Skip to content
This repository


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Python data validation library

branch: master

Fix js

latest commit 7890a5f5c5
Mikhail Krivushin authored October 30, 2011
Octocat-spinner-32 procrustes Changed README and flatten func August 20, 2011
Octocat-spinner-32 sample Fix js October 30, 2011
Octocat-spinner-32 .gitignore January 31, 2011
Octocat-spinner-32 LICENSE Add license January 31, 2011
Octocat-spinner-32 Add October 17, 2011
Octocat-spinner-32 README.rst Changed README and flatten func August 20, 2011
Octocat-spinner-32 TODO Add TODO February 02, 2011
Octocat-spinner-32 version October 30, 2011
Octocat-spinner-32 Big rework August 19, 2011


Procrustes is a validation library, suitable for validating user input data, i.e. web forms or API calls. With procrustes you can validate any structure.

For doing this procrustes has three container classes - Tuple, List and Dict. Tuple come from FP, has fixed count of elements with fixed types. List can include any amount of values of one type. And Dict is a dict.

Procrustes consist of validators and forms. You can use validators without forms functionality. Forms adds mainly widgets and form data parsing.


Most simple what you can to do with procrustes is build validator and use it:

>>> from procrustes import validators as v
>>> two_strings_v = v.Tuple(v.String(), v.String())
>>> auth = two_strings_v(['login', 'password'])
('login', 'password')
>>> auth.errors
>>> auth = two_strings_v(['login'])
>>> auth.errors
['Must be iterable of length 2']

We have very powerful List:

>>> list_of_pairs_v = v.List(two_strings_v)
>>> pairs = list_of_pairs_v([(str(x), str(x)) for x in xrange(10)])
[('0', '0'), ('1', '1'), ('2', '2'), ('3', '3'), ('4', '4'), ('5', '5'), ('6', '6'), ('7', '7'), ('8', '8'), ('9', '9')]
>>> pairs.errors

List, Tuple and Dict are recursive, so you can build any data structure you want.

Dict example:

>>> dict_v = v.Dict({'pair': two_strings_v, 'pairs': list_of_pairs_v})
>>> data = dict_v({'pair': ['a', 'b'], 'pairs': [['a', 'b'], ['c', 'd']]})
{'pair': ('a', 'b'), 'pairs': [('a', 'b'), ('c', 'd')]}
>>> data.errors

Validators have required keyword argument:

>>> two_strings_v = v.Tuple(v.String(), v.String(), required=False)
>>> dict_v = v.Dict({'pair': two_strings_v, 'pairs': list_of_pairs_v})
>>> data = dict_v({'pairs': [['a', 'b'], ['c', 'd']]})
{'pair': None, 'pairs': [('a', 'b'), ('c', 'd')]}
>>> data.errors


Procrustes can flat data:

>>> list(data.flatten())
[('pairs__0__0', 'a'), ('pairs__0__1', 'b'), ('pairs__1__0', 'c'), ('pairs__1__1', 'd')]

And unflat it back:

>>> dict_v.deepen(dict([('pairs__0__0', 'a'), ('pairs__0__1', 'b'), ('pairs__1__0', 'c'), ('pairs__1__1', 'd')]))
{'pair': (None, None), 'pairs': [('c', 'd'), ('a', 'b')]}


Forms derived from validators and implemented as mixins to them. Every Field can work as form. Your form may consist from one forms.String(). In addition to validators forms adds methods widgets, template_widgets, unflat and is_valid.

Something went wrong with that request. Please try again.