statictypes 0.0.2

Enforce type annotations

statictypes

statictypes helps your Python functions do exactly what you asked them to do, and not anything else. This package delivers three simple decorators:

• @statictypes.enforce: Raise an error if the incorrect argument or return type is given.
• @statictypes.warn: Give a warning if the incorrect argument or return type is given.
• @statictypes.convert: Try to convert into the given argument type, or return an error.

Example 1

Let's define a simple function, and enforce Python's type annotations.

import statictypes

@statictypes.enforce
def myfunc(text: str, number: int) -> str:
    return text + " " + str(number)

myfunc("my number is", 1)  # This works as intended
myfunc("my number is", 1.1)  # This raises an error

Calling myfunc("my number is", 1) is valid, but myfunc("my number is", 1.1) results in:

statictypes.StaticTypeError: Argument 'number' got incorrect type <class'float'>, expected <class'int'>.

Example 2

Let's instead choose to convert the arguments to the given type annotations.

import statictypes

@statictypes.convert
def myfunc(text: str, number: int) -> str:
    return text + " " + str(number)

myfunc("my number is", 1)  # This works as intended
myfunc("my number is", 1.1)  # This gives the same output as above

This time, both myfunc("my number is", 1) and myfunc("my number is", 1.1) is valid. Note, however, that both expressions give the output "my number is 1", since number is in this case always converted to an integer.

Limitations

1. @statictypes.convert only works for simple types from where a constructor method can be called, e.g. str, int. The decorator does not work on e.g. List[float] -> List[int] since the following incorrect conversion is attempted: List[int](list_of_floats).
2. Only the builtin generic types Any, Union, Optional, Dict, List and Tuple are currently supported.
3. When using @statictypes.convert, if a list is given to a function expecting a numpy.ndarray , an empty numpy.ndarray is returned with the shape of the list. This is because the numpy.ndarray has shape as its first positional argument, leading to a working but arguably unexpected result.

Requirements

• Python 3.6 or above.

Installation

pip install statictypes