PragTob / benchee / 1155

95%

master: 94%

Pull #244

travis-ci

Remove any bindings to any data allocated during measurement

This will ensure that we're collecting _only_ data that is allocated as
part of the execution of the function we're trying to benchmark. The
downside is that we end up running the function twice - once to
capture the return value and once to measure the memory usage - but
this way we're sure that we're not accidentally counting terms that
weren't directly a result of the function we're measuring.

8 of 8 new or added lines in 1 file covered. (100.0%)

574 of 603 relevant lines covered (95.19%)

15750.17 hits per line