fc, lc, uc, and tc (tc is titlecase, known in Perl as ucfirst) change the case of their operands. It would be relatively simple to move their full Unicode definitions into core, like we did in 5.20 for things like \w, [:xdigit:], etc,. Part of this is a no-brainer. It turns out that the tc definition is identical to the uc definition except for 89 of the million+ possible code points, and this number is unlikely to grow. By separating those 89 out, we can combine the remaining data structures for the two ops. It's only a little extra text-segment data to compile-in those 89 definitions. But then it occurred to me that we might want to compile in the three remaining properties. This would increase the static text segment size, by roughly what \w did, for each one. And it would decrease the data segment should these definitions actually get used. (The definitions for the code points 0-255 are already compiled-in.) I just want to get a general sense from people of whether this might be worth pursuingThread Next