Skip to content
/ freddy Public template

Provides random fake data valid for a given schema

License

Notifications You must be signed in to change notification settings

lferran/freddy

Repository files navigation

freddy

Provides randomized json data (samples) that complies with a given schema.

Works both for json schema and pydantic models.

Usage

pydantic

import datetime
from pprint import pprint
from typing import List, Optional
from pydantic import BaseModel, Field
import freddy


class User(BaseModel):
    id: int
    name = 'John Doe'
    signup_ts: Optional[datetime.datetime] = None
    friends: List[int] = []
    pattern_field: str = Field(..., regex=r"^[-_a-zA-Z0-9]+$")


sample = freddy.sample(User)
pprint(sample)
{'id': 452, 'signup_ts': '1903-03-12T20:20:00', 'friends': [675, 408], 'pattern_field': 'EUvKs7BIK-Ne', 'name': 'xfphlync'}
User.validate(sample)
User(id=452, signup_ts=datetime.datetime(1903, 3, 12, 20, 20), friends=[675, 408], pattern_field='EUvKs7BIK-Ne', name='xfphlync')

jsonschema

from pprint import pprint
import jsonschema
import freddy

family_schema = {
    "type": "array",
    "items": {
        "properties": {
            "member": {"$ref": "#/definitions/person"},
            "role": {"$ref": "#/definitions/role"},
        },
        "type": "object",
    },
    "maxItems": 5,
    "minItems": 1,
    "definitions": {
        "person": {
            "properties": {
                "age": {"type": "integer"},
                "name": {"type": "string"},
                "pets": {
                    "items": {"$ref": "#/definitions/pet"},
                    "maxItems": 2,
                    "type": "array",
                },
            },
            "type": "object",
        },
        "pet": {
            "properties": {
                "kind": {"enum": ["dog", "cat"], "type": "string"},
                "name": {"type": "string"},
            },
            "type": "object",
        },
        "role": {
            "enum": [
                "father",
                "mather",
                "son",
                "daughter",
                "aunt",
                "grandma",
                "grandpa",
            ],
            "type": "string",
        },
    }
}

# Get 10 random samples
for i in range(10):
    sample_family = freddy.sample(family_schema)

    # Validate against schema
    jsonschema.validate(sample_family, family_schema)

pprint(sample_family)
[
    {"member": {"age": 77, "name": "k", "pets": []}, "role": "grandma"},
    {"member": {"age": 64, "name": "naifvxf", "pets": []}, "role": "grandpa"},
    {
        "member": {
            "age": 23,
            "name": "itruydotrj",
            "pets": [{"kind": "cat", "name": "o"}, {"kind": "cat", "name": "uonmvfgd"}],
        },
        "role": "son",
    },
]

Install

pip install freddy

Development

# Clone the repo
[email protected]:lferran/freddy.git
cd freddy

make develop

# Run tests
make tests

JSON Schema support

Conforms to JSON Schema Draft 7. The following features are supported:

  • boolean type

  • null type

  • string type

  • number type

  • integer type

  • array type

  • object type

  • definitions/references

  • Boolean type

  • consts

  • exclusiveMinimum and exclusiveMaximum in integers and numbers.

  • number multipleOf keyword

  • string pattern regex keyword

  • required keyword

  • additionalProperties

  • all string built-in formats

  • be able to provide custom basic type factories

  • multiple types: {"type": ["string", "array"]}

  • look into allOf: generate multiple objects + merge

Does not support:

  • ID referencing
  • allOf and not keywords
  • conditional keywords if, then and else
  • patternProperties on objects
  • property and schema dependencies on objects.