Rank Files by Technical Debt (Hotspots)

debt_index() and quality_score() give a single project-wide number, but they don't tell you where to start fixing things. sonar_hotspots() fills that gap: it breaks the technical debt down per file and ranks files from the most to the least costly to remediate – the R equivalent of SonarQube's "Code Smells" / hotspots view, which developers use to prioritize their first pull request after an audit.

Usage

sonar_hotspots(
  x,
  n = 10,
  cost_lint_error = 30,
  cost_lint_warning = 10,
  cost_lint_style = 2,
  cost_style = 5
)

Arguments

x

An rsonar_result object returned by sonar_analyse().

n

Maximum number of files to return. Default 10.

cost_lint_error

Cost in minutes per lint issue of type error. Default 30.

cost_lint_warning

Cost in minutes per lint warning. Default 10.

cost_lint_style

Cost in minutes per lint style violation. Default 2.

cost_style

Cost in minutes per improperly formatted file (styler). Default 5.

Value

An rsonar_hotspots object (a data frame) with one row per file, ordered by decreasing debt, containing the columns:

rank

Rank position, 1 = highest debt

file

File path, relative to the analyzed project

lint_errors, lint_warnings, lint_style

Issue counts by severity, from lintr

style_issue

TRUE if the file needs re-formatting (styler)

coverage_pct

Line coverage for the file, or NA if coverage was not computed or the file has no coverage data

debt_minutes

Estimated remediation effort, in minutes

See also

debt_index(), sonar_analyse(), quality_gate()

Examples

if (FALSE) { # \dontrun{
res <- sonar_analyse(".")

# The 10 files that concentrate the most technical debt
hotspots <- sonar_hotspots(res)
print(hotspots)

# Only the top 3, with custom weights
sonar_hotspots(res, n = 3, cost_lint_error = 60)
} # }