Ran into some errors when trying to create new redis tests, a lot of inconsistent results. I believe the majority of it was due to not clearing redis before every test.
Sorry about not submitting separate pull requests. I can if you wish, but I also fixed a few other tiny errors. Thanks!
correct a spelling error
add .rvmrc and .ds_store files to .gitignore
flush the entire redis database before each example
Why 10%, this seems like an arbitrary limit? You can push as much data as you want, you'll simply tradeoff against collisions.
Theoretically, you should have 0 collisions at this point.
And really, if this isn't a good test, that's fine. I was also trying to test the clearing of redis did something. If you ramp this up from 10%, then more than this test would fail if you didn't clear redis. So whatever you think is best.
There is no guarantee that there will be 0 collisions. Bloomfilters are probabilistic data structures.. It is highly unlikely that you will get a collision with your configuration, but it is possible -- which makes for some great heisen failures in your test suite. :)
fix assertions and comments for change to BloomFilter::ConfigurationM…
Okay got rid of those tests :)