« first day (2122 days earlier)      last day (2738 days later) » 

11:02 AM
Is anyone on?
 
Can we please close this question: codegolf.stackexchange.com/questions/100538/…
 
11:25 AM
I did the final close vote.
 
Times (and docs) have changed:
 
11:38 AM
oh well, I just gave a question eight answers before anyone else had given any
 
I'm in the process of moving over to a new laptop, and while I want to continue with C++, I don't like using visual studio, it just feels clumsy and slow. Anyone know any lightweight IDEs for C++?
 
C++ IDEs tend not to be that lightweight, it takes a lot of effort to integrate the IDE
 
@Flp.Tkc vim + gcc + gdb
 
if you merely want to develop C++, you could use a programming editor of your choice + a command-line compiler of your choice; that's not the same thing as an IDE
but it's sufficient to write programs
 
11:39 AM
7 hours ago, by Doorknob
@ais523 Unrelated, but do you happen to be the ais523 who is now on the NetHack devteam?
:O
you work on NH
 
my favoured choice is mingw (installed via Strawberry Perl) plus Emacs, but that has a pretty steep learning curve; there are plenty of choices which work almost as well and are easier to learn
 
I'd prefer an IDE over command line tbh
 
@ais523 You typoed vim
 
thanks for the suggestions everyone! I'll take a look
 
@TuxCopter: I use a wide range of editors (I have three open right now), Emacs and vim are suited for different tasks though
 
11:41 AM
¯\_(ツ)_/¯ vim is pretty versatile
It's even used in this site
 
vim is a better at editing text once you're fully used to it; Emacs is better at everything you do in an editor which isn't editing text
 
What do you do in an editor that is not editing... ?
 
@TuxCopter: compiling programs and jumping to the errors; integrating with external syntax-checkers to get informed of syntax errors as soon as you type them; quickly jumping around between multiple files to follow the control flow (this is particularly useful with NetHack)
the basic difference between Emacs and vim is that Emacs has a much better understanding of what it is that you're editing, whereas vim just sees the bytes
btw, I'm not crazy enough to learn both to a high enough level to make this useful, but it seems widely accepted at this point that the "perfect editor" is to run a vim implementation inside Emacs, so you get vim behaviour for the editing and Emacs behaviour for everything else
 
evil-mode in emacs AFAIK
 
yes, that's the recommended method nowadays IIRC
 
11:46 AM
atom doesn't look too bad
 
I tried it. 7 seconds startup time, too slow for a text editor.
 
gcc looks better though
 
My new hobby: Finding Mathematica solutions written by newer golfers and then writing a ~40% shorter solution
 
@Flp.Tkc Yeah if you have a 12-core CPU and 3 TB of RAM, Atom is fine.
 
:)
 
11:47 AM
^^ Exactly
 
@mınxomaτ lol, if only my home PC had that much powah
 
Emacs gets round this problem by recommending you never close it ;-)
GNU Emacs it was actually originally intended as a desktop environment, sort-of like Gnome or KDE, but became best known for its editing facilities
 
When I code I have my text editor on the left and a term on the right, so I pratically never close my editor
 
When I code, it's a compiled language, so it makes no sense to have a term
 
wat
 
11:49 AM
That statement doesn't really make any sense
 
IDK, just use MonoDevelop for C#
 
same except my screen is smaller, so I either have the terminal behind Emacs while editing (my usual setup on Linux), or I have the shell as a buffer inside Emacs itself (my usual setup in Windows, whose default terminal sucks)
 
(of course, it opens a term for I/O while it's running)
 
I use SciTe everywhere, which automatically opens one term instance as a buffer.
 
I use IntelliJ for Java
 
11:51 AM
ConEmu + git bash = Linux console in Windows
 
@KritixiLithos I'm basic. I use eclipse for java
 
Pretty much what I'm looking at all day (looks the same on linux).
 
one of the things I've looked into heavily is terminal implementations (this is important for a terminal-based game like NetHack)
what ConEmu is doing is completely unsupported by the Windows APIs
 
hm?
 
11:53 AM
@Flp.Tkc IntelliJ has better code auto-completion
 
I looked into how it works and then wished I hadn't
 
@mınxomaτ XCode is better than that
 
@ais523 What does it do badly?
 
@KritixiLithos That's not comparable. First off, XCode is proprietary. Second SciTe is an editor, not an IDE
 
it's not so much to do with badly
 
11:54 AM
@KritixiLithos to be honest, auto-completion messes me up. I just never really got into using it. I'm sure if I wasn't so stubborn, I'd be coding much faster in IntelliJ
 
but that implementing a terminal isn't something Windows provides a sane way to do
so it has to do it an insane way instead
 
The only problem I ever had with ConEmu is WSL's bash making weird things with the colors
 
ConEmu doesn't support non-legacy console hosts (like WSL). It uses bodged client-windows to make it look like it's handling the terminal, but it isn't.
 
@mınxomaτ Again, XCode is an IDE and not just an editor
 
I just said that.
I don't need an IDE.
 
11:57 AM
That is why XCode is better
 
wat?
Sometimes you only need an editor, not an entire IDE
 
But XCode takes up a LOT of space
@TuxCopter Why?
 
Example: You need to modify a line in a file
You will not use VS for it
 
@Flp.Tkc Trust me, IntelliJ is worth it
 
Sure, an IDE that doesn't run on any other platform is clearly "better". Sorry, but I'd like to edit code in the same environment everywhere. Emacs, vim, SciTe can all run everywhere and only need a few MB of disk. XCode is way too bloated and unportable for any of my use cases.
 
11:59 AM
Fun fact: When I use vim, I enter insert mode, write whatever changes I need to do, exit insert mode, write to file, and exit
 
I hate Xcode
 
3 mins ago, by Kritixi Lithos
That is why XCode is better
WTF
 
I didn't say Xcode is the best
 
Kritixi = crazy confirmed. Case closed.
 
12:00 PM
I just said that it is better
 
gtg
 
I've used a lot of editors, some are really terrible
 
I tried deleting it, but I couldn't and I'm stuck with it. I just deleted Xcode
 
I was using sixth edition UNIX ed a few minutes ago, it's painful
 
12:01 PM
Ouch, just cringed reading that
 
then after I'd got used to not having a working backspace key
I came back onto PPCG chat and suddenly couldn't figure out how to fix my typos
 
-1
Q: Where Stenganography is using in real life?

IsrarIt's totally understandable that in world-war II Steganography was quit efficiently but the question is that in this era where we are using Steganography in our life? if anyone can explain that would be better.

 
huh, new post and it's already at -1? that was fast
 
That's nothing
 
do new posts normally get downvoted here?
 
12:04 PM
It's what happens when NewMainPosts takes 15 minutes to post
 
(also, that question clearly should be voted-to-close rather than downvoted)
 
@ais523 The question does indeed not show any research effort, which is the usual reason for a downvote.
 
oh right, downvotes speed up deletes, and a delete is highly likely to be appropriate here
the main confusion I have is as to how the question ended up on this particular stackexchange site
given that it isn't even about programming
 
@ais523 More importantly, unclosed questions can benefit from downvotes, as anything below -8 will vanish from the frontpage.
 
I don't want to vote to migrate because I can't immediately think of somewhere which would be a better fit
and it may well be offtopic everywhere
 
12:12 PM
Infosec's the most appropriate IMO, but I don't think they need a new question on it, so I've just left a comment
 
Nov 11 at 21:02, by mınxomaτ
There's a codegolf answer in the HN frontpage btw.
@Sp3000 Is there a crypto.SE?
 
Yes, apparently: link
 
> https://http//crypto.stackexchange.com/
 
(I always forget it exists)
(oops :P)
It's confusing when Chrome doesn't show the http:// extension
 
Then its should probably migrate there, it's fairly on topic. Probably a dupe there, though.
@Sp3000 Paste first, then look at it :P
 
12:19 PM
steganography isn't really crypto though, especially not the sort used during the wars
 
Hmm to be honest I find the Crypto/Infosec pair slightly confusing since there's a lot of overlap, and they both have steg tags
And the cryptocurrency Stacks and I really do wonder what goes where sometimes :P
 
In a nutshell, Crypto is more about theory, Infosec more about practice.
 
back
 
Many questions on infosec that can't be answered with one sentence using hard facts are way too speculative for my taste. It's borderline world-building, but for attack vectors.
 
12:25 PM
@Dennis wow, you actually followed through with the showcase thing.
 
@NewMainPosts I have absolutely no idea how he could think a site called 'Programming Puzzle and Code Golf' is about steganography/encryption/whatever
 
@DLosc My main concern was the lack of a tie-break, which has been added now. There were a few other things that seemed contradictory but I wasn't sure if it was just me being tired, so I thought it safest to close vote to draw attention and let other people decide what to change. Looking today I can see I simply misunderstood some parts. Yes I agree the spec is excellent.
The main problem with X without Y is defining what it means in different languages, but in this case a lot of effort has gone into making this objective, so this seems like one of the "good X without Y" questions that the meta post mentions.
As for looking very difficult, there are so many challenges that are "obviously impossible" until the answers start coming in...
 
I think the trick is to store code for one language in a string literal in the others, and then do a calculation whose result depends on the content of the literal in question
that way it can contribute to the result whilst still being fairly easy to change
however, trying to write the start and end of the code under those conditions is fairly difficult, I'm stuck at five languages as a result
/me reads the list
 
Wait, GNU C has a builtin to SIGILL?
 
Stuck at 5?? That would be the winner so far - you can just go ahead and post... :)
 
12:34 PM
@TuxCopter has
Please.
 
the programs aren't fully written yet, I've just reached the bit where it'd be easy
 
Sorry
 
I want to get the framework in place before working on the details
as that's the hard part
 
You don't see GNU C as plural?
 
I don't know when to use 'have' and when to use 'has' :/
 
12:36 PM
@TuxCopter he/she/it -> s
 
kthx
gtg
 
I see GNU C as singular, language names normally are
but it has a builtin to crash the program
it's not necessarily implemented in terms of an illegal instruction but that's the usual implementation, because it's small, fast, and does the job
and saving executable size is often important
 
0
A: Sandbox for Proposed Challenges

BlueEyedBeastProcedurally Generate a Dungeon popularity-contestgridgamegenerationmaze It's time to make the new Rogue! Given no input, output a (randomly) procedurally generated dungeon with a start and an end. It must be possible to get from the start of the dungeon to the end. Elements may include: (but...

 
12:50 PM
@ais523 main[]="/"; just errors for me (wide character array initialized from non-wide string), but {6} should work.
 
@Dennis: on a modern C compiler it does, but that answer is targeted against older OSes
main[]={6} segfaults on a remotely modern OS
so you need to compile for pre-C94, which would mean no L
I should probably clarify that
was thinking about too many things at once
 
@ais523 Not with the switches you provided.
 
like, the assumption is that you'd need to find the perfect OS to make it work, so it might well have the perfect C compiler too
{6} is the same length though
may I mention that in the answer? (alternatively, you could mention it in a comment)
 
Answer would be better imho.
main=6; seems to work as well.
 
now you're completely outgolfing me :-)
have you found an OS where it works, or is it just a guess?
 
12:58 PM
back
 
I edited the post, anyway
 
@ais523 With -Wl,-z,execstack, it works on my machine.
 
what, main=6?
or the final program?
 
main=6.
 
I think you get to claim a score of 6+15 there if you want it
because a) I didn't think of that, b) even if I did I wouldn't have expected it to work
that said, my 1 byte solution is going to be hard to beat
btw, is there a way to see what happened to recently expired bounties?
I'm often interested in which answer ended up getting the bounty
 
1:06 PM
If you have a particular challenge in mind, the bounty amount will be displayed next to the answer that received it. I'm not sure if there's a way to search for recently awarded bounties across all challenges.
 
right, I was hoping there would be
 
incidentally, it's really liberating to be able to do perl -p and the like at only a 1 byte penalty
on anagolf that was a 9 byte penalty, and even then it was still sometimes cheaper to take the penalty than work around the lack of the flag
so that's one thing I'm already enjoying about the site
 
What does the p flag do?
 
it puts the program in a loop that reads input a line at a time, and automatically prints output at the end of the loop
and the input goes into $_, a variable which is used by most commands implicitly
 
1:09 PM
Ok
 
so it lets you avoid worrying about (or more importantly, spending bytes on) writing I/O code
 
@ais523 BTW, the UD2 solution should also work for ICC for Windows.
 
You might be able to find/write a query on SEDE for bounties.
 
if your program has to take input, it's nearly always cheaper (which is why most Perl submissions here tend to eat at least a 1 byte penalty for -p or -n, which is an input-only version of -p)
I've never found a question for which -n is shorter than -p the way I write programs, but some other Perl golfers seem to like it
 
@ais523 Feel free to use that yourself. I just did what your answer told me. :P
@ais523 If you see a bountied question, you could favorite it to visit it later.
 
1:20 PM
I mentioned it in my answer but didn't claim it in the title, it seems like the best option (especially as none of the answers can win unless the PDP-11 answer is ruled invalid for some reason)
 
1:46 PM
@mınxomaτ Again thank you, just used it the first time today for something other than playing around=)
 
@flawr Multi-editing or SciTE?
 
2:01 PM
Multi-cursor editing=)
I'm not very used to SciTE yet, so when I want to quickly get something done I still use Npp which is what I am used ot
 
Oooh, multi-editing seems like it could be very useful in some situations.
 
Especially vertical multi-editing
 
This and knowing some regex can save you heaps of time:)
(And touch typing. I'm still amazed by the number of people working in buissnesses earning a good amount of money who don't know either.)
 
I wrote a LUA extension for SciTE to use full PCRE in the F&R dialog but lost the code :(
 
(Who have jobs that could be automated with a weeks worth of programming or even less.)
@mınxomaτ Is there no find and replace?
What is PCRE? Perl compatible regular expressions, found it.
 
2:13 PM
@flawr Of course there is. It's the same as in NPP
Since it comes from Scintilla :)
 
Ironically I cannot find it!
 
The F&R function
 
In NPP?
 
No in SciTE
Ah found it.
 
2:15 PM
Search->Replace
 
Why would anyone make [crtl]+[R] "revert"
 
Or CTRL-H (which also loads the current selection into replace)
 
Where can you change the shortcuts?
 
@flawr Global properties file (open as admin) or user properties (Options -> Open User Options)
E.g. keyText=Shift+F13
 
I opened both of them but cannot find those settings=/
 
2:21 PM
You need to add it. These are the built-ins: scintilla.org/CommandValues.html
User shortcuts can be defined like this:
# User defined key commands
user.shortcuts=\
Ctrl+Shift+V|IDM_PASTEANDDOWN|\
Ctrl+PageUp|IDM_PREVFILE|\
Ctrl+PageDown|IDM_NEXTFILE|\
KeypadPlus|IDM_EXPAND|\
KeypadMinus|IDM_BLOCK_COMMENT|\
Ctrl+F1|IDM_HELP_SCITE|
 
That's a point where a GUI would be very helpful for me.
@mınxomaτ thanks
 
0
Q: Japanese Mahjong Score Calculator

Link NgI'm new to Japanese Mahjong and have trouble calculating the scores. Please help me write a program to calculate it. Introduction (aka you can skip this boring part) Mahjong is a popular tile game in Asia. Different rules have evolved and systematically established in various regions. Japan, b...

 
2:37 PM
@flawr BTW, if you really need it, there are some options at scintilla.org/ScintillaRelated.html. But then again, NPP is the most GUI-y of all Scintilla-based editors.
 
Anyone know pyth here?
I'm getting
s_push: parser stack overflow
Traceback (most recent call last):
  File "pyth.py", line 752, in <module>
    exec(code_to_remove_tools + py_code_line, environment)
MemoryError
 
What are you trying to run dare I ask
 
Hi guys, what's going on ?
 
@Serg @BlueEyedBeast needs some help with Python
 
Pyth not python
 
2:43 PM
No it's Pyth
 
oh Pyth
 
My polyglot
#s[\s\w\e\e\t\p\o\t\a\t\o\p\i\e)w.q"biscuits")\m\a\c\a\r\o\n\i\a\n\d\c\h\e\e\s\esp00/;oooooooo'stuffing;oooooo~'`turkey'
print'corn'#)))))))]+[--------->++<]>+w.q++[->+++<]>++w.q+++++++++++w.q-----------w.q+++w.q;\o;
'rolls'#)))\é's'l'l'o'r+++++))))))))))))))))))))))))))))))))))))))))))))))))))))))
 
looks like Python :p
 
ok . .. in that case, i'll go back to AU room . . .
 
2:56 PM
@ais523 are you also the guy trying to make a TAS for nethack?
 
3:12 PM
This is either a bomb or a physics experiment: 137.193.65.98:8080
 
@EᴀsᴛᴇʀʟʏIʀᴋ oh man that's so cool
 
@quartata Tell that @Mego
 
3:29 PM
@Downgoat Riker wanted this plugin by the way
 
Who's @Riker?
 
@KritixiLithos easterly irk
 
3:59 PM
yeah
it's my IRL name, idc about my first name being known
 
@ais523 Found something even better:
1
A: Shortest code to throw SIGILL

DennisC, 13 bytes const main=6; Inspired by this answer. Verification $ gcc --version gcc (SUSE Linux) 4.8.3 20140627 [gcc-4_8-branch revision 212064] Copyright (C) 2013 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for...

Such a nice, round number.
 
@Dennis meh, come back when you have 111,111
 
4:21 PM
That's impressively horrible.
 
@mınxomaτ and it's 32-bit o_O
 
Yeah, that's what electron defaults to.
There's nothing GitHub can do (keep in mind that this is the frontend, git is doing the heavy lifting and it isn't even on that list), that would benefit from larger addresses et al
 
I was more expecting 32-bit not to take up so much resources
 
Apart from smaller executable sizes, 32bit can consume all the resources it wants. There's no correlation here
 
@mınxomaτ Use GitUp - it's basically something to go with the command line. The best thing is unlimited undo. And being able to commit specific lines. Lightweight.
 
4:31 PM
I use git. This was just a test to see if the app is remotely usable now (because some of my users would like to use it). Turns out it isn't.
 
@mınxomaτ I see you're running Tor Browser... and Google Chrome at the same time! Do you use Tor for TNB and personal stuff?
 
That's not my PC
 
Oh.
GitHub Desktop (and most other GUIs) are useless because of the command line simplicity. They also take a lot of CPU.
@Bob why are you in so many rooms?
 
Bob
@noɥʇʎPʎzɐɹC why not?
 
I think I'm in more rooms :P
 
4:43 PM
@mınxomaτ o_O
 
This is just a 400k files commit
 
On my machine GitHub desktop crash when it start lel
 
It's still going
Aaaaaand, GH crashed
 
I don't use GH Desktop
 
Does anyone know what I'm missing here?
 
4:54 PM
@Adnan An upvote, I guess?
 
hahaha
I got a downvote, yet it works for all test cases
 
I HATE GOLFING LANGUAGES. I MUST DOWNVOTE SHORTEST ANSWER
 
@Adnan No idea. It seems to work good!
 
(but only if not mine)
 
4:56 PM
i'll try something. can you give me 7 to 13 up votes?
:p
 
@BlueEyedBeast That is exactly how I feel.
 
Hey, it's raining upvotes! :-D @Adnan
 
hahaha, thanks :p
 
the new meta effect: the chat effect
 
Jelly? Downvote. Pyth? Downvote. CJam? Downvote. MATL? Downvote. 05AB1E? Downvote.
I bet @Adnan used a sock to downvote his own answer and beg for pity votes in the chat.
XD
 
4:58 PM
That's a genius idea
 
why not just get 4 socks to make it look lite 4 pity votes?
 
@BlueEyedBeast that is cheating
 
Will be back after creating 3 more socks :P
 

« first day (2122 days earlier)      last day (2738 days later) »