Channels
#advent-of-code
- k
karelpeeters
12/23/2017, 10:22 PMFor me it's
versus73ms
, that's a way bigger difference.8ms - k
karelpeeters
12/23/2017, 10:23 PMAh, but if you look in the sources of
there's an upper limit of 50 rounds.BigInteger - t
todd.ginsberg
12/23/2017, 10:24 PMOh, I was just running my unit test, so probably some overhead. Let me use time nanos... - t
todd.ginsberg
12/23/2017, 10:27 PMYes, 31ms vs 16ms. - k
karelpeeters
12/23/2017, 10:34 PMMakes sense! - m
menegatti
12/24/2017, 11:59 AMToday’s solution got me a bit of trouble because I was using a flatMap in my recursion and that increased the run time by a lot! - a
andyb
12/24/2017, 12:32 PMThink today's problem really lends itself to recursion. Only change between Part 1 & Part 2 was in the
function.maxBy - k
karelpeeters
12/24/2017, 12:59 PMRecursion for me too. I'm really liking the
DSL that kotlin has.Comparator - k
karelpeeters
12/24/2017, 1:00 PM@andyb you could have used the elvis operator instead of line 13. - a
andyb
12/24/2017, 1:04 PM@karelpeeters Nice, I always forget the
function as wellmaxWith - t
todd.ginsberg
12/24/2017, 2:12 PMRecursion for me as well. I still need to clean it up a bit. - t
todd.ginsberg
12/24/2017, 2:13 PMThis feels like a variation on the change making problem. - t
todd.ginsberg
12/24/2017, 2:14 PMOh!
! I didn't know that existed. I'm going to use that. 🙂maxWith - t
todd.ginsberg
12/24/2017, 2:19 PM@andyb That's not actually tail recursive, is it? - t
todd.ginsberg
12/24/2017, 2:19 PMI banged my head against a wall for a few minutes trying to figure out how to make it tailrec but I didn't end up needing it so I stopped focusing on it. - a
andyb
12/24/2017, 2:23 PM@todd.ginsberg I think the compiler does actually treat it as tail recursion as the IDE shows it as tail recursive & it doesn't complain if the function is marked astailrec - t
todd.ginsberg
12/24/2017, 2:24 PMWow, interesting. Mine complains if I mark it and it's not tailrec. IntelliJ says that one is? The recursive function call isn't the last call... - a
andyb
12/24/2017, 2:25 PMI had a look at the numbers though and figured that the stack depth probably wouldn't exceed ~40 so you could get away without tail recusion - t
todd.ginsberg
12/24/2017, 2:26 PMTotally. - t
todd.ginsberg
12/24/2017, 2:28 PMYou could also make your function a bit cleaner to call with default arguments. 🙂 - a
andyb
12/24/2017, 2:28 PMAh, you are right. The function is not tail recursive, hadn't noticed the warning 😢 - t
todd.ginsberg
12/24/2017, 2:30 PMMeh. Like you said, we don't need it in this case. I was worried the second part would be some way to permute the inputs so we're building super long splindly bridges and would end up needing it. - t
todd.ginsberg
12/24/2017, 2:31 PMOne of the things I want to do is blog about what I learned during this AoC. One of them is that requirements change and it's almost always not worth anticipating how (for me). - a
andyb
12/24/2017, 2:33 PMI always have a look at the Stats page before attempting the AoC problem as it often shows if Part 2 is much more difficult than Part 1. Today there were v few people who had solved Part 1 & not Part 2 and so figured that it would be pretty straightforward - t
todd.ginsberg
12/24/2017, 2:45 PMHuh, that's clever. - t
todd.ginsberg
12/24/2017, 2:47 PMIf I wake up in the middle of the night, I'll glance at the problem on my phone (it comes out at 23:00 my time) and usually wake up with a way I want to do it. Usually I know right away how I want to do it. Like today, I knew it was a variant on the change counting problem, so it would be recursive. - m
menegatti
12/25/2017, 10:35 AMToday’s an easy one to finish off… what a ride - m
menegatti
12/25/2017, 10:36 AMGlad to have shared this with you guys 😄 Enjoy the holidays