DotCov.Tool 4.3.0

DotCov.Tool

dotcov gates any Cobertura report: it tells you, your CI, or your coding agent whether coverage passes, and exactly which lines and branches to test next.

dotnet tool install -g DotCov.Tool

Quick start

dotcov test --min-line 90 --min-branch 80                 # .NET: run the tests, then gate them
dotcov check TestResults/ --min-line 90 --min-branch 80   # any Cobertura report

FAIL: line 81.8% (min 90%), branch 50.0% (min 80%) - line and branch coverage below threshold
files below line or branch threshold:
  src/Orders.cs  77.7% lines  33.3% branches
    uncovered 14-15 · branches 12 (1/2), 18 (1/4)

The verdict is the first stderr line and matches the exit code. Below it is every file under a threshold, worst first, with the lines no test runs and the branches only partly taken.

For agents

• Streams. Stdout carries only data: the table, markdown, or JSON. The verdict of check, test and crap is always the first stderr line; warnings, excluded files and merged reports follow it.
• JSON. --format json prints one document that opens with "schemaVersion": 1, "command", "inputs" and "warnings". check, test and crap add gate: outcome, exitCode, reason (the verdict line), the thresholds, and for check/test the offenders, each with path, lineRate, branchRate, uncovered ("14-15,21") and partialBranches.
• Size. --top N caps the offenders (default 20) or the files of report; uncovered lines are ranges. --no-color or NO_COLOR=1 removes ANSI.
• Help. dotcov <command> --help prints that command's usage, options and exit codes. A mistyped option answers with the closest valid one and the full option list, exit 2.
• dotcov test prints reports: TestResults/<run> on stderr after the verdict. With --format json or md it writes the dotnet test log to TestResults/<run>/dotnet-test.log and shows it on stderr only when the run fails.

Commands

dotcov test   [<project>] --min-line 80 [-- <dotnet test arguments>]
dotcov check  TestResults/ --min-line 80 --min-branch 60 --format json
dotcov check  coverage/cobertura-coverage.xml --min-line 90   # Vitest, c8, Jest, coverage.py
dotcov report TestResults/ --top 10                       # the ten worst files
dotcov crap   TestResults/ --max-crap 30                  # per-method risk gate
dotcov diff   before.xml after.xml --format md
dotcov snapshot TestResults/ --commit SHA --branch main --project MyApp

check and test need --min-line, --min-branch, or both (0 to 100); a dimension left out is not checked, and 0 disables one. A path is a file or a directory; a directory is searched for **/*cobertura*.xml and every match is merged. Name other files with --pattern (--pattern "**/coverage.xml" for gcovr and coverage.py). A directory without a match gets a warning naming the .xml files it holds.

dotcov test passes the coverage options of the runner dotnet test uses here (DOTNET_TEST_RUNNER, else test.runner in global.json, else VSTest). Microsoft.Testing.Platform projects need Microsoft.Testing.Extensions.CodeCoverage (TUnit includes it); VSTest projects get the collector from Microsoft.NET.Test.Sdk.

Exit codes

crap

crap scores every method comp² · (1 − cov)³ + comp, worst first, with each method's file:line. A fully covered method scores its complexity, so a high score on covered code means split it, not test it. crap is for .NET: Coverlet embeds complexity in its report, and for other .NET emitters --metrics from dotnet msbuild /t:Metrics (Microsoft.CodeAnalysis.Metrics) supplies it. Istanbul (Vitest, c8, Jest) and coverage.py write none; gate their reports with check. Every method must be scored: one without complexity or lines makes the gate NODATA, and so does a metrics member no covered method matches.

Options

A command rejects an option it does not take, --name=value, and a second path; it never ignores one. An active exclusion is listed on stderr and in the JSON as excluded and exclusionRules. Percentages read 62.0% on every host.

Also in this family

DotCov — the parser and gates as a library · DotCov.Fallout — coverage targets for Fallout builds

Documentation · Issues · MIT