Modify Column.keys in projection query

[Александр Петраки]

Hello Everyone,

I am having a simple projection query which ends with

project('time','capacity).by(keys).by(values)

Query itself works fine, but I need to make a substring of each key. Is there a way to do this?

[HadoopMarc]

Hi Alexander,

See:

https://groups.google.com/g/gremlin-users/c/OqpJwd2WWdk/m/3cKVp8iYDQAJ

Best wishes,    Marc

Op donderdag 15 oktober 2020 om 13:26:36 UTC+2 schreef Александр Петраки:

[Александр Петраки]

But map() can not be applied to Column.keys structure

вс, 18 окт. 2020 г. в 13:21, HadoopMarc <bi...@xs4all.nl>:

--
You received this message because you are subscribed to a topic in the Google Groups "JanusGraph users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/janusgraph-users/G1WUbfpPyHg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to janusgraph-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/janusgraph-users/7e06d78d-faca-4e94-b2bf-f1b3cf5bed70n%40googlegroups.com.

[HadoopMarc]

Hi Alexander,

Readers of this list can help you better if you provide the query you tried on one of the example graphs or a sample graph you provide yourself.

The Column.keys and Column.values are not useful for the project() step. If the properties of your vertices are called time and capacity you can do something like (not tested):

project('time','capacity).by(values('time).map{it.get().substring(1,4)}).by(values('capacity).map{it.get().substring(1,4)})

The time and capacity args in the project step are used as keys in the resulting Map.

Best wishes,   Marc

Op maandag 19 oktober 2020 om 06:42:41 UTC+2 schreef Александр Петраки: