With NCrunch as a free tool, it is easy to justify shortcomings, but I can't justify a $150 tool that doesn't do EXACTLY what I need it to.
Don't get me wrong, NCrunch does SO MUCH right, but so do a lot of $50-ish tools like Beyond Compare, Regex Buddy, and Sublime Text. $150 demands a very high level of polish.
My three big features that would make the cost justifiable:
[list=1]
Better hybrid operating modes, like: "Run all tests for saved files, but impacted tests for files that have un-saved edits in the editor."
Data-driven test support for MSTest.
Real support for "inconclusive" tests. (Don't just mask inconclusive tests as pass or fail.)
These are the only things that actually get in my way on a daily basis. To be perfectly honest, I would probably buy it if just two out of these three were to be fixed.
Real support for "inconclusive" tests. (Don't just mask inconclusive tests as pass or fail.)
I second that, the third state is a must.
But i have bougth it anyway ;)
Likewise - my company got me a licence, but I'd really like better support for inconclusives. When TDDing I like to be able to stick an inconclusive test as a placeholder against new classes that fall out as a reminder to get back to it. I don't want them to pass as they need to show up, but I don't want em to fail as that gives an incorrect view of the true state, so I need to keep looking into the details to see if my code is passing its tests.
maybe just adding an additional filter to the NCrunch Tests windowalongside the show passing, failing, unexecuted & ignored filters would be enough. keep all the rest of the reporting, passing, failing and whatnot as it is, but just give that view to show that some of the tests got where they are because of their inconclusiveness.