Feature: User-defined dataset driven tests

[Snaipe]

From the discussion at #40, a use-case has come to light:

[...] to implement parameterized tests, that would iterate over a user-generated dataset rather than the cartesian product of many datasets.

Also, this will run a separate test for each member of dataset and hence yield better result stats.

e.g. current approach is to enumerate over TheoryDataSet placeholder data array, which results in:

[====] Synthesis: Tested: 1 | Passing: 0 | Failing: 1 | Crashing: 0

for 1276 data items tested out of which (for example) only 26 are failing.

With ParameterizedTest feature, we expect to get:

[====] Synthesis: Tested: 1276 | Passing: 1250 | Failing: 26 | Crashing: 0

[Snaipe]

It seems the hub utility chocked on the number of commits included for the merge, the pull request has been recreated at #44.

[am11]

Today is a mystery day. This is my first day to learn that it is possible to convert an existing issue to a pull request on GitHub. "How?" is still a mystery. It took me sometime to confirm that I indeed opened an issue, not a PR. 😄

[Snaipe]

@am11 you can convert any issue to a pull request with the hub CLI utility:

hub pull-request -b <base branch> -i <issue number>

[am11]

Excellent! Will give it a try. Thanks for sharing git + hub = GitHub. :)