flask-shell-turbo 1.0.1

Flask shell with steroids

Flask Shell Turbo

A drop-in utility that supercharges the Flask interactive shell, inspired by Django's shell_plus (from django-extensions). It auto-discovers your SQLAlchemy models, injects common utilities, and gives you a productive REPL (Read-Eval-Print Loop) for debugging, data exploration, and quick fixes — no repetitive imports required.

Description

Working with flask shell out of the box means manually importing every model and helper you need each time you open a session. Flask Shell Turbo solves this by:

• Auto-discovering all SQLAlchemy models registered on db.Model, so every model in your project is available in the shell immediately — no imports.
• Injecting common utilities like datetime, date, timedelta, timezone, and core SQLAlchemy query helpers (func, select, text, and_, or_, not_, desc, asc).
• Using IPython automatically when available, with %autoreload enabled so code changes are picked up without restarting the shell.
• Providing safe helpers — commit() and rollback() — that wrap db.session operations with clear success/error feedback.
• Enhancing the standard flask shell too, via a shell_context_processor, so you get the benefits even without the new command.
• Adding a new CLI command, flask shell-turbo, with flags for a plain Python fallback and SQL query logging.

Requirements

• Flask
• Flask-SQLAlchemy (or a project using the db.Model declarative pattern)
• SQLAlchemy 1.4+ (a legacy fallback is included for older versions)
• IPython (optional, but recommended)

Installation

1. Install flask-shell-turbo

   pip install flask-shell-turbo
   

Integration into a Flask Project

Call init_shell_turbo(app, db) right after your db is initialized — typically in your app factory, after db.init_app(app).

from flask import Flask
from your_project.extensions import db
from flask_shell_turbo import init_shell_turbo

def create_app():
    app = Flask(__name__)
    # ... your config ...
    db.init_app(app)
    # Register shell_turbo after db is initialized
    init_shell_turbo(app, db)

    return app

That's it — no further configuration is needed. Model discovery happens automatically each time the shell starts, so newly added models are picked up without any manual registration.

Usage

Launch the enhanced shell

FLASK_APP=app flask shell-turbo

This drops you into an IPython session (or the standard Python shell if IPython isn't installed) with everything pre-loaded:

Flask shell-turbo
Models: Account, Order, User
Extras : db, session, func, select, and_, or_, desc, asc, datetime, date, timedelta, text
Helpers: commit(), rollback()
──────────────────────────────

Work with your models directly — no imports

user = User.query.filter_by(username="ab").first()
user.name = "Alejandro"
commit()

If something goes wrong, commit() automatically rolls back and prints the error instead of leaving your session in a broken transaction state:

✘ rollback — (psycopg2.errors.UniqueViolation) duplicate key value violates unique constraint...

CLI flags

Flag	Description
--plain	Forces the standard Python shell (code.interact) even if IPython is installed.
--print-sql	Enables SQL echo (db.engine.echo = True) so every generated query is printed to the console.

Example:

flask shell-turbo --print-sql

Standard flask shell also benefits

Even without using shell-turbo, the regular command is enriched automatically:

flask shell

Because init_shell_turbo registers a shell_context_processor, all discovered models and utilities are available there too — shell-turbo simply adds the nicer IPython experience, autoreload, and the extra CLI flags on top.

Benefits

• Zero-import debugging — jump straight into querying and modifying data without hunting down import paths.
• Faster iteration — %autoreload picks up code changes to your models and helpers without restarting the shell.
• Safer transactions — commit()/rollback() prevent half-finished transactions from silently blocking further queries.
• Query visibility on demand — --print-sql lets you inspect the exact SQL Flask-SQLAlchemy generates, useful for debugging N+1 queries or unexpected joins.
• Works everywhere — enhances both the new shell-turbo command and the existing flask shell, so teammates who haven't adopted it yet still benefit.
• No dependencies required — IPython is optional; the tool gracefully falls back to a standard shell with tab-completion via readline/rlcompleter.
• Version-safe model discovery — supports both the modern SQLAlchemy registry.mappers API and the legacy _decl_class_registry, so it works across SQLAlchemy versions.