mongotic

The concept of MongoDB, SQLAlchemy, and Pydantic combined together in one simple and effective solution. It enables you to use SQLAlchemy v2 query syntax with MongoDB, and lets you define data models with Pydantic.

Documentation: https://allen2c.github.io/mongotic/

v0.6.0 — Breaking change ⚠️

mongotic v0.6.0 introduces a new field declaration style with full IDE and pyright support for query operators (.in_, .like, .between, .is_, .contains, etc.).

# v0.5 and earlier — still works at runtime, emits DeprecationWarning
name: str = Field(...)

# v0.6 and later — recommended; static-type-checks every operator
name: Mapped[str] = mapped_field()

• Legacy Field() declarations continue to work in v0.6 with a DeprecationWarning at class creation.
• The compatibility shim will be removed in v0.7.0.
• See the migration guide for the full substitution table.

Overview

mongotic is designed to make working with MongoDB feel familiar by reusing patterns from the SQLAlchemy and Pydantic ecosystems. It gives you a consistent and expressive way to interact with MongoDB collections, and uses Pydantic for validation and schema definition.

Features

• SQLAlchemy v2 API — select(), session.scalars(), ScalarResult; familiar patterns without a SQL database.
• Typed query expressions — Mapped[T] descriptor makes User.name == "x", User.age.between(18, 65), User.name.in_([...]), and friends fully IDE-aware.
• Rich query operators — logical combinators (or_, and_, not_), null checks, string matching, range, and distinct.
• Session management — refresh(), merge(), expunge(), expire(), state inspection (.new, .dirty, .deleted).
• Declarative indexes — define __indexes__ on the model and apply with create_indexes(); per-field index= / unique= / sparse= shorthand.
• Bulk operations — insert(), update(), and delete() statement builders via session.execute(), returning a Result with .rowcount and .inserted_ids.
• Column projection — select(User.name, User.email) returns lightweight Row results; single-column projection unwraps to plain values via session.scalars().
• Full async API — mongotic.asyncio mirrors the sync session on top of pymongo.AsyncMongoClient.
• Pydantic validation — schema definitions, JSON schema generation, and every Pydantic field constraint (min_length, ge/le, pattern, etc.) flow through mapped_field().
• Type checking — IDE autocomplete and pyright basic mode with zero warnings on idiomatic mongotic code.
• Works on standalone MongoDB — no replica set required and no multi-document transaction dependency.

Installation

pip install mongotic

Usage

from typing import Optional

from mongotic import (
    Mapped,
    MongoBaseModel,
    MultipleResultsFound,
    NotFound,
    create_engine,
    delete,
    mapped_field,
    select,
    update,
)
from mongotic.orm import sessionmaker


class User(MongoBaseModel):
    __databasename__ = "test_database"
    __tablename__ = "user"

    name:    Mapped[str]           = mapped_field(max_length=50)
    email:   Mapped[str]           = mapped_field()
    company: Mapped[Optional[str]] = mapped_field(default=None, max_length=50)
    age:     Mapped[Optional[int]] = mapped_field(default=None, ge=0, le=200)


engine = create_engine("mongodb://localhost:27017")
Session = sessionmaker(bind=engine)

# ── Add ──────────────────────────────────────────────────────────────────────
session = Session()
session.add(User(name="Allen Chou", email="allen@example.com", company="Acme", age=30))
session.add_all([
    User(name="Bob",   email="bob@example.com",   company="Acme", age=25),
    User(name="Carol", email="carol@example.com", company="Acme", age=28),
])
session.commit()

# ── Query ────────────────────────────────────────────────────────────────────
session = Session()

# Fetch all / first
users = session.scalars(select(User)).all()
users = session.scalars(select(User).where(User.age > 18)).all()
users = session.scalars(
    select(User)
    .where(User.company == "Acme")
    .order_by(-User.age)      # descending; use User.age for ascending
    .limit(10)
    .offset(0)
).all()

user = session.scalars(select(User).where(User.email == "allen@example.com")).first()
user = session.get(User, "<object_id_string>")   # PK lookup; returns None if not found

# Rich operators (all fully IDE-typed)
guests = session.scalars(select(User).where(User.company.in_(["Acme", "Acme Corp"]))).all()
matches = session.scalars(select(User).where(User.email.like("%@example.com"))).all()
ranged = session.scalars(select(User).where(User.age.between(18, 65))).all()

# Strict single-result fetch
try:
    user = session.scalars(select(User).where(User.email == "allen@example.com")).one()
    # raises NotFound if 0 results; raises MultipleResultsFound if 2+ results
except NotFound:
    ...
except MultipleResultsFound:
    ...

user = session.scalars(
    select(User).where(User.email == "allen@example.com")
).one_or_none()
# returns None if 0 results; raises MultipleResultsFound if 2+ results

# Count and existence check
count = session.scalars(select(User).where(User.company == "Acme")).count()
exists = session.scalars(select(User).where(User.company == "Acme")).exists()

# ── Update ───────────────────────────────────────────────────────────────────
session = Session()
user = session.scalars(select(User).where(User.email == "allen@example.com")).first()
user.email = "new.allen@example.com"   # tracked automatically
session.commit()

# ── Delete ───────────────────────────────────────────────────────────────────
session = Session()
user = session.scalars(select(User).where(User.email == "new.allen@example.com")).first()
session.delete(user)
session.commit()

# ── Bulk Operations ──────────────────────────────────────────────────────────
session = Session()
# Bulk update: returns Result with .rowcount
modified = session.execute(
    update(User).where(User.company == "Acme").values(company="Acme Corp")
)
# Bulk delete: returns Result with .rowcount
deleted = session.execute(
    delete(User).where(User.age < 18)
)

# ── Context manager + flush ──────────────────────────────────────────────────
with Session() as session:
    new_user = User(name="Dave", email="dave@example.com", age=35)
    session.add(new_user)
    session.flush()          # writes immediately; new_user._id is now available
    print(new_user._id)
    session.commit()         # alias for flush()

Async usage

mongotic.asyncio mirrors the sync API on pymongo.AsyncMongoClient. See the async documentation for the full reference.

import asyncio
from mongotic import insert, select, update, delete
from mongotic.asyncio import create_async_engine, async_sessionmaker

async_engine = create_async_engine("mongodb://localhost:27017")
AsyncSession = async_sessionmaker(bind=async_engine)

async def main():
    async with AsyncSession() as session:
        # Bulk insert
        r = await session.execute(
            insert(User).values([
                {"name": "Alice", "email": "alice@example.com", "age": 30},
            ])
        )
        print(r.inserted_ids)   # ["<ObjectId>"]

        # Query
        adults = await session.scalars(select(User).where(User.age >= 18)).all()

        # Column projection — returns Row objects
        names = await session.scalars(select(User.name)).all()

        # Scalar shortcut
        age = await session.scalar(select(User.age).where(User.name == "Alice"))

        # Bulk update / delete
        await session.execute(update(User).where(User.company == "Acme").values(company="Acme Corp"))
        await session.execute(delete(User).where(User.age < 18))

asyncio.run(main())

Contributing

Pull requests and issues are welcome. Please run make fmt && make test before opening a PR.

License

This project is licensed under the MIT License — see the LICENSE file for details.

Support

If you encounter any problems or have suggestions, please open an issue or feel free to reach out directly.