• 单机离线调用模式；
    • Client/Server请求模式；

    由于lib目录下已经包含了HugeCore的jar包，且HugeGraph已经作为插件注册到Console中，用户可以直接写groovy脚本调用HugeGraph-Core的代码，然后交由Gremlin-Console内的解析引擎执行，就能在不启动Server的情况下操作图。

    这种模式便于用户快速上手体验，但是不适合大量数据插入和查询的场景。下面给一个示例：

    在script目录下有一个示例脚本：example.groovy

    其实这一段groovy脚本几乎就是Java代码，不同之处仅在于变量的定义可以不写类型声明，以及每一行末尾的分号可以去掉。

    2. objc[5038]: Class JavaLaunchHelper is implemented in both /Library/Java/JavaVirtualMachines/jdk1.8.0_121.jdk/Contents/Home/bin/java (0x10137a4c0) and /Library/Java/JavaVirtualMachines/jdk1.8.0_121.jdk/Contents/Home/jre/lib/libinstrument.dylib (0x102bbb4e0). One of the two will be used. Which one is undefined.
    4.          \,,,/
    5.          (o o)
    6. -----oOOo-(3)-oOOo-----
    7. plugin activated: com.baidu.hugegraph
    8. plugin activated: tinkerpop.server
    9. plugin activated: tinkerpop.utilities
    10. plugin activated: tinkerpop.tinkergraph
    11. 2018-01-15 14:36:19 7516  [main] [WARN ] com.baidu.hugegraph.config.HugeConfig [] - The config option 'rocksdb.data_path' is redundant, please ensure it has been registered
    12. 2018-01-15 14:36:19 7523  [main] [WARN ] com.baidu.hugegraph.config.HugeConfig [] - The config option 'rocksdb.wal_path' is redundant, please ensure it has been registered
    13. 2018-01-15 14:36:19 7604  [main] [INFO ] com.baidu.hugegraph.HugeGraph [] - Opening backend store 'cassandra' for graph 'hugegraph'
    14. >>>> query all vertices: size=6
    15. >>>> query all edges: size=6

    可以看到，插入了6个顶点、6条边，并查询出来了。进入console之后，还可继续输入groovy语句对图做操作：

    1. gremlin> g.V()
    2. ==>v[2:ripple]
    3. ==>v[1:vadas]
    4. ==>v[1:peter]
    5. ==>v[1:josh]
    6. ==>v[1:marko]
    7. ==>v[2:lop]
    8. ==>e[S1:josh>2>>S2:ripple][1:josh-created->2:ripple]
    9. ==>e[S1:marko>1>20160110>S1:vadas][1:marko-knows->1:vadas]
    10. ==>e[S1:peter>2>>S2:lop][1:peter-created->2:lop]
    12. ==>e[S1:marko>1>20130220>S1:josh][1:marko-knows->1:josh]
    13. ==>e[S1:marko>2>>S2:lop][1:marko-created->2:lop]

    更多的Gremlin语句请参考

    2 Client/Server请求模式

    因为Gremlin-Console只能通过WebSocket连接HugeGraph-Server，默认HugeGraph-Server是对外提供HTTP连接的，所以先修改gremlin-server的配置。

    注意：将连接方式修改为WebSocket后，HugeGraph-Client、HugeGraph-Loader、HugeGraph-Studio等配套工具都不能使用了。

    将channelizer: org.apache.tinkerpop.gremlin.server.channel.HttpChannelizer修改成channelizer: org.apache.tinkerpop.gremlin.server.channel.WebSocketChannelizer或直接注释，然后按照步骤启动Server。

    1. bin/gremlin-console.sh 
    2. objc[5761]: Class JavaLaunchHelper is implemented in both /Library/Java/JavaVirtualMachines/jdk1.8.0_121.jdk/Contents/Home/bin/java (0x10ec584c0) and /Library/Java/JavaVirtualMachines/jdk1.8.0_121.jdk/Contents/Home/jre/lib/libinstrument.dylib (0x10ecdc4e0). One of the two will be used. Which one is undefined.
    4.          \,,,/
    5.          (o o)
    6. -----oOOo-(3)-oOOo-----
    7. plugin activated: com.baidu.hugegraph
    8. plugin activated: tinkerpop.server
    9. plugin activated: tinkerpop.utilities
    10. plugin activated: tinkerpop.tinkergraph

    连接server，需在配置文件中指定连接参数，在conf目录下有一个默认的remote.yaml

    1. # cat conf/remote.yaml
    2. hosts: [localhost]
    3. port: 8182
    4. serializer: {
    5.   className: org.apache.tinkerpop.gremlin.driver.ser.GryoMessageSerializerV1d0,
    6.   config: {
    7.     serializeResultToString: true,
    8.     ioRegistries: [com.baidu.hugegraph.io.HugeGraphIoRegistry]
    9.   }
    10. }

    连接成功之后，在console的上下文中能使用的变量只有hugegraph和hugegraph1两个图对象（在gremlin-server.yaml中配置），如果想拥有更多的变量，可以在scripts/empty-sample.groovy中添加，如:

    2. // an init script that returns a Map allows explicit setting of global bindings.
    3. def globals = [:]
    5. // defines a sample LifeCycleHook that prints some output to the Gremlin Server console.
    6. // note that the name of the key in the "global" map is unimportant.
    7. globals << [hook: [
    8.         onStartUp : { ctx ->
    9.             ctx.logger.info("Executed once at startup of Gremlin Server.")
    10.         },
    11.         onShutDown: { ctx ->
    12.             ctx.logger.info("Executed once at shutdown of Gremlin Server.")
    13.         }
    14. ] as LifeCycleHook]
    16. // define schema manger for hugegraph
    17. schema = hugegraph.schema()
    18. // define the default TraversalSource to bind queries to - this one will be named "g".
    19. g = hugegraph.traversal()

    这样在console中便可以直接使用schema和g这两个对象，做元数据的管理和图的查询了。

    不定义了也没关系，因为所有的对象都可以通过graph获得，例如：

    1. gremlin> :> hugegraph.traversal().V()
    2. ==>v[2:ripple]
    3. ==>v[1:vadas]
    4. ==>v[1:peter]
    5. ==>v[1:josh]
    6. ==>v[1:marko]

    在Client/Server模式下，所有跟Server有关的操作都要加上:>，如果不加，表示在console本地操作。

    更多关于gremlin-console的使用，请参考