AsyncOM

Async OM it's a super basic Object mapper based almost all on sqlalchemy ORM layer.

We use the declarative extension for building the classes, and also, factor instances of them on querys. (*Limited support)

There is still no support for relations.

Depens on encode/databases dependency.

Motivation

I don't like the asyncpgsa approach where they are just using, the core layer, to build run the sqlgenerator. I like to build, around the declarative layer of sqlachemy, and later found that I can patch some of the methods on the session.query, to use it's own sql generator, and turn async the query system.

Object persistent is minimal, and needs some love.

Usage

from sqlalchemy.ext.declarative import declarative_base
import sqlalchemy as sa
from databases import DatabaseURL

Base = declarative_base()


class OrmTest(Base):
    __tablename__ = 'orm_test'

    id = sa.Column(sa.Integer, primary_key=True)
    name = sa.Column(sa.String(100), index=True)
    value = sa.Column(sa.Text)

# Instead of usign Database from databases, you can use:
db = OMDatabase(DatabaseURL('postgres://root@postgres:{port}/guillotina'))

# instances of the object can be created with:
test = OrmTest(name='xx', value='yy')
await db.add(test)

print(test.id)  # pk column is correct set

# get an instance
ins = await db.query(OrmTest).get(1)

total = await db.query(OrmTest).count()
assert total == 1

# remove it
await db.delete(ins)

# filter them
res = await db.query(OrmTest).filter(
    OrmTest.name.like('xx')).all()

# Or just iterate over the results with a cursor:
async for row in db.query(OrmTest).filter(OrmTest.name.like('xx')):
    print(f'Row {row.name}: {row.value}')


# There is basic support for table inheritance query OneToOne


# Look at tests

Changelog

0.1.5

• Load inherited models.

0.1.4

• query(Object).get should return None if not found

0.1.3

• Fixes and improvements. Lay out basic API

0.1.0 (2019-04-07)

• First release on PyPI.