Talk:Fun With ZCharacter
Oh hey, I get it now. Neat hack!
One catch with this is that you've got to remember that the ZCharacter directive affects the dictionary, as well as text strings for output. For Z5/8 you've only got nine Z-characters for a dictionary entry. Any lowercase letter you move to the third row will take up four Z-characters in the dictionary entry, which could lead to problems: two characters in a dictionary word from the third row use up almost all the space there is. Template:Unsigned
OK, and yes I have thought about these kind of thing too. The corresponding syntax for ZILF, ZAPF, Zasm, and Frolg might also be worth mentioning (as well as where in an Inform 7 file you have to mention this, if you are doing so). Since I made Frolg, I know how it would work, but the standard library does not currently contain a macro to do this (although it could easily be written). One thing is that in Frolg, " marks are used only for packed strings, so for the Z-characters table you need to use ' marks around the list of characters instead. Another thing to note is that outside of COMPLIANT mode, it may be worth it to put characters commonly used together in the same row, and that a string of characters from row 2 with a single character from row 1 in the middle is a more efficient encoding than a string of characters from row 1 with a single character from row 2 in the middle. Characters not used in the vocabulary may even be available in multiple rows, if that would help. (The Frolg assembler does take all of these things into account.) --Zzo38 06:12, 7 August 2013 (UTC)