« first day (3623 days earlier)      last day (176 days later) » 

12:25 AM
CMQ: Do you have a question you think was underappreciated (in terms of votes/# of answers/etc.)? If so, tell me which one, and I'll be happy to give it a try, see if a new answer can draw some more attention
 
10
Q: Implement Multiplicative Fuzzy Logic

AdámInspired by this excellent challenge (from which the bulk of this text is blatantly duct-taped) – and my highschool philosophy project... I define the following operators: Fuzzy Conjunction a ×F b is a × b Fuzzy Division a ÷F b is a ÷ b Fuzzy Negation –F b is 1 – b Fuzzy Disjunction a +F b i...

 
@Adám Does "extend the language of your choice to include the notation." mean that I can define 4 functions that take the necessary arguments and return the correct results?
 
I generally like interpreter questions, they're the part of programming that appeals to me
 
@Wezl I both like and dislike most good questions, as my go-to language for golfing is Jelly :/
 
@cairdcoinheringaahing That'd only be valid if you can get the required infix syntax into your language. Otherwise the whole thing becomes rather trivial.
 
12:37 AM
I'm thinking about redoing my dotcomma interpreter challenge, I ended up deleting it for some reason but I think it'd be an interesting challenge
 
@Adám I think I've got a Jelly answer that should work, let me know on my (incoming) answer if not
 
Will do.
Heh, we need a "Sandbox for proposed answers"…
5
 
12:49 AM
@Adám Should be posted
 
Looking at it already.
@cairdcoinheringaahing I'd say it is almost acceptable. Can you insert some string substitution preprocessing before you eval-as-Jelly? Then you can actually use ×F etc. in the code.
 
@Adám Shouldn't be too hard, I'll get on that
 
Cool
 
@Adám Given that it's getting late here and that Jelly hates 2-letter substitution more than I expected, I might have to give it a proper go tomorrow. I'll delete my answer until I get it tho
 
1:07 AM
No worries. I hope you're having fun at least.
 
@Adám Oh, bending Jelly to my will is always fun, until I realise I've been copy-pasting the F character the entire time :P
 
haha, yeah, that's just a normal F.
 
Well, it's disgustingly hacked together, but it works :D (editing + undeleting now)
 
Disgustingly hacked together? That's the spirit we all want to see!
 
I'm including my old version because I quite like it (and it almost meets the challenge spec), but I'll use my disgusting version for my score :)
@Adám Kinda weird question, but can we take input with a trailing space?
 
1:25 AM
@cairdcoinheringaahing That's an interesting question. I've also not specified internal spacing, though all my examples have a single space between tokens. I guess I'll allow it, though I can't imagine it takes a lot of Jelly to append a space, does it?
 
@Adám It's +2 bytes in Jelly, so not a massive issue if not, but every byte helps :P
 
@cairdcoinheringaahing Added explicit rule about spacing, and also that you can use _ for minus if you're using Jelly.
 
@Adám Oh yeah, I did kinda assume _ was allowed, but thanks for making it explicit :)
 
2:03 AM
@Adám Still disgustingly hacked together, but now 13 bytes shorter than an hour ago, I've undeleted my answer :P
 
Upvoted, but ugh, Jelly needs regex support.
 
Well, if a regex does the job, you just wouldn't use Jelly.
 
And if a regex does part of the job?
 
QuadR...?
 
i feel like regex wouldn't fit in jelly's style of doing things
 
2:12 AM
@Bubbler That's just for specific tasks in golfing, and only uses regex as preprocessing too.
@Razetime Could pretty much be implemented as in APL, no? Last two links would make the pattern(s) and the transformation(s)
 
@Adám yeah but it's just gonna be another two byte builtin which only makes string based tasks less interesting
 
@Adám that’s something that’s been hit hard by Dennis’ hiatus. There’s been a PR for regex atoms since 2018 that was being worked on
 
IC.
 
2:30 AM
(the QuadR port of APL answer happens to be 21 bytes)
 
@Bubbler That's very elegant. Post it!
 
It works great and handles multiple test cases just fine, except that it can't handle newlines in the input properly
 
@Bubbler Right, but for the "try all test cases" link, you can use ⍎⍤1⎕FMT⍵
 
 
9 hours later…
11:33 AM
CMQ: What is the longest Jelly answer you have seen/written?
Just saw this one
 
A very space consuming 14 bytes
 
11:48 AM
 
12:05 PM
@Lyxal Not at all, but I’m kinda busy today, so I’ll take a look when I can :)
 
ngn
12:30 PM
@fomin i'm thinking of mounting a second attack on your challenge
if only i knew how to dfa an nfa, the dfa could be used to generate its language
 
 
3 hours later…
3:39 PM
I know how to do that :)
I think there was a challenge on NFA to DFA too
@ngn ^^^
 
4:34 PM
@Adám Do we have to handle regular negation, or can we change the -500 test case to 0 - 500?
 
4:49 PM
@cairdcoinheringaahing No, negation isn't on the list of required functionality.
 
@Adám Ok, good. Now I just need to find a way to remove another byte so I can tie with QuadR :P
 
5:57 PM
@ngn the cheesy way to convert nfa to dfa is to have 2^n states where each DFA state is a sum of powers of 2 of the states that the nfa might be in at the time
 
Any way to golf this HTML/CSS/JS?
 
ngn
@Neil right, that can work. for this challenge it's easy to construct of an nfa with 3*(n+1) states, where n is at most 16. so bitmasks can fit in 64 bits.
 
6:15 PM
@Adám all I see is the default template
 
ngn
@Adám rm " around attribute values where possible
 
@JohnDvorak What? Isn't it populated with some HTML?
 
ngn
@JohnDvorak try reload. i got the default template the first time too.
 
... how does that (fail to) work?
font:monospace should work?
 
Nope.
 
 
1 hour later…
7:18 PM
@ngn How did you get in with the automata?
 
ngn
@Anush i'm still struggling :)
 
Ah.. Is it the conversion to DFA that is the problem?
 
ngn
@Anush i have a dfa (not necessarily correct) already. now i'm trying to count the possible outputs without generating them.
but i'm getting fewer than expected for "obenquillo" :(
 
@Adám can you get away with onload or at least onloadend instead of onreadystagechange?
<code><input style="font:inherit" is slightly shorter I guess
 
@ngn ah. Can you try it with the empty string?
 
7:25 PM
join`\n` ("template string")
 
@Neil No.
 
ngn
@Anush it prints "1"
 
If it doesn't work for that you know you are in trouble :)
@ngn that's bad
It sounds be 95 + 95*95
 
huh, why not, need to support IE9 or something?
 
I am talking about the number of strings within edit distance 2 of the empty string
 
7:28 PM
(side note: is the result HTML, if not, use innerText)
 
ngn
@Anush i got so carried away with these automata, i forgot what challenge i'll use them for :)
 
with(new XMLHttpRequest()){...} might save some bytes
 
Ooh, "with" exists. Nice.
@Neil Actually, it does. Something had gone stuck.
 
ngn
@Anush +1 for the empty string itself, i think
 
well that's all I can think of offhand
 
7:34 PM
@Neil Such golf, very hack.
5
 
ngn
@Neil @Adám are the parens after new XMLHttpRequest required?
 
@ngn Apparently not :-)
 
ngn
@Anush now i'm much closer but still missing some
 
8:11 PM
@ngn :)
 
ngn
@Anush can you give me another test case?
 
@Neil Thank you so much. You inspired me to even more golfs.
 
 
2 hours later…
9:57 PM
@ngn does the empty string work now? You could try the single letter a. I can make some more complicated ones in about 12 hours
 

« first day (3623 days earlier)      last day (176 days later) »