Little model objects using the dataset framework: https://dataset.readthedocs.org/en/latest/index.html

dataset-models.py

# Imports
import dataset
 
###
#   CONFIG
###
DATABASE_URL = 'sqlite:///'  # In memory database
 
 
###
#   GLOBALS
###
# This is considered a bad practice... :P
engine = dataset.connect(DATABASE_URL)
 
 
###
#   OBJECTS
###
class DDBB(object):
    "Main database object that creates engine connection."
    def __init__(self, *args, **kwargs):
        global engine
        self._ddbb = engine
 
 
class DatasetObject(DDBB):
 
    # PK is added only for scalation purposes
    # DO NOT MODIFY
    _primary_key = 'id'
    _table = ''
    _fields = []  # List of tuples
    #_indexes = []  # List of tuples, containing a list of fields and a string
    _indexes = []  # List of fields
 
    def __init__(self, *args, **kwargs):
        # Calling superclass __init__ (stablishes ddbb connection)
        super(DatasetObject, self).__init__(*args, **kwargs)
        self._connection = self._ddbb[self._table]
 
        # Check if table is set
        if not self._table:
            raise Exception('%s _table parameter is not defined!' % (
                    self.__class__.__name__
                )
            )
 
        # Create table indexes
        # This block is for indexes using custom naming, its really needed?
        # for index in self._indexes:
        #     fields = index[0]
        #     if not isinstance (fields, list):
        #         fields = [index[0]]
        #     name = None
        #     try:
        #         name = index[1]
        #     except:
        #         pass
        #     if name:
        #         result = self._connection.create_index(fields, name=name)
        #     else:
        #         result = self._connection.create_index(fields)
        if len(self._indexes):
            result = self._connection.create_index(self._indexes)
 
        # Set object fields according with the configured fields
        self._fields.append((self._primary_key, None))  # Adding primary key
        for field in self._fields:
            value = field[1]
            if field[0] in kwargs:
                value = kwargs[field[0]]
            setattr(self, field[0], value)
 
    def save(self):
        data = dict()
        for field in self._fields:
            data[field[0]] = getattr(self, field[0])
 
        try:
            # Update!
            data['id'] = self.id
            result = self._connection.update(data, ['id'])
            # result -> BOOL if update went ok
        except AttributeError:
            # Insert
            row_id = self._connection.insert(data)
            self.id = row_id
 
    def delete(self):
        try:
            object_id = self.id
            result = self._connection.delete(id=object_id)
            self.__init__()
        except AttributeError:
            result = False
 
        return result
 
    def find(self, *args, **kwargs):
        result = []
        items = self._connection.find(*args, **kwargs)
        for item in items:
            obj = self.__class__(**item)
            result.append(obj)
 
        return result
 
    def find_one(self, *args, **kwargs):
        result = None
        item = self._connection.find_one(**kwargs)
        if item:
            result = self.__class__(**item)
        return result

 
class Person(DatasetObject):
    _table = 'people'
    _indexes = ['name', 'id']
    _fields = [
        ('name', None),
        ('age', None),
    ]
 
 
# Creating objects:
p = Person(name="Felipe", age=26)
 
# Saving!
p.save() 
 
# Find objects aka. dataset[table].find()
p2 = Person().find(port=7000)
 
# Find one object aka dataset[table].find_one()
p3 = Person().find_one(id=1)
 
# Remove object
p.delete()