Queries are represented as JSON dictionaries.

id

The query id. This has to be an integer (signed)

query

A dictionary of condition arrays. The dictionary may contain the keys and, or and not. These keys can be used to model complex logical relationships between conditions. The arrays can also contain additional objects with one of these keys to combine logical relationships.

Example

...
"query": {
    "and": [
        {
            "or": [
                {
                    "field": "category",
                    "condition": "eq",
                    "value": "bargain"
                },
                {
                    "field": "price",
                    "condition": "lt",
                    "value": 500000
                }
            ]
        },
        {
            "field": "description",
            "condition": "match",
            "value": "haunted house"
        }
    ]
}
...

This example could also be written like this:

(category == "bargain" || price < 500000) && match(description, "haunted house")

Caution

Be careful using OR and NOT logic. This will usually impact performance negatively.

Condition dictionary

field

Name of the field to match in document. This field has to be available in the schema.

condition

How the value has to relate to the document field.

Condition Description
match Used for matching FullText fields. This is the only condition that can be used with that field type.
radius Used for bounding box searches.
eq Equals: value in document field has to be the exact same value.
gt Greater than: value in document field must be greater than this.
lt Less than: value in document field must be less than this.
gte Greater than, equals: value in document field must be greater than this or equal.
lte Less than, equals: value in document field must be greater than this or equal.

value

The document field will be matched against this value. Depending on field type, the data type of value varies.

String: “strvalue”

Integer, Long: 12345

Double: 1.0

FullText: “foo bar”

Date: “2014-08-10T19:22:51Z” (default) or custom format

GeoDistance:

{
    "lat": 53.55,
    "lng": 10,
    "distance": 100000
}

Latitude, longitude and radius in Meters.

not

Let’s you invert a condition. If not is set to true on a condition only documents that do not match this condition will be matched.

Example

{
    "id": 123456,
    "query": {
        "and": [
            {
                "field": "description",
                "condition": "match",
                "value": "haunted house"
            },
            {
                "field": "price",
                "condition": "gte",
                "value": 500000
            },
            {
                "field": "location",
                "condition": "radius",
                "value": {
                    "lat": 53.55,
                    "lng": 10,
                    "distance": 100000
                }
            }
        ]
    }
}