04:16 <FromGitter> <incognitorobito> Hi all, I want to check if the type used to initialize my class is comparable ⏎ Would that just be T.is_a?(Comparable) or should I use responds_to just check that the methods I need are there?

04:55 <FromGitter> <3n-k1> is there a crystal library for runtime-configured plugins, or do i have to load the so and grab a fn pointer manually?

06:32 <Andriamanitra> how do i check or is it even possible to check if ARGF got anything to read? ideally i would like both "cat input.txt | my_program" and "my_program input.txt" to do work on ARGF but "my_program" should do something else (maybe print a helpful message about usage)

06:33 <Andriamanitra> the api docs don't really list the methods for ARGV object

06:34 <FromGitter> <HertzDevil> `ARGV` is an `Array(String)`

06:34 <Andriamanitra> ARGF i mean but i guess the same is true for argv too :p

06:35 <Andriamanitra> oh i should be looking at IO methods?

06:35 <Andriamanitra> got it, i'm stupid

06:48 <Andriamanitra> (ok i read through the methods and source code in argf.cr, still can't figure it out, i guess i'll have to manage without helpful messages)

11:37 <FromGitter> <3n-k1> event doesn't have that method, but the subclass does

11:38 <FromGitter> <3n-k1> oh wait, ignore me

11:43 <FromGitter> <3n-k1> i was scrolled way far up lol

13:32 <FromGitter> <ImAHopelessDev_gitlab> hello

13:32 <FromGitter> <ImAHopelessDev_gitlab> happy december first

13:35 <FromGitter> <tenebrousedge> same atcha

13:35 <FromGitter> <ImAHopelessDev_gitlab> just got off work, it's crystal time baby!!

13:36 <FromGitter> <tenebrousedge> yay!

13:36 <FromGitter> <tenebrousedge> 💎

13:36 <FromGitter> <ImAHopelessDev_gitlab> :ruby:

13:36 <FromGitter> <ImAHopelessDev_gitlab> 💎

13:37 <FromGitter> <ImAHopelessDev_gitlab> opening up notepad++ with crystal code, and waiting on a doordash order for food. life can't get better

13:37 <FromGitter> <tenebrousedge> spacevim :plus1:

13:38 <FromGitter> <tenebrousedge> what noms are you nomming?

13:38 <FromGitter> <ImAHopelessDev_gitlab> jack in the box. value fries and jumbo jack

13:39 <FromGitter> <tenebrousedge> cool

13:39 <FromGitter> <tenebrousedge> oh shit it's Advent of Code

13:39 <FromGitter> <ImAHopelessDev_gitlab> ary on it

13:41 <erdnaxeli> does people are doing it on crystal and share their solutions?

13:42 <FromGitter> <Blacksmoke16> @incognitorobito no `is_a?` would be how you do it

13:42 <erdnaxeli> it's always a good way to learn

13:43 <FromGitter> <ImAHopelessDev_gitlab> learning? what's dat

13:49 <kevinsjoberg> I did it this morning (GMT+1), https://github.com/kevinsjoberg/aoc-2020.

13:58 <FromGitter> <Dan-Do> new UI for https://crystal-lang.org/reference is nice :)

13:59 <erdnaxeli> kevinsjoberg, cool! I try to optimize it a bit so it's not as compact: https://github.com/erdnaxeli/adventofcode/blob/master/2020/src/01.cr

14:00 <kevinsjoberg> erdnaxeli: I was thinking of writing an optimised solution where I subtract one of the pairs, but I decided not to pursue it given how small the actual puzzle input was.

14:00 <erdnaxeli> yep the input is really small

14:01 <kevinsjoberg> @Dan-Do when did they release that? Just now? Look slick!

14:02 <erdnaxeli> yesterday iirc

14:14 <FromGitter> <naqvis> @Dan-Do Added `CSS Selector` support to crystal-html5 (https://github.com/naqvis/crystal-html5) shard

14:16 <FromGitter> <Dan-Do> that's great, I will look at it

14:16 <FromGitter> <naqvis> 👍

14:22 <FromGitter> <Dan-Do> btw, is there setter `node.inner_html = "<div id='foo'>bar</div>"`?

14:26 <FromGitter> <naqvis> you can set the textual part by updating the `data`, but not the node markup

14:26 <FromGitter> <naqvis> to create new node, you better construct a `Node` instance

14:40 <Andriamanitra> i did it in a couple different ways https://github.com/Andriamanitra/adventofcode2020/blob/main/day01/benchmarks.cr

14:42 <kevinsjoberg> Andriamanitra: Could you share the benchmark results. Not at the computer right now but interested in the output. :)

14:42 <Andriamanitra> kevinsjoberg: https://github.com/Andriamanitra/adventofcode2020/blob/main/day01/benchmark-results.txt

14:43 <Andriamanitra> i'm sure there are even smarter ways for the part two but at least mine is better than O(n^3)

14:45 <erdnaxeli> the smart one is … smart :p

14:46 <Andriamanitra> yeah, i wish i could say i came up with it myself :(

14:48 <kevinsjoberg> This is how my friend implemented p1 and p2 in Ruby.

14:48 <kevinsjoberg> https://github.com/Istanful/adventofcode2020/tree/main/day1

14:48 <kevinsjoberg> Kind of like how he went about it.

14:50 <Andriamanitra> i was thinking there's probably a way to re-use part 1 solution, that's nice

14:55 <straight-shoota> hint: Crystal has lots of useful collection methods for these little AoC tasks

14:57 <Andriamanitra> straight-shoota: did i miss something obvious in mine?

15:02 <straight-shoota> hm

15:02 <straight-shoota> that smart solution looks interesting

15:03 <straight-shoota> but I'd figure it would probably be faster if you use an array instead of a set. The numbers aren't that big.

15:03 <kevinsjoberg> straight-shoota: I think my implementation is very readable but it's basically "brute force". https://github.com/KevinSjoberg/aoc-2020/blob/main/01/day01.cr

15:04 <straight-shoota> And you save on hashing

15:05 <straight-shoota> kevinsjoberg, yeah it's brute force but it's also concise. And performance doesn't need to be a constraint for such tasks.

15:05 <straight-shoota> I did it that way, too. Just with less boilerplate

15:05 <kevinsjoberg> Pretty much my thinking, the puzzle input was so small I thought I'd favour readability over performance here.

15:07 <straight-shoota> readability is always top priority

15:07 <Andriamanitra> i thought combinations(2).find is too boring so i did it the fun way first

15:08 <straight-shoota> I used each_combination for being a bit less boring :D

15:08 <kevinsjoberg> Haha. I like it! 🤘

15:09 <kevinsjoberg> Anyone attending Raw Crystal by the way?

15:10 <straight-shoota> of course

15:15 <FromGitter> <Blacksmoke16> theres also that `reuse` argument

15:15 <FromGitter> <Blacksmoke16> granted idk what the challenge/problem is so no idea if its useful or not tho 😆

15:15 <straight-shoota> yeah, reuse should be passed

15:19 <kevinsjoberg> What does `reuse` do more specifically?

15:23 <FromGitter> <Blacksmoke16> ```code paste, see link``` [https://gitter.im/crystal-lang/crystal?at=5fc65fe28abfba5d471fc301]

15:23 <FromGitter> <ImAHopelessDev_gitlab> falsey, i hate that word

15:25 <straight-shoota> if you don't pass reuse: true the method will allocate a new array for each invocation, adding a lot of overhead

15:26 <straight-shoota> that's the safe default because it allows you to modify the array and keep a reference to it

15:27 <straight-shoota> with reuse its contents would change on the next invocation

16:12 <yxhuvud> I can recommend sorting the input before using it btw - it gives me a factor 10 in performance :)

16:13 <FromGitter> <tenebrousedge> I love the gh app

16:13 <FromGitter> <tenebrousedge> apparently

16:13 <FromGitter> <tenebrousedge> `gh repo create`

16:37 <kevinsjoberg> straight-shoota: didn’t know about that. Thanks!

19:20 <FromGitter> <wontruefree> https://www.meetup.com/ChicagoRuby/events/pjfxvrybcqbcb/

19:21 <FromGitter> <wontruefree> I am presenting at Chicago Ruby tonight if anyone is interested https://www.meetup.com/ChicagoRuby/events/pjfxvrybcqbcb/

19:52 <FromGitter> <RespiteSage> Is there a Crystal leaderboard for AoC?

20:51 <FromGitter> <tenebrousedge> oh do we do leaderboards for AoC?

21:22 <FromGitter> <RespiteSage> Who can say? :P It's my first year trying to do AoC as the puzzles come out. I thought there was a Crystal leaderboard or something last year, but I might be misremembering.

21:24 <erdnaxeli> there can be private leaderboard, accessible with a code

21:24 <erdnaxeli> we could do a crystal one and share the code here

21:24 <erdnaxeli> I don't know if there is a user count limit