Tips on cassiobotaro

The recipes hidden in the itertools documentation

Mon, 13 Jul 2026 00:00:00 +0000

At the end of the itertools page there is a section called Itertools Recipes: a collection of useful functions built by combining the module’s building blocks. Flattening a list of lists, finding the first element that passes a test, walking a sequence in windows: a lot of what you would write by hand is already solved there. The recipes do not come ready to use in itertools, the idea is to copy them into your code. We have already used one of them here, consume, in the article about consuming iterables without allocating memory. I picked three others that are worth knowing.

Invoking modules with python -m

Mon, 06 Jul 2026 00:00:00 +0000

Python has the -m flag, which locates a module and runs it as a script. And the standard library is full of modules that turn into command line utilities ready to use, with nothing to install.

Serve the current folder over HTTP:

python -m http.server

Format (and validate) a JSON:

echo '{"name": "cassio", "language": "python"}' | python -m json.tool

See the calendar for the year in your terminal:

python -m calendar 2026

You have probably already used the flag without noticing, in commands like python -m venv .venv or python -m pip install. It is the same mechanism: -m locates the module on the import path and executes it as the main program, with __name__ set to "__main__" (when the target is a package, what runs is its __main__.py).

Fully comparable objects with a single method

Wed, 01 Jul 2026 00:00:00 +0000

The problem Link to heading

For an object to respond to <, <=, >, >=, == and !=, Python expects you to implement __lt__, __le__, __gt__, __ge__, __eq__ and __ne__. That is six nearly identical methods, tedious to write and easy to leave inconsistent (one says a < b and another disagrees).

The tip Link to heading

Two pieces cover everything:

Running a .zip as a Python application

Wed, 24 Jun 2026 00:00:00 +0000

The problem Link to heading

You have a Python application spread across several files and you want to ship everything as a single file. What if you could package it into a .zip and run it with python app.zip?

The tip Link to heading

You can, and it is simpler than it looks. Python knows how to execute a .zip directly, as long as there is a __main__.py inside it. Let’s start from scratch.

Consuming iterables without allocating memory

Wed, 17 Jun 2026 00:00:00 +0000

The problem Link to heading

When you just want to “run” a map/generator pipeline for its side effects (writing each item to a stream, for example), the reflex is a for or a list(...). The list allocates an entire list that gets thrown away right after, wasting memory for nothing. The for solves the memory issue, but the loop runs in pure Python.

The tip Link to heading

collections.deque accepts a maxlen parameter. With maxlen=0, it pulls each element and discards it immediately, never growing: that is the itertools consume recipe. Instead of materializing a list only to throw it away, you drain the pipeline in constant memory.

The write protocol

Fri, 15 May 2026 00:00:00 +0000

The problem Link to heading

When a function only needs to write to a stream, typing the argument as IO[str] is overkill, since you are demanding read, seek, close and the rest of the interface, even though you will never use them.

The tip Link to heading

Starting with Python 3.14, the io module exposes the Writer[T] and Reader[T] protocols. They describe only the minimum contract (write() or read()), so any object that implements just that method already satisfies the type.

The Money Object

Sun, 01 Mar 2026 00:00:00 +0000

The Story Link to heading

A while ago I took part in the hiring process of a fairly well known purple bank. I did not get in, but I got valuable feedback about the code I wrote. Putting that feedback together with things I learned afterwards, I decided to revisit the challenge and improve the implementation.

One of the most significant improvements was creating a Money object, based on the pattern of the same name described by Martin Fowler in Patterns of Enterprise Application Architecture (PEAA).

The count and index pitfall in NamedTuple

Sat, 17 Jan 2026 00:00:00 +0000

The Problem Link to heading

You are using NamedTuple to create your data classes and decide to add a field called count or index. Everything works perfectly at runtime, but when you run mypy or another type checker, you get strange errors. What is going on?

1
2
3
4
5
6
7
8


from typing import NamedTuple

class Item(NamedTuple):
 name: str
 count: int # Danger!

item = Item(name="apple", count=5)
print(item.count) # Works at runtime

Running mypy:

Tracking progress with IntFlag

Tue, 06 Jan 2026 00:00:00 +0000

The Problem Link to heading

Imagine you are integrating a new product into your system. This product must have complete details and specifications before being published. But that information is added in separate steps, and not necessarily in order. How do you check that every step has been completed?

A naive solution would be to use a boolean variable for each step:

class ProductTrack(models.Model):
 product = models.ForeignKey(Product, on_delete=models.CASCADE)
 has_specification = models.BooleanField(default=False)
 has_detail = models.BooleanField(default=False)
 
 def is_complete(self):
 return self.has_specification and self.has_detail

But what if you need to add more steps? Or remove one? Every change means touching the model, running database migrations and adjusting the logic.