Closed tyilo closed 9 years ago
I really don't like this system. Easy printability is more important than those extra couple of bytes.
A core concept of a programming language is that it's programmable by humans. As soon as you introduce non-printable characters I'd argue that aspect is lost or severely reduced.
In Pyth5 you can still have arbitrary binary data in binary strings (.""
). But that's data, not code.
Normally in codegolf your program is ranked by its number of bytes, so it would make sense to make all 256 different byte values have different meaning. Currently all pyth tokens are in the printable ascii-range (
0x20-0x7e
), but I think we should consider adding functionality to characters out of this range, especially in the range0x80-0xff
.Maybe we could just transform
0x80 + x
to"." + x
(where0 <= x <= 0x7f
), so that for instance.aQ
could be shortened by one byte by using\xe1Q
. (0xe1 == hex(ord('a') + 0x80)
).