[
https://issues.apache.org/jira/browse/MADLIB-1081?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15954393#comment-15954393
]
Frank McQuillan commented on MADLIB-1081:
-----------------------------------------
Another approach would be to create a summary table in _graph_sssp_ in which
case the user would not need to repeat the _grouping_cols_ parameter again in
the _graph_sssp_get_path_ call. That might be a bit cleaner.
So it would be
{code}
madlib.graph_sssp_get_path(sssp_table, dest_vertex, path_table)
{code}
where the _path_table_ can be optional with a default of _sssp_table_path_
The only downside I see is that _graph_sssp_get_path_ in 1.11 will not be
backward compatible with 1.10, but actually I don't think this is a big deal
since we are planning a 1.11 release pretty soon so likely little if any
re-work by the customer base.
> Graph - add grouping to shortest path
> -------------------------------------
>
> Key: MADLIB-1081
> URL: https://issues.apache.org/jira/browse/MADLIB-1081
> Project: Apache MADlib
> Issue Type: Improvement
> Components: Module: Graph
> Reporter: Frank McQuillan
> Assignee: Orhan Kislal
> Priority: Minor
> Fix For: v1.11
>
>
> * Add a GROUP BY column to the edge table
> * Because wants to run SSSP on the different server graphs defined for users,
> i.e., group by userID
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
