GitLocker: The Coding Marketplace

Description:

prettyrecord 0.2

prettyrecord allows you to define your data structures in declarative way,
similar to SQLAlchemy’s declarative_base or Django’s models. To achieve the goal,
prettyrecord uses Python’s descriptor and metaclass mechanisms.

Usage

to define new structure, subclass Record or derived class and fill with fields, eg.:
from prettyrecord import Record

class DummyRecord(Record, some_trait=False):
field1 = Field(required=True, default_value='')
field2 = Field(default_value=0)
field3 = ListField(constraint=lambda value: -5 < value < 5)

dr = DummyRecord(field1='foo', field3=[-2, 3, 1, 0, 4)
print(dr.field1) # shows 'foo'
print(dr.__some_trait__) # shows 'False'

to define new field, subclass Field, RecordField, ListField or derived class, eg.:
from prettyrecord import Field

class Integer(Field):
def normalize(self, value):
value = super(Integer, self).normalize(value)
return int(value)

to change creation process of your structures, subclass MetaRecord and define new base class for them, eg.:
from prettyrecord.record import Record, MetaRecord

class MetaMyRecord(MetaRecord):
def __new__(mcs, name, bases, attrs, **kwargs):
# your code here - remember to call super().__new__

def __init__(cls, name, bases, attrs, **kwargs):
# your code here - remember to call super().__init__

class MyRecord(Record, metaclass=MetaMyRecord):
pass