convert Options from string

Summary: Allow accepting Options as a string of key/value pairs

Test Plan: unit test

Reviewers: yhchiang, sdong, igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D24597
This commit is contained in:
Lei Jin 2014-10-10 10:00:12 -07:00
parent f18b4a4847
commit cd0d581ff5
5 changed files with 285 additions and 103 deletions

View file

@ -1,5 +1,11 @@
# Rocksdb Change Log
## Unreleased
### Public API changes
* Introduce 4 new convenient functions for converting Options from string: GetColumnFamilyOptionsFromMap(), GetColumnFamilyOptionsFromString(), GetDBOptionsFromMap(), GetDBOptionsFromString()
## 3.6.0 (10/7/2014)
### Disk format changes
* If you're using RocksDB on ARM platforms and you're using default bloom filter, there is a disk format change you need to be aware of. There are three steps you need to do when you convert to new release: 1. turn off filter policy, 2. compact the whole database, 3. turn on filter policy

View file

@ -1012,11 +1012,6 @@ extern Options GetOptions(size_t total_write_buffer_limit,
int write_amplification_threshold = 32,
uint64_t target_db_size = 68719476736 /* 64GB */);
bool GetOptionsFromStrings(
const Options& base_options,
const std::unordered_map<std::string, std::string>& options_map,
Options* new_options);
} // namespace rocksdb
#endif // STORAGE_ROCKSDB_INCLUDE_OPTIONS_H_

View file

@ -4,6 +4,7 @@
// of patent rights can be found in the PATENTS file in the same directory.
#include <cassert>
#include <cctype>
#include <unordered_set>
#include "rocksdb/options.h"
#include "util/options_helper.h"
@ -161,13 +162,61 @@ bool GetMutableOptionsFromStrings(
return true;
}
bool GetOptionsFromStrings(
const Options& base_options,
const std::unordered_map<std::string, std::string>& options_map,
Options* new_options) {
namespace {
std::string trim(const std::string& str) {
size_t start = 0;
size_t end = str.size() - 1;
while (isspace(str[start]) != 0 && start <= end) {
++start;
}
while (isspace(str[end]) != 0 && start <= end) {
--end;
}
if (start <= end) {
return str.substr(start, end - start + 1);
}
return std::string();
}
bool 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"
size_t pos = 0;
std::string opts = trim(opts_str);
while (pos < opts.size()) {
size_t eq_pos = opts.find('=', pos);
if (eq_pos == std::string::npos) {
return false;
}
std::string key = trim(opts.substr(pos, eq_pos - pos));
size_t sc_pos = opts.find(';', eq_pos + 1);
if (sc_pos == std::string::npos) {
(*opts_map)[key] = trim(opts.substr(eq_pos + 1));
// It either ends with a trailing semi-colon or the last key-value pair
break;
} else {
(*opts_map)[key] = trim(opts.substr(eq_pos + 1, sc_pos - eq_pos - 1));
}
pos = sc_pos + 1;
}
return true;
}
} // anonymous namespace
bool GetColumnFamilyOptionsFromMap(
const ColumnFamilyOptions& base_options,
const std::unordered_map<std::string, std::string>& opts_map,
ColumnFamilyOptions* new_options) {
assert(new_options);
*new_options = base_options;
for (const auto& o : options_map) {
for (const auto& o : opts_map) {
try {
if (ParseMemtableOptions(o.first, o.second, new_options)) {
} else if (ParseCompactionOptions(o.first, o.second, new_options)) {
@ -247,7 +296,36 @@ bool GetOptionsFromStrings(
new_options->bloom_locality = ParseUint32(o.second);
} else if (o.first == "min_partial_merge_operands") {
new_options->min_partial_merge_operands = ParseUint32(o.second);
} else if (o.first == "create_if_missing") {
} else {
return false;
}
} catch (std::exception) {
return false;
}
}
return true;
}
bool GetColumnFamilyOptionsFromString(
const ColumnFamilyOptions& base_options,
const std::string& opts_str,
ColumnFamilyOptions* new_options) {
std::unordered_map<std::string, std::string> opts_map;
if (!StringToMap(opts_str, &opts_map)) {
return false;
}
return GetColumnFamilyOptionsFromMap(base_options, opts_map, new_options);
}
bool GetDBOptionsFromMap(
const DBOptions& base_options,
const std::unordered_map<std::string, std::string>& opts_map,
DBOptions* new_options) {
assert(new_options);
*new_options = base_options;
for (const auto& o : opts_map) {
try {
if (o.first == "create_if_missing") {
new_options->create_if_missing = ParseBoolean(o.first, o.second);
} else if (o.first == "create_missing_column_families") {
new_options->create_missing_column_families =
@ -325,4 +403,15 @@ bool GetOptionsFromStrings(
return true;
}
bool GetDBOptionsFromString(
const DBOptions& base_options,
const std::string& opts_str,
DBOptions* new_options) {
std::unordered_map<std::string, std::string> opts_map;
if (!StringToMap(opts_str, &opts_map)) {
return false;
}
return GetDBOptionsFromMap(base_options, opts_map, new_options);
}
} // namespace rocksdb

View file

@ -17,6 +17,7 @@
#include "rocksdb/options.h"
#include "util/testharness.h"
#include "utilities/options/convenience.h"
using GFLAGS::ParseCommandLineFlags;
DEFINE_bool(enable_print, false, "Print options generated to console.");
@ -77,8 +78,8 @@ TEST(OptionsTest, LooseCondition) {
PrintAndGetOptions(128 * 1024 * 1024, 4, 8);
}
TEST(OptionsTest, GetOptionsFromStringsTest) {
std::unordered_map<std::string, std::string> options_map = {
TEST(OptionsTest, GetOptionsFromMapTest) {
std::unordered_map<std::string, std::string> cf_options_map = {
{"write_buffer_size", "1"},
{"max_write_buffer_number", "2"},
{"min_write_buffer_number_to_merge", "3"},
@ -120,7 +121,10 @@ TEST(OptionsTest, GetOptionsFromStringsTest) {
{"memtable_prefix_bloom_huge_page_tlb_size", "28"},
{"bloom_locality", "29"},
{"max_successive_merges", "30"},
{"min_partial_merge_operands", "31"},
{"min_partial_merge_operands", "31"}
};
std::unordered_map<std::string, std::string> db_options_map = {
{"create_if_missing", "false"},
{"create_missing_column_families", "true"},
{"error_if_exists", "false"},
@ -154,98 +158,146 @@ TEST(OptionsTest, GetOptionsFromStringsTest) {
{"bytes_per_sync", "47"},
};
Options base_opt;
Options new_opt;
ASSERT_TRUE(GetOptionsFromStrings(base_opt, options_map, &new_opt));
ASSERT_EQ(new_opt.write_buffer_size, 1U);
ASSERT_EQ(new_opt.max_write_buffer_number, 2);
ASSERT_EQ(new_opt.min_write_buffer_number_to_merge, 3);
ASSERT_EQ(new_opt.compression, kSnappyCompression);
ASSERT_EQ(new_opt.compression_per_level.size(), 6U);
ASSERT_EQ(new_opt.compression_per_level[0], kNoCompression);
ASSERT_EQ(new_opt.compression_per_level[1], kSnappyCompression);
ASSERT_EQ(new_opt.compression_per_level[2], kZlibCompression);
ASSERT_EQ(new_opt.compression_per_level[3], kBZip2Compression);
ASSERT_EQ(new_opt.compression_per_level[4], kLZ4Compression);
ASSERT_EQ(new_opt.compression_per_level[5], kLZ4HCCompression);
ASSERT_EQ(new_opt.compression_opts.window_bits, 4);
ASSERT_EQ(new_opt.compression_opts.level, 5);
ASSERT_EQ(new_opt.compression_opts.strategy, 6);
ASSERT_EQ(new_opt.num_levels, 7);
ASSERT_EQ(new_opt.level0_file_num_compaction_trigger, 8);
ASSERT_EQ(new_opt.level0_slowdown_writes_trigger, 9);
ASSERT_EQ(new_opt.level0_stop_writes_trigger, 10);
ASSERT_EQ(new_opt.max_mem_compaction_level, 11);
ASSERT_EQ(new_opt.target_file_size_base, static_cast<uint64_t>(12));
ASSERT_EQ(new_opt.target_file_size_multiplier, 13);
ASSERT_EQ(new_opt.max_bytes_for_level_base, 14U);
ASSERT_EQ(new_opt.max_bytes_for_level_multiplier, 15);
ASSERT_EQ(new_opt.max_bytes_for_level_multiplier_additional.size(), 3U);
ASSERT_EQ(new_opt.max_bytes_for_level_multiplier_additional[0], 16);
ASSERT_EQ(new_opt.max_bytes_for_level_multiplier_additional[1], 17);
ASSERT_EQ(new_opt.max_bytes_for_level_multiplier_additional[2], 18);
ASSERT_EQ(new_opt.expanded_compaction_factor, 19);
ASSERT_EQ(new_opt.source_compaction_factor, 20);
ASSERT_EQ(new_opt.max_grandparent_overlap_factor, 21);
ASSERT_EQ(new_opt.soft_rate_limit, 1.1);
ASSERT_EQ(new_opt.hard_rate_limit, 2.1);
ASSERT_EQ(new_opt.arena_block_size, 22U);
ASSERT_EQ(new_opt.disable_auto_compactions, true);
ASSERT_EQ(new_opt.purge_redundant_kvs_while_flush, true);
ASSERT_EQ(new_opt.compaction_style, kCompactionStyleLevel);
ASSERT_EQ(new_opt.verify_checksums_in_compaction, false);
ASSERT_EQ(new_opt.compaction_options_fifo.max_table_files_size,
ColumnFamilyOptions base_cf_opt;
ColumnFamilyOptions new_cf_opt;
ASSERT_TRUE(GetColumnFamilyOptionsFromMap(
base_cf_opt, cf_options_map, &new_cf_opt));
ASSERT_EQ(new_cf_opt.write_buffer_size, 1U);
ASSERT_EQ(new_cf_opt.max_write_buffer_number, 2);
ASSERT_EQ(new_cf_opt.min_write_buffer_number_to_merge, 3);
ASSERT_EQ(new_cf_opt.compression, kSnappyCompression);
ASSERT_EQ(new_cf_opt.compression_per_level.size(), 6U);
ASSERT_EQ(new_cf_opt.compression_per_level[0], kNoCompression);
ASSERT_EQ(new_cf_opt.compression_per_level[1], kSnappyCompression);
ASSERT_EQ(new_cf_opt.compression_per_level[2], kZlibCompression);
ASSERT_EQ(new_cf_opt.compression_per_level[3], kBZip2Compression);
ASSERT_EQ(new_cf_opt.compression_per_level[4], kLZ4Compression);
ASSERT_EQ(new_cf_opt.compression_per_level[5], kLZ4HCCompression);
ASSERT_EQ(new_cf_opt.compression_opts.window_bits, 4);
ASSERT_EQ(new_cf_opt.compression_opts.level, 5);
ASSERT_EQ(new_cf_opt.compression_opts.strategy, 6);
ASSERT_EQ(new_cf_opt.num_levels, 7);
ASSERT_EQ(new_cf_opt.level0_file_num_compaction_trigger, 8);
ASSERT_EQ(new_cf_opt.level0_slowdown_writes_trigger, 9);
ASSERT_EQ(new_cf_opt.level0_stop_writes_trigger, 10);
ASSERT_EQ(new_cf_opt.max_mem_compaction_level, 11);
ASSERT_EQ(new_cf_opt.target_file_size_base, static_cast<uint64_t>(12));
ASSERT_EQ(new_cf_opt.target_file_size_multiplier, 13);
ASSERT_EQ(new_cf_opt.max_bytes_for_level_base, 14U);
ASSERT_EQ(new_cf_opt.max_bytes_for_level_multiplier, 15);
ASSERT_EQ(new_cf_opt.max_bytes_for_level_multiplier_additional.size(), 3U);
ASSERT_EQ(new_cf_opt.max_bytes_for_level_multiplier_additional[0], 16);
ASSERT_EQ(new_cf_opt.max_bytes_for_level_multiplier_additional[1], 17);
ASSERT_EQ(new_cf_opt.max_bytes_for_level_multiplier_additional[2], 18);
ASSERT_EQ(new_cf_opt.expanded_compaction_factor, 19);
ASSERT_EQ(new_cf_opt.source_compaction_factor, 20);
ASSERT_EQ(new_cf_opt.max_grandparent_overlap_factor, 21);
ASSERT_EQ(new_cf_opt.soft_rate_limit, 1.1);
ASSERT_EQ(new_cf_opt.hard_rate_limit, 2.1);
ASSERT_EQ(new_cf_opt.arena_block_size, 22U);
ASSERT_EQ(new_cf_opt.disable_auto_compactions, true);
ASSERT_EQ(new_cf_opt.purge_redundant_kvs_while_flush, true);
ASSERT_EQ(new_cf_opt.compaction_style, kCompactionStyleLevel);
ASSERT_EQ(new_cf_opt.verify_checksums_in_compaction, false);
ASSERT_EQ(new_cf_opt.compaction_options_fifo.max_table_files_size,
static_cast<uint64_t>(23));
ASSERT_EQ(new_opt.filter_deletes, false);
ASSERT_EQ(new_opt.max_sequential_skip_in_iterations,
ASSERT_EQ(new_cf_opt.filter_deletes, false);
ASSERT_EQ(new_cf_opt.max_sequential_skip_in_iterations,
static_cast<uint64_t>(24));
ASSERT_EQ(new_opt.inplace_update_support, true);
ASSERT_EQ(new_opt.inplace_update_num_locks, 25U);
ASSERT_EQ(new_opt.memtable_prefix_bloom_bits, 26U);
ASSERT_EQ(new_opt.memtable_prefix_bloom_probes, 27U);
ASSERT_EQ(new_opt.memtable_prefix_bloom_huge_page_tlb_size, 28U);
ASSERT_EQ(new_opt.bloom_locality, 29U);
ASSERT_EQ(new_opt.max_successive_merges, 30U);
ASSERT_EQ(new_opt.min_partial_merge_operands, 31U);
ASSERT_EQ(new_opt.create_if_missing, false);
ASSERT_EQ(new_opt.create_missing_column_families, true);
ASSERT_EQ(new_opt.error_if_exists, false);
ASSERT_EQ(new_opt.paranoid_checks, true);
ASSERT_EQ(new_opt.max_open_files, 32);
ASSERT_EQ(new_opt.max_total_wal_size, static_cast<uint64_t>(33));
ASSERT_EQ(new_opt.disableDataSync, false);
ASSERT_EQ(new_opt.use_fsync, true);
ASSERT_EQ(new_opt.db_log_dir, "/db_log_dir");
ASSERT_EQ(new_opt.wal_dir, "/wal_dir");
ASSERT_EQ(new_opt.delete_obsolete_files_period_micros,
static_cast<uint64_t>(34));
ASSERT_EQ(new_opt.max_background_compactions, 35);
ASSERT_EQ(new_opt.max_background_flushes, 36);
ASSERT_EQ(new_opt.max_log_file_size, 37U);
ASSERT_EQ(new_opt.log_file_time_to_roll, 38U);
ASSERT_EQ(new_opt.keep_log_file_num, 39U);
ASSERT_EQ(new_opt.max_manifest_file_size, static_cast<uint64_t>(40));
ASSERT_EQ(new_opt.table_cache_numshardbits, 41);
ASSERT_EQ(new_opt.table_cache_remove_scan_count_limit, 42);
ASSERT_EQ(new_opt.WAL_ttl_seconds, static_cast<uint64_t>(43));
ASSERT_EQ(new_opt.WAL_size_limit_MB, static_cast<uint64_t>(44));
ASSERT_EQ(new_opt.manifest_preallocation_size, 45U);
ASSERT_EQ(new_opt.allow_os_buffer, false);
ASSERT_EQ(new_opt.allow_mmap_reads, true);
ASSERT_EQ(new_opt.allow_mmap_writes, false);
ASSERT_EQ(new_opt.is_fd_close_on_exec, true);
ASSERT_EQ(new_opt.skip_log_error_on_recovery, false);
ASSERT_EQ(new_opt.stats_dump_period_sec, 46U);
ASSERT_EQ(new_opt.advise_random_on_open, true);
ASSERT_EQ(new_opt.use_adaptive_mutex, false);
ASSERT_EQ(new_opt.bytes_per_sync, static_cast<uint64_t>(47));
ASSERT_EQ(new_cf_opt.inplace_update_support, true);
ASSERT_EQ(new_cf_opt.inplace_update_num_locks, 25U);
ASSERT_EQ(new_cf_opt.memtable_prefix_bloom_bits, 26U);
ASSERT_EQ(new_cf_opt.memtable_prefix_bloom_probes, 27U);
ASSERT_EQ(new_cf_opt.memtable_prefix_bloom_huge_page_tlb_size, 28U);
ASSERT_EQ(new_cf_opt.bloom_locality, 29U);
ASSERT_EQ(new_cf_opt.max_successive_merges, 30U);
ASSERT_EQ(new_cf_opt.min_partial_merge_operands, 31U);
options_map["write_buffer_size"] = "hello";
ASSERT_TRUE(!GetOptionsFromStrings(base_opt, options_map, &new_opt));
options_map["write_buffer_size"] = "1";
ASSERT_TRUE(GetOptionsFromStrings(base_opt, options_map, &new_opt));
options_map["unknown_option"] = "1";
ASSERT_TRUE(!GetOptionsFromStrings(base_opt, options_map, &new_opt));
cf_options_map["write_buffer_size"] = "hello";
ASSERT_TRUE(!GetColumnFamilyOptionsFromMap(
base_cf_opt, cf_options_map, &new_cf_opt));
cf_options_map["write_buffer_size"] = "1";
ASSERT_TRUE(GetColumnFamilyOptionsFromMap(
base_cf_opt, cf_options_map, &new_cf_opt));
cf_options_map["unknown_option"] = "1";
ASSERT_TRUE(!GetColumnFamilyOptionsFromMap(
base_cf_opt, cf_options_map, &new_cf_opt));
DBOptions base_db_opt;
DBOptions new_db_opt;
ASSERT_TRUE(GetDBOptionsFromMap(base_db_opt, db_options_map, &new_db_opt));
ASSERT_EQ(new_db_opt.create_if_missing, false);
ASSERT_EQ(new_db_opt.create_missing_column_families, true);
ASSERT_EQ(new_db_opt.error_if_exists, false);
ASSERT_EQ(new_db_opt.paranoid_checks, true);
ASSERT_EQ(new_db_opt.max_open_files, 32);
ASSERT_EQ(new_db_opt.max_total_wal_size, static_cast<uint64_t>(33));
ASSERT_EQ(new_db_opt.disableDataSync, false);
ASSERT_EQ(new_db_opt.use_fsync, true);
ASSERT_EQ(new_db_opt.db_log_dir, "/db_log_dir");
ASSERT_EQ(new_db_opt.wal_dir, "/wal_dir");
ASSERT_EQ(new_db_opt.delete_obsolete_files_period_micros,
static_cast<uint64_t>(34));
ASSERT_EQ(new_db_opt.max_background_compactions, 35);
ASSERT_EQ(new_db_opt.max_background_flushes, 36);
ASSERT_EQ(new_db_opt.max_log_file_size, 37U);
ASSERT_EQ(new_db_opt.log_file_time_to_roll, 38U);
ASSERT_EQ(new_db_opt.keep_log_file_num, 39U);
ASSERT_EQ(new_db_opt.max_manifest_file_size, static_cast<uint64_t>(40));
ASSERT_EQ(new_db_opt.table_cache_numshardbits, 41);
ASSERT_EQ(new_db_opt.table_cache_remove_scan_count_limit, 42);
ASSERT_EQ(new_db_opt.WAL_ttl_seconds, static_cast<uint64_t>(43));
ASSERT_EQ(new_db_opt.WAL_size_limit_MB, static_cast<uint64_t>(44));
ASSERT_EQ(new_db_opt.manifest_preallocation_size, 45U);
ASSERT_EQ(new_db_opt.allow_os_buffer, false);
ASSERT_EQ(new_db_opt.allow_mmap_reads, true);
ASSERT_EQ(new_db_opt.allow_mmap_writes, false);
ASSERT_EQ(new_db_opt.is_fd_close_on_exec, true);
ASSERT_EQ(new_db_opt.skip_log_error_on_recovery, false);
ASSERT_EQ(new_db_opt.stats_dump_period_sec, 46U);
ASSERT_EQ(new_db_opt.advise_random_on_open, true);
ASSERT_EQ(new_db_opt.use_adaptive_mutex, false);
ASSERT_EQ(new_db_opt.bytes_per_sync, static_cast<uint64_t>(47));
}
TEST(OptionsTest, GetOptionsFromStringTest) {
ColumnFamilyOptions base_cf_opt;
ColumnFamilyOptions new_cf_opt;
ASSERT_TRUE(GetColumnFamilyOptionsFromString(base_cf_opt, "", &new_cf_opt));
ASSERT_TRUE(GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=5", &new_cf_opt));
ASSERT_EQ(new_cf_opt.write_buffer_size, 5);
ASSERT_TRUE(GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=6;", &new_cf_opt));
ASSERT_EQ(new_cf_opt.write_buffer_size, 6);
ASSERT_TRUE(GetColumnFamilyOptionsFromString(base_cf_opt,
" write_buffer_size = 7 ", &new_cf_opt));
ASSERT_EQ(new_cf_opt.write_buffer_size, 7);
ASSERT_TRUE(GetColumnFamilyOptionsFromString(base_cf_opt,
" write_buffer_size = 8 ; ", &new_cf_opt));
ASSERT_EQ(new_cf_opt.write_buffer_size, 8);
ASSERT_TRUE(GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=9;max_write_buffer_number=10", &new_cf_opt));
ASSERT_EQ(new_cf_opt.write_buffer_size, 9);
ASSERT_EQ(new_cf_opt.max_write_buffer_number, 10);
ASSERT_TRUE(GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=11; max_write_buffer_number = 12 ;",
&new_cf_opt));
ASSERT_EQ(new_cf_opt.write_buffer_size, 11);
ASSERT_EQ(new_cf_opt.max_write_buffer_number, 12);
// Wrong name "max_write_buffer_number_"
ASSERT_TRUE(!GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=13;max_write_buffer_number_=14;",
&new_cf_opt));
// Wrong key/value pair
ASSERT_TRUE(!GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=13;max_write_buffer_number;", &new_cf_opt));
// Error Paring value
ASSERT_TRUE(!GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=13;max_write_buffer_number=;", &new_cf_opt));
// Missing option name
ASSERT_TRUE(!GetColumnFamilyOptionsFromString(base_cf_opt,
"write_buffer_size=13; =100;", &new_cf_opt));
}
} // namespace rocksdb

View file

@ -0,0 +1,40 @@
// Copyright (c) 2014, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
#pragma once
#include <unordered_map>
#include <string>
#include "rocksdb/options.h"
namespace rocksdb {
// Take a map of option name and option value, apply them into the
// base_options, and return the new options as a result
bool GetColumnFamilyOptionsFromMap(
const ColumnFamilyOptions& base_options,
const std::unordered_map<std::string, std::string>& opts_map,
ColumnFamilyOptions* new_options);
bool GetDBOptionsFromMap(
const DBOptions& base_options,
const std::unordered_map<std::string, std::string>& opts_map,
DBOptions* new_options);
// Take a string representation of option names and values, apply them into the
// base_options, and return the new options as a result. The string has the
// following format:
// "write_buffer_size=1024;max_write_buffer_number=2"
bool GetColumnFamilyOptionsFromString(
const ColumnFamilyOptions& base_options,
const std::string& opts_str,
ColumnFamilyOptions* new_options);
bool GetDBOptionsFromString(
const DBOptions& base_options,
const std::string& opts_str,
DBOptions* new_options);
} // namespace rocksdb