Perform a Gremlin Query that finds Parts that have related Requirements and which groups by the Part

GremlinGroovyPartsRequirementsQuery.groovy

g.V().has( 'ArtifactType', 'name', 'Part' )
  .match( 
     __.as( 't' ).in( 'hasType' ).as( 'p' ),
     __.as( 't' ).out( 'ownedBy' ).as( 'r' ), 
     __.as( 'r' ).out( 'hasType' ).filter{ it.get().value( 'name' ).contains( 'Windchill' ) }.as( 'rt' ),
     __.as( 'p' ).both( 'Relation' ).as( 'o' ),
     __.as( 'o' ).out( 'hasType' ).filter{ it.get().value( 'name' ) ==~ /(?i).*?((requirement)|(task)).*/ }
  )
  .group()
      .by( select( 'p' ) ).unfold().project( 'part', 'requirements' )
         .by( select(keys).map { it.get().value( 'name' ) } )
         .by( select(values).unfold().select( 'o' ).map { it.get().value( 'name' ) }.fold() )

==>{part=Bilge_Water_Subsystem, requirements=[ME-7, Bilgewater Control]}
==>{part=NOx_SCR_Subsystem, requirements=[ME-14]}
==>{part=Cooling_Water_Subsystem, requirements=[ME-10, ME-7]}
==>{part=MetalGlass_Waste_Storage, requirements=[ME-13]}
==>{part=Body, requirements=[Payload Spec]}
==>{part=Incinerator, requirements=[ME-12]}
==>{part=SOx_Closed_Loop_Scrubber, requirements=[ME-15]}
==>{part=Freshwater_Subsystem, requirements=[ME-5]}
==>{part=AAA_20220618, requirements=[ME-7]}
==>{part=Ballast_Water_Subsystem, requirements=[ME-6]}
==>{part=Hazardous_Waste_Storage, requirements=[ME-13]}
==>{part=Aircraft_Platform, requirements=[HELLO-2831, HELLO-3007]}
==>{part=Graywater_Treatment, requirements=[ME-11]}

As the query leaves the match() step, it will have a Traversal that has vertices and edges which are selectable with the matching p, r, rt, and o keys. We only are interested in the pairs (p,o).

We group the match results by their p element (the Part).

We unfold the group result to get back a stream of pairs (key = p, value = the match resultset).

We want to project results into two sets: part and requirement.

For the part, we select the key of the grouped match result and map the Part to its name.

For the requirement, we select the value of the grouped match result, unfold it, select the 'o' entry, and map that vertex to its name. We fold back the unfolded, now subsampled and mapped, requirements value.