The Overdue Update

I should release more often... I mean, 13 months? Really?

General added features

• Added %c and %C character classes (consonants + y and anything else, respectively) (#50).
• Added %p and %P character classes (printables and anything else, respectively).
• Added %q and %Q character classes (printables + newline and anything else, respectively).
• Added 7 shortcuts: Ê (l ), Ë (mDEF{D), Ì (gJ ), Í (n2 ), Î (g ), Ï (XYZ{Y), Ñ (*2).
• Added Math.approx function, which rounds out errors of a magnitude less than 10ⁿ (default n = 5).
• Each line except the last is now assigned to a variable, in order U, V, W, X, Y, Z, A, B, C, ... T (not that anyone will ever get anywhere near that far)

Bug fixes/improvements

• S.e() used the g flag by default, breaking the effect of logging each successive replacement (#31).
• A.o(n) assumed n was an integer, entering infinite loops if it wasn't (#37).
• A.r(f) now starts with the first element if no starting value is given.
• A.x() returned NaN if any of its elements couldn't be parsed as a number.
• A.z(n) broke somehow. I think the fix broke it in a different way.
• Auto-functions in A.ä(f) and A.å(f) didn't use both inputs.
• A.ç(a) both modified A and filled the array with references rather than copies of a.
• A.í(f,a) didn't swap arguments.
• N.k(), N.ç(), and N.î() used this as a Number object rather than a pure number (#35).
• N.v() now acts as if all numbers are divisible by 0.
• Switched from String.fromCharCode to String.fromCodePoint (#42).
• Added more implicit commas before decimal points.
• General fixes with implicit Us.
• General fixes with $interpolated JS$.
• Balance parentheses inside ?:, adding an extra layer if necessary.
• Make ++ and -- take precedence over anything else, adding a layer of parentheses.
• Remove many unnecessary layers of parentheses to improve readability.
• Add spacing to improve readability. (Only permanent side-effect is breaking The Polyglot)
• More I haven't yet mentioned and don't remember.

New String methods

• S.k(s); removes chars in s (opposite of S.o(s)).
• S.k(f); removes chars that satisfy f (opposite of S.o(f)).
• S.n(s); converts from base s.length to integer using s as the alphabet.
• S.n(a); same as above, but with an array of items making up the alphabet.
• S.o(f); keeps only the chars that satisfy f (similar to A.f(f)).
• S.y(f); maps each column through f.
• S.í(s,f?); pairs each char with the corresponding one in s, optionally mapping each pair through f.
• S.ù/ú/û(n,s=" "); left/right/center-pads each line to length n with s.
• S.ù/ú/û(s=" "); left/right/center-pads each line to form a rectangle.

New Array methods

• A.e(a); returns whether the two arrays are equal (#42)
• A.k(f); removes items that satisfy f (opposite of A.f(f)).
• A.y(f); maps each column through f.
• A.ì(s); converts from base s.length to integer using s as the alphabet.
• A.ì(a); same as above, but with an array of items making up the alphabet.
• A,ï(a,f?); Cartesian product, mapping each pair through f. If no a is given, uses A again.
• A.ù/ú/û(n,s=" "); left/right/center-pads each line to length n with s.
• A.ù/ú/û(s=" "); left/right/center-pads each line to form a rectangle.

New Number methods

• N.a(n); absolute difference between N and n.
• N.d(n); gets character at codepoint N+n.
• N.j(n); whether N is coprime to n (they share no prime factors).
• N.s(f); converts to string, maps through f, then converts back to number.
• N.s(s); converts from integer to base s.length using s as the alphabet.
• N.s(a); same as above, but with an array of items making up the alphabet.
• N.y(n=2); GCD. (Even works on decimals, thanks to Math.approx!)
• N.z(n=2); integer division.
• N.ì(s); converts from integer to base s.length using s as the alphabet.
• N.ì(a); same as above, but with an array of items making up the alphabet.

I should keep better track of my changes to make it easier to write these changelogs. Or maybe I could just release an update every few weeks or months so I don't forget what I've done...