https://github.com/shikokuchuo/testbot

Bring package test coverage to 100%
https://github.com/shikokuchuo/testbot

r testing-tools

Last synced: 3 months ago
JSON representation

Bring package test coverage to 100%

• Host: GitHub
• URL: https://github.com/shikokuchuo/testbot
• Owner: shikokuchuo
• License: other
• Created: 2025-06-18T11:43:53.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2025-06-19T14:52:22.000Z (4 months ago)
• Last Synced: 2025-06-19T15:40:59.628Z (4 months ago)
• Topics: r, testing-tools
• Language: R
• Homepage: https://shikokuchuo.net/testbot/
• Size: 1.01 MB
• Stars: 3
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.Rmd
  • Changelog: NEWS.md
  • License: LICENSE

Awesome Lists containing this project

README

          
---

output: github_document

---

```{r}

#| include: false

knitr::opts_chunk$set(

  collapse = TRUE,

  comment = "#>",

  fig.path = "man/figures/README-",

  out.width = "100%"

)

```

# testbot

[![R-CMD-check](https://github.com/shikokuchuo/testbot/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/shikokuchuo/testbot/actions/workflows/R-CMD-check.yaml)

[![Codecov test coverage](https://codecov.io/gh/shikokuchuo/testbot/graph/badge.svg)](https://app.codecov.io/gh/shikokuchuo/testbot)

[![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html#experimental)

Testbot aims to bring package test coverage up to 100%.

This allows for packages with compiled code:

(i) To catch segfaults along error paths / in edge cases

(ii) To catch memory leaks, memory access errors and undefined behaviour by giving diagnostic tooling such as Valgrind and ASAN/UBSAN a wider surface to work on

> Important! This package is highly experimental and its API is subject to change at any time.

## Installation

Install the development version of testbot:

``` r

pak::pak("shikokuchuo/testbot")

```

Testbot is built on the btw and ellmer packages and leverages the current state of the art in LLM agentic tool-calling. Hence, a pre-requisite is an Anthropic or OpenAI API key in a location that can be picked up by ellmer.

## Instructions

Just run `testbot()` in your package working directory.

```{r}

#| label: instructions

#| eval: false

r <- testbot::testbot()

```

An ellmer chat object is returned.

The test file is generated as "tests/testthat/test-bot.R".

## Example

Example output of testbot running on itself:

```{r}

#| label: example

testbot::testbot()

```