Closed area closed 7 years ago
@@ master #52 diff @@
==========================================
Files 6 6
Lines 291 291
Methods 47 47
Messages 0 0
Branches 39 39
==========================================
Hits 286 286
Misses 5 5
Partials 0 0
Powered by Codecov. Last update 93a7f41...a1a4aeb
Oh no, that's really bad that those tests were doing nothing. I'm into using zepellin as a baseline. They actively work on the contracts and cover a lot of ground. It would be cool to be able to swap the zepellin source out every three months or so and run against the new version and see if everything's good.
Ultimately I'd like to remove them, but they're certainly a valuable canary in the coalmine for now!
Well, now they are. Before they were like a plastic canary floating in a canal somewhere.
I've done two things here, one more controversial than the other:
Make sure the Zeppelin tests are actually doing something! For multi-contract compilations, we were passing the wrong object to solc, so an empty object was being returned, which obviously had no errors in it. Once that was corrected, some tests were also missing some files to be included
Instrument all contracts in these tests, not just the one notionally being compiled. This would have caught the problems with upgrading to solc 0.4.8, whereas the previous tests, even if they had worked as intended, would not have done. We are (currently) always instrumenting all contracts, so it makes sense that that is what should be tested here.