On Tue, 24 Apr 2018 12:09:50 +0200 Stefan Schmidt <ste...@osg.samsung.com> wrote:
> https://sonarcloud.io/project/issues?branch=devs%2Fstefan%2Fsonar-test&id=efl&resolved=false > > 34441 issues found looks rather noisy to me, but with some filtering > we might be able to find get the interesting parts out of it. Another way to look at false positives or noise. If the code throws off some static analyzers. How human readable/parseable is the code? If a static analyzer cannot follow a code path. Maybe humans will have a hard time just the same. At least something worth considering IMHO. Does not apply in all situations, but IMHO it is an interesting perspective. I have modified some code for even Coverity to be able to follow it better, and eliminate false positives that way. Which I think helps humans with the code just the same as static analyzers. Along those same lines. One aspect of Sonar I find interesting is the Cognitive Complexity[1][2]. Beyond the Coverage aspect I very much like others ignore, assuming you run all code, via tests, usage, etc. I am not sure many take into consideration the Cognitive Complexity of code they write. Which makes it friendly to other humans reading the code. IMHO easier code is for humans to follow, easier for new devs to get involved, others to help fix or improve things, etc. I think it also is like programming along the Unix philosophy[3]. Where programs are small and do limited things and do them well. Where functions are kept small for the same purposes. It will not always help an analyzer or human, but worth taking into consideration IMHO. https://en.wikipedia.org/wiki/Computational_complexity_theory https://www.sonarsource.com/resources/white-papers/cognitive-complexity.html https://en.wikipedia.org/wiki/Unix_philosophy -- William L. Thomson Jr.
pgpBQr9mkmtnT.pgp
Description: OpenPGP digital signature
------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
