : return nil
656 : }
657 :
658 : // Layout returns the layout (block organization) for an sstable.
659 1 : func (r *Reader) Layout() (*Layout, error) {
660 1 : if r.err != nil {
661 0 : return nil, r.err
662 0 : }
663 :
664 1 : l := &Layout{
665 1 : Data: make([]block.HandleWithProperties, 0, r.Properties.NumDataBlocks),
666 1 : RangeDel: r.rangeDelBH,
667 1 : RangeKey: r.rangeKeyBH,
668 1 : ValueIndex: r.valueBIH.Handle,
669 1 : Properties: r.propertiesBH,
670 1 : MetaIndex: r.metaindexBH,
671 1 : Footer: r.footerBH,
672 1 : Format: r.tableFormat,
673 1 : }
674 1 : if r.filterBH.Length > 0 {
675 1 : l.Filter = []NamedBlockHandle{{Name: "fullfilter." + r.tableFilter.policy.Name(), Handle: r.filterBH}}
676 1 : }
677 1 : ctx := context.TODO()
678 1 :
679 1 : indexH, err := r.readTopLevelIndexBlock(ctx, noEnv, noReadHandle)
680 1 : if err != nil {
681 1 : return nil, err
682 1 : }
683 1 : defer indexH.Release()
684 1 :
685 1 : var alloc bytealloc.A
686 1 :
687 1 : if r.Properties.IndexPartitions == 0 {
688 1 : l.Index = append(l.Index, r.indexBH)
689 1 : iter := r.tableFormat.newIndexIter()
690 1 : err := iter.Init(r.Compare, r.Split, indexH.BlockData(), NoTransforms)
691 1 : if err != nil {
692 0 : return nil, errors.Wrap(err, "reading index block")
693 0 : }
694 1 : for valid := iter.First(); valid; valid = iter.Next() {
695 1 : dataBH, err := iter.BlockHandleWithProperties()
696 1 : if err != nil {
697 0 : return nil, errCorruptIndexEntry(err)
698 0 : }
699 1 : if len(dataBH.Props) > 0 {
700 1 : alloc, dataBH.Props = alloc.Copy(dataBH.Props)
701 1 : }
702 1 : l.Data = append(l.Data, dataBH)
703 : }
704 1 : } else {
705 1 : l.TopIndex = r.indexBH
706 1 : topIter := r.tableFormat.newIndexIter()
707 1 : err := topIter.Init(r.Compare, r.Split, indexH.BlockData(), NoTransforms)
708 1 : if err != nil {
709 0 : return nil, errors.Wrap(err, "reading index block")
710 0 : }
711 1 : iter := r.tableFormat.newIndexIter()
712 1 : for valid := topIter.First(); valid; valid = topIter.Next() {
713 1 : indexBH, err := topIter.BlockHandleWithProperties()
714 1 : if err != nil {
715 0 : return nil, errCorruptIndexEntry(err)
716 0 : }
717 1 : l.Index = append(l.Index, indexBH.Handle)
718 1 :
719 1 : subIndex, err := r.readIndexBlock(ctx, noEnv, noReadHandle, indexBH.Handle)
720 1 : if err != nil {
721 1 : return nil, err
722 1 : }
723 1 : err = func() error {
724 1 : defer subIndex.Release()
725 1 : // TODO(msbutler): figure out how to pass virtualState to layout call.
726 1 : if err := iter.Init(r.Compare, r.Split, subIndex.BlockData(), NoTransforms); err != nil {
727 0 : return err
728 0 : }
729 1 : for valid := iter.First(); valid; valid = iter.Next() {
730 1 : dataBH, err := iter.BlockHandleWithProperties()
731 1 : if err != nil {
732 0 : return errCorruptIndexEntry(err)
733 0 : }
734 1 : if len(dataBH.Props) > 0 {
735 1 : alloc, dataBH.Props = alloc.Copy(dataBH.Props)
736 1 : }
737 1 : l.Data = append(l.Data, dataBH)
738 : }
739 1 : return nil
740 : }()
741 1 : if err != nil {
742 0 : return nil, err
743 0 : }
744 : }
745 : }
746 1 : if r.valueBIH.Handle.Length != 0 {
747 1 : vbiH, err := r.readValueBlock(context.Background(), noEnv, noReadHandle, r.valueBIH.Handle)
748 1 : if err != nil {
749 0 : return nil, err
750 0 : }
751 1 : defer vbiH.Release()
752 1 : l.ValueBlock, err = decodeValueBlockIndex(vbiH.BlockData(), r.valueBIH)
753 1 : if err != nil {
754 0 : return nil, err
755 0 : }
756 : }
757 :
758 1 : return l, nil
759 : }
760 :
761 : // ValidateBlockChecksums validates the checksums for each block in the SSTable.
762 1 : func (r *Reader) ValidateBlockChecksums() error {
763 1 : // Pre-compute the BlockHandles for the underlying file.
764 1 : l, err := r.Layout()
765 1 : if err != nil {
766 1 : return err
767 1 : }
768 :
769 1 : type blk struct {
770 1 : bh block.Handle
771 1 : readFn func(context.Context, readBlockEnv, objstorage.ReadHandle, block.Handle) (block.BufferHandle, error)
772 1 : }
773 1 : // Construct the set of blocks to check. Note that the footer is not checked
774 1 : // as it is not a block with a checksum.
775 1 : blocks := make([]blk, 0, len(l.Data)+6)
776 1 : for i := range l.Data {
777 1 : blocks = append(blocks, blk{
778 1 : bh: l.Data[i].Handle,
779 1 : readFn: r.readDataBlock,
780 1 : })
781 1 : }
782 1 : for _, h := range l.Index {
783 1 : blocks = append(blocks, blk{
784 1 : bh: h,
785 1 : readFn: r.readIndexBlock,
786 1 : })
787 1 : }
788 1 : blocks = append(blocks, blk{
789 1 : bh: l.TopIndex,
790 1 : readFn: r.readIndexBlock,
791 1 : })
792 1 : for _, bh := range l.Filter {
793 1 : blocks = append(blocks, blk{
794 1 : bh: bh.Handle,
795 1 : readFn: r.readFilterBlock,
796 1 : })
797 1 : }
798 1 : blocks = append(blocks, blk{
799 1 : bh: l.RangeDel,
800 1 : readFn: r.readRangeDelBlock,
801 1 : })
802 1 : blocks = append(blocks, blk{
803 1 : bh: l.RangeKey,
804 1 : readFn: r.readRangeKeyBlock,
805 1 : })
806 1 : readNoInit := func(ctx context.Context, env readBlockEnv, rh objstorage.ReadHandle, bh block.Handle) (block.BufferHandle, error) {
807 1 : return r.readBlockInternal(ctx, env, rh, bh, noInitBlockMetadataFn)
808 1 : }
809 1 : blocks = append(blocks, blk{
810 1 : bh: l.Properties,
811 1 : readFn: readNoInit,
812 1 : })
813 1 : blocks = append(blocks, blk{
814 1 : bh: l.MetaIndex,
815 1 : readFn: readNoInit,
816 1 : })
817 1 :
818 1 : // Sorting by offset ensures we are performing a sequential scan of the
819 1 : // file.
820 1 : slices.SortFunc(blocks, func(a, b blk) int {
821 1 : return cmp.Compare(a.bh.Offset, b.bh.Offset)
822 1 : })
823 :
824 1 : ctx := context.Background()
825 1 : for _, b := range blocks {
826 1 : // Certain blocks may not be present, in which case we skip them.
827 1 : if b.bh.Length == 0 {
828 1 : continue
829 : }
830 1 : h, err := b.readFn(ctx, noEnv, noReadHandle, b.bh)
831 1 : if err != nil {
832 1 : return err
833 1 : }
834 1 : h.Release()
835 : }
836 :
837 1 : return nil
838 : }
839 :
840 : // CommonProperties implemented the CommonReader interface.
841 1 : func (r *Reader) CommonProperties() *CommonProperties {
842 1 : return &r.Properties.CommonProperties
843 1 : }
844 :
845 : // EstimateDiskUsage returns the total size of data blocks overlapping the range
846 : // `[start, end]`. Even if a data block partially overlaps, or we cannot
847 : // determine overlap due to abbreviated index keys, the full data block size is
848 : // included in the estimation.
849 : //
850 : // This function does not account for any metablock space usage. Assumes there
851 : // is at least partial overlap, i.e., `[start, end]` falls neither completely
852 : // before nor completely after the file's range.
853 : //
854 : // Only blocks containing point keys are considered. Range deletion and range
855 : // key blocks are not considered.
856 : //
857 : // TODO(ajkr): account for metablock space usage. Perhaps look at the fraction of
858 : // data blocks overlapped and add that same fraction of the metadata blocks to the
859 : // estimate.
860 1 : func (r *Reader) EstimateDiskUsage(start, end []byte) (uint64, error) {
861 1 : if !r.tableFormat.BlockColumnar() {
862 1 : return estimateDiskUsage[rowblk.IndexIter, *rowblk.IndexIter](r, start, end)
863 1 : }
864 1 : return estimateDiskUsage[colblk.IndexIter, *colblk.IndexIter](r, start, end)
865 : }
866 :
867 : func estimateDiskUsage[I any, PI indexBlockIterator[I]](
868 : r *Reader, start, end []byte,
869 1 : ) (uint64, error) {
870 1 : if r.err != nil {
871 0 : return 0, r.err
872 0 : }
873 1 : ctx := context.TODO()
874 1 :
875 1 : indexH, err := r.readTopLevelIndexBlock(ctx, noEnv, noReadHandle)
876 1 : if err != nil {
877 1 : return 0, err
878 1 : }
879 : // We are using InitHandle below but we never Close those iterators, which
880 : // allows us to release the index handle ourselves.
881 : // TODO(radu): clean this up.
882 1 : defer indexH.Release()
883 1 :
884 1 : // Iterators over the bottom-level index blocks containing start and end.
885 1 : // These may be different in case of partitioned index but will both point
886 1 : // to the same blockIter over the single index in the unpartitioned case.
887 1 : var startIdxIter, endIdxIter PI
888 1 : if r.Properties.IndexPartitions == 0 {
889 1 : startIdxIter = new(I)
890 1 : if err := startIdxIter.InitHandle(r.Compare, r.Split, indexH, NoTransforms); err != nil {
891 0 : return 0, err
892 0 : }
893 1 : endIdxIter = startIdxIter
894 1 : } else {
895 1 : var topIter PI = new(I)
896 1 : if err := topIter.InitHandle(r.Compare, r.Split, indexH, NoTransforms); err != nil {
897 0 : return 0, err
898 0 : }
899 1 : if !topIter.SeekGE(start) {
900 0 : // The range falls completely after this file.
901 0 : return 0, nil
902 0 : }
903 1 : startIndexBH, err := topIter.BlockHandleWithProperties()
904 1 : if err != nil {
905 0 : return 0, errCorruptIndexEntry(err)
906 0 : }
907 1 : startIdxBlock, err := r.readIndexBlock(ctx, noEnv, noReadHandle, startIndexBH.Handle)
908 1 : if err != nil {
909 1 : return 0, err
910 1 : }
911 1 : defer startIdxBlock.Release()
912 1 : startIdxIter = new(I)
913 1 : err = startIdxIter.InitHandle(r.Compare, r.Split, startIdxBlock, NoTransforms)
914 1 : if err != nil {
915 0 : return 0, err
916 0 : }
917 :
918 1 : if topIter.SeekGE(end) {
919 1 : endIndexBH, err := topIter.BlockHandleWithProperties()
920 1 : if err != nil {
921 0 : return 0, errCorruptIndexEntry(err)
922 0 : }
923 1 : endIdxBlock, err := r.readIndexBlock(ctx, noEnv, noReadHandle, endIndexBH.Handle)
924 1 : if err != nil {
925 1 : return 0, err
926 1 : }
927 1 : defer endIdxBlock.Release()
928 1 : endIdxIter = new(I)
929 1 : err = endIdxIter.InitHandle(r.Compare, r.Split, endIdxBlock, NoTransforms)
930 1 : if err != nil {
931 0 : return 0, err
932 0 : }
933 : }
934 : }
935 : // startIdxIter should not be nil at this point, while endIdxIter can be if the
936 : // range spans past the end of the file.
937 :
938 1 : if !startIdxIter.SeekGE(start) {
939 1 : // The range falls completely after this file.
940 1 : return 0, nil
941 1 : }
942 1 : startBH, err := startIdxIter.BlockHandleWithProperties()
943 1 : if err != nil {
944 0 : return 0, errCorruptIndexEntry(err)
945 0 : }
946 :
947 1 : includeInterpolatedValueBlocksSize := func(dataBlockSize uint64) uint64 {
948 1 : // INVARIANT: r.Properties.DataSize > 0 since startIdxIter is not nil.
949 1 : // Linearly interpolate what is stored in value blocks.
950 1 : //
951 1 : // TODO(sumeer): if we need more accuracy, without loading any data blocks
952 1 : // (which contain the value handles, and which may also be insufficient if
953 1 : // the values are in separate files), we will need to accumulate the
954 1 : // logical size of the key-value pairs and store the cumulative value for
955 1 : // each data block in the index block entry. This increases the size of
956 1 : // the BlockHandle, so wait until this becomes necessary.
957 1 : return dataBlockSize +
958 1 : uint64((float64(dataBlockSize)/float64(r.Properties.DataSize))*
959 1 : float64(r.Properties.ValueBlocksSize))
960 1 : }
961 1 : if endIdxIter == nil {
962 0 : // The range spans beyond this file. Include data blocks through the last.
963 0 : return includeInterpolatedValueBlocksSize(r.Properties.DataSize - startBH.Offset), nil
964 0 : }
965 1 : if !endIdxIter.SeekGE(end) {
966 1 : // The range spans beyond this file. Include data blocks through the last.
967 1 : return includeInterpolatedValueBlocksSize(r.Properties.DataSize - startBH.Offset), nil
968 1 : }
969 1 : endBH, err := endIdxIter.BlockHandleWithProperties()
970 1 : if err != nil {
971 0 : return 0, errCorruptIndexEntry(err)
972 0 : }
973 1 : return includeInterpolatedValueBlocksSize(
974 1 : endBH.Offset + endBH.Length + block.TrailerLen - startBH.Offset), nil
975 : }
976 :
977 : // TableFormat returns the format version for the table.
978 1 : func (r *Reader) TableFormat() (TableFormat, error) {
979 1 : if r.err != nil {
980 0 : return TableFormatUnspecified, r.err
981 0 : }
982 1 : return r.tableFormat, nil
983 : }
984 :
985 : // NewReader returns a new table reader for the file. Closing the reader will
986 : // close the file.
987 : //
988 : // The context is used for tracing any operations performed by NewReader; it is
989 : // NOT stored for future use.
990 1 : func NewReader(ctx context.Context, f objstorage.Readable, o ReaderOptions) (*Reader, error) {
991 1 : if f == nil {
992 1 : return nil, errors.New("pebble/table: nil file")
993 1 : }
994 1 : o = o.ensureDefaults()
995 1 : r := &Reader{
996 1 : readable: f,
997 1 : cacheOpts: o.internal.CacheOpts,
998 1 : loadBlockSema: o.LoadBlockSema,
999 1 : deniedUserProperties: o.DeniedUserProperties,
1000 1 : filterMetricsTracker: o.FilterMetricsTracker,
1001 1 : logger: o.LoggerAndTracer,
1002 1 : }
1003 1 : if r.cacheOpts.Cache == nil {
1004 1 : r.cacheOpts.Cache = cache.New(0)
1005 1 : } else {
1006 1 : r.cacheOpts.Cache.Ref()
1007 1 : }
1008 1 : if r.cacheOpts.CacheID == 0 {
1009 1 : r.cacheOpts.CacheID = r.cacheOpts.Cache.NewID()
1010 1 : }
1011 :
1012 1 : var preallocRH objstorageprovider.PreallocatedReadHandle
1013 1 : rh := objstorageprovider.UsePreallocatedReadHandle(
1014 1 : r.readable, objstorage.ReadBeforeForNewReader, &preallocRH)
1015 1 : defer rh.Close()
1016 1 :
1017 1 : footer, err := readFooter(ctx, f, rh, r.logger, r.cacheOpts.FileNum)
1018 1 : if err != nil {
1019 1 : r.err = err
1020 1 : return nil, r.Close()
1021 1 : }
1022 1 : r.checksumType = footer.checksum
1023 1 : r.tableFormat = footer.format
1024 1 : r.indexBH = footer.indexBH
1025 1 : r.metaindexBH = footer.metaindexBH
1026 1 : r.footerBH = footer.footerBH
1027 1 : // Read the metaindex and properties blocks.
1028 1 : if err := r.readMetaindex(ctx, rh, o.Filters); err != nil {
1029 1 : r.err = err
1030 1 : return nil, r.Close()
1031 1 : }
1032 :
1033 1 : if r.Properties.ComparerName == "" || o.Comparer.Name == r.Properties.ComparerName {
1034 1 : r.Comparer = o.Comparer
1035 1 : r.Compare = o.Comparer.Compare
1036 1 : r.Equal = o.Comparer.Equal
1037 1 : r.Split = o.Comparer.Split
1038 1 : } else if comparer, ok := o.Comparers[r.Properties.ComparerName]; ok {
1039 1 : r.Comparer = o.Comparer
1040 1 : r.Compare = comparer.Compare
1041 1 : r.Equal = comparer.Equal
1042 1 : r.Split = comparer.Split
1043 1 : } else {
1044 1 : r.err = errors.Errorf("pebble/table: %d: unknown comparer %s",
1045 1 : errors.Safe(r.cacheOpts.FileNum), errors.Safe(r.Properties.ComparerName))
1046 1 : }
1047 :
1048 1 : if mergerName := r.Properties.MergerName; mergerName != "" && mergerName != "nullptr" {
1049 1 : if o.Merger != nil && o.Merger.Name == mergerName {
1050 1 : // opts.Merger matches.
1051 1 : } else if _, ok := o.Mergers[mergerName]; ok {
1052 1 : // Known merger.
1053 1 : } else {
1054 1 : r.err = errors.Errorf("pebble/table: %d: unknown merger %s",
1055 1 : errors.Safe(r.cacheOpts.FileNum), errors.Safe(r.Properties.MergerName))
1056 1 : }
1057 : }
1058 :
1059 1 : if r.tableFormat.BlockColumnar() {
1060 1 : if ks, ok := o.KeySchemas[r.Properties.KeySchemaName]; ok {
1061 1 : r.keySchema = ks
1062 1 : } else {
1063 0 : var known []string
1064 0 : for name := range o.KeySchemas {
1065 0 : known = append(known, fmt.Sprintf("%q", name))
1066 0 : }
1067 0 : slices.Sort(known)
1068 0 :
1069 0 : r.err = errors.Newf("pebble/table: %d: unknown key schema %q; known key schemas: %s",
1070 0 : errors.Safe(r.cacheOpts.FileNum), errors.Safe(r.Properties.KeySchemaName), errors.Safe(known))
1071 0 : panic(r.err)
1072 : }
1073 : }
1074 :
1075 1 : if r.err != nil {
1076 1 : return nil, r.Close()
1077 1 : }
1078 :
1079 1 : return r, nil
1080 : }
1081 :
1082 : // ReadableFile describes the smallest subset of vfs.File that is required for
1083 : // reading SSTs.
1084 : type ReadableFile interface {
1085 : io.ReaderAt
1086 : io.Closer
1087 : Stat() (vfs.FileInfo, error)
1088 : }
1089 :
1090 : // NewSimpleReadable wraps a ReadableFile in a objstorage.Readable
1091 : // implementation (which does not support read-ahead)
1092 1 : func NewSimpleReadable(r ReadableFile) (objstorage.Readable, error) {
1093 1 : info, err := r.Stat()
1094 1 : if err != nil {
1095 1 : return nil, err
1096 1 : }
1097 1 : res := &simpleReadable{
1098 1 : f: r,
1099 1 : size: info.Size(),
1100 1 : }
1101 1 : res.rh = objstorage.MakeNoopReadHandle(res)
1102 1 : return res, nil
1103 : }
1104 :
1105 : // simpleReadable wraps a ReadableFile to implement objstorage.Readable.
1106 : type simpleReadable struct {
1107 : f ReadableFile
1108 : size int64
1109 : rh objstorage.NoopReadHandle
1110 : }
1111 :
1112 : var _ objstorage.Readable = (*simpleReadable)(nil)
1113 :
1114 : // ReadAt is part of the objstorage.Readable interface.
1115 1 : func (s *simpleReadable) ReadAt(_ context.Context, p []byte, off int64) error {
1116 1 : n, err := s.f.ReadAt(p, off)
1117 1 : if invariants.Enabled && err == nil && n != len(p) {
1118 0 : panic("short read")
1119 : }
1120 1 : return err
1121 : }
1122 :
1123 : // Close is part of the objstorage.Readable interface.
1124 1 : func (s *simpleReadable) Close() error {
1125 1 : return s.f.Close()
1126 1 : }
1127 :
1128 : // Size is part of the objstorage.Readable interface.
1129 1 : func (s *simpleReadable) Size() int64 {
1130 1 : return s.size
1131 1 : }
1132 :
1133 : // NewReadHandle is part of the objstorage.Readable interface.
1134 : func (s *simpleReadable) NewReadHandle(
1135 : readBeforeSize objstorage.ReadBeforeSize,
1136 1 : ) objstorage.ReadHandle {
1137 1 : return &s.rh
1138 1 : }
1139 :
1140 0 : func errCorruptIndexEntry(err error) error {
1141 0 : err = base.CorruptionErrorf("pebble/table: corrupt index entry: %v", err)
1142 0 : if invariants.Enabled {
1143 0 : panic(err)
1144 : }
1145 0 : return err
1146 : }
1147 :
1148 : type deterministicStopwatchForTesting struct {
1149 : startTime crtime.Mono
1150 : }
1151 :
1152 1 : func makeStopwatch() deterministicStopwatchForTesting {
1153 1 : return deterministicStopwatchForTesting{startTime: crtime.NowMono()}
1154 1 : }
1155 :
1156 1 : func (w deterministicStopwatchForTesting) stop() time.Duration {
1157 1 : dur := w.startTime.Elapsed()
1158 1 : if deterministicReadBlockDurationForTesting {
1159 1 : dur = slowReadTracingThreshold
1160 1 : }
1161 1 : return dur
1162 : }
|
