C++ and QML Code Coverage | Tools | #QtWS21

Are you testing enough? This old question is hard, if not impossible, to answer for certain. The measurement of code coverage, i.e. the percentage of code that is being exercised by your test is a very strong indicator. This kind of analysis can even be a must for certification of safety-critical software. In this talk you will learn how to discover untested parts of your C++ and QML source code. Different coverage levels like Decision Coverage and MC/DC will be explained and compared. And in cases of untestable sections of code, a manual validation can be performed. Coverage data collected for older versions of software can act as an input for the review of changes in new versions. A tool for analysis of patches will be shown.

Speaker: Harri Porten (Director of Product QA Tools, The Qt Company)