Project: codecoverage/crawlerΒΆ

contact:Marco Castelluccio, (backup Release Management)

This project uses the Code Coverage Crawler to browse pages from a list of websites and publishes reports of code that is exercised by interacting with those pages, but not exercised by our automated tests.

First, a code coverage build of Firefox is downloaded from Taskcluster, along with code coverage artifacts that resulted from running the automated tests. This is done using firefox-code-coverage. The downloaded Firefox build is used to navigate a list of websites, generating coverage data on the way. The coverage data resulting from browsing websites is compared with the data resulting from running the automated tests, and a report of these differences is generated using genhtml.