ORDER BY adjusts the order of rows in the data stream.

Syntax: ORDER BY <expression> <string>, <expression> <string>, ...
Input:

• <expression>: Sorting basis; multiple bases must be homologous and are operated from left to right
• <alias>: Fashion of sorting, either asc or desc (case insensitive), or asc when omitted

find().nodes([4, 2]) as nodes
n(nodes).e().n(as n) as path
order by n.radius desc, n
return path

In the UQL above, ORDER BY adjusts the order of the one-step paths resulted from template query, first by radius of terminal nodes n in descending order, then by ID of terminal nodes n that share the same radius in ascending order.

ATTR (Numeric value)

Example: Find 10 cards that have the highest balances, i.e, sorts the card balances in a descending order

find().nodes({@card}) as n
order by n.balance desc 
return n{*} limit 10

ATTR (Time)

Example: Find cards of Customer CU001, return cards' info in an ascending time order

n({_id == "CU001"}).re({@has}).n({@card} as n)
order by n.open_date
return n{*}

ATTR (String)

Example: Find 10 email addresses, sort them in a descending order of letters from addresses

find().nodes({@email}) as n
limit 10
order by n.address desc 
return n{*}

NODE

Example: Find 10 two-step transfer paths, sort paths in a descending order of 2nd nodes' UUIDs

n({@card}).e({@transfer}).n({@card} as n)
  .e({@transfer}).n({@card}) as p
limit 10
order by n desc
return p{*}

EDGE

Example: Find 10 two-step transfer paths, sort paths in a descending order of the 2nd edges' UUIDs

n({@card}).e({@transfer}).n({@card})
  .e({@transfer} as e).n({@card}) as p
limit 10
order by e desc
return p{*}

GROUP BY and ORDER BY

Example: Sort all cards in groups by their levels, count the number of cards at each level, return the 3 levels with the most cards and card numbers at these levels

find().nodes({@card}) as n
group by n.level as level
return level, count(n) as total limit 3
order by total desc