Vinicius Rocha

The Zen of Python

Tue, 12 May 2026 00:00:00 +0000

The Easter Egg

Run import this in any Python interpreter and you get 19 aphorisms written by Tim Peters in 1999, baked into the language as an Easter egg:

Python 3.14.5 (tags/v3.14.5:5607950, May 10 2026, 10:43:50)
Type "help", "copyright", "credits" or "license" for more information.
>>> import this
The Zen of Python, by Tim Peters

Beautiful is better than ugly.
Explicit is better than implicit.
Simple is better than complex.
Complex is better than complicated.
Flat is better than nested.
Sparse is better than dense.
Readability counts.
Special cases aren't special enough to break the rules.
Although practicality beats purity.
Errors should never pass silently.
Unless explicitly silenced.
In the face of ambiguity, refuse the temptation to guess.
There should be one-- and preferably only one --obvious way to do it.
Although that way may not be obvious at first unless you're Dutch.
Now is better than never.
Although never is often better than *right* now.
If the implementation is hard to explain, it's a bad idea.
If the implementation is easy to explain, it may be a good idea.
Namespaces are one honking great idea -- let's do more of those!

Behind the Poem

The poem predates PEP 20; it circulated on the comp.lang.python mailing list years before Guido van Rossum formalized it. The source for the this module is itself a wink: the poem is stored as a ROT-13-encoded string and decoded at import time.

Spinel: Matz's Native Compiler for Ruby

Fri, 24 Apr 2026 00:00:00 +0000

Yukihiro “Matz” Matsumoto, the creator of Ruby, is working on Spinel — an ahead-of-time (AOT) compiler that transforms Ruby source code into standalone native executables. He presented it at RubyKaigi 2026 with a live demonstration, and the results are impressive: 11.6x faster than miniruby across benchmarks, with computation-heavy workloads reaching 24–87x speedups.

How it works

Spinel’s compilation pipeline has three stages:

Parsing — a C binary (spinel_parse) uses libprism to parse Ruby and serialize the AST.
Code generation — a self-hosted Ruby binary (spinel_codegen) performs whole-program type inference and generates optimized C.
Compilation — a standard C compiler produces a standalone binary with no runtime dependencies beyond libc and libm.

The self-hosting part is the most interesting detail. The compiler backend is a 21,000-line Ruby file that Spinel itself compiles into a native binary. The compiler compiles itself.

Debugging TurboTax with Claude Code

Sun, 05 Apr 2026 00:00:00 +0000

TL;DR

Claude Code can analyze HAR files — browser network traffic exports — to debug real-world problems quickly.
TurboTax was detecting my browser timezone and inferring I was in Canada, redirecting me to the Canadian login page.
One prompt was enough to identify the root cause and get actionable steps to fix it.

Claude Code as a daily tool

Claude Code can be a powerful companion for everyday tasks, especially ones that involve large amounts of data to analyze. I use it every day as part of my job, and I have been finding more ways to bring it into my daily life beyond work.

Notes on: Ruby Is the Best Language for Building AI Apps

Fri, 27 Mar 2026 00:00:00 +0000

Carmine Paolino makes a compelling case that Ruby is the best language for building AI applications in 2026 — not because of ML training (that’s Python’s domain), but because modern AI development is just HTTP calls, and everything around it is web engineering.

The core argument: LLM integration is an HTTP call. What matters is everything around it — streaming responses, persisting conversations, tracking costs, switching providers. That’s where Rails shines.

Buffered vs Unbuffered I/O on Unix

Thu, 30 Oct 2025 00:00:00 +0000

TL;DR

Buffered I/O batches data in user-space (or library) buffers and performs fewer, larger system calls. Great for throughput.
Unbuffered I/O sends data directly (or more directly) to the kernel/device, offering lower-latency and more predictable timing — at a throughput cost for many small ops.
Use buffered I/O for normal file processing and high-throughput tasks. Use unbuffered or synchronous I/O when you need immediate visibility, determinism, or strict durability.

What ‘buffered’ and ‘unbuffered’ mean

Buffered I/O: The runtime (for example, the C standard library’s FILE*) keeps an intermediate buffer and only calls read()/write() when the buffer is full, empty, or flushed. This reduces the number of system calls and generally improves throughput.
Unbuffered I/O: Your operations are passed to the kernel (or device) with minimal user-space aggregation. Examples: calling POSIX read()/write() repeatedly with small sizes.

Two layers of buffering on Unix

User-space buffering (stdio, language runtime) — what most people mean by “buffered I/O” at the application level.
Kernel page cache — even after write() returns, the kernel often holds the content in RAM (the page cache) until it writes to disk. To make data durable to stable storage you need fsync()/fdatasync() or O_SYNC/O_DSYNC semantics.

Understanding both layers matters: you may fflush() (user-space) and still need fsync() (kernel/disk) if you require persistence.

Unix time values (the epoch)

Thu, 16 Oct 2025 00:00:00 +0000

Unix time (often called “the epoch”) is a simple way to represent a point in time: the number of seconds that have elapsed since 1970-01-01 00:00:00 UTC, not counting leap seconds. It’s compact, language-agnostic, and easy to compare and sort.

Comic: xkcd #376 — used under the Creative Commons Attribution-NonCommercial 2.5 License. Credit: Randall Munroe / xkcd.

TL;DR

Unix time is seconds elapsed since 1970-01-01 00:00:00 UTC, typically stored as a 64-bit integer.
Use it for internal storage, comparisons, rate limiting, TTLs, and compact binary formats.
Don’t use it for display, calendar/timezone logic, or precise elapsed intervals — use human-readable timestamps (ISO 8601) and monotonic clocks instead.
Always use 64-bit integers; 32-bit overflows in 2038.

Quick examples

Shell: date +%s prints the current Unix time in seconds.
Python: import time; int(time.time()) — seconds, or time.time() for fractional seconds.
JavaScript: Math.floor(Date.now() / 1000) — Date.now() returns milliseconds.

When you should use Unix time

Machine-to-machine timestamps: compact, fast to compare, and well supported across platforms.
Databases and indexes where you only need chronological ordering or durations.
Protocols or binary formats that need a timezone-independent, compact numeric representation.
Short-lived logging or metrics where space and parsing speed matter.

Practical rule: use Unix time for internal storage and calculations (sorting, rate limiting, TTLs), especially when you also store the timezone or a canonical representation elsewhere.

tmux cheat sheet

Wed, 05 Jun 2024 00:00:00 +0000

This is a live document that I’m planning to maintain as I get more familiar with tmux.

Key bindings

All keybindings should be prefixed with a prefix key, C-b (CTRL+b) by default, followed by a command key.

Category	Key binding	Description
Window	`c`	Create new window
Window	`,`	Rename window
Window	`w`	Choose window from list
Window	`0` to `9`	Go from window 0 to 9
Window	`n`	Go to next window
Pane	`%`	Split into two panes, left and right
Pane	`"`	Split into two panes, top and bottom
Pane	`Left`	Go to left pane
Pane	`Right`	Go to right pane
Pane	`Down`	Go to pane at the bottom
Pane	`Up`	Go to pane at the top
Pane	`o`	Go to next pane
Pane	`x`	Close/kill current pane
Pane	`q`	Briefly display pane indexes.
System	`?`	List all key bindings
System	`d`	Detach
System	`:`	Enter tmux command prompt

Resources

man 1 tmux (web version)
GitHub.com/tmux/tmux
[video] Tmux will SKYROCKET your productivity - here’s how
[video] Tmux has forever changed the way I write code.

Installing Ruby using chruby and ruby-install

Thu, 18 Apr 2024 00:00:00 +0000

Every Ruby developer will most likely be using a Ruby version manager. The most popular ones are: RVM, rbenv, and chruby. You can use whatever makes you happy. I went with chruby, and I’m happy with it.

What is chruby

chruby is a command line tool that lets you switch between different versions of Ruby. It allows you to change between them manually, or it can change the version for you automatically, as you navigate between project folders.

Setup Rails using chruby and ruby-install

Wed, 15 Dec 2021 00:00:00 +0000

Note: Check the updated version of this post: Installing Ruby using chruby and ruby-install.

Here are the steps to install and set up Ruby and Ruby on Rails on macOS using chruby and ruby-install. Both are very light tools written by Hal Brodigan. A similar approach can also be used on Linux or Windows/WSL.

Setup

Let’s start by installing Homebrew. Follow the official installation instructions on their website.

With that in place, we can now get chruby and ruby-install.

The Z shell

Tue, 03 Aug 2021 00:00:00 +0000

What it is

The shell is what sits between the user and the core of the operating system, usually in a form of a command line interface (CLI). The most popular shell on Linux distributions is bash. In 2019, Apple started releasing the MacOS with Zsh (Z shell) by default. I believe that is one the reasons that made zsh so popular.

Installing it on Ubuntu

Ubuntu, one of the most popular Linux distributions, comes with bash pre-installed. To install Zsh on it, just run the following commands:

Add a JSON configuration file to a .NET Core console application

Tue, 02 Mar 2021 00:00:00 +0000

A .NET Core application can handle configuration using different formats (providers) like memory, command line and environment. Here, we will be using a file provider, more specifically, a JsonConfigurationProvider. Click on this link to learn more about the different types of configuration providers.

Project Setup

We will start by creating a console application and adding the NuGet package for the provider. In our case, it will be a JSON provider.

dotnet new console -o sample-configuration -n SampleConfiguration
cd SampleConfiguration
dotnet add package Microsoft.Extensions.Configuration.Json

Let’s also add an appsettings.json file that we can use for testing.

Handling WPF Window Dialog events, the elegant way

Fri, 26 Jul 2019 00:00:00 +0000

Here are two elegant ways that you can use to handle WPF Dialogs from the ViewModel. I will demonstrate both examples by closing a WPF Window Dialog from the ViewModel depending on some logic.

Using Events

The first solution is by using an event that notifies the window dialog to close. Consider the following example:

public class DialogViewModel
{
    public event EventHandler Saved;

    public DialogViewModel()
    {
        SaveCommand = new DelegateCommand(OnSave);
    }

    public ICommand SaveCommand { get; }

    public void Save()
    {
        //
        // SOME LOGIC TO SAVE GOES HERE
        //
        Saved?.Invoke(this, EventArgs.Empty);
    }
}

The dialog window can listen to that event and close itself accordingly.

ASP.NET Conditional Validation

Wed, 20 Sep 2017 00:00:00 +0000

Conditional validation can be very useful, and I’m glad that I found an elegant way of doing it. You can use the same View-Model on two different actions of your controller and have different validations for each of them.

Sample Scenario

To help illustrate the issue and the solution, let’s describe a simple scenario for user management. We will have a controller that adds and edits users. I will abstract the business layer so we can focus only on the user validation.

Unit Testing NHibernate Mappings

Wed, 21 Jun 2017 00:00:00 +0000

What is the best approach to map your domain classes to the database using NHibernate? After doing some tests, I believe that the best answer is: it doesn’t matter. As long as you have unit tests in place for it, you are covered.

The same mapping issue that you will face using HBM files, you will encounter using FHN or Mapping By Code. Because of that, I did some experiments on what would be a good approach to unit test NHibernate mappings.

About

Mon, 01 Jan 0001 00:00:00 +0000

Hey, I’m Vinicius Rocha — a software developer based in the United States, originally from Brazil.

I’ve been writing code for over two decades now. Started back in the early .NET days and have been going ever since, picking up new things along the way. I mostly work on backend stuff, but I’ve always been into the whole picture — tooling, infrastructure, the little things that make day-to-day dev life less painful.

Quotes

Mon, 01 Jan 0001 00:00:00 +0000

“Simplicity is prerequisite for reliability.”

— Edsger W. Dijkstra

“Make it work, make it right, make it fast.”

— Kent Beck

“Talk is cheap. Show me the code.”

— Linus Torvalds

“Premature optimization is the root of all evil.”

— Donald Knuth

“The best way to predict the future is to invent it.”

— Alan Kay

“Good code is its own best documentation.”

— Steve McConnell

Vinicius Rocha

The Zen of Python

The Easter Egg

Behind the Poem

Spinel: Matz's Native Compiler for Ruby

How it works

Debugging TurboTax with Claude Code

Claude Code as a daily tool

Notes on: Ruby Is the Best Language for Building AI Apps

Buffered vs Unbuffered I/O on Unix

What ‘buffered’ and ‘unbuffered’ mean

Two layers of buffering on Unix

Unix time values (the epoch)

Quick examples

When you should use Unix time

tmux cheat sheet

Key bindings

Resources

Installing Ruby using chruby and ruby-install

What is chruby

Setup Rails using chruby and ruby-install

Setup

The Z shell

What it is

Installing it on Ubuntu

Add a JSON configuration file to a .NET Core console application

Project Setup

Handling WPF Window Dialog events, the elegant way

Using Events

ASP.NET Conditional Validation

Sample Scenario

Unit Testing NHibernate Mappings

About

Archives

Quotes