跳转至

WITHGraph

WITH子句可以获取并处理查询前半部分的结果,并将处理结果作为输入传递给查询的后半部分。

openCypher兼容性Graph

本文操作仅适用于openCypher方式。

Note

在原生nGQL中,有与WITH类似的Graph,但它们的工作方式不同。不要在openCypher方式中使用管道符,也不要在原生nGQL中使用WITH子句。

组成复合查询Graph

使用WITH子句可以组合语句,将一条语句的输出转换为另一条语句的输入。

示例1Graph

  1. 匹配一个路径。
  2. 通过nodes()函数将路径上的所有点输出到一个列表。
  3. 将列表拆分为行。
  4. 去重后返回点的信息。
nebula> MATCH p=(v:player{name:"Tim Duncan"})--() \
        WITH nodes(p) AS n \
        UNWIND n AS n1 \
        RETURN DISTINCT n1;
+-----------------------------------------------------------+
| n1                                                        |
+-----------------------------------------------------------+
| ("player100" :player{age: 42, name: "Tim Duncan"})        |
| ("player101" :player{age: 36, name: "Tony Parker"})       |
| ("team204" :team{name: "Spurs"})                          |
| ("player102" :player{age: 33, name: "LaMarcus Aldridge"}) |
| ("player125" :player{age: 41, name: "Manu Ginobili"})     |
| ("player104" :player{age: 32, name: "Marco Belinelli"})   |
| ("player144" :player{age: 47, name: "Shaquille O'Neal"})  |
| ("player105" :player{age: 31, name: "Danny Green"})       |
| ("player113" :player{age: 29, name: "Dejounte Murray"})   |
| ("player107" :player{age: 32, name: "Aron Baynes"})       |
| ("player109" :player{age: 34, name: "Tiago Splitter"})    |
| ("player108" :player{age: 36, name: "Boris Diaw"})        |
+-----------------------------------------------------------+

示例2Graph

  1. 匹配点ID为player100的点。
  2. 通过labels()函数将点的所有Tag输出到一个列表。
  3. 将列表拆分为行。
  4. 返回结果。
nebula> MATCH (v) \
        WHERE id(v)=="player100" \
        WITH labels(v) AS tags_unf \
        UNWIND tags_unf AS tags_f \
        RETURN tags_f;
+----------+
| tags_f   |
+----------+
| "player" |
+----------+

过滤聚合查询Graph

WITH可以在聚合查询中作为过滤器使用。

nebula> MATCH (v:player)-->(v2:player) \
        WITH DISTINCT v2 AS v2, v2.age AS Age \
        ORDER BY Age \
        WHERE Age<25 \
        RETURN v2.name AS Name, Age;
+----------------------+-----+
| Name                 | Age |
+----------------------+-----+
| "Luka Doncic"        | 20  |
| "Ben Simmons"        | 22  |
| "Kristaps Porzingis" | 23  |
+----------------------+-----+

collect()之前处理输出Graph

collect()函数将输出结果转换为列表之前,可以使用WITH子句排序和限制输出结果。

nebula> MATCH (v:player) \
        WITH v.name AS Name \
        ORDER BY Name DESC \
        LIMIT 3 \
        RETURN collect(Name);
+-----------------------------------------------+
| collect(Name)                                 |
+-----------------------------------------------+
| ["Yao Ming", "Vince Carter", "Tracy McGrady"] |
+-----------------------------------------------+

结合RETURN语句使用Graph

WITH子句中设置别名,并通过RETURN子句输出结果。

nebula> WITH [1, 2, 3] AS list  RETURN 3 IN list AS r;
+------+
| r    |
+------+
| true |
+------+

nebula> WITH 4 AS one, 3 AS two RETURN one > two AS result;
+--------+
| result |
+--------+
| true   |
+--------+

最后更新: November 1, 2021
Back to top