LocalRunner的疑问

[ganhb]

你们好：
我在看LocalRunnerTest的testCreateConceptInstance 有几个问题，望解答。

问题一

我尝试把代码定义的整理成schema文件，内容如下：

namespace Test

TaxOfUserFeature(用户角色分类):EntityType

UserFeature(用户角色):EntityType
	properties:
		newedge(某种边):TaxOfUserFeature

User(用户)：EntityType
	properties:
		nightTrader(某个标识):Text
		belongTo(属于): UserFeature
	relations:
		belongTo(白领):UserFeature
			rule：[[
			    Define (s:Test.User)-[p:belongTo]->(o:`Test.UserFeature`/`白领`) {
			    GraphStructure {
			        (s)
			    }
			    Rule {
			        r1("属于教师") = s.nightTrader == 1
			        r2("属于程序员") = s.nightTrader == 2
			        r3("属于医生") = s.nightTrader == 3
			        s_r1 = rule_value(r1, "教师", "")
			        s_r2 = rule_value(r2, "程序员", s_r1)
			        s_r3 = rule_value(r3, "医生", s_r2)
			        R: s_r3 != ""
			    }
			    Action {
			        sub_concept = createNodeInstance(
			            type=Test.UserFeature,
			            value={
			                id=concat("白领-", s_r3)
			            }
			        )
			        createEdgeInstance(
			            src=s,
			            dst=sub_concept,
			            type=belongTo,
			            value={
			                __to_id_type__='Test.UserFeature'
			                __from_id_type__='Test.User'
			            }
			        )
			    }
			}
			]]
		belongTo(学生):UserFeature
			rule:[[
			    Define (s:Test.User)-[p:belongTo]->(o:`Test.UserFeature`/`学生`) {
			    GraphStructure {
			        (s)
			    }
			    Rule {
			        r1("就读幼儿园") = s.nightTrader == 1
			        r2("就读小学") = s.nightTrader == 2
			        r3("就读中学") = s.nightTrader == 3
			        s_r1 = rule_value(r1, "就读幼儿园", "")
			        s_r2 = rule_value(r2, "就读小学", s_r1)
			        s_r3 = rule_value(r3, "就读中学", s_r2)
			        R: s_r3 != ""
			    }
			    Action {
			        sub_concept = createNodeInstance(
			            type=Test.UserFeature,
			            value={
			                id=concat("学生-", s_r3)
			            }
			        )
			        createEdgeInstance(
			            src=s,
			            dst=sub_concept,
			            type=belongTo,
			            value={
			                __to_id_type__='Test.UserFeature'
			                __from_id_type__='Test.User'
			            }
			        )
			    }
			}
			]]	

其中LocalRunnerTest的testCreateConceptInstance 对于

    schema.put(
        "Test.User_belongTo_Test.UserFeature",
        Convert2ScalaUtil.toScalaImmutableSet(Sets.newHashSet()));

  xxx.addEdge(
            "Test.User",
            "belongTo",
            "Test.UserFeature/白领",
            Direction.OUT,
            Convert2ScalaUtil.toScalaImmutableSet(Sets.newHashSet())

xxx.addEdge(
            "Test.User",
            "belongTo",
            "Test.UserFeature/学生",
            Direction.OUT,
            Convert2ScalaUtil.toScalaImmutableSet(Sets.newHashSet())

我分别给它映射成（感觉有问题）

	properties:
		belongTo(属于): UserFeature
	relations:
		belongTo(白领):UserFeature
	        belongTo(学生):UserFeature

我看完文档声明式Schema，我感觉不对，希望指导下正确格式应该是什么？

问题二

在实例mock数据中TestFanxiqianGraphLoader

constructionVertex(
      String bizId, String type, Object... kvs) {

type是否是实例和概念的联系，上述例子相当于
id:张三 instanceOf Test:User

问题三

在进行规则推理时候，能否可以在没有实例图的情况下（仅有概念图）进行的推理？