[jira] [Commented] (TINKERPOP-1482) has(x).has(y) chains should be has(x.and(y))

["Marko A. Rodriguez (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/TINKERPOP-1482?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15533295#comment-15533295 ] 

Marko A. Rodriguez commented on TINKERPOP-1482:
-----------------------------------------------

Note that this work should also address {{HasContainer.makeHasContainers}} where {{AndP}}-predicate are split apart into individual {{HasContainers}}. This is hacky and bad.

> has(x).has(y) chains should be has(x.and(y))
> --------------------------------------------
>
>                 Key: TINKERPOP-1482
>                 URL: https://issues.apache.org/jira/browse/TINKERPOP-1482
>             Project: TinkerPop
>          Issue Type: Improvement
>          Components: process
>    Affects Versions: 3.2.2
>            Reporter: Marko A. Rodriguez
>
> Lots of work has been done to inline filters for 3.2.3. A final application should be to inline {{has(x).has(y)}} into {{has(x.and(y))}}. If this is done, providers may have significant breaking changes as they will need to search {{HasContainers}} beyond {{P.eq()}} to walking an and/or-tree. However, the benefit of this grouping is two fold:
> 1. Its a clean, consistent composition.
> 2. It reduces the number of {{HasContainers}} and {{HasSteps}} in areas where provider "fold-ins" are not being used.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)