CherrypyREST

A minimal framework inspired by Django

Cherrypyrest is a minimal framework that provides all basic ORM functionalities and more such as Models, Serializers, REST API Controllers and some basic common utilities.

How to use

from cherrypyrest import models
from cherrypyrest import fields as base_fields

class User(models.Model):

    NAME = 'name'
    EMAIL = 'email'
    AGE = 'age'
    
    fields = [NAME, EMAIL, AGE]

    name = base_fields.String(null=True)
    email = base_fields.Email(required=True)
    age = base_fields.Number()

user = User()
user.set_value({"name": "abc", "email": "abc@xyz.com"})
print(user.db_repr())  # db representation of object
print(user.serialize()) # json serializable form


class Address(models.Model):

    CITY = 'city'
    STATE = 'state'
    COUNTRY = 'country'
    POSTAL_CODE = 'postal_code'  # internal representation of a field
    USER = 'user'
    alias = {
          POSTAL_CODE: 'pin_code',  # external or api reppresentation of a field
    }
    fields = [CITY, STATE, COUNTRY, POSTAL_CODE, USER]
    
    city = base_fields.String()
    state = base_fields.String()
    country = base_fields.String()
    postal_code = base_fields.Number(required=True)
    user = base_fields.RelatedField(child=User(), required=True)


address = Address()

address.set_value({"state": "ABC", "postal_code": 12345, "user": {"email": "abc@xyz.com"}})

print(address.serialize())
#{'city': '',
# 'state': 'ABC',
# 'country': '',
# 'pin_code': 12345,   # use of alias 
# 'user': {'name': None, 'email': 'abc@xyz.com', 'age': None}}

You can add a lot of attributes in the model class such as public_fields, read_only_fields etc to make model response more flexible.

The model works best with MongoDB as the ObjectID field is already provided in the fields module. Add a properly setup manager class obejct in the model.

manager = UserManager()

NOTE: Manager object should have a valid database client object to connect to. You can conect with any database but mongodb works well without any changes in the code.

Add an attribute in the models class to identify the database fields db_fields which gets a list of fields that will be fetched from the database when required. So in our Address class

db_fields = [USER]

and then

user = User()
user.set_value({"email": "abc@xyz.com"})
user.save()
user_id = user.pk

address = Address()
address.set_value({"postal_code": 234325, "user": user_id})  # no call to the database
print(address.user.pk)
# $ ObjectId("23487h2i374x2748bbksjedhskf")
print(address.user.email)  
# a call will be initiated to the database to fetch the user object by id and will set the whole object to the user attribute of addres object

So a call to db will only happend when we really need the related data.

I created this small package during my work on a project of one of my previous company. I haven't really paid a lot of attention to the design patterns and structure but it worked for me at that time. You can update the code according to your needs.