dirty-validators

Agnostic validators for python 3

Freely based on [WTF-Forms](https://github.com/wtforms/wtforms) validators.

Warning

Be aware about incompatible changes on Version 0.6.0.

Features

• Python 3 package.
• Easy to create a validator.
• Chained validations.
• Conditional validations.
• Specific error control messages.
• Dirty model integration (https://github.com/alfred82santa/dirty-models)
• No database dependent.
• Asynchronous validators.
• Stateless validators.
• Deferred validators.

Changelog

Version 0.6.0

The stateless refactor. Dirty validators now are full stateless. It means now there is no messages on validator instance. The is_valid method returns a validation context. It is evaluated as True if there were no errors and false otherwise. So, it is possible to still to use checks like if validator.is_valid(my_data): but we recomend to use the Walruf operator (https://peps.python.org/pep-0572/): if result := validator.is_valid(my_data):. That is important because messages are stored on context.

# Python +3.8
if result := validator.is_valid(my_data):
    # some code
else:
    print(result.error_messages)

#or

if not (result := validator.is_valid(my_data)):
    print(result.error_messages)

On the other hand, validation error messages now are an object. It stores error code, error message template and context values. There is a helper function from_context_to_legacy_message in order to convert a context to legacy message format.

• Stateless refactor
• New Context class stores state of validation process.
• New ValidationErrorMessage class to store validation error messages.
• New Deferred validator in order to build validator on runtime depending on context.
• New from_context_to_legacy_message helper.

Version 0.5.4

• Allow model validator inheritance.

Version 0.5.2

• Remove hard dependency from Dirty Models.
• Fix bug iterating ListModels.

Version 0.5.1

• Added value validators for mappings.

Version 0.5.0

• Added asynchronous validators.

Version 0.4.0

• Added <root> keyword in order to look up a field from root model of context.
• Added key_validator argument for spec validators in order to validate keys on hashmaps.

Installation

$ pip install dirty-validators

Basic usage

from dirty_validators.basic import EqualTo, Length, Regexp, Email
from dirty_validators.complex import Optional, ModelValidate

validator = Optional(validators=[EqualTo(comp_value="test")])

assert validator.is_valid("test") is True

# Chained validation
validator_chain = Chain(validators=[Length(min=14, max=16), Regexp(regex='^abc'), Email()])

assert validator_chain.is_valid('abcdefg@test.com')

# Model validation

class MyModelValidator(ModelValidate):
    fieldName1 = Optional(validators=[Length(min=4, max=6)])
    fieldName2 = Optional(validators=[Length(min=1, max=2)])
    fieldName3 = Required(validators=[Length(min=7, max=8)])

validator_model = MyModelValidator()

data = {
    "fieldName1": "1234",
    "fieldName1": "12",
    "fieldName3": "123456qw"
 }

assert validator_model.is_valid(FakeModel(data)) is True

Note

Look at tests for more examples