2016-02-09 23:12:00 +00:00
|
|
|
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
2013-11-20 00:29:42 +00:00
|
|
|
// 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.
|
2017-04-28 00:50:56 +00:00
|
|
|
// This source code is also licensed under the GPLv2 license found in the
|
|
|
|
// COPYING file in the root directory of this source tree.
|
2013-11-20 00:29:42 +00:00
|
|
|
|
|
|
|
#include "db/table_properties_collector.h"
|
|
|
|
|
|
|
|
#include "db/dbformat.h"
|
|
|
|
#include "util/coding.h"
|
2014-11-25 04:44:49 +00:00
|
|
|
#include "util/string_util.h"
|
2013-11-20 00:29:42 +00:00
|
|
|
|
|
|
|
namespace rocksdb {
|
|
|
|
|
A new call back to TablePropertiesCollector to allow users know the entry is add, delete or merge
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
2015-04-06 17:04:30 +00:00
|
|
|
Status InternalKeyPropertiesCollector::InternalAdd(const Slice& key,
|
|
|
|
const Slice& value,
|
|
|
|
uint64_t file_size) {
|
2013-11-20 00:29:42 +00:00
|
|
|
ParsedInternalKey ikey;
|
|
|
|
if (!ParseInternalKey(key, &ikey)) {
|
|
|
|
return Status::InvalidArgument("Invalid internal key");
|
|
|
|
}
|
|
|
|
|
Support for SingleDelete()
Summary:
This patch fixes #7460559. It introduces SingleDelete as a new database
operation. This operation can be used to delete keys that were never
overwritten (no put following another put of the same key). If an overwritten
key is single deleted the behavior is undefined. Single deletion of a
non-existent key has no effect but multiple consecutive single deletions are
not allowed (see limitations).
In contrast to the conventional Delete() operation, the deletion entry is
removed along with the value when the two are lined up in a compaction. Note:
The semantics are similar to @igor's prototype that allowed to have this
behavior on the granularity of a column family (
https://reviews.facebook.net/D42093 ). This new patch, however, is more
aggressive when it comes to removing tombstones: It removes the SingleDelete
together with the value whenever there is no snapshot between them while the
older patch only did this when the sequence number of the deletion was older
than the earliest snapshot.
Most of the complex additions are in the Compaction Iterator, all other changes
should be relatively straightforward. The patch also includes basic support for
single deletions in db_stress and db_bench.
Limitations:
- Not compatible with cuckoo hash tables
- Single deletions cannot be used in combination with merges and normal
deletions on the same key (other keys are not affected by this)
- Consecutive single deletions are currently not allowed (and older version of
this patch supported this so it could be resurrected if needed)
Test Plan: make all check
Reviewers: yhchiang, sdong, rven, anthony, yoshinorim, igor
Reviewed By: igor
Subscribers: maykov, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D43179
2015-09-17 18:42:56 +00:00
|
|
|
// Note: We count both, deletions and single deletions here.
|
|
|
|
if (ikey.type == ValueType::kTypeDeletion ||
|
|
|
|
ikey.type == ValueType::kTypeSingleDeletion) {
|
2013-11-20 00:29:42 +00:00
|
|
|
++deleted_keys_;
|
2016-05-19 21:24:48 +00:00
|
|
|
} else if (ikey.type == ValueType::kTypeMerge) {
|
|
|
|
++merge_operands_;
|
2013-11-20 00:29:42 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
Status InternalKeyPropertiesCollector::Finish(
|
2013-12-05 21:09:13 +00:00
|
|
|
UserCollectedProperties* properties) {
|
2013-11-20 00:29:42 +00:00
|
|
|
assert(properties);
|
|
|
|
assert(properties->find(
|
|
|
|
InternalKeyTablePropertiesNames::kDeletedKeys) == properties->end());
|
2016-05-19 21:24:48 +00:00
|
|
|
assert(properties->find(InternalKeyTablePropertiesNames::kMergeOperands) ==
|
|
|
|
properties->end());
|
2013-11-20 00:29:42 +00:00
|
|
|
|
2016-05-19 21:24:48 +00:00
|
|
|
std::string val_deleted_keys;
|
|
|
|
PutVarint64(&val_deleted_keys, deleted_keys_);
|
|
|
|
properties->insert(
|
|
|
|
{InternalKeyTablePropertiesNames::kDeletedKeys, val_deleted_keys});
|
|
|
|
|
|
|
|
std::string val_merge_operands;
|
|
|
|
PutVarint64(&val_merge_operands, merge_operands_);
|
|
|
|
properties->insert(
|
|
|
|
{InternalKeyTablePropertiesNames::kMergeOperands, val_merge_operands});
|
2013-11-20 00:29:42 +00:00
|
|
|
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
2013-12-05 21:09:13 +00:00
|
|
|
UserCollectedProperties
|
2013-11-20 00:29:42 +00:00
|
|
|
InternalKeyPropertiesCollector::GetReadableProperties() const {
|
2016-05-19 21:24:48 +00:00
|
|
|
return {{"kDeletedKeys", ToString(deleted_keys_)},
|
|
|
|
{"kMergeOperands", ToString(merge_operands_)}};
|
2013-11-20 00:29:42 +00:00
|
|
|
}
|
|
|
|
|
A new call back to TablePropertiesCollector to allow users know the entry is add, delete or merge
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
2015-04-06 17:04:30 +00:00
|
|
|
namespace {
|
Support for SingleDelete()
Summary:
This patch fixes #7460559. It introduces SingleDelete as a new database
operation. This operation can be used to delete keys that were never
overwritten (no put following another put of the same key). If an overwritten
key is single deleted the behavior is undefined. Single deletion of a
non-existent key has no effect but multiple consecutive single deletions are
not allowed (see limitations).
In contrast to the conventional Delete() operation, the deletion entry is
removed along with the value when the two are lined up in a compaction. Note:
The semantics are similar to @igor's prototype that allowed to have this
behavior on the granularity of a column family (
https://reviews.facebook.net/D42093 ). This new patch, however, is more
aggressive when it comes to removing tombstones: It removes the SingleDelete
together with the value whenever there is no snapshot between them while the
older patch only did this when the sequence number of the deletion was older
than the earliest snapshot.
Most of the complex additions are in the Compaction Iterator, all other changes
should be relatively straightforward. The patch also includes basic support for
single deletions in db_stress and db_bench.
Limitations:
- Not compatible with cuckoo hash tables
- Single deletions cannot be used in combination with merges and normal
deletions on the same key (other keys are not affected by this)
- Consecutive single deletions are currently not allowed (and older version of
this patch supported this so it could be resurrected if needed)
Test Plan: make all check
Reviewers: yhchiang, sdong, rven, anthony, yoshinorim, igor
Reviewed By: igor
Subscribers: maykov, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D43179
2015-09-17 18:42:56 +00:00
|
|
|
|
A new call back to TablePropertiesCollector to allow users know the entry is add, delete or merge
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
2015-04-06 17:04:30 +00:00
|
|
|
EntryType GetEntryType(ValueType value_type) {
|
|
|
|
switch (value_type) {
|
|
|
|
case kTypeValue:
|
|
|
|
return kEntryPut;
|
|
|
|
case kTypeDeletion:
|
|
|
|
return kEntryDelete;
|
Support for SingleDelete()
Summary:
This patch fixes #7460559. It introduces SingleDelete as a new database
operation. This operation can be used to delete keys that were never
overwritten (no put following another put of the same key). If an overwritten
key is single deleted the behavior is undefined. Single deletion of a
non-existent key has no effect but multiple consecutive single deletions are
not allowed (see limitations).
In contrast to the conventional Delete() operation, the deletion entry is
removed along with the value when the two are lined up in a compaction. Note:
The semantics are similar to @igor's prototype that allowed to have this
behavior on the granularity of a column family (
https://reviews.facebook.net/D42093 ). This new patch, however, is more
aggressive when it comes to removing tombstones: It removes the SingleDelete
together with the value whenever there is no snapshot between them while the
older patch only did this when the sequence number of the deletion was older
than the earliest snapshot.
Most of the complex additions are in the Compaction Iterator, all other changes
should be relatively straightforward. The patch also includes basic support for
single deletions in db_stress and db_bench.
Limitations:
- Not compatible with cuckoo hash tables
- Single deletions cannot be used in combination with merges and normal
deletions on the same key (other keys are not affected by this)
- Consecutive single deletions are currently not allowed (and older version of
this patch supported this so it could be resurrected if needed)
Test Plan: make all check
Reviewers: yhchiang, sdong, rven, anthony, yoshinorim, igor
Reviewed By: igor
Subscribers: maykov, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D43179
2015-09-17 18:42:56 +00:00
|
|
|
case kTypeSingleDeletion:
|
|
|
|
return kEntrySingleDelete;
|
A new call back to TablePropertiesCollector to allow users know the entry is add, delete or merge
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
2015-04-06 17:04:30 +00:00
|
|
|
case kTypeMerge:
|
|
|
|
return kEntryMerge;
|
|
|
|
default:
|
|
|
|
return kEntryOther;
|
|
|
|
}
|
|
|
|
}
|
Support for SingleDelete()
Summary:
This patch fixes #7460559. It introduces SingleDelete as a new database
operation. This operation can be used to delete keys that were never
overwritten (no put following another put of the same key). If an overwritten
key is single deleted the behavior is undefined. Single deletion of a
non-existent key has no effect but multiple consecutive single deletions are
not allowed (see limitations).
In contrast to the conventional Delete() operation, the deletion entry is
removed along with the value when the two are lined up in a compaction. Note:
The semantics are similar to @igor's prototype that allowed to have this
behavior on the granularity of a column family (
https://reviews.facebook.net/D42093 ). This new patch, however, is more
aggressive when it comes to removing tombstones: It removes the SingleDelete
together with the value whenever there is no snapshot between them while the
older patch only did this when the sequence number of the deletion was older
than the earliest snapshot.
Most of the complex additions are in the Compaction Iterator, all other changes
should be relatively straightforward. The patch also includes basic support for
single deletions in db_stress and db_bench.
Limitations:
- Not compatible with cuckoo hash tables
- Single deletions cannot be used in combination with merges and normal
deletions on the same key (other keys are not affected by this)
- Consecutive single deletions are currently not allowed (and older version of
this patch supported this so it could be resurrected if needed)
Test Plan: make all check
Reviewers: yhchiang, sdong, rven, anthony, yoshinorim, igor
Reviewed By: igor
Subscribers: maykov, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D43179
2015-09-17 18:42:56 +00:00
|
|
|
|
2016-05-19 21:24:48 +00:00
|
|
|
uint64_t GetUint64Property(const UserCollectedProperties& props,
|
|
|
|
const std::string property_name,
|
|
|
|
bool* property_present) {
|
|
|
|
auto pos = props.find(property_name);
|
|
|
|
if (pos == props.end()) {
|
|
|
|
*property_present = false;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
Slice raw = pos->second;
|
|
|
|
uint64_t val = 0;
|
|
|
|
*property_present = true;
|
|
|
|
return GetVarint64(&raw, &val) ? val : 0;
|
|
|
|
}
|
|
|
|
|
A new call back to TablePropertiesCollector to allow users know the entry is add, delete or merge
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
2015-04-06 17:04:30 +00:00
|
|
|
} // namespace
|
2013-11-20 00:29:42 +00:00
|
|
|
|
A new call back to TablePropertiesCollector to allow users know the entry is add, delete or merge
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
2015-04-06 17:04:30 +00:00
|
|
|
Status UserKeyTablePropertiesCollector::InternalAdd(const Slice& key,
|
|
|
|
const Slice& value,
|
|
|
|
uint64_t file_size) {
|
2013-11-20 00:29:42 +00:00
|
|
|
ParsedInternalKey ikey;
|
|
|
|
if (!ParseInternalKey(key, &ikey)) {
|
|
|
|
return Status::InvalidArgument("Invalid internal key");
|
|
|
|
}
|
|
|
|
|
A new call back to TablePropertiesCollector to allow users know the entry is add, delete or merge
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
2015-04-06 17:04:30 +00:00
|
|
|
return collector_->AddUserKey(ikey.user_key, value, GetEntryType(ikey.type),
|
|
|
|
ikey.sequence, file_size);
|
2013-11-20 00:29:42 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
Status UserKeyTablePropertiesCollector::Finish(
|
2013-12-05 21:09:13 +00:00
|
|
|
UserCollectedProperties* properties) {
|
2013-11-20 00:29:42 +00:00
|
|
|
return collector_->Finish(properties);
|
|
|
|
}
|
|
|
|
|
2013-12-05 21:09:13 +00:00
|
|
|
UserCollectedProperties
|
2013-11-20 00:29:42 +00:00
|
|
|
UserKeyTablePropertiesCollector::GetReadableProperties() const {
|
|
|
|
return collector_->GetReadableProperties();
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
const std::string InternalKeyTablePropertiesNames::kDeletedKeys
|
|
|
|
= "rocksdb.deleted.keys";
|
2016-05-19 21:24:48 +00:00
|
|
|
const std::string InternalKeyTablePropertiesNames::kMergeOperands =
|
|
|
|
"rocksdb.merge.operands";
|
2013-11-20 00:29:42 +00:00
|
|
|
|
|
|
|
uint64_t GetDeletedKeys(
|
2013-12-05 21:09:13 +00:00
|
|
|
const UserCollectedProperties& props) {
|
2016-05-19 21:24:48 +00:00
|
|
|
bool property_present_ignored;
|
|
|
|
return GetUint64Property(props, InternalKeyTablePropertiesNames::kDeletedKeys,
|
|
|
|
&property_present_ignored);
|
|
|
|
}
|
|
|
|
|
|
|
|
uint64_t GetMergeOperands(const UserCollectedProperties& props,
|
|
|
|
bool* property_present) {
|
|
|
|
return GetUint64Property(
|
|
|
|
props, InternalKeyTablePropertiesNames::kMergeOperands, property_present);
|
2013-11-20 00:29:42 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
} // namespace rocksdb
|