Hello, and welcome to the twenty-eigth Learn You a Language. Today's language being learned for a greater good is Fig, made by everyone's favourite JVM optimising compiler maker Seggan. Fig is an irrational byte lisp-like golflang. Fig CMCs (FCMCs) will be posted throughout the next 24 hours, and y'all are free to ask questions about the language.
12
Seggan isn't here to kick things off, but he's left a few pre-written notes for LYAL
Hello people in the future, I see LYAL has started. As I am probably not here, I'll quickly explain scoring to you. You see, Fig only uses the 96 printable ASCII characters (not including tab). How is it that it is so golfy, you ask? Well, on the basis that you can theoretically make a computer architecture using any log_256(n) bytes, Fig is actually scored in irrational bytes. A Fig "byte" is just big enough to contain the 96 characters. Therefore, any Fig score is nlog_256(96).
String compression works very differently from Vyxal. As you (probably) saw, compressed strings start with D and end with ". First, the string is treated as a base-94 integer (the codepage minus quote and newline). Then it is modded with 8, to get a number 0-7. If the 1st bit is set, the next thing is obtained from the dictionary. Otherwise, the next thing is a char.
If the second bit is set, a space is added before the just-obtained string. If the 3rd bit is set, capitalise the first character of the obtained string. Finally, append the string to the result. If you are brave enough to read me terrible atom-based-while-looping decompression algo, it is here
For reference, the online interpreter for Fig is here, and the operator list is here
FCMC: Given a list of positive integers, double the integers at even indices and halve the integers at odd indices (0-indexed)
@DialFrost Because several of us realized that we like learning new languages, and you know what's more fun than learning a new language? Learning a new language together. :D
Seems to me it's also undocumented behavior--the documentation for n only gives (any, num, fun), it doesn't say you can also do (any, fun, num). I figured it was worth a try, though, and it worked.
@DialFrost I enjoyed a few LYAL languages enough to keep using them (mainly BQN and Haskell). For others, yeah, I forget a lot of the specifics, but I usually remember at least the basics of how the language works, so if I want to use it again, it's easier to re-learn.
Ah, yes. I do find the irrational byte-count concept a bit off-putting, but that's how everyone else is counting it, so for the purposes of comparing figs to figs (heh) I'm doing the same.
to be clear, i'm not suggesting disallowing Fig and other alternate bit languages
there just needs to be a better way of representing the score
maybe rounding up to the nearest bit rather than byte? that way it actually makes some sense as something that can actually be represented as information
I guess if we're willing to entertain counting tokens or trytes or whatever, that calls into question the whole "Answers are scored in bytes" paradigm. Which, in theory, is fine because of the whole "Different languages don't compete with each other" philosophy.
I'm conflicted, personally. On the one hand, I do like the consistency of "score is measured in bytes." On the other hand, some languages have traditionally been scored in something other than bytes (e.g. Piet being scored in codels), and that does seem to make more sense than forcing every language to define an encoding in terms of bytes.
There's probably something wrong with this, but if your lang has N symbols, what about converting your code from bijective base N to binary and counting bits?
With my dwindling activity over the last few months, this probably doesn't come as a surprise... but I've decided to hand in my diamond. My reasons are similar to Alex's, half a year ago, in that there are various things (none bad, thankfully) these days that take up my time otherwise, and recent...
Remove digits or increment
code-golf
Given a number n tell how many step to reach 0 by removing every occurrence of the length of that number from it or, in case no changes happened, increment it by one.
Example
n = 2510105225
Step 1: 10 digits
remove 10's
25xxXX5225 -> 255225
Step 2: 255225 ha...
typically these languages have a 2D field of commands that is navigated by a code pointer, where each character is a command that can affect either the pointer, a data structure or the code field itself
Remove digits or increment
code-golf
Given a number n tell how many step to reach 0 by removing every occurrence of the length of that number from it or, in case no changes happened, increment it by one.
Example
n = 2510105225
Step 1: 10 digits
remove 10's
25xxXX5225 -> 255225
Step 2: 255225 ha...
