HugeGraph Plugin机制及插件扩展流程

费辰阳
2023-12-01

HugeGraph是一个Java语言实现的开源图数据库,内置支持多种后端存储(Cassandra、RocksDB等),此外还支持通过插件的方式扩展存储后端。插件实现者无需更改HugeGraph本身源代码,插件使用者无需关注代码细节,就能轻松增加插件扩展功能。这篇文章主要讲述了HugeGraph扩展自定义插件的流程。

可扩展维度

目前插件方式提供如下几个维度的扩展项:

  • 后端存储:HugeGraph支持多种内置存储后端,也允许用户无需更改现有源码的情况下扩展自定义后端。
  • 系列化器:配合自定义后端实现系列化器,或者扩展内置后端的系列化器。
  • 自定义配置项:配合自定义插件扩展配置项。
  • 分词器:HugeGraph支持全文检索,全文检索功能涉及到各语言分词,目前已内置8种中文分词器,也允许用户无需更改现有源码的情况下扩展自定义分词器。

插件实现机制

  1. HugeGraph提供插件接口HugeGraphPlugin,通过Java SPI机制支持插件化
  2. HugeGraph提供了4个扩展项注册函数:registerOptions()registerBackend()registerSerializer()registerAnalyzer()
  3. 插件实现者实现相应的Options、Backend、Serializer或Analyzer的接口
  4. 插件实现者实现HugeGraphPlugin接口的register()方法,在register()中调用上述第2步中的函数,来注册第3步列出的具体实现类,并打成jar包
  5. 插件使用者将jar包放在HugeGraph-Server安装目录的plugins目录下,修改相关配置项为插件自定义值,重启即可生效

插件实现流程实例

1 新建一个maven项目

1.1 项目名称取名:hugegraph-plugin-demo
1.2 添加hugegraph-core Jar包依赖

maven pom.xml详细内容如下:

<?xml version="1.0" encoding="UTF-8"?>

<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">

    <modelVersion>4.0.0</modelVersion>
    <groupId>com.baidu.hugegraph</groupId>
    <artifactId>hugegraph-plugin-demo</artifactId>
    <version>1.0.0</version>
    <packaging>jar</packaging>

    <name>hugegraph-plugin-demo</name>

    <dependencies>
        <dependency>
            <groupId>com.baidu.hugegraph</groupId>
            <artifactId>hugegraph-core</artifactId>
            <version>${project.version}</version>
        </dependency>
    </dependencies>
</project>

2 实现扩展功能

扩展功能一般包括:扩展自定义后端存储(下文2.1讲述)和扩展自定义分词器(下文2.2讲述)。插件实现者关注其中之一即可,若有需要当然也可两者同时进行扩展,或者只扩展下文2.1.3所述系列化器等。

2.1 扩展自定义后端

2.1.1 实现接口BackendStoreProvider
  • 可实现接口:com.baidu.hugegraph.backend.store.BackendStoreProvider
  • 或者继承抽象类:com.baidu.hugegraph.backend.store.AbstractBackendStoreProvider

以RocksDB后端RocksDBStoreProvider为例:

public class RocksDBStoreProvider extends AbstractBackendStoreProvider {

    protected String database() {
        return this.graph().toLowerCase();
    }

    @Override
    protected BackendStore newSchemaStore(String store) {
        return new RocksDBSchemaStore(this, this.database(), store);
    }

    @Override
    protected BackendStore newGraphStore(String store) {
        return new RocksDBGraphStore(this, this.database(), store);
    }

    @Override
    public String type() {
        return "rocksdb";
    }

    @Override
    public String version() {
        return "1.0";
    }
}
2.1.2 实现接口BackendStore

BackendStore接口定义如下:

public interface BackendStore {
    // Store name
    public String store();

    // Database name
    public String database();

    // Get the parent provider
    public BackendStoreProvider provider();

    // Open/close database
    public void open(HugeConfig config);
    public void close();

    // Initialize/clear database
    public void init();
    public void clear();

    // Add/delete data
    public void mutate(BackendMutation mutation);

    // Query data
    public Iterator<BackendEntry> query(Query query);

    // Transaction
    public void beginTx();
    public void commitTx();
    public void rollbackTx();

    // Get metadata by key
    public <R> R metadata(HugeType type, String meta, Object[] args);

    // Backend features
    public BackendFeatures features();

    // Generate an id for a specific type
    public Id nextId(HugeType type);
}
2.1.3 扩展自定义系列化器

系列化器必须继承抽象类:com.baidu.hugegraph.backend.serializer.AbstractSerializer(implements GraphSerializer, SchemaSerializer)
主要接口的定义如下:

public interface GraphSerializer {
    public BackendEntry writeVertex(HugeVertex vertex);
    public BackendEntry writeVertexProperty(HugeVertexProperty<?> prop);
    public HugeVertex readVertex(HugeGraph graph, BackendEntry entry);
    public BackendEntry writeEdge(HugeEdge edge);
    public BackendEntry writeEdgeProperty(HugeEdgeProperty<?> prop);
    public HugeEdge readEdge(HugeGraph graph, BackendEntry entry);
    public BackendEntry writeIndex(HugeIndex index);
    public HugeIndex readIndex(HugeGraph graph, ConditionQuery query, BackendEntry entry);
    public BackendEntry writeId(HugeType type, Id id);
    public Query writeQuery(Query query);
}

public interface SchemaSerializer {
    public BackendEntry writeVertexLabel(VertexLabel vertexLabel);
    public VertexLabel readVertexLabel(HugeGraph graph, BackendEntry entry);
    public BackendEntry writeEdgeLabel(EdgeLabel edgeLabel);
    public EdgeLabel readEdgeLabel(HugeGraph graph, BackendEntry entry);
    public BackendEntry writePropertyKey(PropertyKey propertyKey);
    public PropertyKey readPropertyKey(HugeGraph graph, BackendEntry entry);
    public BackendEntry writeIndexLabel(IndexLabel indexLabel);
    public IndexLabel readIndexLabel(HugeGraph graph, BackendEntry entry);
}
2.1.4 扩展自定义配置项

增加自定义后端时,可能需要增加新的配置项,实现流程主要包括:

  • 增加配置项容器类,并实现接口com.baidu.hugegraph.config.OptionHolder
  • 提供单例方法public static OptionHolder instance(),并在对象初始化时调用方法OptionHolder.registerOptions()
  • 增加配置项声明,单值配置项类型为ConfigOption、多值配置项类型为ConfigListOption

以RocksDB配置项定义为例:

public class RocksDBOptions extends OptionHolder {

    private RocksDBOptions() {
        super();
    }

    private static volatile RocksDBOptions instance;

    public static synchronized RocksDBOptions instance() {
        if (instance == null) {
            instance = new RocksDBOptions();
            instance.registerOptions();
        }
        return instance;
    }

    public static final ConfigOption<String> DATA_PATH =
            new ConfigOption<>(
                    "rocksdb.data_path",
                    "The path for storing data of RocksDB.",
                    disallowEmpty(),
                    "rocksdb-data"
            );

    public static final ConfigOption<String> WAL_PATH =
            new ConfigOption<>(
                    "rocksdb.wal_path",
                    "The path for storing WAL of RocksDB.",
                    disallowEmpty(),
                    "rocksdb-data"
            );

    public static final ConfigListOption<String> DATA_DISKS =
            new ConfigListOption<>(
                    "rocksdb.data_disks",
                    false,
                    "The optimized disks for storing data of RocksDB. " +
                    "The format of each element: `STORE/TABLE: /path/to/disk`." +
                    "Allowed keys are [graph/vertex, graph/edge_out, graph/edge_in, " +
                    "graph/secondary_index, graph/range_index]",
                    null,
                    String.class,
                    ImmutableList.of()
            );
}

2.2 扩展自定义分词器

分词器需要实现接口com.baidu.hugegraph.analyzer.Analyzer,以实现一个SpaceAnalyzer空格分词器为例。

package com.baidu.hugegraph.plugin;

import java.util.Arrays;
import java.util.HashSet;
import java.util.Set;

import com.baidu.hugegraph.analyzer.Analyzer;

public class SpaceAnalyzer implements Analyzer {

    @Override
    public Set<String> segment(String text) {
        return new HashSet<>(Arrays.asList(text.split(" ")));
    }
}

3 实现插件接口,并进行注册

插件注册入口为HugeGraphPlugin.register(),自定义插件必须实现该接口方法,在其内部注册上述定义好的扩展项。
接口com.baidu.hugegraph.plugin.HugeGraphPlugin定义如下:

public interface HugeGraphPlugin {

    public String name();

    public void register();

    public String supportsMinVersion();

    public String supportsMaxVersion();
}

并且HugeGraphPlugin提供了4个静态方法用于注册扩展项:

  • registerOptions(String name, String classPath):注册配置项
  • registerBackend(String name, String classPath):注册后端(BackendStoreProvider)
  • registerSerializer(String name, String classPath):注册系列化器
  • registerAnalyzer(String name, String classPath):注册分词器

下面以注册SpaceAnalyzer分词器为例:

package com.baidu.hugegraph.plugin;

public class DemoPlugin implements HugeGraphPlugin {

    @Override
    public String name() {
        return "demo";
    }

    @Override
    public void register() {
        HugeGraphPlugin.registerAnalyzer("demo", SpaceAnalyzer.class.getName());
    }
}

4 配置SPI入口

  1. 确保services目录存在:hugegraph-plugin-demo/resources/META-INF/services
  2. 在services目录下建立文本文件:com.baidu.hugegraph.plugin.HugeGraphPlugin
  3. 文件内容如下:com.baidu.hugegraph.plugin.DemoPlugin

5 打Jar包

通过maven打包,在项目目录下执行命令mvn package,在target目录下会生成Jar包文件。
使用插件时将该Jar包拷到plugins目录,重启服务即可生效。

HugeGraph扩展Stanford分词器完整示例: https://github.com/hugegraph/hugegraph-plugin-stanford-analyzer
HugeGraph GitHub地址: https://github.com/hugegraph/hugegraph

 类似资料: