A developer rewrites a feature-rich 1996 Expression Calculator in Rust, producing a CLI and an MCP server. The port reproduces decades-old math capabilities, ships to crates.io, Homebrew, and Windows MSI, and documents the effort and measure of effort.
The Rust port (excalc-rs) reproduces the original calculator’s extensive math capabilities and exposes them as a CLI and an MCP server.
The port was produced quickly: one Copilot CLI session over 24 hours, 59 commits, 5,060 lines of Rust, 217 tests, and a v0.2.0 release.
Approximate token usage and cost for the AI-assisted port are reported: 732 API calls, ~83 million input tokens (mostly cache reads), ~262,000 output tokens, estimated near $50.
# What happened In a single development effort, the author ported a 1996 Pascal/Delphi expression evaluator (Expression Calculator) to Rust as excalc-rs. The resulting project is both a CLI and an MCP server so other tools—such as coding agents—can call into it for reliable arithmetic instead of guessing or hallucinating results.
# Why this matters
# What the port reproduces
The post shows a numeric check: gamma(10) computed by numeric integration equals 9! to about seven decimal places (result: 0.000000054249539971351624 difference).
# How the port was done The author used Copilot CLI interactively across one session. Key milestones and CI/packaging work included:
Timeline highlights show the first v0.1.0 shipped seven minutes after the initial scaffold. By the morning the project reached v0.2.0 with extensive features and packaging.
# Project metrics
# The original and provenance
# Practical takeaway If you need a tested, standalone evaluator for complex math, statistics, number theory, financial calculations, or numeric integration, excalc-rs is a modern Rust implementation of a long-lived codebase. It is available as a crate, a CLI, and an MCP server suitable for programmatic use.
The Rust compiler remains unperturbed by the antics of the LLM
Every time you validate data with Pydantic v2, the data-validation library most Python apps reach for, a Rust extension does the work. Its core, pydantic-core, is built with PyO3, the same toolchain we'll use here. This post builds that same kind of bridge, small enough to read in one sitting: a JSON parser written in
Rust Coreutils 0.11 was just released as this alternative to the GNU Coreutils. Besides the long-running task of catching up the GNU Coreutils compatibility with the GNU test suite and never-ending work on optimizing the performance, Rust Coreutils 0.11 is quite a nice release for its improved error messages...
Привет, Хабр! Некоторое время назад у меня образовалось немного (вернее много) свободного времени, и я между моментами серфинга на известном сайте HH решил посмотреть в сторону освоения Rust. Кроме всего прочего мне очень интересует тема библиотек графических интерфейсов. В Rust есть несколько подобных библиотек, я реш
TL;DR — Mark an async fn with @rpc and call it from a .fitzv component as if it were local : let u = get_user(42).await? . The compiler generates both halves from that one declaration — a POST /__rpc/get_user endpoint on the server (which can touch the DB, auth, secrets) and a fetch -based stub on the client — with the
Originally appeared on avdi.codes . Making a Claude plugin named Bill Gates that makes Claude put a dollar in the swear jar every time it says “gate”.
Loading more related stories...
Open the app view to save this story, compare related coverage, and continue from the same source.