type-bridge 0.6.4

A modern, Pythonic ORM for TypeDB with an Attribute-based API

TypeBridge

A modern, Pythonic ORM for TypeDB with an Attribute-based API that aligns with TypeDB's type system.

Features

• True TypeDB Semantics: Attributes are independent types that entities and relations own
• Complete Type Support: All TypeDB value types - String, Integer, Double, Decimal, Boolean, Date, DateTime, DateTimeTZ, Duration
• Flag System: Clean API for @key, @unique, and @card annotations
• Flexible Cardinality: Express any cardinality constraint with Card(min, max)
• Pydantic Integration: Built on Pydantic v2 for automatic validation, serialization, and type safety
• Type-Safe: Full Python type hints and IDE autocomplete support
• Declarative Models: Define entities and relations using Python classes
• Automatic Schema Generation: Generate TypeQL schemas from your Python models
• Schema Conflict Detection: Automatic detection of breaking schema changes to prevent data loss
• Data Validation: Automatic type checking and coercion via Pydantic, including keyword validation
• JSON Support: Seamless JSON serialization/deserialization
• CRUD Operations: Full CRUD with fetching API (get, filter, all, update) for entities and relations
• Chainable Operations: Filter, delete, and bulk update with method chaining and lambda functions
• Query Builder: Pythonic interface for building TypeQL queries
• Multi-player Roles: A single role can accept multiple entity types via Role.multi(...)

Installation

# Clone the repository
git clone https://github.com/ds1sqe/type-bridge.git
cd type_bridge

# Install with uv
uv sync

# Or with pip
pip install -e .

# Or add to project with uv
uv add type-bridge

Quick Start

1. Define Attribute Types

TypeBridge supports all TypeDB value types:

from type_bridge import String, Integer, Double, Decimal, Boolean, Date, DateTime, DateTimeTZ, Duration

class Name(String):
    pass

class Age(Integer):
    pass

class Balance(Decimal):  # High-precision fixed-point numbers
    pass

class BirthDate(Date):  # Date-only values
    pass

class UpdatedAt(DateTimeTZ):  # Timezone-aware datetime
    pass

Configuring Attribute Type Names:

from type_bridge import AttributeFlags, TypeNameCase

# Option 1: Explicit name override
class Name(String):
    flags = AttributeFlags(name="person_name")
# TypeDB: attribute person_name, value string;

# Option 2: Case formatting
class UserEmail(String):
    flags = AttributeFlags(case=TypeNameCase.SNAKE_CASE)
# TypeDB: attribute user_email, value string;

2. Define Entities

from type_bridge import Entity, TypeFlags, Flag, Key, Card

class Person(Entity):
    flags = TypeFlags(name="person")  # Optional, defaults to lowercase class name

    # Use Flag() for key/unique markers and Card for cardinality
    name: Name = Flag(Key)                   # @key (implies @card(1..1))
    age: Age | None = None                   # @card(0..1) - optional field (explicit default)
    email: Email                             # @card(1..1) - default cardinality
    tags: list[Tag] = Flag(Card(min=2))      # @card(2..) - two or more (unordered set)

Note: list[Type] represents an unordered set in TypeDB. TypeDB has no list type - order is never preserved.

3. Create Instances

# Create entity instances with attribute values (keyword arguments required)
alice = Person(
    name=Name("Alice"),
    age=Age(30),
    email=Email("[e-mail]")
)

# Pydantic handles validation and type coercion automatically
print(alice.name.value)  # "Alice"

4. Work with Data

from type_bridge import Database, SchemaManager

# Connect to database
db = Database(address="localhost:1729", database="mydb")
db.connect()
db.create_database()

# Define schema
schema_manager = SchemaManager(db)
schema_manager.register(Person, Company, Employment)
schema_manager.sync_schema()

# Insert entities - use typed instances
alice = Person(
    name=Name("Alice"),
    age=Age(30),
    email=Email("[e-mail]")
)
Person.manager(db).insert(alice)

# Or use PUT for idempotent insert (safe to run multiple times!)
Person.manager(db).put(alice)  # Won't create duplicates

# Insert relations - use typed instances
employment = Employment(
    employee=alice,
    employer=techcorp,
    position=Position("Engineer"),
    salary=Salary(100000)
)
Employment.manager(db).insert(employment)

5. Cardinality Constraints

from type_bridge import Card, Flag

class Person(Entity):
    flags = TypeFlags(name="person")

    # Cardinality options:
    name: Name                              # @card(1..1) - exactly one (default)
    age: Age | None = None                  # @card(0..1) - zero or one (explicit default)
    tags: list[Tag] = Flag(Card(min=2))     # @card(2..) - two or more (unbounded)
    skills: list[Skill] = Flag(Card(max=5)) # @card(0..5) - zero to five
    jobs: list[Job] = Flag(Card(1, 3))      # @card(1..3) - one to three

6. Define Relations

from type_bridge import Relation, TypeFlags, Role

class Employment(Relation):
    flags = TypeFlags(name="employment")

    # Define roles with type-safe Role[T] syntax
    employee: Role[Person] = Role("employee", Person)
    employer: Role[Company] = Role("employer", Company)

    # Relations can own attributes
    position: Position                   # @card(1..1)
    salary: Salary | None = None         # @card(0..1) - explicit default

# Multi-player role example (one role, multiple entity types)
class Document(Entity):
    flags = TypeFlags(name="document")
    name: Name = Flag(Key)

class Email(Entity):
    flags = TypeFlags(name="email")
    name: Name = Flag(Key)

class Trace(Relation):
    flags = TypeFlags(name="trace")
    origin: Role[Document | Email] = Role.multi("origin", Document, Email)

7. Using Python Inheritance

class Animal(Entity):
    flags = TypeFlags(abstract=True)  # Abstract entity
    name: Name

class Dog(Animal):  # Automatically: dog sub animal in TypeDB
    breed: Breed

Documentation

• CLAUDE.md - Project guidance for development, TypeDB concepts, and quick reference
• docs/DEVELOPMENT.md - Development setup, commands, and code quality standards
• docs/TESTING.md - Testing strategy, patterns, and execution
• docs/TYPEDB.md - TypeDB concepts, driver API, and TypeQL syntax
• docs/ABSTRACT_TYPES.md - Abstract types and interface hierarchies in TypeDB
• docs/INTERNALS.md - Internal type system and architecture

API Reference

• docs/api/README.md - API overview and quick reference
• docs/api/attributes.md - Attribute types and value types
• docs/api/entities.md - Entity definition and ownership
• docs/api/relations.md - Relations, roles, and role players
• docs/api/abstract_types.md - Abstract types implementation and patterns
• docs/api/cardinality.md - Card API and Flag system
• docs/api/crud.md - CRUD operations and managers
• docs/api/queries.md - Query expressions and aggregations
• docs/api/schema.md - Schema management and conflict detection
• docs/api/validation.md - Pydantic integration and type safety

Pydantic Integration

TypeBridge is built on Pydantic v2, giving you powerful features:

class Person(Entity):
    flags = TypeFlags(name="person")
    name: Name = Flag(Key)
    age: Age

# Automatic validation and type coercion
alice = Person(name=Name("Alice"), age=Age(30))

# JSON serialization
json_data = alice.model_dump_json()

# JSON deserialization
bob = Person.model_validate_json('{"name": "Bob", "age": 25}')

# Model copying
alice_copy = alice.model_copy(update={"age": Age(31)})

Running Examples

TypeBridge includes comprehensive examples organized by complexity:

# Basic CRUD examples (start here!)
uv run python examples/basic/crud_01_define.py  # Schema definition
uv run python examples/basic/crud_02_insert.py  # Data insertion
uv run python examples/basic/crud_03_read.py    # Fetching API
uv run python examples/basic/crud_04_update.py  # Update operations

# Advanced examples
uv run python examples/advanced/schema_01_manager.py     # Schema operations
uv run python examples/advanced/schema_02_comparison.py  # Schema comparison
uv run python examples/advanced/schema_03_conflict.py    # Conflict detection
uv run python examples/advanced/pydantic_features.py     # Pydantic integration
uv run python examples/advanced/type_safety.py           # Literal types
uv run python examples/advanced/reserved_words_validation.py  # Keyword validation

Running Tests

TypeBridge uses a two-tier testing approach with 100% test pass rate:

# Unit tests (fast, no external dependencies) - DEFAULT
uv run pytest                              # Run 309 unit tests (0.3s)
uv run pytest tests/unit/attributes/ -v   # Test all 9 attribute types
uv run pytest tests/unit/core/ -v         # Test core functionality
uv run pytest tests/unit/flags/ -v        # Test flag system
uv run pytest tests/unit/expressions/ -v  # Test query expressions

# Integration tests (requires running TypeDB server)
# Option 1: Use Docker (recommended)
./test-integration.sh                     # Starts Docker, runs tests, stops Docker

# Option 2: Use existing TypeDB server
USE_DOCKER=false uv run pytest -m integration -v  # Run 249 integration tests (~60s)

# Run specific integration test categories
uv run pytest tests/integration/crud/entities/ -v      # Entity CRUD tests
uv run pytest tests/integration/crud/relations/ -v    # Relation CRUD tests
uv run pytest tests/integration/queries/ -v           # Query expression tests
uv run pytest tests/integration/schema/ -v            # Schema operation tests

# All tests
uv run pytest -m "" -v                    # Run all tests
./test.sh                                 # Run full test suite with detailed output
./check.sh                                # Run linting and type checking

Requirements

• Python 3.13+
• TypeDB 3.7.0-rc0 server (fully compatible)
• typedb-driver>=3.7.0rc3
• pydantic>=2.0.0
• isodate==0.7.2 (for Duration type support)

Release Notes

See the CHANGELOG.md for detailed release notes and version history.

License

MIT License