On 7/17/06, Ovid <[EMAIL PROTECTED]> wrote:
Many people would prefer that $description always be the last thing you type and always be something you can provide.
IMO if I were to write a replacement for Test::More id put the description argument first, and therby make it mandatory. And then you wouldnt have the consistancy issues as it wouldnt matter if the routine took a list or not, the description would always be out of the way. Test names shouldnt be optional. Finding a particular test in a file by its number can be quite difficult, especially in test files where you dont have stuff like 'ok 26'. When ok() and is() are silently incrementing the counter and test names arent used how is one supposed to find the failing test? As you probably know it can be quite difficult. If every test has to have a unique string associated to it then it becomes a LOT easier to find whats failed. When a test can report the line of code that failed reliably this isnt such a big deal, but all to often in my experience the line number reports are not correct, as befits a heuristic process such as we use. However even when you are mechanistically generating the tests the description can be part of the reference data used to do so and therefore make tracking down the actual code failing much easier. Producing a "next generation" test framework where test descriptions are optional IMO would be a disappointing lost opportunity. Yves -- perl -Mre=debug -e "/just|another|perl|hacker/"
