continuwuation/rocksdb
5
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
rocksdb/options/options_helper.cc
Peter Dillinger 9d490593d0 Preliminary support for custom compression algorithms (#13659) 
Summary:
This change builds on https://github.com/facebook/rocksdb/issues/13540 and https://github.com/facebook/rocksdb/issues/13626 in allowing a CompressionManager / Compressor / Decompressor to use a custom compression algorithm, with a distinct CompressionType. For background, review the API comments on CompressionManager and its CompatibilityName() function.

Highlights:
* Reserve and name 127 new CompressionTypes that can be used for custom compression algorithms / schemas. In many or most cases I expect the enumerators such as `kCustomCompression8F` to be used in user code rather than casting between integers and CompressionTypes, as I expect the supported custom compression algorithms to be identifiable / enumerable at compile time.
* When using these custom compression types, a CompressionManager must use a CompatibilityName() other than the built-in one AND new format_version=7 (see below).
* When building new SST files, track the full set of CompressionTypes actually used (usually just one aside from kNoCompression), using our efficient bitset SmallEnumSet, which supports fast iteration over the bits set to 1. Ideally, to support mixed or non-mixed compression algorithms in a file as efficiently as possible, we would know the set of CompressionTypes as SST file open time.
* New schema for `TableProperties::compression_name` in format_version=7 to represent the CompressionManager's CompatibilityName(), the set of CompressionTypes used, and potentially more in the future, while keeping the data relatively human-readable.
  * It would be possible to do this without a new format_version, but then the only way to ensure incompatible versions fail is with an unsupported CompressionType tag, not with a compression_name property. Therefore, (a) I prefer not to put something misleading in the `compression_name` property (a built-in compression name) when there is nuance because of a CompressionManager, and (b) I prefer better, more consistent error messages that refer to either format_version or the CompressionManager's CompatibilityName(), rather than an unrecognized custom CompressionType value (which could have come from various CompressionManagers).
* The current configured CompressionManager is passed in to TableReaders so that it (or one it knows about) can be used if it matches the CompatibilityName() used for compression in the SST file. Until the connection with ObjectRegistry is implemented, the only way to read files generated with a particular CompressionManager using custom compression algorithms is to configure it (or a known relative; see FindCompatibleCompressionManager()) in the ColumnFamilyOptions.
* Optimized snappy compression with BuiltinDecompressorV2SnappyOnly, to offset some small added overheads with the new tracking. This is essentially an early part of the planned refactoring that will get rid of the old internal compression APIs.
* Another small optimization in eliminating an unnecessary key copy in flush (builder.cc).
* Fix some handling of named CompressionManagers in CompressionManager::CreateFromString() (problem seen in https://github.com/facebook/rocksdb/issues/13647)

Smaller things:
* Adds Name() and GetId() functions to Compressor for debugging/logging purposes. (Compressor and Decompressor are not expected to be Customizable because they are only instantiated by a CompressionManager.)
* When using an explicit compression_manager, the GetId() of the CompressionManager and the Compressor used to build the file are stored as bonus entries in the compression_options table property. This table property is not parsed anywhere, so it is currently for human reading, but still could be parsed with the new underscore-prefixed bonus entries. IMHO, this is preferable to additional table properties, which would increase memory fragmentation in the TableProperties objects and likely take slightly more CPU on SST open and slightly more storage.
* ReleaseWorkingArea() function from protected to public to make wrappers work, because of a quirk in C++ (vs. Java) in which you cannot access protected members of another instance of the same class (sigh)
* Added `CompressionManager:: SupportsCompressionType()` for early options sanity checking.

Follow-up before release:
* Make format_version=7 official / supported
* Stress test coverage

Sooner than later:
* Update tests for RoundRobinManager and SimpleMixedCompressionManager to take advantage of e.g. set of compression types in compression_name property
* ObjectRegistry stuff
* Refactor away old internal compression APIs

Pull Request resolved: https://github.com/facebook/rocksdb/pull/13659

Test Plan:
Basic unit test added.

## Performance

### SST write performance
```
SUFFIX=`tty | sed 's|/|_|g'`; for ARGS in "-compression_type=none" "-compression_type=snappy" "-compression_type=zstd" "-compression_type=snappy -verify_compression=1" "-compression_type=zstd -verify_compression=1" "-compression_type=zstd -compression_max_dict_bytes=8180"; do echo $ARGS; (for I in `seq 1 20`; do BIN=/dev/shm/dbbench${SUFFIX}.bin; rm -f $BIN; cp db_bench $BIN; $BIN -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 $ARGS 2>&1 | grep micros/op; done) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```

Ops/sec, Before -> After, both fv=6:
-compression_type=none
1894386 -> 1858403 (-2.0%)
-compression_type=snappy
1859131 -> 1807469 (-2.8%)
-compression_type=zstd
1191428 -> 1214374 (+1.9%)
-compression_type=snappy -verify_compression=1
1861819 -> 1858342 (+0.2%)
-compression_type=zstd -verify_compression=1
979435 -> 995870 (+1.6%)
-compression_type=zstd -compression_max_dict_bytes=8180
905349 -> 940563 (+3.9%)

Ops/sec, Before fv=6 -> After fv=7:
-compression_type=none
1879365 -> 1836159 (-2.3%)
-compression_type=snappy
1865460 -> 1830916 (-1.9%)
-compression_type=zstd
1191428 -> 1210260 (+1.6%)
-compression_type=snappy -verify_compression=1
1866756 -> 1818989 (-2.6%)
-compression_type=zstd -verify_compression=1
982640 -> 997129 (+1.5%)
-compression_type=zstd -compression_max_dict_bytes=8180
912608 -> 937248 (+2.7%)

### SST read performance
Create DBs
```
for COMP in none snappy zstd; do echo $ARGS; ./db_bench -db=/dev/shm/dbbench-7-$COMP --benchmarks=fillseq,flush -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -compression_type=$COMP -format_version=7; done
```
And test
```
for COMP in none
snappy zstd none; do echo $COMP; (for I in `seq 1 8`; do ./db_bench -readonly -db=/dev/shm/dbbench
-7-$COMP --benchmarks=readrandom -num=10000000 -duration=20 -threads=8 2>&1 | grep micros/op; done
) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```

Ops/sec, Before -> After (both fv=6)
none
1491732 -> 1500209 (+0.6%)
snappy
1157216 -> 1169202 (+1.0%)
zstd
695414 -> 703719 (+1.2%)
none (again)
1491787 -> 1528789 (+2.4%)

Ops/sec, Before fv=6 -> After fv=7:
none
1492278 -> 1508668 (+1.1%)
snappy
1140769 -> 1152613 (+1.0%)
zstd
696437 -> 696511 (+0.0%)
none (again)
1500585 -> 1512037 (+0.7%)

Overall, I think we can take the read CPU improvement in exchange for the hit (in some cases) on background write CPU

Reviewed By: hx235

Differential Revision: D76520739

Pulled By: pdillinger

fbshipit-source-id: e73bd72502ff85c8779cba313f26f7d1fd50be3a
2025-06-16 14:19:03 -07:00

1587 lines
66 KiB
C++
Raw Permalink Blame History

// 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).
#include "options/options_helper.h"
#include <atomic>
#include <cassert>
#include <cctype>
#include <cstdlib>
#include <set>
#include <unordered_set>
#include <vector>
#include "options/cf_options.h"
#include "options/db_options.h"
#include "rocksdb/cache.h"
#include "rocksdb/compaction_filter.h"
#include "rocksdb/convenience.h"
#include "rocksdb/filter_policy.h"
#include "rocksdb/flush_block_policy.h"
#include "rocksdb/memtablerep.h"
#include "rocksdb/merge_operator.h"
#include "rocksdb/options.h"
#include "rocksdb/rate_limiter.h"
#include "rocksdb/slice_transform.h"
#include "rocksdb/table.h"
#include "rocksdb/utilities/object_registry.h"
#include "rocksdb/utilities/options_type.h"
#include "util/string_util.h"
namespace ROCKSDB_NAMESPACE {
ConfigOptions::ConfigOptions() : registry(ObjectRegistry::NewInstance()) {
env = Env::Default();
}
ConfigOptions::ConfigOptions(const DBOptions& db_opts) : env(db_opts.env) {
registry = ObjectRegistry::NewInstance();
}
Status ValidateOptions(const DBOptions& db_opts,
const ColumnFamilyOptions& cf_opts) {
Status s;
auto db_cfg = DBOptionsAsConfigurable(db_opts);
auto cf_cfg = CFOptionsAsConfigurable(cf_opts);
s = db_cfg->ValidateOptions(db_opts, cf_opts);
if (s.ok()) {
s = cf_cfg->ValidateOptions(db_opts, cf_opts);
}
return s;
}
DBOptions BuildDBOptions(const ImmutableDBOptions& immutable_db_options,
const MutableDBOptions& mutable_db_options) {
DBOptions options;
BuildDBOptions(immutable_db_options, mutable_db_options, options);
return options;
}
void BuildDBOptions(const ImmutableDBOptions& immutable_db_options,
const MutableDBOptions& mutable_db_options,
DBOptions& options) {
options.create_if_missing = immutable_db_options.create_if_missing;
options.create_missing_column_families =
immutable_db_options.create_missing_column_families;
options.error_if_exists = immutable_db_options.error_if_exists;
options.paranoid_checks = immutable_db_options.paranoid_checks;
options.flush_verify_memtable_count =
immutable_db_options.flush_verify_memtable_count;
options.compaction_verify_record_count =
immutable_db_options.compaction_verify_record_count;
options.track_and_verify_wals_in_manifest =
immutable_db_options.track_and_verify_wals_in_manifest;
options.track_and_verify_wals = immutable_db_options.track_and_verify_wals;
options.verify_sst_unique_id_in_manifest =
immutable_db_options.verify_sst_unique_id_in_manifest;
options.env = immutable_db_options.env;
options.rate_limiter = immutable_db_options.rate_limiter;
options.sst_file_manager = immutable_db_options.sst_file_manager;
options.info_log = immutable_db_options.info_log;
options.info_log_level = immutable_db_options.info_log_level;
options.max_open_files = mutable_db_options.max_open_files;
options.max_file_opening_threads =
immutable_db_options.max_file_opening_threads;
options.max_total_wal_size = mutable_db_options.max_total_wal_size;
options.statistics = immutable_db_options.statistics;
options.use_fsync = immutable_db_options.use_fsync;
options.db_paths = immutable_db_options.db_paths;
options.db_log_dir = immutable_db_options.db_log_dir;
options.wal_dir = immutable_db_options.wal_dir;
options.delete_obsolete_files_period_micros =
mutable_db_options.delete_obsolete_files_period_micros;
options.max_background_jobs = mutable_db_options.max_background_jobs;
options.max_background_compactions =
mutable_db_options.max_background_compactions;
options.max_subcompactions = mutable_db_options.max_subcompactions;
options.max_background_flushes = mutable_db_options.max_background_flushes;
options.max_log_file_size = immutable_db_options.max_log_file_size;
options.log_file_time_to_roll = immutable_db_options.log_file_time_to_roll;
options.keep_log_file_num = immutable_db_options.keep_log_file_num;
options.recycle_log_file_num = immutable_db_options.recycle_log_file_num;
options.max_manifest_file_size = immutable_db_options.max_manifest_file_size;
options.table_cache_numshardbits =
immutable_db_options.table_cache_numshardbits;
options.WAL_ttl_seconds = immutable_db_options.WAL_ttl_seconds;
options.WAL_size_limit_MB = immutable_db_options.WAL_size_limit_MB;
options.manifest_preallocation_size =
immutable_db_options.manifest_preallocation_size;
options.allow_mmap_reads = immutable_db_options.allow_mmap_reads;
options.allow_mmap_writes = immutable_db_options.allow_mmap_writes;
options.use_direct_reads = immutable_db_options.use_direct_reads;
options.use_direct_io_for_flush_and_compaction =
immutable_db_options.use_direct_io_for_flush_and_compaction;
options.allow_fallocate = immutable_db_options.allow_fallocate;
options.is_fd_close_on_exec = immutable_db_options.is_fd_close_on_exec;
options.stats_dump_period_sec = mutable_db_options.stats_dump_period_sec;
options.stats_persist_period_sec =
mutable_db_options.stats_persist_period_sec;
options.persist_stats_to_disk = immutable_db_options.persist_stats_to_disk;
options.stats_history_buffer_size =
mutable_db_options.stats_history_buffer_size;
options.advise_random_on_open = immutable_db_options.advise_random_on_open;
options.db_write_buffer_size = immutable_db_options.db_write_buffer_size;
options.write_buffer_manager = immutable_db_options.write_buffer_manager;
options.compaction_readahead_size =
mutable_db_options.compaction_readahead_size;
options.writable_file_max_buffer_size =
mutable_db_options.writable_file_max_buffer_size;
options.use_adaptive_mutex = immutable_db_options.use_adaptive_mutex;
options.bytes_per_sync = mutable_db_options.bytes_per_sync;
options.wal_bytes_per_sync = mutable_db_options.wal_bytes_per_sync;
options.strict_bytes_per_sync = mutable_db_options.strict_bytes_per_sync;
options.listeners = immutable_db_options.listeners;
options.enable_thread_tracking = immutable_db_options.enable_thread_tracking;
options.delayed_write_rate = mutable_db_options.delayed_write_rate;
options.enable_pipelined_write = immutable_db_options.enable_pipelined_write;
options.unordered_write = immutable_db_options.unordered_write;
options.allow_concurrent_memtable_write =
immutable_db_options.allow_concurrent_memtable_write;
options.enable_write_thread_adaptive_yield =
immutable_db_options.enable_write_thread_adaptive_yield;
options.max_write_batch_group_size_bytes =
immutable_db_options.max_write_batch_group_size_bytes;
options.write_thread_max_yield_usec =
immutable_db_options.write_thread_max_yield_usec;
options.write_thread_slow_yield_usec =
immutable_db_options.write_thread_slow_yield_usec;
options.skip_stats_update_on_db_open =
immutable_db_options.skip_stats_update_on_db_open;
options.skip_checking_sst_file_sizes_on_db_open =
immutable_db_options.skip_checking_sst_file_sizes_on_db_open;
options.wal_recovery_mode = immutable_db_options.wal_recovery_mode;
options.allow_2pc = immutable_db_options.allow_2pc;
options.row_cache = immutable_db_options.row_cache;
options.wal_filter = immutable_db_options.wal_filter;
options.dump_malloc_stats = immutable_db_options.dump_malloc_stats;
options.avoid_flush_during_recovery =
immutable_db_options.avoid_flush_during_recovery;
options.avoid_flush_during_shutdown =
mutable_db_options.avoid_flush_during_shutdown;
options.allow_ingest_behind = immutable_db_options.allow_ingest_behind;
options.two_write_queues = immutable_db_options.two_write_queues;
options.manual_wal_flush = immutable_db_options.manual_wal_flush;
options.wal_compression = immutable_db_options.wal_compression;
options.background_close_inactive_wals =
immutable_db_options.background_close_inactive_wals;
options.atomic_flush = immutable_db_options.atomic_flush;
options.avoid_unnecessary_blocking_io =
immutable_db_options.avoid_unnecessary_blocking_io;
options.write_dbid_to_manifest = immutable_db_options.write_dbid_to_manifest;
options.write_identity_file = immutable_db_options.write_identity_file;
options.prefix_seek_opt_in_only =
immutable_db_options.prefix_seek_opt_in_only;
options.log_readahead_size = immutable_db_options.log_readahead_size;
options.file_checksum_gen_factory =
immutable_db_options.file_checksum_gen_factory;
options.best_efforts_recovery = immutable_db_options.best_efforts_recovery;
options.max_bgerror_resume_count =
immutable_db_options.max_bgerror_resume_count;
options.bgerror_resume_retry_interval =
immutable_db_options.bgerror_resume_retry_interval;
options.db_host_id = immutable_db_options.db_host_id;
options.allow_data_in_errors = immutable_db_options.allow_data_in_errors;
options.checksum_handoff_file_types =
immutable_db_options.checksum_handoff_file_types;
options.lowest_used_cache_tier = immutable_db_options.lowest_used_cache_tier;
options.enforce_single_del_contracts =
immutable_db_options.enforce_single_del_contracts;
options.daily_offpeak_time_utc = mutable_db_options.daily_offpeak_time_utc;
options.follower_refresh_catchup_period_ms =
immutable_db_options.follower_refresh_catchup_period_ms;
options.follower_catchup_retry_count =
immutable_db_options.follower_catchup_retry_count;
options.follower_catchup_retry_wait_ms =
immutable_db_options.follower_catchup_retry_wait_ms;
options.metadata_write_temperature =
immutable_db_options.metadata_write_temperature;
options.wal_write_temperature = immutable_db_options.wal_write_temperature;
options.compaction_service = immutable_db_options.compaction_service;
options.calculate_sst_write_lifetime_hint_set =
immutable_db_options.calculate_sst_write_lifetime_hint_set;
}
ColumnFamilyOptions BuildColumnFamilyOptions(
const ColumnFamilyOptions& options,
const MutableCFOptions& mutable_cf_options) {
ColumnFamilyOptions cf_opts(options);
UpdateColumnFamilyOptions(mutable_cf_options, &cf_opts);
// TODO(yhchiang): find some way to handle the following derived options
// * max_file_size
return cf_opts;
}
void UpdateColumnFamilyOptions(const MutableCFOptions& moptions,
ColumnFamilyOptions* cf_opts) {
// Memtable related options
cf_opts->write_buffer_size = moptions.write_buffer_size;
cf_opts->max_write_buffer_number = moptions.max_write_buffer_number;
cf_opts->arena_block_size = moptions.arena_block_size;
cf_opts->memtable_prefix_bloom_size_ratio =
moptions.memtable_prefix_bloom_size_ratio;
cf_opts->memtable_whole_key_filtering = moptions.memtable_whole_key_filtering;
cf_opts->memtable_huge_page_size = moptions.memtable_huge_page_size;
cf_opts->max_successive_merges = moptions.max_successive_merges;
cf_opts->strict_max_successive_merges = moptions.strict_max_successive_merges;
cf_opts->inplace_update_num_locks = moptions.inplace_update_num_locks;
cf_opts->prefix_extractor = moptions.prefix_extractor;
cf_opts->experimental_mempurge_threshold =
moptions.experimental_mempurge_threshold;
cf_opts->memtable_protection_bytes_per_key =
moptions.memtable_protection_bytes_per_key;
cf_opts->block_protection_bytes_per_key =
moptions.block_protection_bytes_per_key;
cf_opts->paranoid_memory_checks = moptions.paranoid_memory_checks;
cf_opts->bottommost_file_compaction_delay =
moptions.bottommost_file_compaction_delay;
// Compaction related options
cf_opts->disable_auto_compactions = moptions.disable_auto_compactions;
cf_opts->table_factory = moptions.table_factory;
cf_opts->soft_pending_compaction_bytes_limit =
moptions.soft_pending_compaction_bytes_limit;
cf_opts->hard_pending_compaction_bytes_limit =
moptions.hard_pending_compaction_bytes_limit;
cf_opts->level0_file_num_compaction_trigger =
moptions.level0_file_num_compaction_trigger;
cf_opts->level0_slowdown_writes_trigger =
moptions.level0_slowdown_writes_trigger;
cf_opts->level0_stop_writes_trigger = moptions.level0_stop_writes_trigger;
cf_opts->max_compaction_bytes = moptions.max_compaction_bytes;
cf_opts->target_file_size_base = moptions.target_file_size_base;
cf_opts->target_file_size_multiplier = moptions.target_file_size_multiplier;
cf_opts->max_bytes_for_level_base = moptions.max_bytes_for_level_base;
cf_opts->max_bytes_for_level_multiplier =
moptions.max_bytes_for_level_multiplier;
cf_opts->ttl = moptions.ttl;
cf_opts->periodic_compaction_seconds = moptions.periodic_compaction_seconds;
cf_opts->preclude_last_level_data_seconds =
moptions.preclude_last_level_data_seconds;
cf_opts->preserve_internal_time_seconds =
moptions.preserve_internal_time_seconds;
cf_opts->max_bytes_for_level_multiplier_additional.clear();
for (auto value : moptions.max_bytes_for_level_multiplier_additional) {
cf_opts->max_bytes_for_level_multiplier_additional.emplace_back(value);
}
cf_opts->compaction_options_fifo = moptions.compaction_options_fifo;
cf_opts->compaction_options_universal = moptions.compaction_options_universal;
// Blob file related options
cf_opts->enable_blob_files = moptions.enable_blob_files;
cf_opts->min_blob_size = moptions.min_blob_size;
cf_opts->blob_file_size = moptions.blob_file_size;
cf_opts->blob_compression_type = moptions.blob_compression_type;
cf_opts->enable_blob_garbage_collection =
moptions.enable_blob_garbage_collection;
cf_opts->blob_garbage_collection_age_cutoff =
moptions.blob_garbage_collection_age_cutoff;
cf_opts->blob_garbage_collection_force_threshold =
moptions.blob_garbage_collection_force_threshold;
cf_opts->blob_compaction_readahead_size =
moptions.blob_compaction_readahead_size;
cf_opts->blob_file_starting_level = moptions.blob_file_starting_level;
cf_opts->prepopulate_blob_cache = moptions.prepopulate_blob_cache;
// Misc options
cf_opts->max_sequential_skip_in_iterations =
moptions.max_sequential_skip_in_iterations;
cf_opts->paranoid_file_checks = moptions.paranoid_file_checks;
cf_opts->report_bg_io_stats = moptions.report_bg_io_stats;
cf_opts->compression = moptions.compression;
cf_opts->compression_opts = moptions.compression_opts;
cf_opts->bottommost_compression = moptions.bottommost_compression;
cf_opts->bottommost_compression_opts = moptions.bottommost_compression_opts;
cf_opts->compression_manager = moptions.compression_manager;
cf_opts->sample_for_compression = moptions.sample_for_compression;
cf_opts->compression_per_level = moptions.compression_per_level;
cf_opts->last_level_temperature = moptions.last_level_temperature;
cf_opts->default_write_temperature = moptions.default_write_temperature;
cf_opts->memtable_max_range_deletions = moptions.memtable_max_range_deletions;
cf_opts->uncache_aggressiveness = moptions.uncache_aggressiveness;
cf_opts->memtable_op_scan_flush_trigger =
moptions.memtable_op_scan_flush_trigger;
cf_opts->memtable_avg_op_scan_flush_trigger =
moptions.memtable_avg_op_scan_flush_trigger;
}
void UpdateColumnFamilyOptions(const ImmutableCFOptions& ioptions,
ColumnFamilyOptions* cf_opts) {
cf_opts->compaction_style = ioptions.compaction_style;
cf_opts->compaction_pri = ioptions.compaction_pri;
cf_opts->comparator = ioptions.user_comparator;
cf_opts->merge_operator = ioptions.merge_operator;
cf_opts->compaction_filter = ioptions.compaction_filter;
cf_opts->compaction_filter_factory = ioptions.compaction_filter_factory;
cf_opts->min_write_buffer_number_to_merge =
ioptions.min_write_buffer_number_to_merge;
cf_opts->max_write_buffer_size_to_maintain =
ioptions.max_write_buffer_size_to_maintain;
cf_opts->inplace_update_support = ioptions.inplace_update_support;
cf_opts->inplace_callback = ioptions.inplace_callback;
cf_opts->memtable_factory = ioptions.memtable_factory;
cf_opts->table_properties_collector_factories =
ioptions.table_properties_collector_factories;
cf_opts->bloom_locality = ioptions.bloom_locality;
cf_opts->level_compaction_dynamic_level_bytes =
ioptions.level_compaction_dynamic_level_bytes;
cf_opts->num_levels = ioptions.num_levels;
cf_opts->optimize_filters_for_hits = ioptions.optimize_filters_for_hits;
cf_opts->force_consistency_checks = ioptions.force_consistency_checks;
cf_opts->disallow_memtable_writes = ioptions.disallow_memtable_writes;
cf_opts->memtable_insert_with_hint_prefix_extractor =
ioptions.memtable_insert_with_hint_prefix_extractor;
cf_opts->cf_paths = ioptions.cf_paths;
cf_opts->compaction_thread_limiter = ioptions.compaction_thread_limiter;
cf_opts->sst_partitioner_factory = ioptions.sst_partitioner_factory;
cf_opts->blob_cache = ioptions.blob_cache;
cf_opts->persist_user_defined_timestamps =
ioptions.persist_user_defined_timestamps;
cf_opts->default_temperature = ioptions.default_temperature;
// TODO(yhchiang): find some way to handle the following derived options
// * max_file_size
}
std::map<CompactionStyle, std::string>
OptionsHelper::compaction_style_to_string = {
{kCompactionStyleLevel, "kCompactionStyleLevel"},
{kCompactionStyleUniversal, "kCompactionStyleUniversal"},
{kCompactionStyleFIFO, "kCompactionStyleFIFO"},
{kCompactionStyleNone, "kCompactionStyleNone"}};
std::map<CompactionPri, std::string> OptionsHelper::compaction_pri_to_string = {
{kByCompensatedSize, "kByCompensatedSize"},
{kOldestLargestSeqFirst, "kOldestLargestSeqFirst"},
{kOldestSmallestSeqFirst, "kOldestSmallestSeqFirst"},
{kMinOverlappingRatio, "kMinOverlappingRatio"},
{kRoundRobin, "kRoundRobin"}};
std::map<CompactionStopStyle, std::string>
OptionsHelper::compaction_stop_style_to_string = {
{kCompactionStopStyleSimilarSize, "kCompactionStopStyleSimilarSize"},
{kCompactionStopStyleTotalSize, "kCompactionStopStyleTotalSize"}};
std::map<Temperature, std::string> OptionsHelper::temperature_to_string = {
{Temperature::kUnknown, "kUnknown"},
{Temperature::kHot, "kHot"},
{Temperature::kWarm, "kWarm"},
{Temperature::kCold, "kCold"}};
std::unordered_map<std::string, ChecksumType>
OptionsHelper::checksum_type_string_map = {{"kNoChecksum", kNoChecksum},
{"kCRC32c", kCRC32c},
{"kxxHash", kxxHash},
{"kxxHash64", kxxHash64},
{"kXXH3", kXXH3}};
std::unordered_map<std::string, CompressionType>
OptionsHelper::compression_type_string_map = {
{"kNoCompression", kNoCompression},
{"kSnappyCompression", kSnappyCompression},
{"kZlibCompression", kZlibCompression},
{"kBZip2Compression", kBZip2Compression},
{"kLZ4Compression", kLZ4Compression},
{"kLZ4HCCompression", kLZ4HCCompression},
{"kXpressCompression", kXpressCompression},
{"kZSTD", kZSTD},
{"kCustomCompression80", kCustomCompression80},
{"kCustomCompression81", kCustomCompression81},
{"kCustomCompression82", kCustomCompression82},
{"kCustomCompression83", kCustomCompression83},
{"kCustomCompression84", kCustomCompression84},
{"kCustomCompression85", kCustomCompression85},
{"kCustomCompression86", kCustomCompression86},
{"kCustomCompression87", kCustomCompression87},
{"kCustomCompression88", kCustomCompression88},
{"kCustomCompression89", kCustomCompression89},
{"kCustomCompression8A", kCustomCompression8A},
{"kCustomCompression8B", kCustomCompression8B},
{"kCustomCompression8C", kCustomCompression8C},
{"kCustomCompression8D", kCustomCompression8D},
{"kCustomCompression8E", kCustomCompression8E},
{"kCustomCompression8F", kCustomCompression8F},
{"kCustomCompression90", kCustomCompression90},
{"kCustomCompression91", kCustomCompression91},
{"kCustomCompression92", kCustomCompression92},
{"kCustomCompression93", kCustomCompression93},
{"kCustomCompression94", kCustomCompression94},
{"kCustomCompression95", kCustomCompression95},
{"kCustomCompression96", kCustomCompression96},
{"kCustomCompression97", kCustomCompression97},
{"kCustomCompression98", kCustomCompression98},
{"kCustomCompression99", kCustomCompression99},
{"kCustomCompression9A", kCustomCompression9A},
{"kCustomCompression9B", kCustomCompression9B},
{"kCustomCompression9C", kCustomCompression9C},
{"kCustomCompression9D", kCustomCompression9D},
{"kCustomCompression9E", kCustomCompression9E},
{"kCustomCompression9F", kCustomCompression9F},
{"kCustomCompressionA0", kCustomCompressionA0},
{"kCustomCompressionA1", kCustomCompressionA1},
{"kCustomCompressionA2", kCustomCompressionA2},
{"kCustomCompressionA3", kCustomCompressionA3},
{"kCustomCompressionA4", kCustomCompressionA4},
{"kCustomCompressionA5", kCustomCompressionA5},
{"kCustomCompressionA6", kCustomCompressionA6},
{"kCustomCompressionA7", kCustomCompressionA7},
{"kCustomCompressionA8", kCustomCompressionA8},
{"kCustomCompressionA9", kCustomCompressionA9},
{"kCustomCompressionAA", kCustomCompressionAA},
{"kCustomCompressionAB", kCustomCompressionAB},
{"kCustomCompressionAC", kCustomCompressionAC},
{"kCustomCompressionAD", kCustomCompressionAD},
{"kCustomCompressionAE", kCustomCompressionAE},
{"kCustomCompressionAF", kCustomCompressionAF},
{"kCustomCompressionB0", kCustomCompressionB0},
{"kCustomCompressionB1", kCustomCompressionB1},
{"kCustomCompressionB2", kCustomCompressionB2},
{"kCustomCompressionB3", kCustomCompressionB3},
{"kCustomCompressionB4", kCustomCompressionB4},
{"kCustomCompressionB5", kCustomCompressionB5},
{"kCustomCompressionB6", kCustomCompressionB6},
{"kCustomCompressionB7", kCustomCompressionB7},
{"kCustomCompressionB8", kCustomCompressionB8},
{"kCustomCompressionB9", kCustomCompressionB9},
{"kCustomCompressionBA", kCustomCompressionBA},
{"kCustomCompressionBB", kCustomCompressionBB},
{"kCustomCompressionBC", kCustomCompressionBC},
{"kCustomCompressionBD", kCustomCompressionBD},
{"kCustomCompressionBE", kCustomCompressionBE},
{"kCustomCompressionBF", kCustomCompressionBF},
{"kCustomCompressionC0", kCustomCompressionC0},
{"kCustomCompressionC1", kCustomCompressionC1},
{"kCustomCompressionC2", kCustomCompressionC2},
{"kCustomCompressionC3", kCustomCompressionC3},
{"kCustomCompressionC4", kCustomCompressionC4},
{"kCustomCompressionC5", kCustomCompressionC5},
{"kCustomCompressionC6", kCustomCompressionC6},
{"kCustomCompressionC7", kCustomCompressionC7},
{"kCustomCompressionC8", kCustomCompressionC8},
{"kCustomCompressionC9", kCustomCompressionC9},
{"kCustomCompressionCA", kCustomCompressionCA},
{"kCustomCompressionCB", kCustomCompressionCB},
{"kCustomCompressionCC", kCustomCompressionCC},
{"kCustomCompressionCD", kCustomCompressionCD},
{"kCustomCompressionCE", kCustomCompressionCE},
{"kCustomCompressionCF", kCustomCompressionCF},
{"kCustomCompressionD0", kCustomCompressionD0},
{"kCustomCompressionD1", kCustomCompressionD1},
{"kCustomCompressionD2", kCustomCompressionD2},
{"kCustomCompressionD3", kCustomCompressionD3},
{"kCustomCompressionD4", kCustomCompressionD4},
{"kCustomCompressionD5", kCustomCompressionD5},
{"kCustomCompressionD6", kCustomCompressionD6},
{"kCustomCompressionD7", kCustomCompressionD7},
{"kCustomCompressionD8", kCustomCompressionD8},
{"kCustomCompressionD9", kCustomCompressionD9},
{"kCustomCompressionDA", kCustomCompressionDA},
{"kCustomCompressionDB", kCustomCompressionDB},
{"kCustomCompressionDC", kCustomCompressionDC},
{"kCustomCompressionDD", kCustomCompressionDD},
{"kCustomCompressionDE", kCustomCompressionDE},
{"kCustomCompressionDF", kCustomCompressionDF},
{"kCustomCompressionE0", kCustomCompressionE0},
{"kCustomCompressionE1", kCustomCompressionE1},
{"kCustomCompressionE2", kCustomCompressionE2},
{"kCustomCompressionE3", kCustomCompressionE3},
{"kCustomCompressionE4", kCustomCompressionE4},
{"kCustomCompressionE5", kCustomCompressionE5},
{"kCustomCompressionE6", kCustomCompressionE6},
{"kCustomCompressionE7", kCustomCompressionE7},
{"kCustomCompressionE8", kCustomCompressionE8},
{"kCustomCompressionE9", kCustomCompressionE9},
{"kCustomCompressionEA", kCustomCompressionEA},
{"kCustomCompressionEB", kCustomCompressionEB},
{"kCustomCompressionEC", kCustomCompressionEC},
{"kCustomCompressionED", kCustomCompressionED},
{"kCustomCompressionEE", kCustomCompressionEE},
{"kCustomCompressionEF", kCustomCompressionEF},
{"kCustomCompressionF0", kCustomCompressionF0},
{"kCustomCompressionF1", kCustomCompressionF1},
{"kCustomCompressionF2", kCustomCompressionF2},
{"kCustomCompressionF3", kCustomCompressionF3},
{"kCustomCompressionF4", kCustomCompressionF4},
{"kCustomCompressionF5", kCustomCompressionF5},
{"kCustomCompressionF6", kCustomCompressionF6},
{"kCustomCompressionF7", kCustomCompressionF7},
{"kCustomCompressionF8", kCustomCompressionF8},
{"kCustomCompressionF9", kCustomCompressionF9},
{"kCustomCompressionFA", kCustomCompressionFA},
{"kCustomCompressionFB", kCustomCompressionFB},
{"kCustomCompressionFC", kCustomCompressionFC},
{"kCustomCompressionFD", kCustomCompressionFD},
{"kCustomCompressionFE", kCustomCompressionFE},
{"kDisableCompressionOption", kDisableCompressionOption}};
const std::vector<CompressionType>& GetSupportedCompressions() {
static std::vector<CompressionType> supported_compressions = []() {
// std::set internally to deduplicate potential name aliases
std::set<CompressionType> comp_set;
for (const auto& comp_to_name :
OptionsHelper::compression_type_string_map) {
CompressionType t = comp_to_name.second;
if (t != kDisableCompressionOption && CompressionTypeSupported(t)) {
comp_set.insert(t);
}
}
return std::vector<CompressionType>(comp_set.begin(), comp_set.end());
}();
return supported_compressions;
}
const std::vector<CompressionType>& GetSupportedDictCompressions() {
static std::vector<CompressionType> supported_dict_compressions = []() {
std::set<CompressionType> comp_set;
for (const auto& comp_to_name :
OptionsHelper::compression_type_string_map) {
CompressionType t = comp_to_name.second;
if (t != kDisableCompressionOption && DictCompressionTypeSupported(t)) {
comp_set.insert(t);
}
}
return std::vector<CompressionType>(comp_set.begin(), comp_set.end());
}();
return supported_dict_compressions;
}
const std::vector<ChecksumType>& GetSupportedChecksums() {
static std::vector<ChecksumType> supported_checksums = []() {
std::set<ChecksumType> checksum_types;
for (const auto& e : OptionsHelper::checksum_type_string_map) {
checksum_types.insert(e.second);
}
return std::vector<ChecksumType>(checksum_types.begin(),
checksum_types.end());
}();
return supported_checksums;
}
static bool ParseOptionHelper(void* opt_address, const OptionType& opt_type,
const std::string& value) {
switch (opt_type) {
case OptionType::kBoolean:
*static_cast<bool*>(opt_address) = ParseBoolean("", value);
break;
case OptionType::kInt:
*static_cast<int*>(opt_address) = ParseInt(value);
break;
case OptionType::kInt32T:
*static_cast<int32_t*>(opt_address) = ParseInt32(value);
break;
case OptionType::kInt64T:
PutUnaligned(static_cast<int64_t*>(opt_address), ParseInt64(value));
break;
case OptionType::kUInt:
*static_cast<unsigned int*>(opt_address) = ParseUint32(value);
break;
case OptionType::kUInt8T:
*static_cast<uint8_t*>(opt_address) = ParseUint8(value);
break;
case OptionType::kUInt32T:
*static_cast<uint32_t*>(opt_address) = ParseUint32(value);
break;
case OptionType::kUInt64T:
PutUnaligned(static_cast<uint64_t*>(opt_address), ParseUint64(value));
break;
case OptionType::kSizeT:
PutUnaligned(static_cast<size_t*>(opt_address), ParseSizeT(value));
break;
case OptionType::kAtomicInt:
static_cast<std::atomic<int>*>(opt_address)
->store(ParseInt(value), std::memory_order_release);
break;
case OptionType::kString:
*static_cast<std::string*>(opt_address) = value;
break;
case OptionType::kDouble:
*static_cast<double*>(opt_address) = ParseDouble(value);
break;
case OptionType::kCompactionStyle:
return ParseEnum<CompactionStyle>(
compaction_style_string_map, value,
static_cast<CompactionStyle*>(opt_address));
case OptionType::kCompactionPri:
return ParseEnum<CompactionPri>(compaction_pri_string_map, value,
static_cast<CompactionPri*>(opt_address));
case OptionType::kCompressionType:
return ParseEnum<CompressionType>(
compression_type_string_map, value,
static_cast<CompressionType*>(opt_address));
case OptionType::kChecksumType:
return ParseEnum<ChecksumType>(checksum_type_string_map, value,
static_cast<ChecksumType*>(opt_address));
case OptionType::kEncodingType:
return ParseEnum<EncodingType>(encoding_type_string_map, value,
static_cast<EncodingType*>(opt_address));
case OptionType::kCompactionStopStyle:
return ParseEnum<CompactionStopStyle>(
compaction_stop_style_string_map, value,
static_cast<CompactionStopStyle*>(opt_address));
case OptionType::kEncodedString: {
std::string* output_addr = static_cast<std::string*>(opt_address);
(Slice(value)).DecodeHex(output_addr);
break;
}
case OptionType::kTemperature: {
return ParseEnum<Temperature>(temperature_string_map, value,
static_cast<Temperature*>(opt_address));
}
default:
return false;
}
return true;
}
bool SerializeSingleOptionHelper(const void* opt_address,
const OptionType opt_type,
std::string* value) {
assert(value);
switch (opt_type) {
case OptionType::kBoolean:
*value = *(static_cast<const bool*>(opt_address)) ? "true" : "false";
break;
case OptionType::kInt:
*value = std::to_string(*(static_cast<const int*>(opt_address)));
break;
case OptionType::kInt32T:
*value = std::to_string(*(static_cast<const int32_t*>(opt_address)));
break;
case OptionType::kInt64T: {
int64_t v;
GetUnaligned(static_cast<const int64_t*>(opt_address), &v);
*value = std::to_string(v);
} break;
case OptionType::kUInt:
*value = std::to_string(*(static_cast<const unsigned int*>(opt_address)));
break;
case OptionType::kUInt8T:
*value = std::to_string(*(static_cast<const uint8_t*>(opt_address)));
break;
case OptionType::kUInt32T:
*value = std::to_string(*(static_cast<const uint32_t*>(opt_address)));
break;
case OptionType::kUInt64T: {
uint64_t v;
GetUnaligned(static_cast<const uint64_t*>(opt_address), &v);
*value = std::to_string(v);
} break;
case OptionType::kSizeT: {
size_t v;
GetUnaligned(static_cast<const size_t*>(opt_address), &v);
*value = std::to_string(v);
} break;
case OptionType::kDouble:
*value = std::to_string(*(static_cast<const double*>(opt_address)));
break;
case OptionType::kAtomicInt:
*value = std::to_string(static_cast<const std::atomic<int>*>(opt_address)
->load(std::memory_order_acquire));
break;
case OptionType::kString:
*value =
EscapeOptionString(*(static_cast<const std::string*>(opt_address)));
break;
case OptionType::kCompactionStyle:
return SerializeEnum<CompactionStyle>(
compaction_style_string_map,
*(static_cast<const CompactionStyle*>(opt_address)), value);
case OptionType::kCompactionPri:
return SerializeEnum<CompactionPri>(
compaction_pri_string_map,
*(static_cast<const CompactionPri*>(opt_address)), value);
case OptionType::kCompressionType:
return SerializeEnum<CompressionType>(
compression_type_string_map,
*(static_cast<const CompressionType*>(opt_address)), value);
case OptionType::kChecksumType:
return SerializeEnum<ChecksumType>(
checksum_type_string_map,
*static_cast<const ChecksumType*>(opt_address), value);
case OptionType::kEncodingType:
return SerializeEnum<EncodingType>(
encoding_type_string_map,
*static_cast<const EncodingType*>(opt_address), value);
case OptionType::kCompactionStopStyle:
return SerializeEnum<CompactionStopStyle>(
compaction_stop_style_string_map,
*static_cast<const CompactionStopStyle*>(opt_address), value);
case OptionType::kEncodedString: {
const auto* ptr = static_cast<const std::string*>(opt_address);
*value = (Slice(*ptr)).ToString(true);
break;
}
case OptionType::kTemperature: {
return SerializeEnum<Temperature>(
temperature_string_map, *static_cast<const Temperature*>(opt_address),
value);
}
default:
return false;
}
return true;
}
template <typename T>
Status ConfigureFromMap(
const ConfigOptions& config_options,
const std::unordered_map<std::string, std::string>& opt_map,
const std::string& option_name, Configurable* config, T* new_opts) {
Status s = config->ConfigureFromMap(config_options, opt_map);
if (s.ok()) {
*new_opts = *(config->GetOptions<T>(option_name));
}
return s;
}
Status StringToMap(const std::string& opts_str,
std::unordered_map<std::string, std::string>* opts_map) {
assert(opts_map);
// Example:
// opts_str = "write_buffer_size=1024;max_write_buffer_number=2;"
// "nested_opt={opt1=1;opt2=2};max_bytes_for_level_base=100"
size_t pos = 0;
std::string opts = trim(opts_str);
// If the input string starts and ends with "{...}", strip off the brackets
while (opts.size() > 2 && opts[0] == '{' && opts[opts.size() - 1] == '}') {
opts = trim(opts.substr(1, opts.size() - 2));
}
while (pos < opts.size()) {
size_t eq_pos = opts.find_first_of("={};", pos);
if (eq_pos == std::string::npos) {
return Status::InvalidArgument("Mismatched key value pair, '=' expected");
} else if (opts[eq_pos] != '=') {
return Status::InvalidArgument("Unexpected char in key");
}
std::string key = trim(opts.substr(pos, eq_pos - pos));
if (key.empty()) {
return Status::InvalidArgument("Empty key found");
}
std::string value;
Status s = OptionTypeInfo::NextToken(opts, ';', eq_pos + 1, &pos, &value);
if (!s.ok()) {
return s;
} else {
(*opts_map)[key] = value;
if (pos == std::string::npos) {
break;
} else {
pos++;
}
}
}
return Status::OK();
}
Status GetStringFromDBOptions(std::string* opt_string,
const DBOptions& db_options,
const std::string& delimiter) {
ConfigOptions config_options(db_options);
config_options.delimiter = delimiter;
return GetStringFromDBOptions(config_options, db_options, opt_string);
}
Status GetStringFromDBOptions(const ConfigOptions& config_options,
const DBOptions& db_options,
std::string* opt_string) {
assert(opt_string);
opt_string->clear();
auto config = DBOptionsAsConfigurable(db_options);
return config->GetOptionString(config_options, opt_string);
}
Status GetStringFromColumnFamilyOptions(std::string* opt_string,
const ColumnFamilyOptions& cf_options,
const std::string& delimiter) {
ConfigOptions config_options;
config_options.delimiter = delimiter;
return GetStringFromColumnFamilyOptions(config_options, cf_options,
opt_string);
}
Status GetStringFromColumnFamilyOptions(const ConfigOptions& config_options,
const ColumnFamilyOptions& cf_options,
std::string* opt_string) {
const auto config = CFOptionsAsConfigurable(cf_options);
return config->GetOptionString(config_options, opt_string);
}
Status GetStringFromCompressionType(std::string* compression_str,
CompressionType compression_type) {
bool ok = SerializeEnum<CompressionType>(compression_type_string_map,
compression_type, compression_str);
if (ok) {
return Status::OK();
} else {
return Status::InvalidArgument("Invalid compression types");
}
}
Status GetColumnFamilyOptionsFromMap(
const ConfigOptions& config_options,
const ColumnFamilyOptions& base_options,
const std::unordered_map<std::string, std::string>& opts_map,
ColumnFamilyOptions* new_options) {
assert(new_options);
*new_options = base_options;
const auto config = CFOptionsAsConfigurable(base_options);
Status s = ConfigureFromMap<ColumnFamilyOptions>(
config_options, opts_map, OptionsHelper::kCFOptionsName, config.get(),
new_options);
// Translate any errors (NotFound, NotSupported, to InvalidArgument
if (s.ok() || s.IsInvalidArgument()) {
return s;
} else {
return Status::InvalidArgument(s.getState());
}
}
Status GetColumnFamilyOptionsFromString(const ConfigOptions& config_options,
const ColumnFamilyOptions& base_options,
const std::string& opts_str,
ColumnFamilyOptions* new_options) {
std::unordered_map<std::string, std::string> opts_map;
Status s = StringToMap(opts_str, &opts_map);
if (!s.ok()) {
*new_options = base_options;
return s;
}
return GetColumnFamilyOptionsFromMap(config_options, base_options, opts_map,
new_options);
}
Status GetDBOptionsFromMap(
const ConfigOptions& config_options, const DBOptions& base_options,
const std::unordered_map<std::string, std::string>& opts_map,
DBOptions* new_options) {
assert(new_options);
*new_options = base_options;
auto config = DBOptionsAsConfigurable(base_options);
Status s = ConfigureFromMap<DBOptions>(config_options, opts_map,
OptionsHelper::kDBOptionsName,
config.get(), new_options);
// Translate any errors (NotFound, NotSupported, to InvalidArgument
if (s.ok() || s.IsInvalidArgument()) {
return s;
} else {
return Status::InvalidArgument(s.getState());
}
}
Status GetDBOptionsFromString(const ConfigOptions& config_options,
const DBOptions& base_options,
const std::string& opts_str,
DBOptions* new_options) {
std::unordered_map<std::string, std::string> opts_map;
Status s = StringToMap(opts_str, &opts_map);
if (!s.ok()) {
*new_options = base_options;
return s;
}
return GetDBOptionsFromMap(config_options, base_options, opts_map,
new_options);
}
Status GetOptionsFromString(const Options& base_options,
const std::string& opts_str, Options* new_options) {
ConfigOptions config_options(base_options);
config_options.input_strings_escaped = false;
config_options.ignore_unknown_options = false;
return GetOptionsFromString(config_options, base_options, opts_str,
new_options);
}
Status GetOptionsFromString(const ConfigOptions& config_options,
const Options& base_options,
const std::string& opts_str, Options* new_options) {
ColumnFamilyOptions new_cf_options;
std::unordered_map<std::string, std::string> unused_opts;
std::unordered_map<std::string, std::string> opts_map;
assert(new_options);
*new_options = base_options;
Status s = StringToMap(opts_str, &opts_map);
if (!s.ok()) {
return s;
}
auto config = DBOptionsAsConfigurable(base_options);
s = config->ConfigureFromMap(config_options, opts_map, &unused_opts);
if (s.ok()) {
DBOptions* new_db_options =
config->GetOptions<DBOptions>(OptionsHelper::kDBOptionsName);
if (!unused_opts.empty()) {
s = GetColumnFamilyOptionsFromMap(config_options, base_options,
unused_opts, &new_cf_options);
if (s.ok()) {
*new_options = Options(*new_db_options, new_cf_options);
}
} else {
*new_options = Options(*new_db_options, base_options);
}
}
// Translate any errors (NotFound, NotSupported, to InvalidArgument
if (s.ok() || s.IsInvalidArgument()) {
return s;
} else {
return Status::InvalidArgument(s.getState());
}
}
std::unordered_map<std::string, EncodingType>
OptionsHelper::encoding_type_string_map = {{"kPlain", kPlain},
{"kPrefix", kPrefix}};
std::unordered_map<std::string, CompactionStyle>
OptionsHelper::compaction_style_string_map = {
{"kCompactionStyleLevel", kCompactionStyleLevel},
{"kCompactionStyleUniversal", kCompactionStyleUniversal},
{"kCompactionStyleFIFO", kCompactionStyleFIFO},
{"kCompactionStyleNone", kCompactionStyleNone}};
std::unordered_map<std::string, CompactionPri>
OptionsHelper::compaction_pri_string_map = {
{"kByCompensatedSize", kByCompensatedSize},
{"kOldestLargestSeqFirst", kOldestLargestSeqFirst},
{"kOldestSmallestSeqFirst", kOldestSmallestSeqFirst},
{"kMinOverlappingRatio", kMinOverlappingRatio},
{"kRoundRobin", kRoundRobin}};
std::unordered_map<std::string, CompactionStopStyle>
OptionsHelper::compaction_stop_style_string_map = {
{"kCompactionStopStyleSimilarSize", kCompactionStopStyleSimilarSize},
{"kCompactionStopStyleTotalSize", kCompactionStopStyleTotalSize}};
std::unordered_map<std::string, Temperature>
OptionsHelper::temperature_string_map = {
{"kUnknown", Temperature::kUnknown},
{"kHot", Temperature::kHot},
{"kWarm", Temperature::kWarm},
{"kCold", Temperature::kCold}};
std::unordered_map<std::string, PrepopulateBlobCache>
OptionsHelper::prepopulate_blob_cache_string_map = {
{"kDisable", PrepopulateBlobCache::kDisable},
{"kFlushOnly", PrepopulateBlobCache::kFlushOnly}};
Status OptionTypeInfo::NextToken(const std::string& opts, char delimiter,
size_t pos, size_t* end, std::string* token) {
while (pos < opts.size() && isspace(opts[pos])) {
++pos;
}
// Empty value at the end
if (pos >= opts.size()) {
*token = "";
*end = std::string::npos;
return Status::OK();
} else if (opts[pos] == '{') {
int count = 1;
size_t brace_pos = pos + 1;
while (brace_pos < opts.size()) {
if (opts[brace_pos] == '{') {
++count;
} else if (opts[brace_pos] == '}') {
--count;
if (count == 0) {
break;
}
}
++brace_pos;
}
// found the matching closing brace
if (count == 0) {
*token = trim(opts.substr(pos + 1, brace_pos - pos - 1));
// skip all whitespace and move to the next delimiter
// brace_pos points to the next position after the matching '}'
pos = brace_pos + 1;
while (pos < opts.size() && isspace(opts[pos])) {
++pos;
}
if (pos < opts.size() && opts[pos] != delimiter) {
return Status::InvalidArgument("Unexpected chars after nested options");
}
*end = pos;
} else {
return Status::InvalidArgument(
"Mismatched curly braces for nested options");
}
} else {
*end = opts.find(delimiter, pos);
if (*end == std::string::npos) {
// It either ends with a trailing semi-colon or the last key-value pair
*token = trim(opts.substr(pos));
} else {
*token = trim(opts.substr(pos, *end - pos));
}
}
return Status::OK();
}
Status OptionTypeInfo::Parse(const ConfigOptions& config_options,
const std::string& opt_name,
const std::string& value, void* opt_ptr) const {
if (IsDeprecated()) {
return Status::OK();
}
try {
const std::string& opt_value = config_options.input_strings_escaped
? UnescapeOptionString(value)
: value;
if (opt_ptr == nullptr) {
return Status::NotFound("Nullptr option", opt_name);
} else if (parse_func_ != nullptr) {
ConfigOptions copy = config_options;
copy.invoke_prepare_options = false;
void* opt_addr = GetOffset(opt_ptr);
return parse_func_(copy, opt_name, opt_value, opt_addr);
} else if (ParseOptionHelper(GetOffset(opt_ptr), type_, opt_value)) {
return Status::OK();
} else if (IsConfigurable()) {
// The option is <config>.<name>
Configurable* config = AsRawPointer<Configurable>(opt_ptr);
if (opt_value.empty()) {
return Status::OK();
} else if (config == nullptr) {
return Status::NotFound("Could not find configurable: ", opt_name);
} else {
ConfigOptions copy = config_options;
copy.ignore_unknown_options = false;
copy.invoke_prepare_options = false;
if (opt_value.find('=') != std::string::npos) {
return config->ConfigureFromString(copy, opt_value);
} else {
return config->ConfigureOption(copy, opt_name, opt_value);
}
}
} else if (IsByName()) {
return Status::NotSupported("Deserializing the option " + opt_name +
" is not supported");
} else {
return Status::InvalidArgument("Error parsing:", opt_name);
}
} catch (std::exception& e) {
return Status::InvalidArgument("Error parsing " + opt_name + ":" +
std::string(e.what()));
}
}
Status OptionTypeInfo::ParseType(
const ConfigOptions& config_options, const std::string& opts_str,
const std::unordered_map<std::string, OptionTypeInfo>& type_map,
void* opt_addr, std::unordered_map<std::string, std::string>* unused) {
std::unordered_map<std::string, std::string> opts_map;
Status status = StringToMap(opts_str, &opts_map);
if (!status.ok()) {
return status;
} else {
return ParseType(config_options, opts_map, type_map, opt_addr, unused);
}
}
Status OptionTypeInfo::ParseType(
const ConfigOptions& config_options,
const std::unordered_map<std::string, std::string>& opts_map,
const std::unordered_map<std::string, OptionTypeInfo>& type_map,
void* opt_addr, std::unordered_map<std::string, std::string>* unused) {
for (const auto& opts_iter : opts_map) {
std::string opt_name;
const auto* opt_info = Find(opts_iter.first, type_map, &opt_name);
if (opt_info != nullptr) {
Status status =
opt_info->Parse(config_options, opt_name, opts_iter.second, opt_addr);
if (!status.ok()) {
return status;
}
} else if (unused != nullptr) {
(*unused)[opts_iter.first] = opts_iter.second;
} else if (!config_options.ignore_unknown_options) {
return Status::NotFound("Unrecognized option", opts_iter.first);
}
}
return Status::OK();
}
Status OptionTypeInfo::ParseStruct(
const ConfigOptions& config_options, const std::string& struct_name,
const std::unordered_map<std::string, OptionTypeInfo>* struct_map,
const std::string& opt_name, const std::string& opt_value, void* opt_addr) {
assert(struct_map);
Status status;
if (opt_name == struct_name || EndsWith(opt_name, "." + struct_name)) {
// This option represents the entire struct
std::unordered_map<std::string, std::string> unused;
status =
ParseType(config_options, opt_value, *struct_map, opt_addr, &unused);
if (status.ok() && !unused.empty() &&
!config_options.ignore_unknown_options) {
status = Status::InvalidArgument(
"Unrecognized option", struct_name + "." + unused.begin()->first);
}
} else if (StartsWith(opt_name, struct_name + ".")) {
// This option represents a nested field in the struct (e.g, struct.field)
std::string elem_name;
const auto opt_info =
Find(opt_name.substr(struct_name.size() + 1), *struct_map, &elem_name);
if (opt_info != nullptr) {
status = opt_info->Parse(config_options, elem_name, opt_value, opt_addr);
} else if (!config_options.ignore_unknown_options) {
status = Status::InvalidArgument("Unrecognized option", opt_name);
}
} else {
// This option represents a field in the struct (e.g. field)
std::string elem_name;
const auto opt_info = Find(opt_name, *struct_map, &elem_name);
if (opt_info != nullptr) {
status = opt_info->Parse(config_options, elem_name, opt_value, opt_addr);
} else if (!config_options.ignore_unknown_options) {
status = Status::InvalidArgument("Unrecognized option",
struct_name + "." + opt_name);
}
}
return status;
}
Status OptionTypeInfo::Serialize(const ConfigOptions& config_options,
const std::string& opt_name,
const void* const opt_ptr,
std::string* opt_value) const {
// If the option is no longer used in rocksdb and marked as deprecated,
// we skip it in the serialization.
if (opt_ptr == nullptr || IsDeprecated()) {
return Status::OK();
} else if (IsEnabled(OptionTypeFlags::kDontSerialize)) {
return Status::NotSupported("Cannot serialize option: ", opt_name);
} else if (serialize_func_ != nullptr) {
const void* opt_addr = GetOffset(opt_ptr);
return serialize_func_(config_options, opt_name, opt_addr, opt_value);
} else if (IsCustomizable()) {
const Customizable* custom = AsRawPointer<Customizable>(opt_ptr);
opt_value->clear();
if (custom == nullptr) {
// We do not have a custom object to serialize.
// If the option is not mutable and we are doing only mutable options,
// we return an empty string (which will cause the option not to be
// printed). Otherwise, we return the "nullptr" string, which will result
// in "option=nullptr" being printed.
if (IsMutable() || !config_options.mutable_options_only) {
*opt_value = kNullptrString;
} else {
*opt_value = "";
}
} else if (IsEnabled(OptionTypeFlags::kStringNameOnly) &&
!config_options.IsDetailed()) {
if (!config_options.mutable_options_only || IsMutable()) {
*opt_value = custom->GetId();
}
} else {
ConfigOptions embedded = config_options;
embedded.delimiter = ";";
// If this option is mutable, everything inside it should be considered
// mutable
if (IsMutable()) {
embedded.mutable_options_only = false;
}
std::string value = custom->ToString(embedded);
if (!embedded.mutable_options_only ||
value.find('=') != std::string::npos) {
*opt_value = value;
} else {
*opt_value = "";
}
}
return Status::OK();
} else if (IsConfigurable()) {
const Configurable* config = AsRawPointer<Configurable>(opt_ptr);
if (config != nullptr) {
ConfigOptions embedded = config_options;
embedded.delimiter = ";";
*opt_value = config->ToString(embedded);
}
return Status::OK();
} else if (config_options.mutable_options_only && !IsMutable()) {
return Status::OK();
} else if (SerializeSingleOptionHelper(GetOffset(opt_ptr), type_,
opt_value)) {
return Status::OK();
} else {
return Status::InvalidArgument("Cannot serialize option: ", opt_name);
}
}
Status OptionTypeInfo::SerializeType(
const ConfigOptions& config_options,
const std::unordered_map<std::string, OptionTypeInfo>& type_map,
const void* opt_addr, std::string* result) {
Status status;
for (const auto& iter : type_map) {
std::string single;
const auto& opt_info = iter.second;
if (opt_info.ShouldSerialize()) {
status =
opt_info.Serialize(config_options, iter.first, opt_addr, &single);
if (!status.ok()) {
return status;
} else {
result->append(iter.first + "=" + single + config_options.delimiter);
}
}
}
return status;
}
Status OptionTypeInfo::SerializeStruct(
const ConfigOptions& config_options, const std::string& struct_name,
const std::unordered_map<std::string, OptionTypeInfo>* struct_map,
const std::string& opt_name, const void* opt_addr, std::string* value) {
assert(struct_map);
Status status;
if (EndsWith(opt_name, struct_name)) {
// We are going to write the struct as "{ prop1=value1; prop2=value2;}.
// Set the delimiter to ";" so that the everything will be on one line.
ConfigOptions embedded = config_options;
embedded.delimiter = ";";
// This option represents the entire struct
std::string result;
status = SerializeType(embedded, *struct_map, opt_addr, &result);
if (!status.ok()) {
return status;
} else {
*value = "{" + result + "}";
}
} else if (StartsWith(opt_name, struct_name + ".")) {
// This option represents a nested field in the struct (e.g, struct.field)
std::string elem_name;
const auto opt_info =
Find(opt_name.substr(struct_name.size() + 1), *struct_map, &elem_name);
if (opt_info != nullptr) {
status = opt_info->Serialize(config_options, elem_name, opt_addr, value);
} else {
status = Status::InvalidArgument("Unrecognized option", opt_name);
}
} else {
// This option represents a field in the struct (e.g. field)
std::string elem_name;
const auto opt_info = Find(opt_name, *struct_map, &elem_name);
if (opt_info == nullptr) {
status = Status::InvalidArgument("Unrecognized option", opt_name);
} else if (opt_info->ShouldSerialize()) {
status = opt_info->Serialize(config_options, opt_name + "." + elem_name,
opt_addr, value);
}
}
return status;
}
template <typename T>
bool IsOptionEqual(const void* offset1, const void* offset2) {
return (*static_cast<const T*>(offset1) == *static_cast<const T*>(offset2));
}
static bool AreEqualDoubles(const double a, const double b) {
return (fabs(a - b) < 0.00001);
}
static bool AreOptionsEqual(OptionType type, const void* this_offset,
const void* that_offset) {
switch (type) {
case OptionType::kBoolean:
return IsOptionEqual<bool>(this_offset, that_offset);
case OptionType::kInt:
return IsOptionEqual<int>(this_offset, that_offset);
case OptionType::kUInt:
return IsOptionEqual<unsigned int>(this_offset, that_offset);
case OptionType::kInt32T:
return IsOptionEqual<int32_t>(this_offset, that_offset);
case OptionType::kInt64T: {
int64_t v1, v2;
GetUnaligned(static_cast<const int64_t*>(this_offset), &v1);
GetUnaligned(static_cast<const int64_t*>(that_offset), &v2);
return (v1 == v2);
}
case OptionType::kUInt8T:
return IsOptionEqual<uint8_t>(this_offset, that_offset);
case OptionType::kUInt32T:
return IsOptionEqual<uint32_t>(this_offset, that_offset);
case OptionType::kUInt64T: {
uint64_t v1, v2;
GetUnaligned(static_cast<const uint64_t*>(this_offset), &v1);
GetUnaligned(static_cast<const uint64_t*>(that_offset), &v2);
return (v1 == v2);
}
case OptionType::kSizeT: {
size_t v1, v2;
GetUnaligned(static_cast<const size_t*>(this_offset), &v1);
GetUnaligned(static_cast<const size_t*>(that_offset), &v2);
return (v1 == v2);
}
case OptionType::kAtomicInt:
return IsOptionEqual<std::atomic<int>>(this_offset, that_offset);
case OptionType::kString:
return IsOptionEqual<std::string>(this_offset, that_offset);
case OptionType::kDouble:
return AreEqualDoubles(*static_cast<const double*>(this_offset),
*static_cast<const double*>(that_offset));
case OptionType::kCompactionStyle:
return IsOptionEqual<CompactionStyle>(this_offset, that_offset);
case OptionType::kCompactionStopStyle:
return IsOptionEqual<CompactionStopStyle>(this_offset, that_offset);
case OptionType::kCompactionPri:
return IsOptionEqual<CompactionPri>(this_offset, that_offset);
case OptionType::kCompressionType:
return IsOptionEqual<CompressionType>(this_offset, that_offset);
case OptionType::kChecksumType:
return IsOptionEqual<ChecksumType>(this_offset, that_offset);
case OptionType::kEncodingType:
return IsOptionEqual<EncodingType>(this_offset, that_offset);
case OptionType::kEncodedString:
return IsOptionEqual<std::string>(this_offset, that_offset);
case OptionType::kTemperature:
return IsOptionEqual<Temperature>(this_offset, that_offset);
default:
return false;
} // End switch
}
bool OptionTypeInfo::AreEqual(const ConfigOptions& config_options,
const std::string& opt_name,
const void* const this_ptr,
const void* const that_ptr,
std::string* mismatch) const {
auto level = GetSanityLevel();
if (!config_options.IsCheckEnabled(level)) {
return true; // If the sanity level is not being checked, skip it
}
if (this_ptr == nullptr || that_ptr == nullptr) {
if (this_ptr == that_ptr) {
return true;
}
} else if (equals_func_ != nullptr) {
const void* this_addr = GetOffset(this_ptr);
const void* that_addr = GetOffset(that_ptr);
if (equals_func_(config_options, opt_name, this_addr, that_addr,
mismatch)) {
return true;
}
} else {
const void* this_addr = GetOffset(this_ptr);
const void* that_addr = GetOffset(that_ptr);
if (AreOptionsEqual(type_, this_addr, that_addr)) {
return true;
} else if (IsConfigurable()) {
const auto* this_config = AsRawPointer<Configurable>(this_ptr);
const auto* that_config = AsRawPointer<Configurable>(that_ptr);
if (this_config == that_config) {
return true;
} else if (this_config != nullptr && that_config != nullptr) {
std::string bad_name;
bool matches;
if (level < config_options.sanity_level) {
ConfigOptions copy = config_options;
copy.sanity_level = level;
matches = this_config->AreEquivalent(copy, that_config, &bad_name);
} else {
matches = this_config->AreEquivalent(config_options, that_config,
&bad_name);
}
if (!matches) {
*mismatch = opt_name + "." + bad_name;
}
return matches;
}
}
}
if (mismatch->empty()) {
*mismatch = opt_name;
}
return false;
}
bool OptionTypeInfo::TypesAreEqual(
const ConfigOptions& config_options,
const std::unordered_map<std::string, OptionTypeInfo>& type_map,
const void* this_addr, const void* that_addr, std::string* mismatch) {
for (const auto& iter : type_map) {
const auto& opt_info = iter.second;
if (!opt_info.AreEqual(config_options, iter.first, this_addr, that_addr,
mismatch)) {
return false;
}
}
return true;
}
bool OptionTypeInfo::StructsAreEqual(
const ConfigOptions& config_options, const std::string& struct_name,
const std::unordered_map<std::string, OptionTypeInfo>* struct_map,
const std::string& opt_name, const void* this_addr, const void* that_addr,
std::string* mismatch) {
assert(struct_map);
bool matches = true;
std::string result;
if (EndsWith(opt_name, struct_name)) {
// This option represents the entire struct
matches = TypesAreEqual(config_options, *struct_map, this_addr, that_addr,
&result);
if (!matches) {
*mismatch = struct_name + "." + result;
return false;
}
} else if (StartsWith(opt_name, struct_name + ".")) {
// This option represents a nested field in the struct (e.g, struct.field)
std::string elem_name;
const auto opt_info =
Find(opt_name.substr(struct_name.size() + 1), *struct_map, &elem_name);
assert(opt_info);
if (opt_info == nullptr) {
*mismatch = opt_name;
matches = false;
} else if (!opt_info->AreEqual(config_options, elem_name, this_addr,
that_addr, &result)) {
matches = false;
*mismatch = struct_name + "." + result;
}
} else {
// This option represents a field in the struct (e.g. field)
std::string elem_name;
const auto opt_info = Find(opt_name, *struct_map, &elem_name);
assert(opt_info);
if (opt_info == nullptr) {
*mismatch = struct_name + "." + opt_name;
matches = false;
} else if (!opt_info->AreEqual(config_options, elem_name, this_addr,
that_addr, &result)) {
matches = false;
*mismatch = struct_name + "." + result;
}
}
return matches;
}
bool MatchesOptionsTypeFromMap(
const ConfigOptions& config_options,
const std::unordered_map<std::string, OptionTypeInfo>& type_map,
const void* const this_ptr, const void* const that_ptr,
std::string* mismatch) {
for (auto& pair : type_map) {
// We skip checking deprecated variables as they might
// contain random values since they might not be initialized
if (config_options.IsCheckEnabled(pair.second.GetSanityLevel())) {
if (!pair.second.AreEqual(config_options, pair.first, this_ptr, that_ptr,
mismatch) &&
!pair.second.AreEqualByName(config_options, pair.first, this_ptr,
that_ptr)) {
return false;
}
}
}
return true;
}
bool OptionTypeInfo::AreEqualByName(const ConfigOptions& config_options,
const std::string& opt_name,
const void* const this_ptr,
const void* const that_ptr) const {
if (IsByName()) {
std::string that_value;
if (Serialize(config_options, opt_name, that_ptr, &that_value).ok()) {
return AreEqualByName(config_options, opt_name, this_ptr, that_value);
}
}
return false;
}
bool OptionTypeInfo::AreEqualByName(const ConfigOptions& config_options,
const std::string& opt_name,
const void* const opt_ptr,
const std::string& that_value) const {
std::string this_value;
if (!IsByName()) {
return false;
} else if (!Serialize(config_options, opt_name, opt_ptr, &this_value).ok()) {
return false;
} else if (IsEnabled(OptionVerificationType::kByNameAllowFromNull)) {
if (that_value == kNullptrString) {
return true;
}
} else if (IsEnabled(OptionVerificationType::kByNameAllowNull)) {
if (that_value == kNullptrString) {
return true;
}
}
return (this_value == that_value);
}
Status OptionTypeInfo::Prepare(const ConfigOptions& config_options,
const std::string& name, void* opt_ptr) const {
if (ShouldPrepare()) {
if (prepare_func_ != nullptr) {
void* opt_addr = GetOffset(opt_ptr);
return prepare_func_(config_options, name, opt_addr);
} else if (IsConfigurable()) {
Configurable* config = AsRawPointer<Configurable>(opt_ptr);
if (config != nullptr) {
return config->PrepareOptions(config_options);
} else if (!CanBeNull()) {
return Status::NotFound("Missing configurable object", name);
}
}
}
return Status::OK();
}
Status OptionTypeInfo::Validate(const DBOptions& db_opts,
const ColumnFamilyOptions& cf_opts,
const std::string& name,
const void* opt_ptr) const {
if (ShouldValidate()) {
if (validate_func_ != nullptr) {
const void* opt_addr = GetOffset(opt_ptr);
return validate_func_(db_opts, cf_opts, name, opt_addr);
} else if (IsConfigurable()) {
const Configurable* config = AsRawPointer<Configurable>(opt_ptr);
if (config != nullptr) {
return config->ValidateOptions(db_opts, cf_opts);
} else if (!CanBeNull()) {
return Status::NotFound("Missing configurable object", name);
}
}
}
return Status::OK();
}
const OptionTypeInfo* OptionTypeInfo::Find(
const std::string& opt_name,
const std::unordered_map<std::string, OptionTypeInfo>& opt_map,
std::string* elem_name) {
const auto iter = opt_map.find(opt_name); // Look up the value in the map
if (iter != opt_map.end()) { // Found the option in the map
*elem_name = opt_name; // Return the name
return &(iter->second); // Return the contents of the iterator
} else {
auto idx = opt_name.find('.'); // Look for a separator
if (idx > 0 && idx != std::string::npos) { // We found a separator
auto siter =
opt_map.find(opt_name.substr(0, idx)); // Look for the short name
if (siter != opt_map.end()) { // We found the short name
if (siter->second.IsStruct() || // If the object is a struct
siter->second.IsConfigurable()) { // or a Configurable
*elem_name = opt_name.substr(idx + 1); // Return the rest
return &(siter->second); // Return the contents of the iterator
}
}
}
}
return nullptr;
}
} // namespace ROCKSDB_NAMESPACE
Reference in a new issue View git blame Copy permalink