----- Original Message ---- From: Michael G Schwern <[EMAIL PROTECTED]>
> 1. I want to name a group of tests rather than the individuals. ... > Here's what we came up with. > > 1..10 > ..4 - name for this group > ok 1 ... > Pros: > * Its backwards compatible. The ..# lines are currently considered > junk and ignored. Is this behavior documented anywhere? On the off chance I'll be able to finish a TAP grammar, I need to know what to do with lines which don't parse. A parser could throw an exception and this seems a reasonable thing to do, but that means TAP::Parser behavior would not match the current behavior. And since we're on the topic, I'm still wondering how we can disambiguate diag() output (the '#>' proposal seemed workable) and whether version numbers can be included in the TAP output. The latter would allow a simple test of the TAP version and allow the proper grammar to be loaded rather than use heuristics to guess what version is allowed. Cheers, Ovid
