Skip to content

Logical Operators

Edit page

Each query produces a list of documents with a weight according to how good a match each document is for that query. These queries can then be combined to produce a more complex tree-like query structure, with the operators acting as branches within the tree.

The most basic operators are the logical operators: _or, _and and _not - these match documents in the following way:

  • _or - Finds documents which match any of the subqueries.
  • _and - Finds documents which match all of the subqueries.
  • _not - Finds documents which match none of the subqueries: a list is match_all AND_NOT (A OR B ...), and a single subquery is a list of one.
  • _and_not - Finds documents which match the first subquery A but not subquery B.
  • _xor - Finds documents which are matched by subquery A or subquery B, but not both.

Each operator produces a weight for each document it matches, which depends on the weight of one or both subqueries in the following way:

  • _or - Matches documents with the sum of all weights of the subqueries.
  • _and - Matches documents with the sum of all weights of the subqueries.
  • _not - Matches documents with no weight (a match-all weighs nothing).
  • _and_not - Matches documents with the weight from subquery A only.
  • _xor - Matches documents with the sum of the weights of the subqueries that match.

For example, the following matches all of those who either like bananas or are brown-eyed females:

SEARCH /bank/
{
"_query": {
"_or": [
{
"favoriteFruit": "banana"
},
{
"_and": [
{ "gender": "female" },
{ "eyeColor": "brown" }
]
}
]
}
}