uselessvevo / fuse-box Goto Github PK

Just an experiment to port core modules to cython

Append short-methods to use validators without defining Field instance (-s)

For example:

# validators.py

def validate_range(
    value: list[Number, Number], 
    raise_exception: bool = False
) -> bool: 
    ...

At this moment, class Field has allowed_types attribute, but string, integer and etc. classes doesn't implement this attribute

For example:

• UrlValidator - subclass of RegexValidator
• UsernameValidator - subclass of RegexValidator
• PasswordValidator - subclass of RegexValidator
• FileValidator - with these params file: File, method: Union[str, callable]

Remove MinLengthValidator and MaxLengthValidator because we have a CompareValidator

This line doesn't make any sense, because this check requires a value type not value itself.

Because there is a reason to expand exceptions list on demand. And it means, that we need to add this attribute into __init__ method as well.

At this moment, all tests are a little bit messy. For example, test with fields contains validators and etc. Just do a little clean.

Just a thought, but we can add array of operators to handle multiple comparisons. For example:

validators=[CompareValidator(">", "<"), ...]

FloatField.set variable new_value is not used in get_separator method

At this moment, we dont have any built-in cleaner-methods. For example:

def clean_digits(value: str) -> Union[str, None]:
    value = re.search(r'^[\d+\,]+|^[\d+]', value)
    if value:
        value = value.group(0)
        value = value.replace(',', '.')
        return value

    integer_field = IntegerField(method=clean_digits)
    integer_field_result = integer_field.set('123@@@')
    print(f"input value: '123@@@', integer_field value: {integer_field.value}")

    float_field = FloatField(method=clean_digits)
    float_field_result = float_field.set('222,00')
    print(f"input value: '222,00', float_field value: {float_field.value}")

All handlers are named like Mapper, Regex and etc. Just add Handler at the end.

I mean that word new doesn't represent a purpose of this attribute, because this attribute meant to add new into __slots__, not overwrite it.

Converter should serialize and deserialize data from dataclass into basic data structures, dump and load from json and etc.

There is no clear reason to keep ORM part of this project. So, this package will be moved to the independent project.

Add optional parameter to support input data type checking.

Add optional parameter to support input data type checking.

Why? Because of certain conditions we want add or remove handlers or validators.
For example:

if <condition>:
    field.add_handler(Handler(...))
    field.add_validator(Validator(...))

Just a small enhancement in CompareValidator to prevent NoneType error.

• Create setup.py file
• Create repo for pypy