Expose a simple secondary index implementation (facebook#13370)

Summary:
Pull Request resolved: facebook#13370

We have a class called `DefaultSecondaryIndex` in `TransactionTest.SecondaryIndexPutDelete` that contains generally useful functionality. The patch generalizes it a bit to make the column name configurable, renames it to `SimpleSecondaryIndex`, and moves it to the public API so applications can use it.

Reviewed By: jowlyzhang

Differential Revision: D69147890

fbshipit-source-id: 0d2d1cc5adcde01f3978a450ec841c9e990d2170

Author: ltamasi

BUCK

@@ -318,6 +318,7 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
         "utilities/persistent_cache/persistent_cache_tier.cc",
         "utilities/persistent_cache/volatile_tier_impl.cc",
         "utilities/secondary_index/secondary_index_iterator.cc",
+        "utilities/secondary_index/simple_secondary_index.cc",
         "utilities/simulator_cache/cache_simulator.cc",
         "utilities/simulator_cache/sim_cache.cc",
         "utilities/table_properties_collectors/compact_for_tiering_collector.cc",

CMakeLists.txt

@@ -940,6 +940,7 @@ set(SOURCES
         utilities/persistent_cache/persistent_cache_tier.cc
         utilities/persistent_cache/volatile_tier_impl.cc
         utilities/secondary_index/secondary_index_iterator.cc
+        utilities/secondary_index/simple_secondary_index.cc
         utilities/simulator_cache/cache_simulator.cc
         utilities/simulator_cache/sim_cache.cc
         utilities/table_properties_collectors/compact_for_tiering_collector.cc

include/rocksdb/utilities/secondary_index_simple.h

@@ -0,0 +1,57 @@
+//  Copyright (c) Meta Platforms, Inc. and affiliates.
+//
+//  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).
+
+#pragma once
+
+#include <string>
+
+#include "rocksdb/rocksdb_namespace.h"
+#include "rocksdb/utilities/secondary_index.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+// EXPERIMENTAL
+//
+// A simple secondary index implementation that indexes the specified column
+// as-is.
+
+class SimpleSecondaryIndex : public SecondaryIndex {
+ public:
+  explicit SimpleSecondaryIndex(std::string primary_column_name);
+
+  void SetPrimaryColumnFamily(ColumnFamilyHandle* column_family) override;
+  void SetSecondaryColumnFamily(ColumnFamilyHandle* column_family) override;
+
+  ColumnFamilyHandle* GetPrimaryColumnFamily() const override;
+  ColumnFamilyHandle* GetSecondaryColumnFamily() const override;
+
+  Slice GetPrimaryColumnName() const override;
+
+  Status UpdatePrimaryColumnValue(
+      const Slice& primary_key, const Slice& primary_column_value,
+      std::optional<std::variant<Slice, std::string>>* updated_column_value)
+      const override;
+
+  Status GetSecondaryKeyPrefix(
+      const Slice& primary_key, const Slice& primary_column_value,
+      std::variant<Slice, std::string>* secondary_key_prefix) const override;
+
+  Status FinalizeSecondaryKeyPrefix(
+      std::variant<Slice, std::string>* secondary_key_prefix) const override;
+
+  Status GetSecondaryValue(const Slice& primary_key,
+                           const Slice& primary_column_value,
+                           const Slice& original_column_value,
+                           std::optional<std::variant<Slice, std::string>>*
+                               secondary_value) const override;
+
+ private:
+  std::string primary_column_name_;
+  ColumnFamilyHandle* primary_column_family_{};
+  ColumnFamilyHandle* secondary_column_family_{};
+};
+
+}  // namespace ROCKSDB_NAMESPACE

src.mk

@@ -305,6 +305,7 @@ LIB_SOURCES =                                                   \
   utilities/persistent_cache/persistent_cache_tier.cc           \
   utilities/persistent_cache/volatile_tier_impl.cc              \
   utilities/secondary_index/secondary_index_iterator.cc         \
+  utilities/secondary_index/simple_secondary_index.cc           \
   utilities/simulator_cache/cache_simulator.cc                  \
   utilities/simulator_cache/sim_cache.cc                        \
   utilities/table_properties_collectors/compact_for_tiering_collector.cc \

unreleased_history/new_features/simple_secondary_index.md

@@ -0,0 +1 @@
+Experimental feature: added support for simple secondary indices that index the specified column as-is. See `SimpleSecondaryIndex` and `SecondaryIndex` for more details.

utilities/secondary_index/simple_secondary_index.cc

@@ -0,0 +1,79 @@
+//  Copyright (c) Meta Platforms, Inc. and affiliates.
+//  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 <cassert>
+
+#include "rocksdb/utilities/secondary_index_simple.h"
+#include "util/coding.h"
+#include "utilities/secondary_index/secondary_index_helper.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+SimpleSecondaryIndex::SimpleSecondaryIndex(std::string primary_column_name)
+    : primary_column_name_(std::move(primary_column_name)) {}
+
+void SimpleSecondaryIndex::SetPrimaryColumnFamily(
+    ColumnFamilyHandle* column_family) {
+  assert(column_family);
+  primary_column_family_ = column_family;
+}
+
+void SimpleSecondaryIndex::SetSecondaryColumnFamily(
+    ColumnFamilyHandle* column_family) {
+  assert(column_family);
+  secondary_column_family_ = column_family;
+}
+
+ColumnFamilyHandle* SimpleSecondaryIndex::GetPrimaryColumnFamily() const {
+  return primary_column_family_;
+}
+
+ColumnFamilyHandle* SimpleSecondaryIndex::GetSecondaryColumnFamily() const {
+  return secondary_column_family_;
+}
+
+Slice SimpleSecondaryIndex::GetPrimaryColumnName() const {
+  return primary_column_name_;
+}
+
+Status SimpleSecondaryIndex::UpdatePrimaryColumnValue(
+    const Slice& /* primary_key */, const Slice& /* primary_column_value */,
+    std::optional<std::variant<Slice, std::string>>* /* updated_column_value */)
+    const {
+  return Status::OK();
+}
+
+Status SimpleSecondaryIndex::GetSecondaryKeyPrefix(
+    const Slice& /* primary_key */, const Slice& primary_column_value,
+    std::variant<Slice, std::string>* secondary_key_prefix) const {
+  assert(secondary_key_prefix);
+
+  *secondary_key_prefix = primary_column_value;
+
+  return Status::OK();
+}
+
+Status SimpleSecondaryIndex::FinalizeSecondaryKeyPrefix(
+    std::variant<Slice, std::string>* secondary_key_prefix) const {
+  assert(secondary_key_prefix);
+
+  std::string prefix;
+  PutLengthPrefixedSlice(&prefix,
+                         SecondaryIndexHelper::AsSlice(*secondary_key_prefix));
+
+  *secondary_key_prefix = std::move(prefix);
+
+  return Status::OK();
+}
+
+Status SimpleSecondaryIndex::GetSecondaryValue(
+    const Slice& /* primary_key */, const Slice& /* primary_column_value */,
+    const Slice& /* original_column_value */,
+    std::optional<std::variant<Slice, std::string>>* /* secondary_value */)
+    const {
+  return Status::OK();
+}
+
+}  // namespace ROCKSDB_NAMESPACE

utilities/transactions/transaction_test.cc

@@ -19,6 +19,7 @@
 #include "rocksdb/options.h"
 #include "rocksdb/perf_context.h"
 #include "rocksdb/utilities/secondary_index.h"
+#include "rocksdb/utilities/secondary_index_simple.h"
 #include "rocksdb/utilities/transaction.h"
 #include "rocksdb/utilities/transaction_db.h"
 #include "table/mock_table.h"
@@ -8015,75 +8016,9 @@ TEST_P(TransactionTest, SecondaryIndexPutDelete) {
     return;
   }
 
-  // A basic secondary index that indexes the default column.
-  class DefaultSecondaryIndex : public SecondaryIndex {
-   public:
-    void SetPrimaryColumnFamily(ColumnFamilyHandle* cfh) override {
-      primary_cfh_ = cfh;
-    }
-
-    void SetSecondaryColumnFamily(ColumnFamilyHandle* cfh) override {
-      secondary_cfh_ = cfh;
-    }
-
-    ColumnFamilyHandle* GetPrimaryColumnFamily() const override {
-      return primary_cfh_;
-    }
-
-    ColumnFamilyHandle* GetSecondaryColumnFamily() const override {
-      return secondary_cfh_;
-    }
-
-    Slice GetPrimaryColumnName() const override {
-      return kDefaultWideColumnName;
-    }
-
-    Status UpdatePrimaryColumnValue(
-        const Slice& /* primary_key */, const Slice& /* primary_column_value */,
-        std::optional<
-            std::variant<Slice, std::string>>* /* updated_column_value */)
-        const override {
-      return Status::OK();
-    }
-
-    Status GetSecondaryKeyPrefix(
-        const Slice& /* primary_key */, const Slice& primary_column_value,
-        std::variant<Slice, std::string>* secondary_key_prefix) const override {
-      assert(secondary_key_prefix);
-
-      *secondary_key_prefix = primary_column_value;
-
-      return Status::OK();
-    }
-
-    Status FinalizeSecondaryKeyPrefix(
-        std::variant<Slice, std::string>* secondary_key_prefix) const override {
-      assert(secondary_key_prefix);
-
-      std::string prefix;
-      PutLengthPrefixedSlice(
-          &prefix, SecondaryIndexHelper::AsSlice(*secondary_key_prefix));
-
-      *secondary_key_prefix = std::move(prefix);
-
-      return Status::OK();
-    }
-
-    Status GetSecondaryValue(const Slice& /* primary_key */,
-                             const Slice& /* primary_column_value */,
-                             const Slice& /* previous_column_value */,
-                             std::optional<std::variant<Slice, std::string>>*
-                             /* secondary_value */) const override {
-      return Status::OK();
-    }
-
-   private:
-    ColumnFamilyHandle* primary_cfh_{};
-    ColumnFamilyHandle* secondary_cfh_{};
-  };
-
   txn_db_options.secondary_indices.emplace_back(
-      std::make_shared<DefaultSecondaryIndex>());
+      std::make_shared<SimpleSecondaryIndex>(
+          kDefaultWideColumnName.ToString()));
 
   ASSERT_OK(ReOpen());