Minos: A Lightweight Validation Framework
Minos is a library that you can use in your code to create a simple validation framework. Its aim is to provide a lightweight interface of functions that make it simple to quickly build and manipulate models.
Installation
To install Minos locally:
$ python setup.py develop
Minos will soon be on PyPi.
Quickstart
There are a few patterns Minos was designed to support; class validation with validators declared as part of the class definition, class validation where the validators are declared seperately, and as totally independant object validators. See below for a few examples of each pattern.
Class Validators in Class Definition
To declare validators inside of a class, the class needs to inherit from the Minos Mixin, and the validation arguments are passed to each validator on instantiation.
For example, given a class definition like so:
import minos
import pytz
from datetime import datetime
from minos.validators import DatetimeValidator, NumericalityValidator, PresenceValidator
class Toaster(minos.Mixin):
#Attributes
created_at = None
num_slots = None
#Validators
validators = [
DatetimeValidator('created_at'),
PresenceValidator('num_slots'), # every toaster needs to have its number of slots specified
#Toasters can only have an integer number of slots between 2-8
NumericalityValidator(
'num_slots',
integer_only=True,
greater_than_or_equal_to=2,
less_than_or_equal_to=8
)
]
def __init__(self, bread):
self.bread = bread
self.created_at = datetime.now(pytz.utc)
#...class definition continues...When using the ToasterClass in code, you can do stuff like this:
def toast_bread(bread):
toaster = Toaster(num_slots=4)
try:
toaster.validate() # Validates that toaster's attributes are valid
except minos.errors.FormValidationError:
raise ToastError("can't toast bread, toasters busted.")
toasted_bread = toaster.toast(bread)
return toasted_breadClass Validators Outside Class Definition
Similarly, You can also create a validator that validates a single attribute of a python object, and pass objects to be validated to it using the Validator's .validate_wrapper() call.
import minos
from minos.validators import NumericalityValidator
#Create a validator that for a toaster's slots ahead of time
slot_validator = NumericalityValidator(
'num_slots',
integer_only=True,
greater_than_or_equal_to=2,
less_than_or_equal_to=8
)
#Given a list of toasters, validate them all
for toaster in [dorm_toaster, kitchen_toaster, fancy_toaster]:
try:
#Check this toaster using the parameters specified earlier
slot_validator.validate_wrapper(toaster)
except minos.errors.FormValidationError:
print "{} doesn't have the right number of slots.".format(toaster.name)Single-Use Validation
You can also create validators without configuration, and specify the configuration inline using the .validate() call:
from minos.validators import NumericalityValidator
from minos.errors import ValidationError
#Make a validator that checks numeric qualities of objects
accountant = NumericalityValidator()
#Check a bunch of stuff
try:
accountant.validate(bank_account, greater_than=0)
except ValidationError:
print "You're broke!"
try:
accountant.validate(tax_payment, greater_than=0, less_than=5000.00)
except ValidationError:
print "You're tax payment doesn't seem right..."
try:
accountant.validate(num_deductions, integer_only=True)
except ValidationError:
print "Your number of deductions doesn't make sense."
