« first day (80 days earlier)      last day (497 days later) » 
00:00 - 18:0018:00 - 00:00

12:14 AM
@MDXF Yeah, that is the question. lol
 
I'm gonna try to get LOLCODE working :P
Waiting for my cloud computer to boot so I can start messing with it
nvm LCI doesn't like unrecognized tokens
Do we already have Whitespace?
 
1:07 AM
yes
 
1:18 AM
Does C1R count as a language?
 
2:05 AM
@WheatWizard
 
2:29 AM
@MDXF yes but accessing the internet is a loophole
 
Oh darn :/
 
 
11 hours later…
1:30 PM
@MDXF For what it's worth, this is the list of languages I've been looking at. It's either stuff that someone mentioned in chat/in their post, or things I stumbled across that seemed like they might work, though I haven't spent too long on any of them. pastebin.com/53VAg3t0
 
1:41 PM
@SnoringFrog Grass has already been done btw
Might want to remove it
 
2:04 PM
Thanks. Weirdly, I would've noticed that in a moment, it's one of the things my most recent attempt to add Commercial just broke lol
 
2:36 PM
Looks like our TIO link got changed to the in-progress Fortran attempt. Fixed it now.
 
3:06 PM
@SnoringFrog Whats going on with the TIO link for rk-lang?
 
The link we had pointed to an in-progress answer 70, for some reason. So it was broken and had some Fortran in it
 
Are you adding a new link?
 
Already did
 
Oh ok, I didn't see that
Did you?
I still don't see it
 
"edited 35 minutes ago by SnoringFrog"
 
3:13 PM
Ah its under verification
 
Weird. Thought I had enough rep to avoid that now.
On another note, I've got an answer working in everything but Grass, but I have no clue how to fix Grass.
https://tio.run/##zTrLcttIkufGL8ylDLINgCRepEhLAklJtqVZd9strSU/wiQtF8EiCQsPNgCKlGU69robe5j7XPY4/zHX/Yr5hz33ZlaBL5GSuyNmIxaSgAIqKzMrX5WVpS5Nhr/9lsOLnEX@zcCPUpKyJCW92LtmsUGe3pBD6iXVckWSOFQc9cauFw5IOmTEC0fjlPQ9nyWSS1PSJKMMi4ENkiMxoz0BJvWjmOAQElz5xCPuaES6Ph0aLvE9DptEAfanLB7FDO4JCRnrkWTEXK/vuZwOYdOUhYkXhYlBPPeKTDzfJz3mw4AlcbcEyMg4WfkkSBlSL5IIXO7oDquLtzwDmJDNp8tGNObTjciIz30@60lMRyNgUsqRc2A89QKWEC9VEsJo4qEMYUjCwh6X1Ggh3YjQcHWW5NqjhAK/45jjTlicwvQMjpcMWcwAJUXig5gGRJ3EXgoiQEEekTNGY3IRM6YBFVAAS1w6A
 
I have no idea how Grass works
 
well crap
 
Ok well thats not entirely true
 
I know it uses w, and I have to have 3 of those in my answer
 
3:17 PM
It also uses v and W
 
those I don't use at all
 
Ok
Can you make your ws uppercase?
 
nope
 
Commercial (esolangs.org/wiki/Commercial) is annoyingly case-sensitive
 
3:19 PM
I think this can be fixed
Can you add other letters inside your code, e.g. as variable names?
 
In a limited way.
 
https://tio.run/##zTrLcttIkufGL/SlDLINQCRepEhLAklJtqVZd9strSU/wiQtF8EiCQuvBkCRskzHXndjD3Ofyx7nP@a6XzE/0ptZBT4kUnJ3xG7EQhJQQGVlZuWrsrLUo@no998LeJHTyL8e@lFGMpZmpJ94VywxyNNrckC9tFapShKHSqL@2PXCIclGjHhhPM7IwPNZKrk0Iy0S51gMbJACSRjtCzBpECUEh5Dg0iceceOY9Hw6Mlziexw2jQLsz1gSJwzuKQkZ65M0Zq438FxOh7BpxsLUi8LUIJ57SSae75M@82HAkrhbBmRknK58EqQMqR9JBC43vsPq4q3IACZk8@mymCZ8uhGJ@dzns54kNI6BSalAzoDxzAtYSrxMSQmjqYcyhCEpC/tcUvFCuhGh4eosyZVHCQV@xwnHnbIkg@kZHC8ZsYQBSorEhwkNiDpJvAxEgII8JKeMJuQ8YUwDKqAAlro0Bj5ukQRiEVLIcCIGV1SjoRydHCugMDHCoHEmFZLq@b/@LFFiNEnxwoG/piw70tGvz8lNsYNt4AKQgVLjRKUaaX4jqdnpMPib2j1zmMwkwCqBVHx
^ fixed grass
But I think that rk was already broken
so you might want to check that out
No guarantees about Incident either
 
Is that the right rk interpreter? I didn't know it was on tio yet
 
Dennis just added it a couple of minutes ago
@MDXF It looks like rk prints some unprintables, have you checked a hexdump of your answer?
 
@WheatWizard rk automatically prints an extra space after every print: statement. It's unavoidable, as are the >>'s from Deadfish~.
 
3:28 PM
Ok
 
How do we get the test driver to ignore it?
 
I don't know I'm a test driver noob
 
I'm not sure there's an easy way to
 
Oh I think I know
 
but that'd explain why the initial rk answer also fails lol
 
3:31 PM
I'll include in my answer that rk auto-prints a space after printing integers
(I just realized, it doesn't do so after printing strings)
Actually we might be able to golf the rk-lang
 
I have a way to fix
 
rk:start print: "69" rk:end will work
 
Hm...we ran into that with another language and I don't think it was allowed, because that caused me to go on a crazy hunt through the depths of the internet for an interpreter that didn't follow int outputs with a space
for whichever lang that was; but if we golf around it anyways that works
 
/opt/rk/rk polyglot.poly | python -c 'print raw_input()[:-1]'
Or golfier
/opt/rk/rk polyglot.poly|python -c'print raw_input()[:-1]'
 
ok that works for rk, now I have to fix alphuck and incident - _ -
 
3:35 PM
@SnoringFrog What do you mean?
Did I break it?
 
the os variable name was needed for both
 
Oh yeah never mind
Never mind about the print: "69"
it really breaks stuff badly
 
but I think we have to avoid the excess space if it's possible, which it is
if we don't, then I can go trim like 20 bytes out of one of these langs...
 
Ok my fix does not work on TIO
it works locally but TIO is strange
 
@ais523 @Chance do you know how the rules work with an unavoidable space that's only after int prints? I know it came up for us before
 
3:54 PM
Hmm... any way to add PHP?
It uses # as comments
Also there's a programming language called what that only uses groups of three question marks and exclamation points (e.g. !?!, ??!, etc)
I forgot where the repo is, I'll search for it
 
4:10 PM
@snoringfrog it came up with the befunges and minkolang. But we output the number as a string which didn't have trailing spaces. I would argue that if it is possible to output as a string without a trailing space, that should be required. Otherwise, if it is truly impossible to output the right value without a trailing space, it's fine.
 
Ugh, so much activity
 
That's what I thought. And that means our current RK answer is invalidated
 
No it doesn't...
Oh wait - you mean that if it's possible to print it without the trailing space (e.g. rk:start print "69" rk:end) it must be done that way?
 
yep
 
Crap
I can fix the interpreter, it's a bug...
 
4:13 PM
yeah, and that's making an Incident mess for me right now. Namely bc it ends up making `"6` and `"3" tokens

and no, that'd make the interpreter newer than the challenge
 
@MDXF ais523 had an answer invalidated for a similar reason
 
@SnoringFrog Well, we could stick "63" somewhere else in the rk code and it won't care
 
Should 4" and "6 both be able to be tokens? I thought that would count as an overlap
 
@SnoringFrog Are you working on making the updated rk work or should I be doing that?
 
You should be working on one too
Mine will work, but it also adds a new language. Not sure if just my rk changes would work right in your answer
 
4:27 PM
Sounds good
 
4:41 PM
Okay, I fixed mine
Updating answer...
 
I almost have mine
broke Moorhens again though
 
Could you test mine with Moorhens?
https://tio.run/##zTrLcttIkufGL8ylDLINgCRepEhLAklJtqVZd9strSU/wiQtl8AiCQuvBkCRskzHXndjD3Ofyx7nP@a6XzH/sOfezCrwIZGSuyNmIxaSgAIqKzMrX5WVpQuajn77rYAXOYn866EfZSRjaUb6iXfFEoM8vSb71Evr1Zokcagk6o9dLxySbMSIF8bjjAw8n6WSSzPSJnGOxcAGKZCE0b4AkwZRQnAICS594hE3jsmFT0eGS3yPw6ZRgP0ZS@KEwT0lIWN9ksbM9Qaey@kQNs1YmHpRmBrEcy/JxPN90mc@DFgSdyuAjIzTlU@ClCH1I4nA5cZ3WF28FRnAhGw@XRbThE83IjGf@3zWk4TGMTApFcgpMJ55AUuJlykpYTT1UIYwJGVhn0sqXkg3IjRcnSW58iihwO844bhTlmQwPYPjJSOWMEBJkfgwoQFRJ4mXgQhQkAfkhNGEnCWMaUAFFMBSl8bAxy2SQCxCChlOxOCKajaVw@MjBRQmRhg0zqRCUjv7158kSowWKZ478NeSZUc6/OU5uSl2sQ1cADJQapyoVCOtbyQ1u10Gf1P7whwmMwmwSiA
 
works in Moorhens
 
Thanks
Fixed
8
A: Add a language to a polyglot

MD XF69. rk-lang, 1750 1748 1737 1734 1733 1744 bytes #16 "(}o+?23!@)-("//*\Dv;'[af2.q]PkPPX'#CO)"14";n4 #/*0|7//```"` [-'v][!(>77*,;68*,@;'1,@10␉␉11)(22)S␉␉(1 P''53'S^'q #>␉ # 36!@␉` # #_>++++.>++++++::@-----x-.+? #`<` #<]}} +<[<.>>-]>[ #{ #x} # #=x<R+++++[D>+++++++59xL+++<-][pPLEASE,2<-#2DO,2SUB#1<

 
4:57 PM
@MDXF Why do you still have int os = 69?
 
It balances out Alphuck
 
why don't you just remove everything except the alphuck
 
I might be able to remove the int though
Yeah
 
you'll need the int, I believe, for incident
os is used for Incident, alphuck, and Moorhens
 
os is not used by moorhens
ax deflects the ip and prevents os from being run
 
4:58 PM
yes it is, I added it fornhat
I removed the ax
Unless that's back again now
os is a reverse as well
 
You had to add a bunch of xis then?
Ah I see
 
Okay so I won't change anything :P
 
You can certainly remove the non letter characters
e.g. | int os = 69 -> |int os
 
Trying that...
 
also there's not a ton of point to golfing that answer as my answer is already done
 
5:01 PM
@SnoringFrog it's an overlap only if 4"6 appears somewhere in the source
i.e. conceptually being able to overlap isn't enough, there has to be a specific point in the code you can point to and say "see, that's an overlap"
 
@MDXF you should add python and rk to your driver
 
Ah, that makes sense.
 
0
A: Add a language to a polyglot

SnoringFrog70. Commercial, 1802 bytes #16 "(}o+?23!@)-("//*\Dv;'[af2.q]PkPPX'#CO)"14";n4 #/*0|7//```"` [-'v][!(>77*,;68*,@;'1,@10␉␉11)(22)S␉␉(1 P''53'S^'q #>␉ # 36!@␉` # #_>++++.>++++++::@-----x-.+? #`<` #<]}} +<[<.>>-]>[ #{ #x} # #=x<R+++++[D>+++++++59xL+++<-][pPLEASE,2<-#2DO,2SUB#1<-#52DO,2SUB#2<-#32D...

 
@SnoringFrog someone edited the question to say that trailing newlines were acceptable, anything else isn't
that wasn't me, it was actually a voted edit (which are really rare on PPCG because most people have enough rep to be able to edit directly)
 
@ais523 I told him that, because it turned out the Surface answer outputted a null byte when run on windows
 
5:08 PM
given that it's my question I can unilaterally reverse a voted edit on it
but I'm disinclined to unless people strongly feel the voting was wrong
(also, many answers were outputting stray NULs for ages)
(which is why I embellished the polyglot driver to detect them)
we should probably put Commercial on TIO? we have Alphuck there, after all
 
We've already asked Dennis for two languages today. I'm not saying we shouldn't I'm mentioning that.
 
Yeah, I plan on asking, was just more focused on the write up
gotta go pretty soon
 
well, if we aren't diligent about this, some day we may end up with more languages in the polyglot than there are on TIO :-P
 
I think it should be added eventually, languages that are not on TIO are a pain.
 
lol
yeah, about that you two. :P
 
5:11 PM
Have we asked Dennis to put Deadfish~ on?
Or Moorhens, or Surface, or Incident, or any of those?
 
I don't want Moorhens on TIO
Surface needs Wine
Incident requires a library TIO doesn't have
 
Why don't you want Moorhens?
@WheatWizard Done, it has the latest version of the script now
Also, I'm going to add what
 
@MDXF Moorhens is very buggy, I would like to fix all the bugs before adding it to TIO, in which case it wouldn't be very useful because we would have to test it locally anyway to make sure it does not encounter a bug.
 
@WheatWizard Ah okay
I didn't realize it was your language
 
Ah yes, it is. Its not a very good language but it is mine
@MDXF Wise is going to have a fit.
 
5:17 PM
What's the shortest way to print 71 in brainfuck?
 
Currently ! and ? are buggy and do awful things.
 
Crap
Dammit, wise...
 
You could try
its not impossible
its just !!FUN!!
 
Yeah ik
 
@MDXF Esolang keeps a guide to the shortest known way to calculate each individual number, but if you're printing 71 you'll have to do so as a string
and optimizing strings in BF is very hard, even if they're only two bytes long
 
5:19 PM
@ais523 Where is that page? Ive been looking for it for a while
 
@WheatWizard The Alphuck interpreter in the test driver is newer than this challenge. It is considered a test abstraction for the intended interpreter which predates this challenge. It's not perfect, but it is better than not being in Tio. Moorhens could probably fall into this same line of reasoning, if you were so inclined to put a version on Tio.
 
Looks like this prints 71 in what
!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!??!????!????????????????????!
 
@Chance I want to build a Huristic test into the driver sometime, once I've done that I'll fix all the bgs and ask Dennis to add to TIO.
 
note that that guide aims for portability (in particular, it doesn't move the pointer left of its starting position), although it does contain special cases for 8-bit cells
so you can often golf off a leading < if you can make assumptions about the interpreter, which you typically can on PPCG
 
5:23 PM
@MDXF -[>+<-----]>--.++++++. can be made into a shorter one
 
@WheatWizard Prints 17 in what
 
Oh I got it backwards
 
Well, there are no three-character combinations of ! and ? anywhere else in the current code, so that's good
 
@MDXF -[++>+[+<]>]>-.------. same length but 71
 
@WheatWizard Yep, thanks
????!?!??!??!!!!???!?!??!!?!?!!!!!?!!!!?????!????????????????????!
what a weird language (*ba dum crash*)
 
5:25 PM
Would any of the languages choke in an unfixable way with a line that started Something like >
 
@Potato44 no, although you'd have to be very careful in where you placed it
 
> main = print "70" --
 
we have a "window" whose purpose is to allow lines with awkward syntax like that to be placed safely
 
@Potato44 Also, if you're working on a submission, I'd hold off, I'm working on adding What
 
> is also a hard character to have unmatched, you'd need to somehow safeguard a leading < on a previous line
 
5:26 PM
Not working on anything just throwing ideas around
 
btw, if you're thinking about Literate Haskell, it was previously inconsistent with Thutu
however, now that Thutu parserblock technology has improved, Literate Haskell is quite possibly going to be addable
 
Oh, okay, that's exactly what I was thinking
 
Retina doesn't like What, neither does Stringie, nor does trigger. but wise doesn't care
 
there are other literate languages too, some of which use the same syntax (leading >)
 
what do Retina and Stringie and Trigger care about?
(*ba dum crash*)
 
5:28 PM
@MDXF Where are you adding it
 
@MDXF Retina doesn't like ???, but it uses alternating parsers for alternating lines
 
@WheatWizard I tried dumping it at the end again
 
Retina cares about even or odd lines I can't remember which
 
so moving your addition up or down a line will probably help; one of its two parsers is fine with it
 
I didn't add any lines
 
5:28 PM
@MDXF The end is not a great place IMO
 
Where in the program does Thiruvananthapuram get parser blocked?
 
@WheatWizard Ah okay I'll move it
 
I like adding things to the really long python line
 
Where should I move it?
 
Underload (implemented by stringie) will hate all those unbalanced !; your best solution there is to enclose the whole thing in parentheses
 
5:29 PM
@ais523 Hm okay
 
@Potato44 this is the thutu pare block line [ "]56p26q[puts 59][exit]" ,'['];#/s\\/;print"24";exit}}__DATA__/
 
Whoops really bad typo
Damn Google keyboard that I'm not used to
 
Trigger cares about having the same character 2 or 3 times in a row; however, it also has a jump instruction that can be used to jump over awkward code
and it's currently used to jump over the bulk of the program, so moving your line somewhere in the middle seems like the best option there
 
Oh wow I fixed everything just by moving it
 
Although the more you move it towards the front the more of an issue Wise is going to have
 
5:30 PM
you probably got it inside a pair of parens then :-)
 
Ugh but I think I detokenized something :(
@ais523 Yep
 
@chance so after that block is fine for thutu?
 
@Potato44 after that line the thutu compiler isn't even running, that line crashes the compiler
and it never gets to read in the rest of the source
14
Q: If code fails to compile, but runs anyway, does that count as executing without error?

ais523Here's a situation that came up recently in chat, and I realised I'm not sure what our rules are. A few challenges require code to run without crashing/erroring. As such, a runtime error would make the answer ineligible. For example, here's a requirement from the "add a language to a polyglot" c...

luckily, ^ that happens
 
Ok, I think I got the write up finished.
 
(admittedly, we contrived it so that that would happen, but it's still amazing)
 
5:32 PM
Well, no promises since I'm new around here, but I might see if I can get literate Haskell in during the next few days
 
@ais523 @WheatWizard, can one of you double check Incident for me? Manual verification seems solid but I want to be sure
 
@SnoringFrog sure, paste a link-to-TIO or some other method that will let me get at ungarbled source
 
Just so I know, besides brainfuck which languages off the top of you guys head care about >?
 
@SnoringFrog works
@Potato44 Brain-Flak, Brain-Flak Classic, Wise
 
5:34 PM
Okay now all I have to do is make sure the what?!? code starts on a 3rd character
 
brain-flak is easy to screw up in anything that cares about <>()[]{}
 
Brain-Flak and Brian-Flak Classic require that parens are balanced throghout the program, but ignored anything after a #
 
<> specifically will also disrupt Perl 6, but that should be fairly easy to fix
 
Ahh crap evil is getting angry at me
It prints RP instead of 42 wtf?
 
evil cares about lowercase letters, I think?
 
5:36 PM
but...but... I didn't add any
 
however it also jumps over much of the code
RP has the same relative ASCII code pattern as 42 so I'm guessing some internal counter got added to
 
Oh it's not my fault
 
oh whoops
 
@SnoringFrog Your answer prints RP in evil instead of 42
 
yeah that was my bad
working on it
 
5:37 PM
Let me know when it's fixed
 
right, I'll only Incident-check the most recent version
 
@Potato44 SMBF, and perl6 come to mind. But there are spaces where BF and SMBF won't read the >. Just prior to the commercial answer is probably good.
 
Incident is working
 
and might not be able to get to it immediately, I'm going to have to go out to buy food soon
 
I can do incident checks
 
5:40 PM
Can't we just continue to manually balance out the tokens?
 
We should always check
 
Okay
 
the most common mistake when manually balancing tokens is accidentally detokenizing something in the middle of the Incident code
and missing it because the tokens around the edge are still balanced
 
@ais523 I found something interesting with gfortran. While fortran doesn't impliment a block comment, I suspect the preprocessor reads /* block comments .
 
What I do is I always have a Windows Notepad window open with the working Incident tokens pasted into it, then I copy-paste the new tokens under it and see if they're the same
and if they're not the same I screw around with it until they are
 
5:42 PM
https://tio.run/##zVjrUttIFv4dPUXHdmLJlmzZDEkGY4MngSlmkoUN5FIxTmiktt1Bt9EFCwipfYT9vy8w7zF/9yn2RbJfd0u2gcxktmq3agXGre7T53zn0uec5pQmsy9fquIhB6F3MfXClKQsSYkb83MWt8gPF2Sb8mS9u6ZpkioO3czhwZSkM0Z4EGUpmXCPJZpDUzIgUcGlJQakSmJGXUWmTcKYiC3EP/MIJ04UkVOPzloO8Thod79fB30S@oImZXEUM/xNSMCYS5KIOXzCHSmLsDxlQcLDIGkR7pyROfc84jIPG5YAHBPMSJasTClxLc0NNYLHiW7BXbzVGGgCVqrMIhpLlUMSSf1LzecxjSKA1KrkEMBT7rOE8LSeEEYTLuyILQkLXGmtaGHhkNBgVUtyzimhwJvFknfC4hTqtSRfMmMxA0sqhE9j6hN9HvMUJhDGHJIDRmNyFDNmQAqcwBKHRsBxQySEhUJCKhRpSWdtbtZ39nfrcJra0aJRqlXjtaO//qRR0uqT2ocePv1Kpaft/OUZuaodizFQgBkcG8U6NUj/M0nax8cMn7xz2p7G1xq4arCKR9phlLb
the debug on that link mentions an unterminated comment.
 
if I had to guess, that's a consequence of using a capital F in the filename
 
lol there are so many warnings in the debug section
 
gcc (the driver) often interprets a capitalized letter as a request to run the C preprocessor on the file before compiling it
 
yeah, that was my intent.
 
@SnoringFrog how's it coming?
 
5:45 PM
That's my strategy to skip over most of the lines.
 
Should I ask Dennis to add what? it's not my language so idk if it's my job
 
you can ask for other people's languages
the main reason why adding a language to TIO would be a problem would be if the interpreter is somehow problematic
 
Hm, maybe I'll fork and patch up some stuff in the interpreter
 
You might want to just imbedd the what interpreter into the driver
i.e. convert to brainfuck and run the brainfuck
 
I need a 3 char Moorhens token for pop that doesn't include m/s/p or a Moorhens token for roll that includes an f and doesn't have any dangerous-to-evil characters before the f
 
5:53 PM
Ok I have a token finder
 
currently checking every 3 char word in Moorhens dict one by one...
oh thank goodness
 
I'll see if I can find one
 
should have asked before I checked all the 2 char ones
 
@WheatWizard why do you keep removing stuff?
 
Here's a list for category 1:
are
awe
bal
ban
bay
bow
bra
cub
due
eld
ell
fag
fid
fil
fin
fir
fog
ged
ghi
gib
gin
ha
hat
hex
hoy
hun
jar
ken
kif
lit
lox
nah
net
nib
ole
oot
oxy
rag
rah
rub
rut
tie
vee
vig
way
wen
woo
wry
yob
Here's a list for category 2:
fen
fez
 
5:59 PM
Got it. Fir worked. Thanks!
 
@MDXF I misunderstood the question a couple of times
 
00:00 - 18:0018:00 - 00:00

« first day (80 days earlier)      last day (497 days later) »