Michael Gross

Results 58 comments of Michael Gross

> > I think for now it would be best to remove these lines (maybe you could open a separate PR so that we have this well tracked). > >...

> > What do you think? > > I would encourage introducing files generated by other languages. But if we were to generate our own, how would we go about...

@aconrad I think I still need your decision here about the `cobertura.xml`, i.e. correcting the tests to account for the duplicate `Main.java` or removing the corresponding line from the `cobertura.xml`...

> The method `Cobertura.files()` was designed to deduplicate filenames if they appeared multiple times and we should retain that behavior since that seems to be a valid use case (the...

Hi @aconrad After thinking more about it now, I understood that the current implementation is indeed correct, because we measure the stats on a file-based level. The only question one...

> This looks good, thanks! Can we update CHANGES file? Also, should we add a note somewhere in the README about this new option? done, can you please check again...

@loganharbour Is this related: https://github.com/aconrad/pycobertura/issues/117?

Is there a way to formulate bash commands for the individual statistics that would be much faster, e.g. Filename could be extracted via: ``` grep -o 'filename=[^ ]*' coverage_before.xml |...

What seems rather quick is the following code snippet I tried via `python -I` (this though only extracts filename): ``` >>> from xml.dom import minidom >>> import os >>> os.listdir()...

Ah ok I see, the problem is not loading the xml, but deriving the statistics from it...