blob: 3ed76bacec409296e688920b09144461dafc0574 [file] [log] [blame]
// Copyright (C) 2013 The Android Open Source Project
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package com.google.gerrit.index;
import com.google.common.collect.ImmutableList;
import com.google.gerrit.exceptions.StorageException;
import com.google.gerrit.index.query.DataSource;
import com.google.gerrit.index.query.FieldBundle;
import com.google.gerrit.index.query.IndexPredicate;
import com.google.gerrit.index.query.Matchable;
import com.google.gerrit.index.query.Predicate;
import com.google.gerrit.index.query.QueryParseException;
import java.util.Optional;
/**
* Secondary index implementation for arbitrary documents.
*
* <p>Documents are inserted into the index and are queried by converting special {@link
* com.google.gerrit.index.query.Predicate} instances into index-aware predicates that use the index
* search results as a source.
*
* <p>Implementations must be thread-safe and should batch inserts/updates where appropriate.
*/
public interface Index<K, V> {
/** Returns the schema version used by this index. */
Schema<V> getSchema();
/** Close this index. */
void close();
/**
* Insert a document into the index.
*
* <p>Results may not be immediately visible to searchers, but should be visible within a
* reasonable amount of time.
*
* @param obj document object
*/
void insert(V obj);
/**
* Update a document in the index.
*
* <p>Semantically equivalent to deleting the document and reinserting it with new field values. A
* document that does not already exist is created. Results may not be immediately visible to
* searchers, but should be visible within a reasonable amount of time.
*
* @param obj document object
*/
void replace(V obj);
/** Delete a document from the index by value */
void deleteByValue(V value);
/**
* Delete a document from the index by key.
*
* @param key document key
*/
void delete(K key);
/** Delete all documents from the index. */
void deleteAll();
/**
* Convert the given operator predicate into a source searching the index and returning only the
* documents matching that predicate.
*
* <p>This method may be called multiple times for variations on the same predicate or multiple
* predicate subtrees in the course of processing a single query, so it should not have any side
* effects (e.g. starting a search in the background).
*
* @param p the predicate to match. Must be a tree containing only AND, OR, or NOT predicates as
* internal nodes, and {@link IndexPredicate}s as leaves.
* @param opts query options not implied by the predicate, such as start and limit.
* @return a source of documents matching the predicate, returned in a defined order depending on
* the type of documents.
* @throws QueryParseException if the predicate could not be converted to an indexed data source.
*/
DataSource<V> getSource(Predicate<V> p, QueryOptions opts) throws QueryParseException;
/**
* Get a single document from the index.
*
* @param key document key.
* @param opts query options. Options that do not make sense in the context of a single document,
* such as start, will be ignored.
* @return a single document if present.
*/
default Optional<V> get(K key, QueryOptions opts) {
opts = opts.withStart(0).withLimit(2);
ImmutableList<V> results;
try {
results = getSource(keyPredicate(key), opts).read().toList();
} catch (QueryParseException e) {
throw new StorageException("Unexpected QueryParseException during get()", e);
}
if (results.size() > 1) {
throw new StorageException("Multiple results found in index for key " + key + ": " + results);
}
return results.stream().findFirst();
}
/**
* Get a single raw document from the index.
*
* @param key document key.
* @param opts query options. Options that do not make sense in the context of a single document,
* such as start, will be ignored.
* @return an abstraction of a raw index document to retrieve fields from.
*/
default Optional<FieldBundle> getRaw(K key, QueryOptions opts) {
opts = opts.withStart(0).withLimit(2);
ImmutableList<FieldBundle> results;
try {
results = getSource(keyPredicate(key), opts).readRaw().toList();
} catch (QueryParseException e) {
throw new StorageException("Unexpected QueryParseException during get()", e);
}
if (results.size() > 1) {
throw new StorageException("Multiple results found in index for key " + key + ": " + results);
}
return results.stream().findFirst();
}
/**
* Get a predicate that looks up a single document by key.
*
* @param key document key.
* @return a single predicate.
*/
Predicate<V> keyPredicate(K key);
/**
* Mark whether this index is up-to-date and ready to serve reads.
*
* @param ready whether the index is ready
*/
void markReady(boolean ready);
/**
* Returns whether the index is enabled. {@code true} by default, but could be overridden by
* implementations.
*/
default boolean isEnabled() {
return true;
}
/**
* An Optional filter that is invoked right after the results are returned from the index, but
* before any post-filter predicates.
*
* <p>The filter is invoked before any other index predicates. If the filter returns 'true', then
* other index predicates are evaluated. Otherwise, the result from the index is not returned to
* the DataSource.
*/
default Optional<Matchable<V>> getIndexFilter() {
return Optional.empty();
}
}