[
https://issues.apache.org/jira/browse/TINKERPOP-1991?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16573098#comment-16573098
]
stephen mallette commented on TINKERPOP-1991:
---------------------------------------------
Hmm - I'm suddenly realizing that sparql-gremlin doesn't work well for GLVs.
The logic for transforming sparql to gremlin (and thus bytecode) occurs on the
client which means the engine resides on the client. There is no such engine
for doing those things in GLVs.
I'm not sure if that's "wrong" but it certainly doesn't make things easier for
GLVs and maintaining parity with the JVM. I guess a {{SparqlTraversalSource}}
would have to be configured on the server for GLVs to work at all and GLVs
would need a sparql step of some sort .....
> Generalized DSL in GLVs
> -----------------------
>
> Key: TINKERPOP-1991
> URL: https://issues.apache.org/jira/browse/TINKERPOP-1991
> Project: TinkerPop
> Issue Type: Improvement
> Components: process
> Affects Versions: 3.3.3
> Reporter: stephen mallette
> Assignee: stephen mallette
> Priority: Major
> Labels: deprecation
>
> Create an Algorithm DSL so that we don't pollute gremlin-core with too many
> of those types of steps (e.g. {{pageRank()}}, {{peerPressure()}}, etc). Will
> need to deprecate the current algorithm steps in core as part of this.
> As part of this ensure that GLVs have a way to deal with new DSLs in the
> future and apply it to both the Algorithm DSL and the sparql-gremlin DSL.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
