rhosocial-activerecord-postgres ($\rho_{\mathbf{AR}\text{-postgres}}$)

PostgreSQL Backend for rhosocial-activerecord

Native Array & JSONB Support · Advanced Types · Sync & Async

Note: This is a backend implementation for rhosocial-activerecord. It cannot be used standalone.

Why This Backend?

1. PostgreSQL's Unique Type System

2. No Adapter Overhead for Arrays

Unlike databases without native arrays, PostgreSQL stores and queries arrays directly:

# PostgreSQL: Native array - no serialization needed
class Article(ActiveRecord):
    tags: list[str]  # Maps directly to TEXT[]

# MySQL/SQLite: Requires adapter
class Article(ActiveRecord):
    tags: Annotated[list[str], UseAdapter(ListToStringAdapter(), str)]

3. Powerful Array Operators

Query arrays with native PostgreSQL operators:

# Contains
Article.query().where("tags @> ARRAY[?]", ('python',)).all()

# Overlaps (any element matches)
Article.query().where("tags && ARRAY[?, ?]", ('python', 'database')).all()

# Any element equals
Article.query().where("? = ANY(tags)", ('python',)).all()

💡 AI Prompt: "Show me all PostgreSQL array operators with examples"

Quick Start

Installation

pip install rhosocial-activerecord-postgres

Basic Usage

from rhosocial.activerecord.model import ActiveRecord
from rhosocial.activerecord.backend.impl.postgres import PostgresBackend
from rhosocial.activerecord.backend.impl.postgres.config import PostgresConnectionConfig
from typing import Optional
from uuid import UUID

class User(ActiveRecord):
    __table_name__ = "users"
    id: Optional[UUID] = None  # Native UUID type
    name: str
    tags: list[str]  # Native array type
    metadata: dict  # JSONB type

# Configure
config = PostgresConnectionConfig(
    host="localhost",
    port=5432,
    database="myapp",
    username="user",
    password="password"
)
User.configure(config, PostgresBackend)

# Use
user = User(name="Alice", tags=["python", "postgres"])
user.save()

# Query with array operators
python_users = User.query().where("tags @> ARRAY[?]", ('python',)).all()

💡 AI Prompt: "How do I configure connection pooling for PostgreSQL?"

PostgreSQL-Specific Features

Array Types

Native array support without adapters:

class Article(ActiveRecord):
    __table_name__ = "articles"
    title: str
    tags: list[str]      # TEXT[]
    scores: list[int]    # INTEGER[]

# Query with array operators
Article.query().where("tags @> ARRAY[?]", ('python',)).all()
Article.query().where("? = ANY(tags)", ('database',)).all()
Article.query().where("array_length(tags, 1) > ?", (3,)).all()

See Array Type Comparison for differences with other databases.

JSONB Operations

Binary JSON with indexing support:

class Product(ActiveRecord):
    __table_name__ = "products"
    name: str
    attributes: dict  # JSONB

# Query JSONB
Product.query().where("attributes->>'brand' = ?", ("Dell",)).all()

# JSONB contains (@>)
Product.query().where("attributes @> ?", ('{"brand": "Dell"}',)).all()

UUID Type

Native UUID storage and querying:

from uuid import UUID, uuid4

class User(ActiveRecord):
    __table_name__ = "users"
    id: UUID
    name: str

user = User(id=uuid4(), name="Alice")
user.save()

Range Types

Built-in support for range types:

from datetime import date

class Booking(ActiveRecord):
    __table_name__ = "bookings"
    room_id: int
    date_range: tuple  # DATERANGE

# Query with range operators
Booking.query().where("date_range @> ?", (date(2024, 1, 15),)).all()

RETURNING Clause

Retrieve data after INSERT/UPDATE/DELETE:

# INSERT with RETURNING
user = User(name="Alice")
user.save()
print(user.id)  # Populated automatically via RETURNING

# Works for all operations (unlike MySQL)

Requirements

• Python: 3.8+ (including 3.13t/3.14t free-threaded builds)
• Core: rhosocial-activerecord>=1.0.0
• Driver: psycopg>=3.2.12

PostgreSQL Version Compatibility

Supported SQL Protocols:

• ✅ SetOperationSupport (UNION, INTERSECT, EXCEPT) — All versions
• ✅ TruncateSupport (TRUNCATE TABLE) — All versions; RESTART IDENTITY ≥ 8.4

Recommended: PostgreSQL 12+ for optimal feature support.

See PROTOCOL_SUPPORT.md for complete protocol support matrix.

Get Started with AI Code Agents

This project supports AI-assisted development:

git clone https://github.com/rhosocial/python-activerecord-postgres.git
cd python-activerecord-postgres

Example AI Prompts

• "How do I use array operators to query tags?"
• "What's the difference between JSON and JSONB in PostgreSQL?"
• "Show me how to create a GIN index on an array column"
• "How do I use range types for scheduling?"

For Any LLM

Feed the documentation files in docs/ for context-aware assistance.

Testing

⚠️ CRITICAL: Tests MUST run serially. Do NOT use pytest -n auto or parallel execution.

# Run all tests
PYTHONPATH=src pytest tests/

# Run specific feature tests
PYTHONPATH=src pytest tests/rhosocial/activerecord_postgres_test/feature/basic/
PYTHONPATH=src pytest tests/rhosocial/activerecord_postgres_test/feature/query/

See the Testing Documentation for details.

Documentation

• Getting Started — Installation and configuration
• PostgreSQL Features — PostgreSQL-specific capabilities
• Type Adapters — Data type handling
• Array Comparison — Array support across databases
• Transaction Support — Transaction management

Comparison with Other Backends

💡 AI Prompt: "When should I choose PostgreSQL over MySQL for my project?"

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

License

Apache License 2.0 — Copyright © 2026 vistart