Skip to main content

Filters

Filters narrow down which rows are returned. Wrap them in square brackets [ ].

Equality​

shop.customers[status = "active"]

Response (example):

[
  { "id": "c1", "name": "Ada",   "status": "active" },
  { "id": "c2", "name": "Linus", "status": "active" }
]

Comparison​

shop.customers[age > 18]
shop.products[price <= 999]

Multiple conditions​

Combine with and / or in a single filter:

shop.customers[status = "active" and country = "IN"]

Or chain multiple filter brackets — they compose with and:

shop.customers[status = "active"][country = "IN"]
// equivalent to the query above
shop.orders[status = "open" or status = "partial"]

Pattern match​

Use the _like aggregate for SQL-style LIKE:

shop.customers[email._like("%@gmail.com")]

Parameters​

Use $1, $2, ... as placeholders. They're substituted at query time, which avoids string-concatenation bugs and lets you reuse the same query with different inputs.

shop.customers[email = $1]

You pass values via the ctxvalues field of the request.

Combine with projection​

Filter, slice, and project all in the same query — but remember, projection always comes last:

shop.customers[country = "IN"][0:10]{id, name, email}

Reads as: customers where country = IN, take the first 10, return id/name/email.

Response:

[
  { "id": "c1", "name": "Ada",    "email": "ada@example.com" },
  { "id": "c2", "name": "Anjali", "email": "anjali@example.com" }
]

Filters and slices can appear in any order and repeat; projection is terminal.

What's next​

Once you can pick the rows you want, learn how to page and slice results.