如何使用Cypher在Neo4j中返回随机记录？

Question

我有这样的查询：

$query = "MATCH (U:User)
          RETURN U
          ORDER BY RAND()
          LIMIT 100";

但是当我运行它时，这会显示错误。 它说：

捕获异常：无法执行查询[400]：标题：数组（[Content-Type] => application / json; charset = UTF-8; stream = true [Access-Control-Allow-Origin] => * [转移 - 编码] => chunked [Server] => Jetty（9.0.z-SNAPSHOT））正文：数组（[message] => ORDER BY表达式必须是确定性的。例如，你不能在表达式中使用rand（）函数[ exception] => PatternException [fullname] => org.neo4j.cypher.PatternException [stacktrace] => Array（[0] => org.neo4j.cypher.internal.compiler.v2_0.commands.SortItem.apply（SortItem.scala） ：30）[1] => org.neo4j.cypher.internal.compiler.v2_0.pipes.ExecutionContextComparer $ class.compareBy（SortPipe.scala：43）[2] => org.neo4j.cypher.internal.compiler.v2_0 .pipes.TopPipe.compareBy（TopPipe.scala：33）[3] => org.neo4j.cypher.internal.compiler.v2_0.pipes.TopPipe $$ anonfun $ 1 $$ anonfun $ apply $ 1.apply（TopPipe.scala： 38）[4] => org.neo4j.cypher.internal.compiler.v2_0.pipes.TopPipe $$ anonfun $ 1 $$ anonfun $ apply $ 1.apply（TopPipe.scala：38）[5] => sc ala.Option.forall（Option.scala：226）[6] => org.neo4j.cypher.internal.compiler.v2_0.pipes.TopPipe $$ anonfun $ 1.apply（TopPipe.scala：38）[7] => org.neo4j.cypher.internal.compiler.v2_0.pipes.TopPipe $$ anonfun $ 1.apply（TopPipe.scala：38）[8] => org.neo4j.cypher.internal.compiler.v2_0.pipes.TopPipe $$ anonfun $ internalCreateResults $ 1.apply（TopPipe.scala：56）[9] => org.neo4j.cypher.internal.compiler.v2_0.pipes.TopPipe $$ anonfun $ internalCreateResults $ 1.apply（TopPipe.scala：49）[10 ] => scala.collection.Iterator $ class.foreach（Iterator.scala：727）[11] => org.neo4j.cypher.internal.compiler.v2_0.pipes.HeadAndTail.foreach（SlicePipe.scala：72）[12 ] => org.neo4j.cypher.internal.compiler.v2_0.pipes.TopPipe.internalCreateResults（TopPipe.scala：49）[13] => org.neo4j.cypher.internal.compiler.v2_0.pipes.PipeWithSource.createResults（ Pipe.scala：71）[14] => org.neo4j.cypher.internal.compiler.v2_0.pipes.PipeWithSource.createResults（Pipe.scala：68）[15] => org.neo4j.cypher.internal.compiler。 v2_0.executionplan.ExecutionPlanB uilder.org $ neo4j $ cypher $ internal $ compiler $ v2_0 $ executionplan $ ExecutionPlanBuilder $$ prepareStateAndResult（ExecutionPlanBuilder.scala：149）[16] => org.neo4j.cypher.internal.compiler.v2_0.executionplan.ExecutionPlanBuilder $$ anonfun $ 2.apply（ExecutionPlanBuilder.scala：126）[17] => org.neo4j.cypher.internal.compiler.v2_0.executionplan.ExecutionPlanBuilder $$ anonfun $ 2.apply（ExecutionPlanBuilder.scala：125）[18] => org。 neo4j.cypher.internal.compiler.v2_0.executionplan.ExecutionPlanBuilder $$ anon $ 6.execute（ExecutionPlanBuilder.scala：50）[19] => org.neo4j.cypher.internal.ExecutionPlanWrapperForV2_0.execute（CypherCompiler.scala：93）[ 20] => org.neo4j.cypher.ExecutionEngine.execute（ExecutionEngine.scala：61）[21] => org.neo4j.cypher.ExecutionEngine.execute（ExecutionEngine.scala：65）[22] => org.neo4j。 cypher.javacompat.ExecutionEngine.execute（ExecutionEngine.java:78）[23] => org.neo4j.server.rest.web.CypherService.cypher（CypherService.java:100）[24] => java.lang.reflect。 Method.invoke（未知来源）[25] => org.neo4j.server.rest.transactional.TransactionalRequestDispatcher.dispatch（TransactionalRequestDispatcher.java:139)[26] => org.neo4j.server.rest.security.SecurityFilter.doFilter（SecurityFilter.java:112)[27] => java.lang.Thread.run（未知来源）））

请帮我。 谢谢。

Answer 1

您需要按节点属性排序，而不是按函数排序。 您可以执行以下操作（如果您的节点包含例如属性'name'）：

MATCH (u:User)
WITH u, rand() AS number
RETURN u
ORDER BY number
LIMIT 100