pytyped-curry
Function currying that can be statically typed.
Functional tool for currying a function. Currying a function makes
the function callable multiple times before the function is actually
ran. Use curry(n)(func) or @curry(n) to transform a function into
a curried function which takes n calls before running.
Imports
from pytyped_curry import curry
from pytyped_curry import decorator # python >= 3.9
Example
# Transform the function into a curried function that takes
# two function calls before running.
@curry(2)
def add(x: int, y: int) -> int:
return x + y
# Add needs to be called twice to be ran.
add(2)(3) # 5
# Partial evaluation is easy.
increment = add(1)
increment(5) # 6
# The two arguments accept multiple forms.
add(x=2)(y=3)
add(y=3)(x=2)
add(2, 3)()
add()(2, 3)
For Decorators
Often times writing decorators requires writing several nested functions. This is often a hassle, and in many cases completely unnecessary due to currying.
Note: reveal_type is ran using mypy.
from typing import Callable, TypeVar
T = typing.TypeVar("T")
RT = typing.TypeVar("RT")
@curry(2, ...)
def decorator(func: Callable[[T], RT], x: T) -> RT:
print("Start")
y = func(x)
print("Finished")
return y
reveal_type(decorator)
"""
def (def (T`-1) -> RT`-2) -> def (T`-1) -> RT`-2
"""
@decorator
def increment(x: int) -> int:
return x + 1
reveal_type(increment)
"""
def (builtins.int) -> builtins.int
"""
@curry(3, ...)
def rate_limit(timeout: float, func: Callable[[T], RT], x: T) -> RT:
time.sleep(timeout)
return func(x)
reveal_type(rate_limit)
"""
def (builtins.float) -> (def (T`-1) -> RT`-2) -> def (T`-1) -> RT`-2
"""
@rate_limit(5)
def request_data(name: str) -> int:
return len(name)
reveal_type(request_data)
"""
def (builtins.str) -> builtins.int
"""
Documentation
New in Python 3.9
Doc-strings can be applied to arbitrary objects at runtime for runtime use
with the help(...) function. A few additional pieces of metadata are also
accessible at runtime to provide clearer documentation, such as the name of
the result.
@curry(3)
def add(x: int, y: int, z: int) -> int:
"""Returns add(x)(y)(z) = x + y + z."""
return x + y + z
help(add)
"""
Help on Curried in module __main__:
add = curry(3)(add(x: int, y: int, z: int) -> int)
Returns add(x)(y)(z) = x + y + z.
""""
help(add(1))
"""
Help on Curried in module __main__:
add(1) = curry(2)(add(x: int, y: int, z: int) -> int, 1)
Returns add(x)(y)(z) -> x + y + z.
"""
help(add(1)(2))
"""
Help on Curried in module __main__:
add(1, 2) = curry(1)(add(x: int, y: int, z: int) -> int, 1, 2)
Returns add(x)(y)(z) -> x + y + z.
"""
add(1)(2)(3) # 6
Type-Hinting
New in Python 3.8
Type-hints for curried functions are nigh impossible in the general case, as
can be seen by the last example. However, this doesn't stop us from enabling
typing in many common use-cases. Curried functions are hinted as functions
which take any arguments but take n calls, up to n = 3 for Python <
(3, 11) and up to n = 4 otherwise. Although the arguments are not
preserved, the final return type is.
Note: reveal_type is ran using mypy.
@curry(2)
def add(x: int, y: int) -> int:
return x + y
reveal_type(add)
"""
def (*Any, **Any) -> def (*Any, **Any) -> builtins.int
"""
For Python < (3, 11), one can also use curry(n, ...) to hint the curried
function as taking exactly 1 positional argument per call, up to n = 3.
@curry(2, ...)
def add(x: int, y: int) -> int:
return x + y
reveal_type(add)
"""
def (builtins.int) -> def (builtins.int) -> builtins.int
"""
For Python >= (3, 11), one can also use curry(n, ...) to hint the curried
function as taking exactly 1 positional argument per call, up to n = 3,
except for the last call. Notice that the y parameter is preserved as a
positional-or-keyword parameter.
@curry(2, ...)
def add(x: int, y: int) -> int:
return x + y
reveal_type(add)
"""
def (builtins.int) -> def (y: builtins.int) -> builtins.int
"""
For more precise hinting, one must use typing.cast around the currying
function.
from typing import Protocol, overload
class AddEmpty(Protocol):
def __call__(self) -> int:
...
class AddX(Protocol):
def __call__(self, x: int) -> int:
...
class AddY(Protocol):
def __call__(self, y: int) -> int:
...
class Add(Protocol):
@typing.overload
def __call__(self, x: int, y: int) -> AddEmpty:
...
@typing.overload
def __call__(self, x: int) -> AddY:
...
@typing.overload
def __call__(self, *, y: int) -> AddX:
...
def __call__(self, x, y):
...
@typing.cast(Add, curry(2))
def add(x: int, y: int) -> int:
return x + y
reveal_type(add)
"""
__main__.Add
"""
Metadata
Release files for pytyped-curry 1.0.3
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| pytyped-curry-1.0.3.tar.gz | 8.7 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| pytyped_curry-1.0.3-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 24.2 kB
Release files / pytyped-curry-1.0.3.tar.gz
| Download URL | pytyped-curry-1.0.3.tar.gz |
|---|---|
| Size | 8.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
bd7877443c6f6a92f4496903497371dc65700f88412aab5fef59bb36f14b19a0
|
|
BLAKE2b-256 checksum How to use checksums |
439088564b0cf5da9556f73008b931650091201449a083ff5b57577862d94d9f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.10.4
|
Release files / pytyped_curry-1.0.3-py3-none-any.whl
| Download URL | pytyped_curry-1.0.3-py3-none-any.whl |
|---|---|
| Size | 15.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
61a692b698c631a5ffd586dcb371cde0dc06ca4f69c07d07effbab2df8a3411e
|
|
BLAKE2b-256 checksum How to use checksums |
24547cad62c998921a955d9e3af5d8bd560d03c1c7c6b9c88c9ed0c60d77bc6a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.10.4
|
Security for pytyped-curry 1.0.3
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.