rlichess

rlichess is a modern, tidy R client for the Lichess API. It provides tools to fetch, clean, and analyze chess games, opening repertoires, user profiles, rating histories, and puzzles using standard tidyverse conventions.

Architecture

rlichess separates data access, cleaning, and analytics into three distinct layers:

1. Layer 1: API Clients (lic_*): Direct HTTP wrappers corresponding 1-to-1 with official Lichess API endpoints.
2. Layer 2: Tidy Wranglers (lic_tidy_*): Pure offline transformation functions that unnest and standardize chess data into tidy tibbles.
3. Layer 3: Chess Analytics (lic_stats_*): Offline statistics and analysis tools (e.g. opening win rates).

Installation

Install the development version from GitHub:

# install.packages("pak")
pak::pak("h8gi/rlichess")

# or using devtools / remotes
devtools::install_github("h8gi/rlichess")

Quick Example

Download recent games, tidy them into user-centric format, unnest moves into a long tibble, and analyze opening win rates in a single pipeline:

library(rlichess)
library(dplyr)

# 1. Fetch recent games from Lichess API
raw_games <- lic_games_user("h8gi", perf_type = "bullet", max = 50)

# 2. Tidy into user perspective (dates, win/loss, opponent info)
games <- lic_tidy_games(raw_games, username = "h8gi")

# 3. Analyze opening win rates
stats <- lic_stats_openings(games, min_games = 3)
head(stats)

# 4. Unpack 1-row-per-ply moves for move-by-move analysis
moves <- lic_tidy_moves(games)
head(moves)

Documentation & Tutorials

For detailed walkthroughs, authentication setup, and function references, visit the documentation site:

👉 https://h8gi.github.io/rlichess/
👉 Get Started Guide

Development

• devtools::load_all(): Load package into memory
• devtools::document(): Update roxygen documentation and NAMESPACE
• devtools::test(): Run test suite
• devtools::check(): Run comprehensive CRAN check

License

MIT © Hiroki Yagi