📦 Package rsonar

Overview

rsonar is the R equivalent of SonarQube: it centralizes code quality analysis into a single interactive report, estimates technical debt and integrates natively into GitLab CI and GitHub Actions pipelines.

For a quick local check (without CI/forge), run quality_score(".") to display a quality percentage directly in your IDE console.

See this repository that illustrates some features of the rsonar package.

It orchestrates four proven tools:

Tool	Role
lintr	Static analysis — bugs, style, complexity
styler	Code formatting (tidyverse style guide)
covr	Test coverage
goodpractice	R packaging best practices

SonarQube → rsonar Analogies

SonarQube	rsonar
Issues (bugs, code smells)	lintr issues
Style violations	styler violations
Coverage	covr (line + branch)
Maintainability rating	debt_index()
Quality Gate	quality_gate()
HTML report	sonar_report()
Generic Issue Import	SonarQube JSON export
New Code analysis	sonar_diff()
Project history	sonar_trend()
SARIF integration	export_sarif()
Hotspots / files to fix first	sonar_hotspots()

Installation

# From GitHub
remotes::install_github("ddotta/rsonar")

Quick Start

library(rsonar)

# Full analysis of an R package
res <- sonar_analyse("path/to/my/package")

# Console summary
print(res)

# Quick local quality percentage in IDE
quality_score(res)

# Interactive HTML report
sonar_report(res, output = "quality.html")

# Technical debt index (A to E)
debt_index(res)

# Hotspots: the files to fix first, ranked by debt
sonar_hotspots(res)

# Quality Gate (pass/fail like SonarQube)
quality_gate(res, coverage_min = 80, lint_errors_max = 0)

# Compare two analyses (regression detection)
diff <- sonar_diff(current_res, baseline_res)

# Track quality trends over time
sonar_trend(res)

# Export SonarQube Generic Issue Import format
export_sonar_json(res, "sonar-issues.json")

# Export SARIF for GitHub Code Scanning
export_sarif(res, "results.sarif")

# Export JUnit XML (for GitLab CI artifacts)
export_junit(res, "test-results.xml")

print(res) now also displays the quality score (%) alongside the SQALE rating.

Quick Local IDE Check

If you want instant feedback while coding (without CI), run:

library(rsonar)

# Fast local score (by default: no coverage/goodpractice for speed)
quality_score(".")

Typical output:

── Quick Quality Score ─────────────────────────────────────
ℹ Path   : /path/to/project
ℹ Score  : 82.4%
ℹ Rating : B
ℹ Time   : 2026-04-21 10:15

Alternatives and Added Value

See the full documentation.

Auto-Fix with air

rsonar integrates with air, Posit's fast R code formatter, to automatically fix code style issues:

library(rsonar)

# Install air (once per machine)
install_air()

# Check what would be changed (dry-run)
fix <- sonar_fix(".", dry_run = TRUE)
print(fix)

# Auto-fix all R files in the project
fix <- sonar_fix(".")

# Auto-fix and create a Merge Request (GitLab CI)
fix <- sonar_fix(".", create_mr = TRUE)

In CI pipelines, rsonar-fix is an optional manual job that:

1. Installs air via install_air()
2. Runs sonar_fix() to format all R files automatically
3. Creates a Merge Request (GitLab) or Pull Request (GitHub) with the changes

This allows developers to review and merge auto-formatted code easily, without manual intervention.

Function	Purpose
install_air()	Download and install the air formatter binary
sonar_fix()	Run air on R files, optionally create MR/PR

CI Integration

See the CI/CD examples.