Skip to content
This repository has been archived by the owner on Dec 1, 2022. It is now read-only.

[bug] is itlegal to reference a not exist property (a.k.a property name typo) #869

Closed
whitewum opened this issue Mar 24, 2021 · 3 comments
Closed
Assignees

Comments

@whitewum
Copy link
Contributor

match (s1:stocks )-[e:relation]->(s2:stocks) where e.weight > 0.1 return s1.name, e.weight, s2.name

I should type e.weight but typo e.weigth

the return column of e.weigth is NULL.

When I create the edge, I set relation.weight NOT NULL.

SO I confuse a sec, and find out that typo.

But, Is it legal to return a not existed alias or property name?

@Aiee Aiee added the bug Type: something is unexpected label Apr 20, 2021
@CPWstatic
Copy link
Contributor

CPWstatic commented May 13, 2021

As far as i known, it's legal in openCypher/neo4j. Or, we can say that the it was designed to be NULL. For further concern, we should know if the TCK has the relative test cases.

@HarrisChu
Copy link
Contributor

https://discuss.nebula-graph.com.cn/t/topic/4413

It's difficult to find the typo error in a complex query, and the result data set may be confused.

# e.harris is NULL, so these is no data set.
(root@nebula) [nba]>  match (v:player)- [e] -() where e.harris > 1   return v  limit 1
Empty set (time spent 10305/19358 us)

Tue, 15 Jun 2021 14:21:09 CST

@HarrisChu HarrisChu added the med-pri Priority: medium label Jun 15, 2021
@CPWstatic CPWstatic removed bug Type: something is unexpected med-pri Priority: medium labels Jun 24, 2021
@CPWstatic
Copy link
Contributor

As I replied earlier that it's legal in openCypher/neo4j. I'll close this issue first, let us know if you have a further concern.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants