defaultfactory

Defaultfactory, often written as default_factory, is a programming concept describing a callable used to generate default values for missing entries in a data structure. The callable is invoked without arguments to supply a value when a requested key or field does not yet exist. This pattern helps avoid repeated checks for presence and ensures consistent, initialized defaults.

In Python, the term is most commonly encountered in two contexts. The collections.defaultdict class has an attribute named default_factory. When a missing key is accessed, the dictionary calls default_factory to create a value, inserts it under that key, and returns it. If default_factory is None, missing keys raise KeyError. Typical factories include list, set, and int, which create an empty container or a zero value.

The second major use is in dataclasses, where the field function accepts a default_factory keyword argument. This provides a per-instance default value produced by the factory, ensuring that each object gets its own independent mutable value (such as a list). For example, field(default_factory=list) assigns a new empty list to the field for each instance.

Key characteristics: default_factory must be a zero-argument callable; it is invoked only when a new default is needed. It is especially useful for mutable defaults and for lazy initialization. The concept also appears in other libraries and languages with analogous features.

Historically, default factories in Python originated with the collections module (defaultdict) and later expanded to dataclasses in Python 3.7. See also defaultdict, dataclasses.field.