treefmt; TIPT; ghdl

Tis another hodgepodge Drop, today. I’ve been meaning to talk about bookend sections for a few weeks and figured they’d be a decent to the sandwiched thinkpiece.

TL;DR

(This is an AI-generated summary of today’s Drop using Ollama + llama 3.2 and a custom prompt.)

Today’s Drop includes:

• A code formatting multiplexer that runs formatters in parallel, only processes changed files, and supports 74+ formatters through a unified TOML config (https://treefmt.com)
• A framework for evaluating technological progress by comparing innovations against fundamental capability changes like indoor plumbing, providing a reality check on AI hype (https://freddiedeboer.substack.com/p/the-shitting-in-the-yard-test)
• A Golang CLI tool for selectively downloading GitHub repository content, supporting authentication and preserving directory structure (https://codeberg.org/hrbrmstr/ghdl)

treefmt

Treefmt (GH) is an spiffy code formatting multiplexer that solves a common challenge in modern software development — managing multiple code formatters across different programming languages in a single repository.

It executes all configured formatters simultaneously, significantly reducing the total formatting time compared to traditional sequential execution. Instead of processing every file in the project, treefmt intelligently tracks file changes and only formats modified files, optimizing performance. So, a single incantation handles all formatting needs through a standardized configuration and output system, eliminating the need to remember multiple formatter commands.

The system uses a TOML configuration file located in the project root. Here’s an example configuration:

[formatter.gofmt]
command = "gofmt"
excludes = ["vendor/*"]
includes = ["*.go"]
options = ["-w"]

[formatter.prettier]
command = "prettier"
excludes = []
includes = ["*.cjs", "*.css", "*.html", "*.js", "*.json", "*.json5", "*.jsx", "*.md", "*.mdx", "*.mjs", "*.scss", "*.ts", "*.tsx", "*.vue", "*.yaml", "*.yml"]
options = ["--write"]

Basic usage involves two main commands:

# Initialize configuration
treefmt --init

# Run formatting
treefmt

There is support for over 74 different code formatters, including:

• clang-format for C/C++/Java/JavaScript
• gofmt for Golang
• Prettier for JavaScript/HTML/CSS

The project is actively evolving with planned enhancements focusing on expanded IDE integration beyond the current VSCode support, and implementation of pre-commit hooks.

TIPT

The “Indoor Plumbing Test” (NOTE: the article uses a more crass way of describing the “test”) offers a unusually grounded perspective on technological progress that cuts through the noise of AI hype (and tech hype in general). Freddie deBoer presents a deceptively simple yet profound framework for evaluating technological advancement: does it enable humans to do something genuinely new, or does it merely provide a different way to do what we could already accomplish?

In our current moment of breathless AI coverage and grandiose predictions, deBoer’s analysis provides a welcome reality check. While language models and image generators dominate headlines, their actual capabilities often amount to streamlining existing tasks rather than enabling truly novel human achievements.

Consider the transformative impact of indoor plumbing, electricity, or vaccines — technologies that have fundamentally altered human capabilities and quality of life. These innovations didn’t just make existing processes more efficient; they solved problems that were previously unsolvable and enabled entirely new possibilities for human flourishing.

The full piece on deBoer’s Substack goes deeper into this framework, offering compelling examples and clever analysis that may/will change how you think about technological progress. It’s a must-read for anyone interested in separating genuine innovation from marketing hype — and understanding why that distinction matters.

Rather than spoiling the specific examples and insights, I encourage y’all to collect your beverage of choice and head on over to Freddie’s post. This clear-eyed assessment of current AI capabilities against historical technological progress provides solid context for our ongoing conversations about artificial intelligence and technological advancement.

ghdl

I came across this repo, and that made me realize that I never even considered the utility value of being able to selectively retrieve individual files from a remote GH repository right from the CLI. Sure, you could just clone it and surgically cp or mv target files somewhere. But, if you just need one file, that seems needlessly wasteful. And, GitHub’s API makes it easy to target a specific file.

It’s written in Python, so that makes it kind of a non-starter for me. And, it doesn’t figure out the main branch name when using the “download whole repo” method; nor does it support using a GITHUB_TOKEN to bypass rate-limiting.

So, ofc I made a Golang port of it.

The core features include:

• Download entire repositories, specific folders, or single files
• Support for private repositories using GitHub token authentication
• Custom output directory naming
• Automatic default branch detection
• Preserves directory structure when downloading

The “download whole repo” option could use some parallel file download functionality, but I see myself using this primarily for grabbing one or two files, so that modification exercise is left to the reader.

NOTE: I made sure to also show how to use treefmt (see the first section) in the repo’s justfile.

FIN

Remember, you can follow and interact with the full text of The Daily Drop’s free posts on:

• 🐘 Mastodon via @dailydrop.hrbrmstr.dev@dailydrop.hrbrmstr.dev
• 🦋 Bluesky via https://bsky.app/profile/dailydrop.hrbrmstr.dev.web.brid.gy

Also, refer to:

• this post, and
• this post

to see how to access a regularly updated database of all the Drops with extracted links, and full-text search capability. ☮️