Predicate-based selection
Keep records satisfying a validation or eligibility predicate before processing them.
Python · Function
filter
filter() constructs an iterator containing elements from an iterable that pass a function's truth test. With None as the function, it removes false-valued elements.
How it works
filter() lazily iterates over an iterable and yields each element whose predicate produces a truthy result. Passing None uses each element's own truth value, so false elements are excluded; the result can be consumed like any other iterator.
Syntax
filter(function, iterable, /)Parameters
| Name | Type | Description |
|---|---|---|
function | callable or None | A callable tested against each element; use None to test elements by their own truth value. |
iterable | iterable | A sequence, iterable container, or iterator whose elements are examined. |
Returns
An iterator over elements of iterable for which function is true.
Key points
- filter() produces an iterator containing only elements accepted by the supplied function.
- The second argument may be a sequence, an iterable container, or an existing iterator.
- When the function is None, false-valued elements are omitted.
- A non-None function is called for each element to decide whether that element is retained.
- The None case tests each element directly instead of invoking a callable.
- Use itertools.filterfalse() when the desired output is the elements rejected by a predicate.
Common mistakes
- With None, filtering uses each element itself as the truth test rather than calling a predicate.
- The operation is equivalent to a generator expression, so it does not create a list automatically.
Example: Keep matching strings
active_names = list(filter(lambda name: name.startswith("A"), ["Ada", "Grace", "Alan", "Linus"]))
print(active_names)Converts the lazy filter iterator to a list containing names that start with A.
When to use it
Remove false values
Remove empty strings, zeroes, None values, or other false-valued items from an iterable.
Lazy pipeline
Process large or streaming inputs incrementally without immediately materializing a list.
Interview questions
QWhat does filter() do, and what kinds of iterables can it accept?
filter() takes a function and an iterable, then constructs an iterator containing elements for which the function is true. The iterable can be a sequence, container, or iterator.
Follow-up: What happens when the function is None? Each element is tested directly, and false elements are removed.
QDoes filter() return a list?
The result is an iterator, so it is evaluated as it is consumed. Convert it with list() or another consumer when a materialized collection is needed.
Follow-up: How could the same behavior be expressed without filter()? Use a generator expression with the corresponding condition.
QWhat is the effect of passing None as the function?
With function set to None, filter() applies the identity behavior: elements whose truth value is false are omitted. This is equivalent to iterating with an if item condition.
Follow-up: What is the complementary standard-library operation? itertools.filterfalse() selects elements for which the predicate is false.
Related
itertools.filterfalse()map()generator expressionslist comprehensions