A path template n().e()...n() can apply filters on each node and edge of a path independently. It is a very flexible way of path query.

Syntax:

• Statement alias: supported (PATH)
• Prefix: OPTIOANL (returns a pseudo path for any subquery that finds no result, with empty string as _id and 0 as _uuid for all nodes and edges in the path)
• Optional parameters:

1 step

Example: as the path structure shown below, find transactions from a customer's card to another customer's card, return 10 paths and their properties

n({@customer}).re({@has}).n({@card})
  .re({@transfer}).n({@card})
  .le({@has}).n({@customer}) as p
limit 10 
return p{*}

N steps

Example: as the path structure shown below, find 3-step transaction paths from Customer CU001's card, return 10 paths and their properties

n({_id == "CU001"}).re({@has}).n({@card})
  .re({@transfer})[3].n({@card}) as p
limit 10 
return p{*}

1~N steps

Example: find 1 to 3 steps' transaction paths from Customer CU001's card, return 10 paths and their properties

n({_id == "CU001"}).re({@has}).n({@card})
  .re({@transfer})[:3].n({@card})
  .le({@has}).n({@customer}) as p
limit 10 
return p{*}

M~N steps

Example: find 2 to 3 steps' transaction paths from Customer CU001's card, return 10 paths and their properties

n({_id == "CU001"}).re({@has}).n({@card})
  .re({@transfer})[2:3].n({@card})
  .le({@has}).n({@customer}) as p
limit 10 
return p{*}

Non-weighted Shortest Path

Example: as the path structure shown below, find the shortest transaction paths from Customer CU001's card to Customer CU002's card within 3 steps, return 10 paths and their properties.

n({_id == "CU001"}).re({@has}).n({@card} as n1)
n({_id == "CU002"}).re({@has}).n({@card} as n2)
with n1, n2
n(n1).re({@transfer})[*:3].n(n2) as p 
limit 10
return p{*}

Analysis: in this example, two path templates are first applied to find Customer CU001 and Customer CU002's cards respectively, then one card is used as scr and another as dest in a third transaction path template to find the shortest path between them.

Circle

Example: as the path structure shown below, find the circular single-direction transaction paths from Customer CU001's card within 4 steps, return 10 paths and their properties

n({_id == "CU001"}).re({@has}).n({@card} as n)
  .re({@transfer})[:4].n(n) as p
limit 10
return p{*}

no_circle()

Example: find the non-circular single-direction transaction paths from Customer CU001's card within 4 steps, return 10 paths and their properties

n({_id == "CU001"}).re({@has}).n({@card})
  .re({@transfer})[:4].n({@card})
  .no_circle() as p
limit 10
return p{*}

limit

Example: find the transaction paths from Card CA001, CA002, and CA005 to Card CA016 within 3 steps, return 2 paths for each pair and their properties

find().nodes({_id in ["CA001", "CA002", "CA005"]}) as n
n(n).e()[:3].n({_id == "CA016"})
  .limit(2) as p
return p{*}

OPTIONAL

Example: find the transaction paths from Card CA001, CA002, and CA005 to Card CA016 within 3 steps, return 2 paths for each pair and their properties; if no paths returned, return optional paths

find().nodes({_id in ["CA001", "CA002", "CA005"]}) as n
optional n(n).e()[:3].n({_id == "CA016"})
  .limit(2) as p
return p{*}

0~N steps

Example: find customers who viewed or purchased Product P001, then find other customers 0 to 2 steps away from those customers, return all secondary customers' properties.

n({_id == "P001"}).e().n({@customer})
  .e()[0:2].n({@customer}) as p
return p{*}

Inter-step Filter

Example: find the outbound @transfer paths from Card CA001 to Card CA002 within 4 steps, with each @transfer time greater than the previous one within the same path, return 10 paths and their properties.

n({_id == "CA001"}).re({@transfer.time > prev_e.time})[:4].n({_id == "CA002"}) as p
limit 10
return p{*}