Quick reference

  • Quick guide to convert a Python script into a CLI tool

    Reading the Click’s documentation will bring you up to speed on what it can do, but I’m writing a quick list of steps I follow every time I need to make a quick CLI out of a Python script.

    So, let’s start by creating a Python project and installing Click as a dependency.

    $ poetry new cool-tool
$ cd cool-tool
$ poetry add click

    Poetry will create a scaffold directory that looks like this:

    cool-tool/
  cool_tool/
    __init__.py
  poetry.lock
  pyproject.toml

    For this example, we will create a main.py file with a very cool function.

    $ cd cool-tool 
$ touch cool_tool/main.py

    In our file we create a function and some logic.

    # main.py
def cool():
  print("hello cool world")

    OK, now to the part we really care. Let’s convert this script into a CLI

    • create a directory named “console” inside our main module.
    • create a __init__.py file to let Python know this is a module.
    • create a run.py file where we will add our Click logic.
    $ mkdir console
$ touch console/__init__.py
$ touch console/run.py

    Inside our run.py we add:

    import os
import click
from cool_tool.main import cool

@click.group
def cli():
   pass


@click.command
@click.argument("cool")
def show_cool():
   click.echo(cool())

    Then we add our CLI script to our pyproject.toml so we can use it directly inside our virtual environment, e.g., cool --help. Add this section and we should be done:

    # pyproject.toml
[tool.poetry.scripts]
cool = 'cool_tool.console.run:cli'

    Run poetry install, make sure you have the virtual environment activated, and verify everything works by running the help command.

    $ poetry install
$ poetry shell
$ cool --help

Usage: cool [OPTIONS] COMMAND [ARGS]...

Options:
  --help Show this message and exit.

Commands:
  show_cool

    Resources

  • Python function to send a single Telegram bot message

    This is my console.log equivalent when I need a quick observability implementation.

    Usually, when you build a Telegram bot, you have to have a small server listening for updates.

    The server logic is abstracted away as a simple API using the python-telegram-bot package, but we can write a simpler implementation for a single message:

    import os
import telegram

TELEGRAM_BOT_ID = os.getenv("TELEGRAM_BOT_ID")
TELEGRAM_BOT_TOKEN = os.getenv("TELEGRAM_BOT_TOKEN")

async def send_telegra_notification(msg: str, origin: str = "AppName:main"):
    notification_msg = f"*{origin}*: {msg}"
    bot = telegram.Bot(token=TELEGRAM_BOT_TOKEN)
    try:
        await bot.initialize()
        await bot.send_message(
            chat_id=TELEGRAM_BOT_ID,
            text=notification_msg,
            parse_mode=telegram.constants.ParseMode.MARKDOWN,
        )
    except Exception as e:
        # handle exception, e.g., logger.error(e)
        print(e)
    finally:
        await bot.shutdown()

  • Deploy PostgREST to Dokku

    Here is a list of steps to deploy a PostgREST instance to Dokku.

    Create a new app.

    dokku apps:create pgrest-myproject

    Add a port mapping from 3000 (PostgREST container) to 80 (host).

    dokku ports:add pgrest-myproject http:80:3000

    Create a new PostgreSQL database using the dokku-postgres plugin and link it to our project.

    dokku postgres:create pgrest-myproject
dokku posgres:link pgrest-myproject pgrest-myproject

    A new environment variable will be created:

    =====> pgrest-myproject
DATABASE_URL: postgres://postgres:alongpass@dokku-postgres-pgrest-myproject:5432/pgrest_myproject

    But we need to define the same value in a variable named PGRST_DB_URI so PostgREST can access the database.

    dokku config:set pgrest-myproject PGRST_DB_URI="postgres://postgres:alongpass@dokku-postgres-pgrest-myproject:5432/pgrest_myproject"

    Pull and deploy the Docker image from dockerhub.

    docker pull postgrest/postgrest:latest
docker tag postgrest/postgrest:latest dokku/postgrest:latest
dokku git:from-image pgrest-myproject dokku/postgrest:latest

    Add a Let’s Encrypt TLS certificate with the Dokku letsencrypt plugin.

    dokku letsencrypt:set pgrest-myproject email <your-email>
dokku letsencrypt:enable pgrest-myproject

  • Types of UUID

    • v1: mac address + time + random
    • v4: completely random
    • v5: input + seed (consistent, derived from input)
    • v7: time + random (distributed sortable ids)

    Source