functionali package
Submodules
functionali.higher_order_functions module
- class functionali.higher_order_functions.Reduced(x)
Bases:
object
- functionali.higher_order_functions.comp(*fns: Callable)
returns a composed function that takes a variable number of args, and applies it to
fns
passed from right to left.Added in version: 0.1.2
- functionali.higher_order_functions.curry(fn: Callable) Callable
Returns a curried version of the function.
>>> def fn(arg1, arg2, arg3): # test function return [arg1, arg2, arg3] >>> curried_fn = curry(fn) >>> curried_fn(1)(2)(3) [1, 2, 3]
Added in version: 0.1.0
- functionali.higher_order_functions.filter(fn: Callable, iterable: Iterable) Tuple
filter with eager evaluation. Prefer this over
lazyfilter
.>>> filter(is_even, range(10)) (0, 2, 4, 6, 8)
- functionali.higher_order_functions.flip(fn: Callable) Callable
returns a function that takes in a flipped order of args. Usage:
>>> f = lambda a,b : a-b >>> f(1,3) -2 >>> f(3,1) 2 >>> flipped_f = flip(f) >>> flipped_f(3,1) -2 >>> flipped_f(1,3) 2
Added in version: 0.1.0
- functionali.higher_order_functions.foldr(fn: Callable, iterable: Iterable, initial: Optional[Any] = None) Any
Fold right. Stack safe implementation
Added in version: 0.1.0
- class functionali.higher_order_functions.lazyfilter(fn: Callable, iterable: Iterable)
Bases:
object
Similar to python’s filter but returns a new generator everytime
__iter__
is called on it. Prefer using functionali.filter unless you know what you’re doing.>>> res = lazyfilter(is_even, range(10)) >>> take(2, res) (0, 2) >>> take(2, res) (0, 2) # same result
Compare this to python’s implementation
# you get a different result everytime >>> res = filter(is_even, range(10)) >>> take(2, res) (0, 2) >>> take(2, res) (4, 6) >>> take(2, res) (8,)
Also
__repr__
is implemented to make repl-driven development easy :)>>> lazyfilter(is_even, range(10)) (0, 2, 4, 6, 8)
- class functionali.higher_order_functions.lazymap(fn: Callable, *iterables: Iterable)
Bases:
object
Similar to python’s map but returns a new generator everytime
__iter__
is called on it. Prefer using functionali.map unless you know what you’re doing.>>> res = lazymap(lambda x:x+1, range(100)) >>> take(5, res) (1, 2, 3, 4, 5) >>> take(5, res) (1, 2, 3, 4, 5) # same result
Compare this to python’s implementation
>>> res = map(lambda x:x+1, range(100)) >>> take(5, res) (1, 2, 3, 4, 5) >>> take(5, res) (6, 7, 8, 9, 10)
Also
__repr__
is implemented to make repl-driven development easy :)>>> lazymap(lambda x:x+1,range(10)) (1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
- functionali.higher_order_functions.map(fn: Callable, *iterables: Iterable) Tuple
Map with eager evaluation. Prefer this over
lazymap
.>>> map(lambda x:x+1, range(10)) (1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
- functionali.higher_order_functions.reduce(fn, iterable, initial=None)
Similar to python’s reduce, but can be prematurely terminated with
reduced
. Works with dictionaries too.Usage:
>>> # Reducing over dictionaries. >>> def inc_value(result, kv_pair): k = kv_pair[0] v = kv_pair[1] return result[k]= v+1 >>> reduce(inc_value, {"a":1,"b":2}, {}) {'a': 2, 'b': 3}
>>> #premature termination with reduced >>> def inc_while_odd(result, element): if element%2==0: return reduced(result) else: result.append(element+1) return result >>> reduce(inc_while_odd, [1,3,5,6,7,8],[]) [2, 4, 6] # increments uptil 5 (third element) and prematurely terminates.
- functionali.higher_order_functions.reduced(x)
Use with
functionali.reduce
to prematurely terminatereduce
with the value ofx
.Usage:
>>> reduce(lambda acc, el: reduced("!"), [1,3,4]) "!" # reduce is prematurely terminated and returns a value of "!"
- functionali.higher_order_functions.threadf(arg: Any, forms: Iterable[Union[Callable, Iterable]]) Any
Thread first, passes
arg
as the first argument to the first function informs
and passes the result as the first argument to the second form and so on.see also
threadl
.>>> from functionali import identity >>> from operator import add, sub, mul >>> threadf(5, [identity]) >>> 5
>>> threadf(5, [identity, [add, 2]]) >>> 7
>>> threadf(5, [[sub, 2]]) >>> 3 # threadf(5, [[sub, 2]]) -> sub(5, 2) -> 5-2 -> 3
>>> # combining multiple functions >>> threadf(5, [identity, (add, 1), (sub, 1), (mul, 3)]) 15
- functionali.higher_order_functions.threadl(arg: Any, forms: Iterable[Union[Callable, Iterable]]) Any
Thread last, passes
arg
as the last argument to the first function informs
and passes the result as the last argument to the second form and so on.see also
threadf
.>>> from functionali import identity >>> from operator import add, sub, mul >>> threadl(5, [identity]) >>> 5
>>> threadl(5, [identity, [add, 2]]) >>> 7
>>> threadl(5, [[sub, 2]]) >>> -3 # threadl(5, [[sub, 2]]) -> sub(2, 5) -> 2-5 -> -3
>>> # combining multiple functions >>> threadl(5, [identity, (add, 1), (sub, 1), (mul, 3)]) -15
- functionali.higher_order_functions.trampoline(fn: Callable, *args: Any)
takes a function
fn
and calls if with*args
. iffn
returns a function, calls the function until a function is not returned i.e. the base case is reached. functionfn
must return a function in its recursive case. Useful for optimizing tail recursive functions or mutual recursions.>>> def fact(x, curr=1, acc=1): >>> if curr == x: >>> return curr*acc >>> else: >>> return lambda: fact(x, curr+1, acc*curr) >>> trampoline(fact, 3) == 6 >>> trampoline(fact, 100000000000) # does not raise RecursionError
functionali.predicates module
A file containing useful predicates.
- functionali.predicates.all_predicates(*predicates: Callable[[Any], bool]) Callable[[Any], bool]
Takes a set of predicates and returns a function that takes an entity and checks if it satisfies all the predicates.
>>> even_and_prime = all_predicates(is_even, is_prime) >>> even_and_prime(2) True >>> even_and_prime(4) False >>> even_and_prime(3) False
Added in version: 0.1.0
- functionali.predicates.complement(expr: Union[bool, Callable[[Any], bool]]) Union[bool, Callable[[Any], bool]]
Takes in a predicate or a Boolean expression and returns a negated version of the predicate or expression.
>>> complement(True) >>> False
>>> def fn(el): # returns the Boolean of el return bool(el) >>> negated_fn = complement(fn) >>> fn(1) >>> True >>> negated_fn(1) >>> False
Added in version: 0.1.0
- functionali.predicates.contains(entity: Any, collection: Iterable) bool
Checks whether collection contains the given entity. Note, won’t automatically convert a tuple of keys and values to a
dict
.Added in version: 0.1.0
- functionali.predicates.equals(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed to it is equal toa
; else returns True whena
,``b`` and*args
are equal.with one argument
>>> equals_one = equals(1) >>> equals_one(1) True >>> equals_one(2) False
with two or more arguments
>>> equals(1,1,1) True >>> equals(1,1,2) False
Added in version: 0.1.0
- functionali.predicates.greater_than(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed to is greater thana
; else returns True whena
is greater than``b`` and*args
.with one argument
>>> greater_than_one = greater_than(1) >>> greater_than_one(2) True >>> greater_than_one(0) False
with two or more arguments
>>> greater_than(2,1) >>> True >>> greater_than(3,2,1) True >>> greater_than(3,2,1,3) False
Useful to use with filter
>>> list(filter(greater_than(5),range(10))) [6,7,8,9]
Added in version: 0.1.0
- functionali.predicates.greater_than_eq(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
greater than or equal toa
; else returns True whena
is greater than or equal tob
and*args
.with one argument
>>> greater_than_eq_one = greater_than_eq(1) >>> greater_than_eq_one(2) True >>> greater_than_eq_one(1) True
with two or more arguments
>>> greater_than_eq(2,1) >>> True >>> greater_than_eq(3,2,1) True >>> greater_than_eq(3,2,1,3) True
Useful to use with filter
>>> list(filter(greater_than_eq(5),range(10))) [5,6,7,8,9]
Added in version: 0.1.0
- functionali.predicates.identity(x)
Returns its argument as it is.
- functionali.predicates.is_(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed is the same object asa
; else returns True whena
,``b`` and*args
are.with one argument
>>> d1 = {1,2,3} >>> d2 = {1,2,3} >>> is_d1 = is_(d1) >>> is_d1(d2) >>> False >>> d1 == d2 >>> True
with two or more arguments
>>> is_(d1,d1) >>> True >>> is_(d1,d1,d2) >>> False
Added in version: 0.1.0
- functionali.predicates.is_atom(entity: Any) bool
Everything that is NOT an iterable(except strings) are considered atoms.
>>> is_atom("plain string") True >>> is_atom(1) True >>> is_atom([1, 2]) False
Added in version: 0.1.0
- functionali.predicates.is_divisible(divident: Union[int, float], divisor: Union[int, float]) bool
Returns true if dividend is divisible by divisor.
Added in version: 0.1.0
- functionali.predicates.is_divisible_by(divisor: Union[int, float]) Callable[[Union[int, float]], bool]
Takes a
divisor
And returns a function (closure) That expects a dividend. returns true if it passes the divisibility test. for e.g.>>> f = is_divisible_by(5) >>> f(10) True >>> f(7) False
This is particularly useful to use with a filter.
>>> list(filter(is_divisible_by(5), [1,2,3,4,5,6,7,8,9,10])) [5, 10]
Suppose you want to filter out numbers that are divisible by 2 or 3
>>> list(filter(some_predicates([is_divisible_by(2), is_divisible_by(3)]), range(1, 10))) [2, 3, 4, 6, 8, 9, 10]
Added in version: 0.1.0
- functionali.predicates.is_empty(collection: Iterable) bool
Returns true if the collection is empty.
Added in version: 0.1.0
- functionali.predicates.is_even(num: int) bool
Returns true when num is even.
Added in version: 0.1.0
- functionali.predicates.is_nested(collection: Iterable) bool
returns true if a collection is nested. Added in version: 0.1.0
- functionali.predicates.is_numeric(entity: Any) bool
Return True if
entity
Is anint
,float
, or acomplex
.Added in version: 0.1.0
- functionali.predicates.is_odd(num: int) bool
Returns true when num is odd
Added in version: 0.1.0
- functionali.predicates.is_prime(num: int) bool
Returns true when num is prime
Added in version: 0.1.0
- functionali.predicates.less_than(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed to is less thana
; else returns True whena
is less than``b`` and*args
.with one argument
>>> less_than_one = less_than(1) >>> less_than_one(2) False >>> less_than_one(0) True
with two or more arguments
>>> less_than(1,2) >>> True >>> less_than(1,2,3) True >>> less_than(1,2,3,1) False
Useful to use with filter
>>> list(filter(less_than(5),range(10))) [0,1,2,3,4]
Added in version: 0.1.0
- functionali.predicates.less_than_eq(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
less than or equal toa
; else returns True whena
is less than or equal tob
and*args
.with one argument
>>> less_than_or_eq_to_one = less_than_eq(1) >>> less_than_or_eq_to_one(2) False >>> less_than_or_eq_to_one(1) True
with two or more arguments
>>> less_than_eq(1,2) >>> True >>> less_than_eq(1,2,3) True >>> less_than_eq(1,2,3,1) True
Useful to use with filter
>>> list(filter(less_than_eq(5),range(10))) [0,1,2,3,4,5]
Added in version: 0.1.0
- functionali.predicates.some_predicates(*predicates: Callable[[Any], bool]) Callable[[Any], bool]
Takes a set of predicates and returns a function that takes an entity and checks if it satisfies some of the predicates.
>>> even_or_prime = some_predicates(is_even, is_prime) >>> even_or_prime(2) True >>> even_and_prime(4) True >>> even_and_prime(3) True
Added in version: 0.1.0
functionali.seq_transform module
Functions that transform sequences
- functionali.seq_transform.argmap(functions: Iterable[Callable], args: Iterable) Generator
Maps the same argument(s) to multiple functions.
>>> inc = lambda x:x+1 >>> dec = lambda x:x-1 >>> list(argmap([inc, dec],[1])) [2,0]
you can even map multiple arguments
>>> add = lambda a,b: a+b >>> sub = lambda a,b: a-b >>> list(argmap([add, sub], [2, 1])) # two arguments [3, 1]
Added in version: 0.1.0
- functionali.seq_transform.argzip(sequence: Iterable[Callable], *args: Any) Generator
Similar to zip, but instead of zipping iterables, It zips an argument(s) with all the values of the iterable. for example.
>>> list(argzip([1,2,3,4], "number")) [(1, 'number'), (2, 'number'), (3, 'number'), (4, 'number')] >>> list(argzip([1,2,3,4], "number", "int")) [(1, 'number', 'int'), (2, 'number', 'int'), (3, 'number', 'int'), (4, 'number', 'int')]
Added in version: 0.1.0
- functionali.seq_transform.concat(iterable, *args)
Add items to the end of the iterable.
>>> concat([1,2,3,4],5) [1, 2, 3, 4, 5] >>> concat(deque([1,2]), 3,4) deque([1, 2, 3, 4])
Added in version: 0.1.0
- functionali.seq_transform.conj(iterable: Iterable, *args: Any) Iterable
Short for conjoin, adds element to the iterable, at the appropriate end. Adds to the left of a deque.
>>> conj([1,2,3,4],5) [1, 2, 3, 4, 5]
>>> conj(deque([1,2]), 3,4) deque([4, 3, 1, 2])
>>> conj([1,2,3,4],5,6,7,8) [1, 2, 3, 4, 5, 6, 7, 8]
>>> conj([1,2,3,4],[5,6,7]) [1, 2, 3, 4, [5, 6, 7]]
>>> conj((1,2,3,4),5,6,7) (1, 2, 3, 4, 5, 6, 7)
>>> conj(range(10), 11) (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 11)
>>> conj({1:"a", 2:"b"}, {3:"c"}) {1: 'a', 2: 'b', 3: 'c'}
Added in version: 0.1.0
- functionali.seq_transform.cons(arg: Any, iterable: Iterable) collections.deque
Returns a deque with arg as the first element.
Adds to the left of a deque.
>>> cons(5, [1,2,3,4]) deque([5, 1, 2, 3, 4])
>>> cons(3, deque([1,2])) deque([3, 1, 2])
>>> cons((3, "c"), {1:"a", 2: "b"}) deque([(3, "c"), (1, "a"), (2, "b")])
Added in version: 0.1.0
- functionali.seq_transform.flatten(sequence: Iterable) Tuple
Returns the contents of a nested sequence as a flat sequence. Flatten is recursive.
>>> flatten([1,2,[3,[4],5],6,7]) (1, 2, 3, 4, 5, 6, 7)
Added in version: 0.1.0
- functionali.seq_transform.insert(element: Any, iterable: Iterable, *, key: Callable = <function <lambda>>) Tuple
Inserts
element
right before the first element in the iterable that is greater thanelement
>>> insert(3, [1,2,4,2]) (1,2,3,4,2)
>>> insert((2, "b"), {1:"a", 3:"c"}) ((1, "a"), (2, "b"), (3, "c"))
Using the key Parameter
>>> Person = namedtuple("Person", ("name", "age")) >>> person1 = Person("John", 18) >>> person2 = Person("Abe", 50) >>> person3 = Person("Cassy", 25) >>> insert(person3, (person1, person2), key=lambda p:p.age) (person1, person3, person2) >>> insert(person3, (person1, person2), key=lambda p:p.name) (person3, person1, person2)
Added in version: 0.1.0
- functionali.seq_transform.interleave(*seqs: Iterable) Tuple
Similar to clojure’s interleave. returns a flat sequence with the contents of iterables interleaved.
>>> interleave([1,2,3],["a","b","c"]) (1, 'a', 2, 'b', 3, 'c') >>> interleave([1,2,3],["int","int","int"], ["a","b","c"],["str","str","str" ]) (1, 'int', 'a', 'str', 2, 'int', 'b', 'str', 3, 'int', 'c', 'str')
Added in version: 0.1.0
- functionali.seq_transform.remove(predicate: Callable, iterable: Iterable) Tuple
Opposite of filter; Constructs an iterable of elements that falsify the predicate.
>>> remove(lambda x: x==1, [1,1,9,1,1] [9] >>> remove(lambda x: x%2==0, range(10)) [1,3,5,7,9] # filter would return [2,4,6,8]
Added in version: 0.1.0
- functionali.seq_transform.tuplize(iterable: Iterable) Tuple
Recursively converts
iterable
to tuple.>>> tuplize([1,2,[3,4],5]) (1, 2, (3, 4), 5)
- functionali.seq_transform.unzip(sequence: Iterable) Tuple[Any]
Opposite of zip. Unzip is shallow.
>>> unzip([[1,'a'], [2,'b'], [3,'c']]) ((1, 2, 3), ('a', 'b', 'c')) >>> unzip([ [1,'a','A'], [2, 'b','B'], [3,'c','C'] ]) ((1, 2, 3), ('a', 'b', 'c'), ('A', 'B', 'C'))
shallow nature of unzip.
>>> unzip([ [[1,'num'],['a','str']], [[2,'num'],['b','str']] ]) (([1, 'num'], [2, 'num']), (['a', 'str'], ['b', 'str']))
Added in version: 0.1.0
functionali.seq_traverse module
Functions to traverse the sequence
- functionali.seq_traverse.butlast(iterable: Iterable[Any]) Optional[Tuple[Any]]
returns an iterable of all but the last element in the iterable
>>> butlast([1, 2, 3]) (1, 2)
Added in version: 0.1.0
- functionali.seq_traverse.count(iterable: Iterable) int
counts the number of elements in the iterable, works with map objects, filter objets, and iterators.
count
will consume iterators, usecount_
if you want access to the iterators. Added in version: 0.1.2>>> count(iter([1,2,3])) 3
- functionali.seq_traverse.count_(iterable: Iterable) Tuple[int, Iterable]
returns a tuple of the number of elements in the iterable and the iterable itself. This can be used if you wish to find the length of iterators and want to consume the iterators later on. Added in version: 0.1.2 >>> count(iter([1,2,3])) (3,[1,2,3])
- functionali.seq_traverse.drop(n: int, iterable: Iterable) Tuple
Returns All the Elements after the first n number of elements in iterable. Returns an empty tuple if iterable is empty
>>> drop(3, [1,2,3,4,5]) (4,5) >>> drop(2, {1: "a", 2: "b", 3: "c"}) ((3, "c"),)
Added in version: 0.1.0
- functionali.seq_traverse.drop_while(predicate: Callable, iterable: Iterable) Tuple
Drops elements from
iterable
whilepredicate
is true, And returns a tuple of the remaining elements initerable
.>>> drop_while(is_even, [2,4,6,7,8,9,10]) (7,8,9, 10)
>>> def is_even_dict(d): #checks if the key of dict d is even return d[0]%2==0 >>> drop_while(is_even_dict, {2:"a", 4:"b",5:"c"}) ((5, "c"),)
Added in version: 0.1.0
- functionali.seq_traverse.ffirst(iterable: Iterable[Any]) Optional[Any]
same as
first(first(iterable))
expects a nested iterable, returns None if iterable is empty>>> ffirst([[1,2], [3,4], [5,6]]) 1
Added in version: 0.1.0
- functionali.seq_traverse.fifth(iterable: Iterable[Any]) Optional[Any]
Returns the fifth item in iterable, or
None
if length is less than 5>>> fifth([1,2,3,4,5]) 5
Added in version: 0.1.0
- functionali.seq_traverse.first(iterable: Iterable[Any]) Optional[Any]
Returns the first item in an iterable or
None
if iterable is empty. If iterable is a dict, returns a tuple of the First key-value pair>>> first([1,2,3,4,5]) 1 >>> first({1:"a", 2:"b"}) (1, "a")
Added in version: 0.1.0
- functionali.seq_traverse.fourth(iterable: Iterable[Any]) Optional[Any]
Returns the fourth item in iterable, or
None
if length is less than 4>>> fourth([1,2,3,4,5]) 4
Added in version: 0.1.0
- functionali.seq_traverse.iter_(iterable: Iterable) Iterator
Returns appropriate iterator for the given iterable. This is mainly created because python’s
iter
returns an iterable of keys instead of keys and values fordict
.>>> tuple(iter_({1: "a", 2: "b", 3: "c"})) ((1, "a"),(2, "b"), (3, "c"))
Added in version: 0.1.0
- functionali.seq_traverse.last(iterable: Iterable[Any]) Optional[Any]
returns the last element in the iterable.
>>> last([1,2,3,4]) 4 >>> last({1: 'a', 2: 'b', 3: 'c'}) (3, "c")
Added in version: 0.1.0
- functionali.seq_traverse.rest(iterable: Iterable) Iterator
Returns an iterator of all but the first element in the iterable. If iterable is empty it returns an empty iterator.
>>> list(rest([1,2,3,4,5])) [2, 3, 4, 5]
>>> tuple(rest({1:"a", 2:"b", 3:"c"})) ((2,"b"), (3, "c"))
>>> tuple(rest([])) ()
Added in version: 0.1.0
- functionali.seq_traverse.reversed_(iterable: Iterable) Iterator
Returns appropriate reversed iterator for the given iterable. This is mainly created because python’s
reversed
returns an iterable of keys instead of keys and values fordict
.>>> tuple(reversed_({1: "a", 2: "b", 3: "c"})) ((3, 'c'), (2, 'b'), (1, 'a'))
Added in version: 0.1.0
- functionali.seq_traverse.second(iterable: Iterable[Any]) Optional[Any]
Returns the second item in iterable, or
None
if length is less than 2>>> second([1,2,3,4,5]) 2
Added in version: 0.1.0
- functionali.seq_traverse.split_with(predicate: Callable, iterable: Iterable) Tuple[Tuple, Tuple]
Equivalent to
(take_while(predicate, iterable), drop_while(predicate, iterable))
>>> split_with(is_even, [2, 4, 6, 7, 8, 9, 10]) ((2, 4, 6), (7, 8, 9, 10))
Added in version: 0.1.0
- functionali.seq_traverse.take(n: int, iterable: Iterable) Tuple
Returns the first n number of elements in iterable. Returns an empty tuple if iterable is empty
>>> take(3, [1,2,3,4,5]) (1, 2, 3) >>> take(2, {1: "a", 2: "b", 3: "c"}) ((1, "a"), (2, "b"))
Added in version: 0.1.0
- functionali.seq_traverse.take_while(predicate: Callable, iterable: Iterable) Tuple
Constructs a iterable list by taking elements from
iterable
whilepredicate
is true, Stop taking after the first element falsifies the predicate.>>> take_while(is_even, [2,4,6,7,8,9,10]) (2,4,6) # Notice that it does not include 8 and 10
>>> def is_even_dict(d): #checks if the key of dict d is even return d[0]%2==0 >>> take_while(is_even_dict, {2:"a", 4:"b",5:"c"}) ((2, "a"), (4, "b"))
Added in version: 0.1.0
- functionali.seq_traverse.third(iterable: Iterable[Any]) Optional[Any]
Returns the third item in iterable, or
None
if length is less than 3>>> third([1,2,3,4,5]) 3
Added in version: 0.1.0
functionali.statements module
Module contents
- functionali.all_predicates(*predicates: Callable[[Any], bool]) Callable[[Any], bool]
Takes a set of predicates and returns a function that takes an entity and checks if it satisfies all the predicates.
>>> even_and_prime = all_predicates(is_even, is_prime) >>> even_and_prime(2) True >>> even_and_prime(4) False >>> even_and_prime(3) False
Added in version: 0.1.0
- functionali.argmap(functions: Iterable[Callable], args: Iterable) Generator
Maps the same argument(s) to multiple functions.
>>> inc = lambda x:x+1 >>> dec = lambda x:x-1 >>> list(argmap([inc, dec],[1])) [2,0]
you can even map multiple arguments
>>> add = lambda a,b: a+b >>> sub = lambda a,b: a-b >>> list(argmap([add, sub], [2, 1])) # two arguments [3, 1]
Added in version: 0.1.0
- functionali.argzip(sequence: Iterable[Callable], *args: Any) Generator
Similar to zip, but instead of zipping iterables, It zips an argument(s) with all the values of the iterable. for example.
>>> list(argzip([1,2,3,4], "number")) [(1, 'number'), (2, 'number'), (3, 'number'), (4, 'number')] >>> list(argzip([1,2,3,4], "number", "int")) [(1, 'number', 'int'), (2, 'number', 'int'), (3, 'number', 'int'), (4, 'number', 'int')]
Added in version: 0.1.0
- functionali.butlast(iterable: Iterable[Any]) Optional[Tuple[Any]]
returns an iterable of all but the last element in the iterable
>>> butlast([1, 2, 3]) (1, 2)
Added in version: 0.1.0
- functionali.comp(*fns: Callable)
returns a composed function that takes a variable number of args, and applies it to
fns
passed from right to left.Added in version: 0.1.2
- functionali.complement(expr: Union[bool, Callable[[Any], bool]]) Union[bool, Callable[[Any], bool]]
Takes in a predicate or a Boolean expression and returns a negated version of the predicate or expression.
>>> complement(True) >>> False
>>> def fn(el): # returns the Boolean of el return bool(el) >>> negated_fn = complement(fn) >>> fn(1) >>> True >>> negated_fn(1) >>> False
Added in version: 0.1.0
- functionali.concat(iterable, *args)
Add items to the end of the iterable.
>>> concat([1,2,3,4],5) [1, 2, 3, 4, 5] >>> concat(deque([1,2]), 3,4) deque([1, 2, 3, 4])
Added in version: 0.1.0
- functionali.conj(iterable: Iterable, *args: Any) Iterable
Short for conjoin, adds element to the iterable, at the appropriate end. Adds to the left of a deque.
>>> conj([1,2,3,4],5) [1, 2, 3, 4, 5]
>>> conj(deque([1,2]), 3,4) deque([4, 3, 1, 2])
>>> conj([1,2,3,4],5,6,7,8) [1, 2, 3, 4, 5, 6, 7, 8]
>>> conj([1,2,3,4],[5,6,7]) [1, 2, 3, 4, [5, 6, 7]]
>>> conj((1,2,3,4),5,6,7) (1, 2, 3, 4, 5, 6, 7)
>>> conj(range(10), 11) (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 11)
>>> conj({1:"a", 2:"b"}, {3:"c"}) {1: 'a', 2: 'b', 3: 'c'}
Added in version: 0.1.0
- functionali.cons(arg: Any, iterable: Iterable) collections.deque
Returns a deque with arg as the first element.
Adds to the left of a deque.
>>> cons(5, [1,2,3,4]) deque([5, 1, 2, 3, 4])
>>> cons(3, deque([1,2])) deque([3, 1, 2])
>>> cons((3, "c"), {1:"a", 2: "b"}) deque([(3, "c"), (1, "a"), (2, "b")])
Added in version: 0.1.0
- functionali.contains(entity: Any, collection: Iterable) bool
Checks whether collection contains the given entity. Note, won’t automatically convert a tuple of keys and values to a
dict
.Added in version: 0.1.0
- functionali.count(iterable: Iterable) int
counts the number of elements in the iterable, works with map objects, filter objets, and iterators.
count
will consume iterators, usecount_
if you want access to the iterators. Added in version: 0.1.2>>> count(iter([1,2,3])) 3
- functionali.count_(iterable: Iterable) Tuple[int, Iterable]
returns a tuple of the number of elements in the iterable and the iterable itself. This can be used if you wish to find the length of iterators and want to consume the iterators later on. Added in version: 0.1.2 >>> count(iter([1,2,3])) (3,[1,2,3])
- functionali.curry(fn: Callable) Callable
Returns a curried version of the function.
>>> def fn(arg1, arg2, arg3): # test function return [arg1, arg2, arg3] >>> curried_fn = curry(fn) >>> curried_fn(1)(2)(3) [1, 2, 3]
Added in version: 0.1.0
- functionali.drop(n: int, iterable: Iterable) Tuple
Returns All the Elements after the first n number of elements in iterable. Returns an empty tuple if iterable is empty
>>> drop(3, [1,2,3,4,5]) (4,5) >>> drop(2, {1: "a", 2: "b", 3: "c"}) ((3, "c"),)
Added in version: 0.1.0
- functionali.drop_while(predicate: Callable, iterable: Iterable) Tuple
Drops elements from
iterable
whilepredicate
is true, And returns a tuple of the remaining elements initerable
.>>> drop_while(is_even, [2,4,6,7,8,9,10]) (7,8,9, 10)
>>> def is_even_dict(d): #checks if the key of dict d is even return d[0]%2==0 >>> drop_while(is_even_dict, {2:"a", 4:"b",5:"c"}) ((5, "c"),)
Added in version: 0.1.0
- functionali.equals(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed to it is equal toa
; else returns True whena
,``b`` and*args
are equal.with one argument
>>> equals_one = equals(1) >>> equals_one(1) True >>> equals_one(2) False
with two or more arguments
>>> equals(1,1,1) True >>> equals(1,1,2) False
Added in version: 0.1.0
- functionali.ffirst(iterable: Iterable[Any]) Optional[Any]
same as
first(first(iterable))
expects a nested iterable, returns None if iterable is empty>>> ffirst([[1,2], [3,4], [5,6]]) 1
Added in version: 0.1.0
- functionali.fifth(iterable: Iterable[Any]) Optional[Any]
Returns the fifth item in iterable, or
None
if length is less than 5>>> fifth([1,2,3,4,5]) 5
Added in version: 0.1.0
- functionali.filter(fn: Callable, iterable: Iterable) Tuple
filter with eager evaluation. Prefer this over
lazyfilter
.>>> filter(is_even, range(10)) (0, 2, 4, 6, 8)
- functionali.first(iterable: Iterable[Any]) Optional[Any]
Returns the first item in an iterable or
None
if iterable is empty. If iterable is a dict, returns a tuple of the First key-value pair>>> first([1,2,3,4,5]) 1 >>> first({1:"a", 2:"b"}) (1, "a")
Added in version: 0.1.0
- functionali.flatten(sequence: Iterable) Tuple
Returns the contents of a nested sequence as a flat sequence. Flatten is recursive.
>>> flatten([1,2,[3,[4],5],6,7]) (1, 2, 3, 4, 5, 6, 7)
Added in version: 0.1.0
- functionali.flip(fn: Callable) Callable
returns a function that takes in a flipped order of args. Usage:
>>> f = lambda a,b : a-b >>> f(1,3) -2 >>> f(3,1) 2 >>> flipped_f = flip(f) >>> flipped_f(3,1) -2 >>> flipped_f(1,3) 2
Added in version: 0.1.0
- functionali.foldr(fn: Callable, iterable: Iterable, initial: Optional[Any] = None) Any
Fold right. Stack safe implementation
Added in version: 0.1.0
- functionali.fourth(iterable: Iterable[Any]) Optional[Any]
Returns the fourth item in iterable, or
None
if length is less than 4>>> fourth([1,2,3,4,5]) 4
Added in version: 0.1.0
- functionali.greater_than(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed to is greater thana
; else returns True whena
is greater than``b`` and*args
.with one argument
>>> greater_than_one = greater_than(1) >>> greater_than_one(2) True >>> greater_than_one(0) False
with two or more arguments
>>> greater_than(2,1) >>> True >>> greater_than(3,2,1) True >>> greater_than(3,2,1,3) False
Useful to use with filter
>>> list(filter(greater_than(5),range(10))) [6,7,8,9]
Added in version: 0.1.0
- functionali.greater_than_eq(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
greater than or equal toa
; else returns True whena
is greater than or equal tob
and*args
.with one argument
>>> greater_than_eq_one = greater_than_eq(1) >>> greater_than_eq_one(2) True >>> greater_than_eq_one(1) True
with two or more arguments
>>> greater_than_eq(2,1) >>> True >>> greater_than_eq(3,2,1) True >>> greater_than_eq(3,2,1,3) True
Useful to use with filter
>>> list(filter(greater_than_eq(5),range(10))) [5,6,7,8,9]
Added in version: 0.1.0
- functionali.identity(x)
Returns its argument as it is.
- functionali.insert(element: Any, iterable: Iterable, *, key: Callable = <function <lambda>>) Tuple
Inserts
element
right before the first element in the iterable that is greater thanelement
>>> insert(3, [1,2,4,2]) (1,2,3,4,2)
>>> insert((2, "b"), {1:"a", 3:"c"}) ((1, "a"), (2, "b"), (3, "c"))
Using the key Parameter
>>> Person = namedtuple("Person", ("name", "age")) >>> person1 = Person("John", 18) >>> person2 = Person("Abe", 50) >>> person3 = Person("Cassy", 25) >>> insert(person3, (person1, person2), key=lambda p:p.age) (person1, person3, person2) >>> insert(person3, (person1, person2), key=lambda p:p.name) (person3, person1, person2)
Added in version: 0.1.0
- functionali.interleave(*seqs: Iterable) Tuple
Similar to clojure’s interleave. returns a flat sequence with the contents of iterables interleaved.
>>> interleave([1,2,3],["a","b","c"]) (1, 'a', 2, 'b', 3, 'c') >>> interleave([1,2,3],["int","int","int"], ["a","b","c"],["str","str","str" ]) (1, 'int', 'a', 'str', 2, 'int', 'b', 'str', 3, 'int', 'c', 'str')
Added in version: 0.1.0
- functionali.is_(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed is the same object asa
; else returns True whena
,``b`` and*args
are.with one argument
>>> d1 = {1,2,3} >>> d2 = {1,2,3} >>> is_d1 = is_(d1) >>> is_d1(d2) >>> False >>> d1 == d2 >>> True
with two or more arguments
>>> is_(d1,d1) >>> True >>> is_(d1,d1,d2) >>> False
Added in version: 0.1.0
- functionali.is_atom(entity: Any) bool
Everything that is NOT an iterable(except strings) are considered atoms.
>>> is_atom("plain string") True >>> is_atom(1) True >>> is_atom([1, 2]) False
Added in version: 0.1.0
- functionali.is_divisible_by(divisor: Union[int, float]) Callable[[Union[int, float]], bool]
Takes a
divisor
And returns a function (closure) That expects a dividend. returns true if it passes the divisibility test. for e.g.>>> f = is_divisible_by(5) >>> f(10) True >>> f(7) False
This is particularly useful to use with a filter.
>>> list(filter(is_divisible_by(5), [1,2,3,4,5,6,7,8,9,10])) [5, 10]
Suppose you want to filter out numbers that are divisible by 2 or 3
>>> list(filter(some_predicates([is_divisible_by(2), is_divisible_by(3)]), range(1, 10))) [2, 3, 4, 6, 8, 9, 10]
Added in version: 0.1.0
- functionali.is_empty(collection: Iterable) bool
Returns true if the collection is empty.
Added in version: 0.1.0
- functionali.is_even(num: int) bool
Returns true when num is even.
Added in version: 0.1.0
- functionali.is_nested(collection: Iterable) bool
returns true if a collection is nested. Added in version: 0.1.0
- functionali.is_numeric(entity: Any) bool
Return True if
entity
Is anint
,float
, or acomplex
.Added in version: 0.1.0
- functionali.is_odd(num: int) bool
Returns true when num is odd
Added in version: 0.1.0
- functionali.is_prime(num: int) bool
Returns true when num is prime
Added in version: 0.1.0
- functionali.iter_(iterable: Iterable) Iterator
Returns appropriate iterator for the given iterable. This is mainly created because python’s
iter
returns an iterable of keys instead of keys and values fordict
.>>> tuple(iter_({1: "a", 2: "b", 3: "c"})) ((1, "a"),(2, "b"), (3, "c"))
Added in version: 0.1.0
- functionali.last(iterable: Iterable[Any]) Optional[Any]
returns the last element in the iterable.
>>> last([1,2,3,4]) 4 >>> last({1: 'a', 2: 'b', 3: 'c'}) (3, "c")
Added in version: 0.1.0
- class functionali.lazyfilter(fn: Callable, iterable: Iterable)
Bases:
object
Similar to python’s filter but returns a new generator everytime
__iter__
is called on it. Prefer using functionali.filter unless you know what you’re doing.>>> res = lazyfilter(is_even, range(10)) >>> take(2, res) (0, 2) >>> take(2, res) (0, 2) # same result
Compare this to python’s implementation
# you get a different result everytime >>> res = filter(is_even, range(10)) >>> take(2, res) (0, 2) >>> take(2, res) (4, 6) >>> take(2, res) (8,)
Also
__repr__
is implemented to make repl-driven development easy :)>>> lazyfilter(is_even, range(10)) (0, 2, 4, 6, 8)
- class functionali.lazymap(fn: Callable, *iterables: Iterable)
Bases:
object
Similar to python’s map but returns a new generator everytime
__iter__
is called on it. Prefer using functionali.map unless you know what you’re doing.>>> res = lazymap(lambda x:x+1, range(100)) >>> take(5, res) (1, 2, 3, 4, 5) >>> take(5, res) (1, 2, 3, 4, 5) # same result
Compare this to python’s implementation
>>> res = map(lambda x:x+1, range(100)) >>> take(5, res) (1, 2, 3, 4, 5) >>> take(5, res) (6, 7, 8, 9, 10)
Also
__repr__
is implemented to make repl-driven development easy :)>>> lazymap(lambda x:x+1,range(10)) (1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
- functionali.less_than(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
passed to is less thana
; else returns True whena
is less than``b`` and*args
.with one argument
>>> less_than_one = less_than(1) >>> less_than_one(2) False >>> less_than_one(0) True
with two or more arguments
>>> less_than(1,2) >>> True >>> less_than(1,2,3) True >>> less_than(1,2,3,1) False
Useful to use with filter
>>> list(filter(less_than(5),range(10))) [0,1,2,3,4]
Added in version: 0.1.0
- functionali.less_than_eq(a, b=None, *args)
if only
a
is passed, a function is returned that returns True when thearg
less than or equal toa
; else returns True whena
is less than or equal tob
and*args
.with one argument
>>> less_than_or_eq_to_one = less_than_eq(1) >>> less_than_or_eq_to_one(2) False >>> less_than_or_eq_to_one(1) True
with two or more arguments
>>> less_than_eq(1,2) >>> True >>> less_than_eq(1,2,3) True >>> less_than_eq(1,2,3,1) True
Useful to use with filter
>>> list(filter(less_than_eq(5),range(10))) [0,1,2,3,4,5]
Added in version: 0.1.0
- functionali.map(fn: Callable, *iterables: Iterable) Tuple
Map with eager evaluation. Prefer this over
lazymap
.>>> map(lambda x:x+1, range(10)) (1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
- functionali.reduce(fn, iterable, initial=None)
Similar to python’s reduce, but can be prematurely terminated with
reduced
. Works with dictionaries too.Usage:
>>> # Reducing over dictionaries. >>> def inc_value(result, kv_pair): k = kv_pair[0] v = kv_pair[1] return result[k]= v+1 >>> reduce(inc_value, {"a":1,"b":2}, {}) {'a': 2, 'b': 3}
>>> #premature termination with reduced >>> def inc_while_odd(result, element): if element%2==0: return reduced(result) else: result.append(element+1) return result >>> reduce(inc_while_odd, [1,3,5,6,7,8],[]) [2, 4, 6] # increments uptil 5 (third element) and prematurely terminates.
- functionali.reduced(x)
Use with
functionali.reduce
to prematurely terminatereduce
with the value ofx
.Usage:
>>> reduce(lambda acc, el: reduced("!"), [1,3,4]) "!" # reduce is prematurely terminated and returns a value of "!"
- functionali.remove(predicate: Callable, iterable: Iterable) Tuple
Opposite of filter; Constructs an iterable of elements that falsify the predicate.
>>> remove(lambda x: x==1, [1,1,9,1,1] [9] >>> remove(lambda x: x%2==0, range(10)) [1,3,5,7,9] # filter would return [2,4,6,8]
Added in version: 0.1.0
- functionali.rest(iterable: Iterable) Iterator
Returns an iterator of all but the first element in the iterable. If iterable is empty it returns an empty iterator.
>>> list(rest([1,2,3,4,5])) [2, 3, 4, 5]
>>> tuple(rest({1:"a", 2:"b", 3:"c"})) ((2,"b"), (3, "c"))
>>> tuple(rest([])) ()
Added in version: 0.1.0
- functionali.reversed_(iterable: Iterable) Iterator
Returns appropriate reversed iterator for the given iterable. This is mainly created because python’s
reversed
returns an iterable of keys instead of keys and values fordict
.>>> tuple(reversed_({1: "a", 2: "b", 3: "c"})) ((3, 'c'), (2, 'b'), (1, 'a'))
Added in version: 0.1.0
- functionali.second(iterable: Iterable[Any]) Optional[Any]
Returns the second item in iterable, or
None
if length is less than 2>>> second([1,2,3,4,5]) 2
Added in version: 0.1.0
- functionali.some_predicates(*predicates: Callable[[Any], bool]) Callable[[Any], bool]
Takes a set of predicates and returns a function that takes an entity and checks if it satisfies some of the predicates.
>>> even_or_prime = some_predicates(is_even, is_prime) >>> even_or_prime(2) True >>> even_and_prime(4) True >>> even_and_prime(3) True
Added in version: 0.1.0
- functionali.split_with(predicate: Callable, iterable: Iterable) Tuple[Tuple, Tuple]
Equivalent to
(take_while(predicate, iterable), drop_while(predicate, iterable))
>>> split_with(is_even, [2, 4, 6, 7, 8, 9, 10]) ((2, 4, 6), (7, 8, 9, 10))
Added in version: 0.1.0
- functionali.take(n: int, iterable: Iterable) Tuple
Returns the first n number of elements in iterable. Returns an empty tuple if iterable is empty
>>> take(3, [1,2,3,4,5]) (1, 2, 3) >>> take(2, {1: "a", 2: "b", 3: "c"}) ((1, "a"), (2, "b"))
Added in version: 0.1.0
- functionali.take_while(predicate: Callable, iterable: Iterable) Tuple
Constructs a iterable list by taking elements from
iterable
whilepredicate
is true, Stop taking after the first element falsifies the predicate.>>> take_while(is_even, [2,4,6,7,8,9,10]) (2,4,6) # Notice that it does not include 8 and 10
>>> def is_even_dict(d): #checks if the key of dict d is even return d[0]%2==0 >>> take_while(is_even_dict, {2:"a", 4:"b",5:"c"}) ((2, "a"), (4, "b"))
Added in version: 0.1.0
- functionali.third(iterable: Iterable[Any]) Optional[Any]
Returns the third item in iterable, or
None
if length is less than 3>>> third([1,2,3,4,5]) 3
Added in version: 0.1.0
- functionali.threadf(arg: Any, forms: Iterable[Union[Callable, Iterable]]) Any
Thread first, passes
arg
as the first argument to the first function informs
and passes the result as the first argument to the second form and so on.see also
threadl
.>>> from functionali import identity >>> from operator import add, sub, mul >>> threadf(5, [identity]) >>> 5
>>> threadf(5, [identity, [add, 2]]) >>> 7
>>> threadf(5, [[sub, 2]]) >>> 3 # threadf(5, [[sub, 2]]) -> sub(5, 2) -> 5-2 -> 3
>>> # combining multiple functions >>> threadf(5, [identity, (add, 1), (sub, 1), (mul, 3)]) 15
- functionali.threadl(arg: Any, forms: Iterable[Union[Callable, Iterable]]) Any
Thread last, passes
arg
as the last argument to the first function informs
and passes the result as the last argument to the second form and so on.see also
threadf
.>>> from functionali import identity >>> from operator import add, sub, mul >>> threadl(5, [identity]) >>> 5
>>> threadl(5, [identity, [add, 2]]) >>> 7
>>> threadl(5, [[sub, 2]]) >>> -3 # threadl(5, [[sub, 2]]) -> sub(2, 5) -> 2-5 -> -3
>>> # combining multiple functions >>> threadl(5, [identity, (add, 1), (sub, 1), (mul, 3)]) -15
- functionali.trampoline(fn: Callable, *args: Any)
takes a function
fn
and calls if with*args
. iffn
returns a function, calls the function until a function is not returned i.e. the base case is reached. functionfn
must return a function in its recursive case. Useful for optimizing tail recursive functions or mutual recursions.>>> def fact(x, curr=1, acc=1): >>> if curr == x: >>> return curr*acc >>> else: >>> return lambda: fact(x, curr+1, acc*curr) >>> trampoline(fact, 3) == 6 >>> trampoline(fact, 100000000000) # does not raise RecursionError
- functionali.tuplize(iterable: Iterable) Tuple
Recursively converts
iterable
to tuple.>>> tuplize([1,2,[3,4],5]) (1, 2, (3, 4), 5)
- functionali.unzip(sequence: Iterable) Tuple[Any]
Opposite of zip. Unzip is shallow.
>>> unzip([[1,'a'], [2,'b'], [3,'c']]) ((1, 2, 3), ('a', 'b', 'c')) >>> unzip([ [1,'a','A'], [2, 'b','B'], [3,'c','C'] ]) ((1, 2, 3), ('a', 'b', 'c'), ('A', 'B', 'C'))
shallow nature of unzip.
>>> unzip([ [[1,'num'],['a','str']], [[2,'num'],['b','str']] ]) (([1, 'num'], [2, 'num']), (['a', 'str'], ['b', 'str']))
Added in version: 0.1.0