I should have also added, you can test coverage locally. I've posted (and made 
big changes to, twice, including just a few minutes ago---sorry!) instructions 
here:
https://github.com/JuliaLang/julia/issues/9493

I did just try this myself, and analyzed the results as described 
https://github.com/IainNZ/Coverage.jl
3 tests had to be omitted, so this is a bit of an underestimate. (There are 
also several caveats to quantifying coverage, almost no matter how we do it.) 
But for what it's worth, the coverage of base came out to be almost exactly 
65%. That's approximately what I'm seeing for some of the better-tested 
packages, too.

Anyone who would like to see that number higher, please help write tests! 
Since I posted a request for help with this, we've already gotten some awesome 
contributions from @kshyatt and @hayd (and maybe others I've forgotten), who 
have surely bumped that number upward.

Best,
--Tim

On Wednesday, January 07, 2015 02:04:17 PM Don MacMillen wrote:
> Very nice.  Are there recent results for base on Coveralls?  All I see
> is from a run from 25 October.
> 
> On Wednesday, January 7, 2015 4:09:49 AM UTC-8, Tim Holy wrote:
> > On Wednesday, January 07, 2015 12:41:16 AM Ariel Keselman wrote:
> > > great, does this mean untested functions are now actually counted in
> > > coverage?
> > 
> > Yes

Reply via email to