Hi Michael, Thanks for suggesting this. Also with Apache Yetus we can enable plugins like Spotbugs for Java, pylint for Python, hadolint for Docker, shellcheck for shellscripts and many more. Is this what you're looking for? I'm in favor of static analysis, but with so many languages in the repository, I'm not sure if it will work.
Cheers, Fokko Op wo 7 nov. 2018 om 21:02 schreef Zoltan Farkas <[email protected]>: > +1 for static analysis. > > here is what codacy looks like on the avro fork I use: > https://app.codacy.com/project/zolyfarkas/avro/dashboard > > > —Z > > > > On Nov 7, 2018, at 10:01 AM, Michael A. Smith <[email protected]> > wrote: > > > > Now that the Yetus/Travis integration is running (yay), I hope it's not > > premature to talk about static analysis tools, or as they're sometimes > > called "automated code review". > > > > Correct me if I have it wrong, but I believe our Yetus/Travis integration > > is focused on running the handwritten test cases in each lang. Even if we > > turn on a static analyzer, we'd have to pick through its output in > Travis' > > console. "Automated Code Review" tools provide line-based feedback in > your > > PR, and they return results much faster than unit tests. Many of the > > companies that run these automated code review tools are free for open > > source projects. Here are a few that I've worked with before, for > > consideration: > > > > - https://codeclimate.com/oss/ > > - https://codebeat.co/open-source/ > > - http://opensource.codacy.com/ > > - https://scrutinizer-ci.com/ > > > > They are all great, and I don't strongly care which one we use, but I > think > > isolating the static analysis from the unit test runner is worth doing so > > we get feedback faster on little easy-to-fix things. > > > > What do you think? > >
