2020-03-13 04:39:36 +00:00
|
|
|
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
|
|
|
// This source code is licensed under both the GPLv2 (found in the
|
|
|
|
// COPYING file in the root directory) and Apache 2.0 License
|
|
|
|
// (found in the LICENSE.Apache file in the root directory).
|
|
|
|
//
|
|
|
|
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
|
|
|
|
// Use of this source code is governed by a BSD-style license that can be
|
|
|
|
// found in the LICENSE file. See the AUTHORS file for names of contributors.
|
|
|
|
#include "table/block_based/hash_index_reader.h"
|
|
|
|
|
|
|
|
#include "table/block_fetcher.h"
|
|
|
|
#include "table/meta_blocks.h"
|
|
|
|
|
|
|
|
namespace ROCKSDB_NAMESPACE {
|
|
|
|
Status HashIndexReader::Create(const BlockBasedTable* table,
|
2020-06-29 21:51:57 +00:00
|
|
|
const ReadOptions& ro,
|
2020-03-13 04:39:36 +00:00
|
|
|
FilePrefetchBuffer* prefetch_buffer,
|
|
|
|
InternalIterator* meta_index_iter,
|
|
|
|
bool use_cache, bool prefetch, bool pin,
|
|
|
|
BlockCacheLookupContext* lookup_context,
|
|
|
|
std::unique_ptr<IndexReader>* index_reader) {
|
|
|
|
assert(table != nullptr);
|
|
|
|
assert(index_reader != nullptr);
|
|
|
|
assert(!pin || prefetch);
|
|
|
|
|
|
|
|
const BlockBasedTable::Rep* rep = table->get_rep();
|
|
|
|
assert(rep != nullptr);
|
|
|
|
|
|
|
|
CachableEntry<Block> index_block;
|
|
|
|
if (prefetch || !use_cache) {
|
|
|
|
const Status s =
|
2020-06-29 21:51:57 +00:00
|
|
|
ReadIndexBlock(table, prefetch_buffer, ro, use_cache,
|
2020-03-13 04:39:36 +00:00
|
|
|
/*get_context=*/nullptr, lookup_context, &index_block);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (use_cache && !pin) {
|
|
|
|
index_block.Reset();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Note, failure to create prefix hash index does not need to be a
|
|
|
|
// hard error. We can still fall back to the original binary search index.
|
|
|
|
// So, Create will succeed regardless, from this point on.
|
|
|
|
|
|
|
|
index_reader->reset(new HashIndexReader(table, std::move(index_block)));
|
|
|
|
|
|
|
|
// Get prefixes block
|
|
|
|
BlockHandle prefixes_handle;
|
|
|
|
Status s =
|
|
|
|
FindMetaBlock(meta_index_iter, kHashIndexPrefixesBlock, &prefixes_handle);
|
|
|
|
if (!s.ok()) {
|
|
|
|
// TODO: log error
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
// Get index metadata block
|
|
|
|
BlockHandle prefixes_meta_handle;
|
|
|
|
s = FindMetaBlock(meta_index_iter, kHashIndexPrefixesMetadataBlock,
|
|
|
|
&prefixes_meta_handle);
|
|
|
|
if (!s.ok()) {
|
|
|
|
// TODO: log error
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
RandomAccessFileReader* const file = rep->file.get();
|
|
|
|
const Footer& footer = rep->footer;
|
2021-05-05 20:59:21 +00:00
|
|
|
const ImmutableOptions& ioptions = rep->ioptions;
|
2020-03-13 04:39:36 +00:00
|
|
|
const PersistentCacheOptions& cache_options = rep->persistent_cache_options;
|
|
|
|
MemoryAllocator* const memory_allocator =
|
|
|
|
GetMemoryAllocator(rep->table_options);
|
|
|
|
|
|
|
|
// Read contents for the blocks
|
|
|
|
BlockContents prefixes_contents;
|
|
|
|
BlockFetcher prefixes_block_fetcher(
|
2023-04-21 16:07:18 +00:00
|
|
|
file, prefetch_buffer, footer, ro, prefixes_handle, &prefixes_contents,
|
|
|
|
ioptions, true /*decompress*/, true /*maybe_compressed*/,
|
|
|
|
BlockType::kHashIndexPrefixes, UncompressionDict::GetEmptyDict(),
|
|
|
|
cache_options, memory_allocator);
|
2020-03-13 04:39:36 +00:00
|
|
|
s = prefixes_block_fetcher.ReadBlockContents();
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
BlockContents prefixes_meta_contents;
|
|
|
|
BlockFetcher prefixes_meta_block_fetcher(
|
2023-04-21 16:07:18 +00:00
|
|
|
file, prefetch_buffer, footer, ro, prefixes_meta_handle,
|
2020-03-13 04:39:36 +00:00
|
|
|
&prefixes_meta_contents, ioptions, true /*decompress*/,
|
|
|
|
true /*maybe_compressed*/, BlockType::kHashIndexMetadata,
|
|
|
|
UncompressionDict::GetEmptyDict(), cache_options, memory_allocator);
|
|
|
|
s = prefixes_meta_block_fetcher.ReadBlockContents();
|
|
|
|
if (!s.ok()) {
|
|
|
|
// TODO: log error
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
BlockPrefixIndex* prefix_index = nullptr;
|
2022-06-10 15:51:45 +00:00
|
|
|
assert(rep->table_prefix_extractor);
|
|
|
|
s = BlockPrefixIndex::Create(rep->table_prefix_extractor.get(),
|
2020-03-13 04:39:36 +00:00
|
|
|
prefixes_contents.data,
|
|
|
|
prefixes_meta_contents.data, &prefix_index);
|
|
|
|
// TODO: log error
|
|
|
|
if (s.ok()) {
|
|
|
|
HashIndexReader* const hash_index_reader =
|
|
|
|
static_cast<HashIndexReader*>(index_reader->get());
|
|
|
|
hash_index_reader->prefix_index_.reset(prefix_index);
|
|
|
|
}
|
|
|
|
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
InternalIteratorBase<IndexValue>* HashIndexReader::NewIterator(
|
|
|
|
const ReadOptions& read_options, bool disable_prefix_seek,
|
|
|
|
IndexBlockIter* iter, GetContext* get_context,
|
|
|
|
BlockCacheLookupContext* lookup_context) {
|
|
|
|
const BlockBasedTable::Rep* rep = table()->get_rep();
|
|
|
|
const bool no_io = (read_options.read_tier == kBlockCacheTier);
|
|
|
|
CachableEntry<Block> index_block;
|
2023-04-04 23:53:14 +00:00
|
|
|
const Status s = GetOrReadIndexBlock(no_io, get_context, lookup_context,
|
|
|
|
&index_block, read_options);
|
2020-03-13 04:39:36 +00:00
|
|
|
if (!s.ok()) {
|
|
|
|
if (iter != nullptr) {
|
|
|
|
iter->Invalidate(s);
|
|
|
|
return iter;
|
|
|
|
}
|
|
|
|
|
|
|
|
return NewErrorInternalIterator<IndexValue>(s);
|
|
|
|
}
|
|
|
|
|
|
|
|
Statistics* kNullStats = nullptr;
|
|
|
|
const bool total_order_seek =
|
|
|
|
read_options.total_order_seek || disable_prefix_seek;
|
|
|
|
// We don't return pinned data from index blocks, so no need
|
|
|
|
// to set `block_contents_pinned`.
|
|
|
|
auto it = index_block.GetValue()->NewIndexIterator(
|
2020-07-08 00:25:08 +00:00
|
|
|
internal_comparator()->user_comparator(),
|
2020-03-13 04:39:36 +00:00
|
|
|
rep->get_global_seqno(BlockType::kIndex), iter, kNullStats,
|
|
|
|
total_order_seek, index_has_first_key(), index_key_includes_seq(),
|
|
|
|
index_value_is_full(), false /* block_contents_pinned */,
|
Add support to strip / pad timestamp when creating / reading a block based table (#11495)
Summary:
Add support to strip timestamp in block based table builder and pad timestamp in block based table reader.
On the write path, use the per column family option `AdvancedColumnFamilyOptions.persist_user_defined_timestamps` to indicate whether user-defined timestamps should be stripped for all block based tables created for the column family.
On the read path, added a per table `TableReadOption.user_defined_timestamps_persisted` to flag whether the user keys in the table contains user defined timestamps.
This patch is mostly passing the related flags down to the block building/parsing level with the exception of handling the `first_internal_key` in `IndexValue`, which is included in the `IndexBuilder` level. The value part of range deletion entries should have a similar handling, I haven't decided where to best fit this piece of logic, I will do it in a follow up.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11495
Test Plan:
Existing test `BlockBasedTableReaderTest` is parameterized to run with:
1) different UDT test modes: kNone, kNormal, kStripUserDefinedTimestamp
2) all four index types, when index type is `kTwoLevelIndexSearch`, also enables partitioned filters
3) parallel vs non-parallel compression
4) enable/disable compression dictionary.
Also added tests for API `BlockBasedTableReader::NewIterator`.
`PartitionedFilterBlockTest` is parameterized to run with different UDT test modes:kNone, kNormal, kStripUserDefinedTimestamp.
```
make all check
./block_based_table_reader_test
./partitioned_filter_block_test
```
Reviewed By: ltamasi
Differential Revision: D46344577
Pulled By: jowlyzhang
fbshipit-source-id: 93ac8542b19319d1298712b8bed908c8831ba675
2023-06-01 18:10:03 +00:00
|
|
|
user_defined_timestamps_persisted(), prefix_index_.get());
|
2020-03-13 04:39:36 +00:00
|
|
|
|
|
|
|
assert(it != nullptr);
|
|
|
|
index_block.TransferTo(it);
|
|
|
|
|
|
|
|
return it;
|
|
|
|
}
|
|
|
|
} // namespace ROCKSDB_NAMESPACE
|