« first day (664 days earlier)      last day (731 days later) » 

12:47
@scottinet hahaha sorry about that command, that was something I implemented a long time ago but I have never actually used it
If you want, you can remove that command, because right now it's useless
@Adnan then I will remove one and replace the other with round_down(a) (math.floor)
okay
btw I just finished a first pass !
I'm doing a second one to make sure I've added all the needed test cases to test my changes (sadly, I did not), then I'll need to polish things up here and there and the PR should quickly follow
Great! Let me know when you've made the PR and I'll review it right away :)
 
3 hours later…
15:54
@Adnan PR ready: https://github.com/Adriandmen/05AB1E/pull/103
I'll add the list of exhaustive changes, command by command, hopefully tomorrow. This is especially important since some commands have now extended functionalities due to 05AB1E being more weakly typed, and it would be wise if you made sure that the choices I made match what you think is best for the language
 
5 hours later…
20:47
DgsO/ = Arithmetic Mean can we get a 2-byte for mean in the list-commands?
@MagicOctopusUrn ÅA Does Arithmetic mean
21:11
@Riley as soon as I typed "list-commands" I realized I hadn't checked the list-commands lol.
B/c they changed the name of list-commands.py awhile back I often forget they even exist.
Is there an "interesction of a and b"?
21:26
Not that I know of, but ʒå} should work.
Wait, I don't think that works
@Riley I am literally tearing my hair out ahahaha.
It's like 9 bytes for intersection hahaha.
Pairwise and "a in b" doesn't want to work.
Which, admittedly it shouldnt.
ü‚vy0èy1èå} can't be the best though.
shouldn't it work if you deep flatten both lists before using å?
I'm trying this on a 2D list.
I'm trying to find "whether or not there is a chain from element 1 to element n".
So basically a consecutive intersect on adjacent elements.
(which is slightly more confusing than a single intersection).
codegolf.stackexchange.com/questions/138790/… was the question I was working on
I was thinking "use pairwise to group adjacent elements, then iterate through them, performing intersect on the factors of each pair.
Got weird fast (even completely ignoring the restricted source).
@scottinet Thanks! I took a look at the PR, it looks great :)
What does å do if the first argument is a list? I can only get it to return 0 or [0]
21:37
I'll try to do the review tomorrow, but it can take several hours haha :p
@Riley Even confuses me sometimes.
@MagicOctopusUrn à should work
@Adnan Oh god, what obvious command did I miss this time.
hahaha :p
FRICK!!!! a.keep(b) has been there FOREVER.
God I'm so bad sometimes -_-
21:39
I forgot about that too...
Which register is initialized at 0?
R is -1, X is 1, Y is 2
the counter variable
½¼¾ one of these haha
can't remember which one
¾ is the one, the increment returns a 1?
Which is weird.
Does it automatically return the counter provided the last instruction was an increment?
21:45
it prints the counter variable if nothing else has been printed
¼¼¼=3 ¼=1, so I'll assume so, interesting.
and if nothing is on the stack (but I'm not sure on this one)
All of the odd functionalities like this are so cool. I love when I write something in 05AB1E and it works 5 bytes before I think it should but I don't know why.
@Adnan I don't think mean vectorizes correctly.
Yeah, I forgot to ask Dennis to sync it on TIO
Try it online! - 1 mean per element in the list (all the correct means).
21:48
but the offline version should work
if I add uniquify, it works.
Ahhh, you just fixed it then?
I fixed it 17 days ago
@Adnan can TIO automatically pull a GIT repo's commits to update?
Like a docker server?
Errr not docker...
Frig, what's that called.
@MagicOctopusUrn I'm not entirely sure what the mechanism is behind TIO
I believe that the repos are pulled automatically to a different server and tested whether they compile or not, with a manual sync to the main server
but I could be easily wrong on that
22:04
@Adnan that sounds smarter, even with our Jenkins (I think that's the program) we still have to approve deployments, even though it compiles our server code and pushes the WAR from a git repo, we still have to manually deploy and read/approve the cimpilation logs.
22:16
Well, they are tested as well: github.com/TryItOnline/tryitonline/blob/master/usr/share/…, which is a really nice detail
I believe those tests are run after each pull in order to prevent compilation errors

« first day (664 days earlier)      last day (731 days later) »