Cheat Sheet

Everything in one place.

Query basics

Pattern	Example	Result shape
db.table	shop.customers	array of rows
db.table{a, b}	shop.customers{id, name}	array with only those keys
db.table[col = "x"]	shop.orders[status = "paid"]	filtered array
db.table[0:10]	shop.orders[0:10]	first 10 rows
db.table[5]	shop.orders[5]	single row object
db.table._desc(col)	shop.orders._desc(placed_at)	sorted array
db.table._count()	shop.orders._count()	scalar number
db.table.relation	shop.customers.orders	walk to related rows
db.table{relation{a}}	shop.customers{orders{total}}	embed relation

Shape rules

• Projection {…} is always last — nothing at the same level after it.
• Filters […], slices [a:b], and aggregates _fn(…) chain in any order, any number.
• Scalar aggregates (_count, _sum, _avg, _min, _max) end the query and return a value; no {…} needed.

Comparison operators

Op	Meaning
= ==	Equal
!=	Not equal
< <=	Less than
> >=	Greater than
and	Logical AND
or	Logical OR

Arithmetic

Op	Example	Result
+	shop.orders._count() + 5	47
-	shop.orders._sum(total) - 100	1742.50
*	shop.products._count() * 2	84
/	shop.orders._sum(total) / shop.orders._count()	43.87
%	shop.orders._count() % 10	2
**	2 ** 10	1024

Slicing

Form	Meaning
[a:b]	Rows a..b-1
[:b]	First b rows
[a:]	From a to end
[a:b:s]	With step s
[-N:]	Last N rows

Aggregates

Function	Returns	Example output
_count()	number	42
_sum(col)	number	1842.50
_avg(col)	number	43.87
_min(col)	number	9.99
_max(col)	number	299.00
_unique(col)	array	["IN","US","DE"]
_asc(col)	sort	—
_desc(col)	sort	—
_like(p)	bool (in filter)	—
_date(col, fmt)	parsed/formatted date	—

Relationship types

Type	Shape	Example
oto	Object	Customer → Profile
otm	Array	Customer → Orders
mto	Object	Order → Customer
mtm	Array	Product ↔ Categories

Request fields

Field	Purpose
protopass	Protocol identifier (use "default" for the auto-generated one)
query	The ONQL string
ctxkey	Context name (e.g. account)
ctxvalues	Substitutions for $1, $2, ...

Context rule

"orders": {
  "context": {
    "customer": "shop.orders[customer_id = $1]",
    "admin":    "shop.orders"
  }
}