On Mon, May 09, 2005 at 12:14:35PM -0700, Larry Wall wrote:
: On Mon, May 09, 2005 at 02:08:31PM -0500, Patrick R. Michaud wrote:
: : Hmmm, then would $x.$j.2 then be equivalent to $x[$j-1][1] ? : : Ouch.
Maybe that's a good reason to switch from 1-based to 0-based $<digit> vars. Not sure what that would do to the current $0 though. Most of the time $/ can stand in for it, I guess, though s/.../$// is visually problematic. We could maybe resurrect $&.
Actually, I'd be just as happy *not* to have $1.1.1 at all. There's no real win over $1[0][0], and I think it would be better to leave the multi-dot syntax to be visually unambiguous as a version number.
Damian
