Home

Why would you want to use it

Usually coverage report is just one number. It’s not always obvious where to look and what to check if coverage decreased because of your code changes.

Situation gets worse if you haven’t changed code at all. What if you just changed the version of 3rd party library, or changed some test configuration?

This tool helps you to identify the list of files that you must focus on. Also it organizes the links to that files for the fast and simple navigation.

Once you have the list of files you must make manual side-by-side comparison. That is not very exciting part of work, but it’s much better than blind guessing or brute-forcing.

What this tool does:

1. Helps with coverage difference analysis between 2 builds reported at coverall.io.
2. It takes 2 builds as an input, the base build and the analyzed build.
3. It grabs coverage info from coverall.io, analyzes it and produces report in a table form.
4. All the listed in the table files showed the decrease of coverage (in %) in analyzed build.
5. Generated markdown report can be inserted in GitHub as is. It contains all the links you need for detailed analysis.

What it does not:

1. It doesn’t identify the base build for comparison, you have to find it by your own.
2. It doesn’t perform deep analysis of the coverage, you still have to analyse files manually to understand the root cause.
3. It is not suitable for CI process automation, but it is there to help you deal with problems identified by CI

Draft version of the tool has been successfully used in this PR, so it is kind of battle-proven :).