Inspirationally shameless, and a great demonstration of the power of 'and' versus '&&', something that I hadn't fully understood. Until now. I have to say, at first it looks like a drop-in alternative to the presumably scary C-style double-ampersands (as in, an artifact of the use English movement), but when you get right down to it, it binds much more loosely, enhancing its utility vastly.

In the spirit of shameless borrowing, switching to 'pop' and using 'and' nets the following: sub f{for($t=pop;$_=(0,l,r)[$$t{d}cmp$_[0]]and$t=$$t{$_};){}$t} But this is really just converging on the same thing: sub f{for($t=pop;$_=$$t{d}cmp$_[0]and$t=$$t{$_>0?l:r};){}$t} Which is the same length, and functionally the same due to heavy cross-pollination.

Of course, if the spec had indicated that the two branches were labelled '-1' and '1' instead of 'r' and 'l', that would certainly simplify things a whole lot. Or at least it would save 6 precious characters: sub f{for($t=pop;$_=$$t{d}cmp$_[0]and$t=$$t{$_};){}$t}