I like to generate HTML code coverage output for my project, it looks good with color coding and everything.
Lately though it is becoming quite cumbersome because I need to be able to quickly and programmatically access information about code coverage for my run. I've been using lxml to query the output html files and get information about which lines are covered and compare for different runs, but I'd like to see if there is a better and more efficient way to do this.
What underlies the HTML files that are generated? Is there a lower level file that provides the information about which lines are covered from what files from which I can get this information programmatically using python but without having to resort to far fetched solutions like querying the coverage file?