{"id":47239524,"title":"Show HN: Validatedata 0.3.0 – lightweight inline data validation for Python","points":2,"user":"EdwardK1","time":1772574496,"time_ago":"8 hours ago","type":"link","content":"

Validatedata is a Python library for validating dicts, lists, function arguments, and API payloads inline. No model classes, no boilerplate — just rules alongside the data.\npip install validatedata

Three ways to use it\n1. Standalone function — validate a dict\npythonfrom validatedata import validate_data

result = validate_data(\n data={'username': 'alice', 'email': 'alice@example.com', 'age': 25},\n rule={'keys': {\n 'username': 'str|min:3|max:32',\n 'email': 'email',\n 'age': 'int|min:18',\n }}\n)

result.ok # True\nresult.errors # []

# on failure:\n# [['username: invalid string length'], ['age: number out of range']]\n2. Decorator — validate function arguments\npythonfrom validatedata import validate

@validate(['str|min:3|max:32', 'email', 'str|min:8|msg:password too weak'])\ndef signup(username, email, password):\n return 'Account Created'

signup('alice', 'alice@example.com', 'Secure@123') # works\nsignup('alice', 'not-an-email', 'weak') # returns {'errors': [...]}\n3. Type annotation decorator\npythonfrom validatedata import validate_types

@validate_types\ndef create_user(username: str, age: int):\n return f'{username} ({age})'

create_user('alice', 30) # works\ncreate_user('alice', 'thirty') # raises ValidationError\nAll three support async functions transparently.

Transforms and mutation\nApply transforms before validation and get cleaned values back:\npythonresult = validate_data(\n data=[' Alice ', ' bob '],\n rule=['str|strip|lower', 'str|strip|lower'],\n mutate=True\n)

result.ok # True\nresult.data # ['alice', 'bob']

Conditional validation\nValidate a field only when a sibling field meets a condition:\npythonrule = {'keys': {\n 'role': {'type': 'str'},\n 'permissions': {\n 'type': 'str',\n 'depends_on': {'field': 'role', 'value': 'admin'},\n 'options': ('full', 'read', 'none')\n }\n}}

# permissions only validated when role == 'admin'

Custom error messages\nAny rule key accepts a {rule}-message override:\npython{\n 'type': 'int',\n 'range': (18, 'any'),\n 'range-message': 'you must be at least 18',\n 'type-message': 'age must be a number',\n}\nShorthand: 'int|min:18|msg:you must be at least 18'

Supported types\nBasic: str, int, float, bool, email, url, ip, uuid, slug, semver, phone, color, date, even, odd, prime\nExtended: dict, list, tuple, set, regex, object\nPhone validation supports E.164 built-in. National, international, and region-specific formats available via pip install phonenumbers.\nColor validation supports hex, rgb, hsl, and named CSS colors.

Supported rules\nrange, length, contains, excludes, options, expression, startswith, endswith, unique, nullable, strict, transform, depends_on, fields, items

Return value\nvalidate_data returns a SimpleNamespace:

result.ok — True if all validation passed\nresult.errors — list of error groups, one per field\nresult.data — transformed values, only present when mutate=True

MIT licensed. Python 3.7+. No required dependencies.\nRepo: https://github.com/Edward-K1/validatedata\nPyPI: https://pypi.org/project/validatedata/\nFeedback, bug reports, and use-case stories welcome.","url":"https://pypi.org/project/validatedata/","domain":"pypi.org","comments":[],"comments_count":0}