« first day (299 days earlier)      last day (2195 days later) » 

12:22 AM
@WheatWizard 99% sure the results of compilation are not cached, the cache system doesn't even know what Lang something is in, if I understood Dennis' explanation earlier.
 
Ok rip
 
@WheatWizard there are fields for both compiler args and program args. Compiler args are near the top in a drawer called "Compiler Options" for most compiled languages.
 
Ah nice
 
Languages like perl and ruby have runtime options for flags like -n
Java has runtime and compiler and program args
 
12:52 AM
@HyperNeutrino Sure, copy aways.
@WheatWizard No, caching uses a hash of the entire request. No state is maintained between executions, among other reasons because there's no guarantee that the next request will be run on the same arena server.
 
Makes sense, but unfortunate
 
What are you trying to do?
 
I'm writing a faster version of BrainHack but the compilation is very slow
 
user165474
@Dennis Yay, thanks :D
 
I'm basically shoving all of the heavy lifting to compile time in stead of run time
 
12:58 AM
@WheatWizard Not sure we're on the same page. The brainhack executable is only compiled once. I was talking about compiled languages, not languages with compiled interpreters.
 
Oh sorry, I'm writing a BrainFlak compiler in Haskell
BrainHack is interpreter rn, but I'm reworking as a compiler
should compile from Brainflak to Haskell
 
Right, we were on the same page then. :)
Multi-part requests will.make this possible. For now, if you want to run the same compiled program for different inputs, your best bet is a test suite in Bash.
 
1:36 AM
@MDXF Any chance you could print the extra information (Notepad: X and the cube) to STDERR?
 
@Dennis Ah, yes. I fixed that in the fake TIO version but not the actual one, haha. Let me fix my Hello World first...
 
user165474
@Dennis Whenever you have time, could you add Anyfix to TIO? The github link is here: github.com/alexander-liao/anyfix
 
user165474
Wait hold on sorry that won't work because the program is taken as STDIN ಠ_ಠ
 
ಠ_ಠ indeed. ;)
 
user165474
:P Forgot to change that after finishing making it functional :P
 
user165474
1:40 AM
Not sure it's TC actually... I'd need help proving / disproving that
 
That's not important for TIO. Or anything else, really.
 
user165474
Yeah. Never mind that should be in TNB
 
I never understood the hype anyway. brainfuck is Turing complete, but isn't. Any classification that implies that brainfuck is more powerful than C is pretty much useless...
 
@Dennis Fixed. (Cubically interpreter and Hello World)
 
1:47 AM
@Dennis I can do tests on my personal computer, I just wanted to know if the compiler would speed up things on TIO.
 
user165474
@Dennis Should work now! First argument is the program, and then all other arguments are pushed onto the stack.
 
@HyperNeutrino Alright, I'll take a look later.
 
user165474
Thanks! :)
 
2:31 AM
@HyperNeutrino Do you have a Hello World program? Also, you don't seem to have added your .gitignore in time. You'd have to explicitly remove the contents of __pycache__ to scrub it from your repo.
 
user165474
2:44 AM
@Dennis It's rather boring; just “Hello, World!”. And yeah, I will go do that, gimme a sec. Thanks for pointing that out :)
 
user165474
Yup, got it. poof
 
user165474
@Dennis It should be good for uploading now; any other issues?
 
user165474
Anyway I'm going to sleep now so if there are any really bad issues, ping me and I'll fix them when I get to it tomorrow morning. Thanks! :)
 
o/
 
3:05 AM
@HyperNeutrino Can't seem to make it work.
Traceback (most recent call last):
  File "/opt/anyfix/lang.py", line 375, in <module>
    interpreter.next()
  File "/opt/anyfix/lang.py", line 356, in next
    self.update()
  File "/opt/anyfix/lang.py", line 252, in update
    self.push(function(*arguments))
TypeError: <lambda>() takes 0 positional arguments but 2 were given
Oh, code comes from the first CLAs I gave it a filename...
 
 
2 hours later…
4:50 AM
@Dennis Can you pull Charcoal?
 
@ASCII-only Done.
 
5:13 AM
@Dennis Could you pull Charcoal again?
 
 
1 hour later…
6:20 AM
@ASCII-only Done.
 
6:46 AM
Can you add Check to TIO?
 
7:31 AM
@Dennis Could you pull Charcoal again? (Sorry, too many bugs :P)
 
7:42 AM
@ASCII-only DOne.
@Challenger5 I'll take a look tomorrow.
 
Okx
8:17 AM
@Dennis Can you update Fireball?
 
Okx
9:08 AM
Additionally, why is the Hello, World! test for Assembly-as so complicated? You can just use .print "Hello, World!"
 
 
2 hours later…
11:30 AM
@Dennis Could you pull Charcoal? Thanks
 
 
2 hours later…
1:01 PM
@ASCII-only Done.
@Okx Done.
 
user165474
2:23 PM
@Dennis Perfect! Thank you very much! :)
 
5:57 PM
@Dennis Thanks!
 
 
1 hour later…
7:10 PM
@Dennis Could you pull and rebuild Cubically? If you run make tio it will generate a new executable rubiks-tio that is actually compatible with TryItOnline
Because right now the debugging output looks like terrible due to nonworking ANSI codes
 
Not at my computer right now; I'll pull it later. I can't see the control characters myself, but that probably varies from browser to browser.
 
Oh okay, sounds good
 
 
2 hours later…
9:01 PM
how does TIO terminate a program after 60 seconds? In particular, can the interpreter react to the termination? It strikes me that several languages could be made more usable on TIO by flushing things like stdout buffers upon timeout
 
@ais523 It sends SigInt, and then after 5 more seconds it's nuked completely.
 
oh, perfect, I was hoping it would be something like that :-)
 
9:24 PM
@ais523 It also uses stdbuf -o0 -e0, so unless the interpreyer itself is buffering, all output produced in the first 60 seconds should be received by the browser.
 
 
2 hours later…
11:21 PM
@Dennis Can you pull Charcoal again?
Not that urgent because I just realized I need to fix another bug :(
 
In a while. Still not at a computer...
 

« first day (299 days earlier)      last day (2195 days later) »