According to Larry Wall: > I don't claim to follow all this talk about "stores"
Think about tied values. When does STORE get called, precisely, on a tied target of s///? It's good to be explicit about this, down at the C API level, just so we know what to optimize for. The final answer is probably less interesting than the discussion that illuminates it. > In an example like > ($a = $genome) ~~ s/TTAGGG// > you'd like to capture the idea that there's already a copy being forced > in the context, so it'd be nice to use that copy to do the transformation > without inducing an additional copy. That's true and worth doing, but it's beside my intended point. Mea culpa: I combined the assignment with the substitution to make the example a one-liner, and thus muddied it. The s/// is the topic. > There are also optimization modes where a single substitution like > $genome ~~ s/TTAGGG// > can be done "mostly in place". Indeed, I had that in mind when I designed Topaz's "open the hood" feature for string Scalars, which I mentioned earlier as a possible model for optimizing s/// on normal strings while giving well-defined semantics to s/// on abnormal strings. -- Chip Salzenberg - a.k.a. - <[EMAIL PROTECTED]> "Persistence in one opinion has never been considered a merit in political leaders." --Marcus Tullius Cicero
