I confess I'm not sure I understand what you're asking. Are you asking about "what percentage of coverage constitutes 'done'?" There isn't such a thing, and that places too much faith in the number anyway. (For example, "the number" is currently affected by the fact that some blocks of code are marked @osx_only or @windows_only, and thus aren't run on my linux machine.)
The best strategy is to look in the test/ folder, see how tests are written, try running code-coverage locally, and find some things that aren't tested that you think should be. Here's an example of a currently-open pull request that added some tests: https://github.com/JuliaLang/julia/pull/9674 But like I said, I'm not sure I really understood your question, so maybe my random collection of statements doesn't help. --Tim On Wednesday, January 07, 2015 08:52:14 PM cdm wrote: > for the benefit of those whom might help to "write tests" ... > > could the masses have an example of an test objective and > a test written to either satisfy, or NOT satisfy said objective. > > presumably, examples could be quite simple, or rather > complex ... > > many thanks, > > cdm
