On Wed, Jan 16, 2008 at 09:18:17AM -0800, Larry Wall wrote: : We could reserve all the is_* isnt_* ok_* names for that, I suppose.
I've gone ahead and done that. Let me also reiterate here in public the basic underlying premise that "fudge" is only for fudging the tests, and a truly passing test must do so by running the bare test, without any preprocessor at all. (And without any equivalent grammar fudges, which is why I don't want each implementation's grammar trying to second-guess the tests in different ways. The preprocessor approach pretty much eliminates such accidental and/or intentional duplicity...) Alternately, if we always run the test file through fudge, we could have fudge police the policy all the time, since it would know whether it was asked to modify anything. I suppose we could go as far as to say that the final test of every test file is that fudge passed the file through unmodified. Or maybe fudge just adds a fail() as the last test of any fudged file. Something like that... Larry
