The Nineteenth Byte

12:27 AM

Any final feedback on this? Thinking about posting today/tomorrow (cc @ngn)

2:29 AM

Does Python have some sort of evil behavior where if it detects "python" in a subprocess command line, it uses its own process rather than resolving "python" according to the user's path settings?

Because the Python I have in my path is Python 2, but I'm getting Python3 when I do subprocess.Popen(["python", ...]) from a Python 3 program.

feersum

2:55 AM

Hmm, it seems to look first for anything in the same directory as the interpreter, not just the interpreter itself.

feersum

3:09 AM

I just added .lnk to %PATHEXT% on Windows. I wonder if anything exciting will happen.

New Main Posts

6:32 AM

0

Q: Stacks and Ques

The 'deque' class from the 'collections' module is preferred to 'list' class for stack implementation. Why is the reason behind it? What is the time complexity for push and pop functions while using lists for stack implementation?

flawr

7:34 AM

@feersum did you make a typo when you named your github "feresum"?

feersum

@flawr Nope

flawr

why

New Main Posts

7:54 AM

-2

Q: Get a function to perform backspace when a keyboard input is provided a backspace button

I am building a keylogger using python but want to make a function which performs backspace when use presses backspace button on this keyboard. Can someone help me with that part

python

Night2

11:01 AM

If I want to add a bounty only for practical language answers (no golfing languages), is a description like "Shortest answer in bytes, in a practical language gets the bounty." enough? Or will it be a hard time deciding what is a practical language?

John Dvorak

11:17 AM

What measure is a practical language? If it's intent, you'll have to accept APL and J.

flawr

@Night2 that is enough, people can ask you if they are not sure, because in the end you're the one selecting answers for the bounty

Night2

12:31 PM

@JohnDvorak That is my issue too, it would be nice to avoid APL, J, etc... too if possible, they aren't golfing languages, but are quite similar in terms of super short syntax. But I'm not sure if there is any better classification other than "practical languages" for this.

Adám

@Night2 You could TIO's practical/recreational distinction.

@Night2 Why do you want to bounty answers in certain languages and not others?

Night2

@Adám That would still include "APL (Dyalog Unicode)" for example.

Adám

@Night2 Absolutely, and J and K. (K makes billions…) But why exactly would you want to exclude these?

Night2

@Adám Basically I want to avoid a super short syntax language get the bounty because it has that super short syntaxt (which is the case for most of golfing language but some others too)

One thing that I can to is to limit the languages to x top language tags from stackoverflow: stackoverflow.com/tags?tab=popular but that will probably leave many practical languages out.

Adám

@Night2 But an interlanguage code length comparison isn't fair, even if you exclude those. E.g. Java tends to be rather verbose.

Fatalize

12:38 PM

@Night2 Just say that you will give a bounty to the answer you find most impressive

Night2

@Adám So maybe my idea from start is wrong? Maybe just leave it to user votes?

Fatalize

You don’t have to justify yourself

Night2

@Fatalize But I want it to have an obvious (specified) and fair condition

Adám

@Night2 Do you have some kind of reason for making a bounty at all? E.g. I want to promote APL, so I bounty APL answers.

Night2

@Adám I want to promote my question which has not got any answers, but I'm not a fan of golf/short syntax languages.

Fatalize

12:39 PM

It can’t be fair if you include multiple languages

Adám

@Night2 I'd say just put a bounty for attention, and don't worry about the answers.

Night2

@Adám So, who gets the bounty then?

Adám

@Night2 Highest voted, I guess.

Night2

@Adám I think I will go with that then, the only issue would be late answers probably won't get much votes even if they are better.

12

Q: Can Jimmy escape the ghosts?

It is Halloween and Jimmy (/o\) has gone into a mysterious neighborhood for trick-or-treating (ask himself why). Now some evil ghosts are chasing him. Can Jimmy escape the ghosts? Challenge: Input: A board showing position of Jimmy, ghosts, solid objects and empty spaces. An example 10x5 bo...

code-golf decision-problem grid path-finding

Hope 200 is enough to get some answers :P

Adám

@Night2 Actually, I don't think it is the bounty amount that matters, as much as the challenge appearing on the "Bountied" tab.

Night2

12:53 PM

I hope so. Thanks for your help and ideas guys 👍

Neil

1:29 PM

on the ball sprite part of the embarrassing mistakes blog post, having a bounding box is better if your coordinates are integers as it allows you both odd and even sized balls

Anush

2:34 PM

hi all. I am a little confused by codegolf.stackexchange.com/questions/195032/totally-blind-chess

why would a chess game ever end in a win under these conditions?

@feersum ^^

DJMcMayhem

2:48 PM

@Anush The main reason it can end in a win is A player who makes 2000 consecutive unsuccessful move attempts shall forfeit the game.

Anush

@DJMcMayhem Oh I see. So really you are always just trying to find a valid move. But given that there are fewer than 2000 possible moves, it seems that all the competitors will be equal

DJMcMayhem

Sure, IMO that makes it easier to get a checkmate. If I have you really close to checkmate, it doesn't have to be a perfect mate, it just has to be close enough that you can't find a valid move

Anush

@DJMcMayhem do you think there will be checkmates? It just seems unlikely from a probabilistic point of view

the proportion of chess positions that are checkmate are just very slow

and everything is being chosen blindly

DJMcMayhem

I agree, it seems unlikely

Anush

maybe there are interesting variants, like the first person to lose a queen loses

DJMcMayhem

2:53 PM

If I had written that challenge, I probably would have allowed capturing the king and moves that leave you in check

Anush

@DJMcMayhem that sounds good

DJMcMayhem

Although IIRC feersum is using the python chess library, so that would take some tweaking

Anush

typo: the proportion of chess positions that are checkmate is just very low
and everything is being chosen blindly

on another topic.. is it now acceptable to pose code-golf questions where the code should run in TIO before it times out

I know this was frowned on at one point

but I feel I have seen more timed code-golf questions in the last few months

DJMcMayhem

@Anush I think that's more acceptable than fastest-code timed on TIO

Because with fastest-code, you need a lot of precision and reliability. Timed code-golf is still scored in bytes with just a restriction that it can't be incredibly slow

Anush

@DJMcMayhem given that acceptability is somewhat binary here.. I will take that as a yes :)

I would really like to encourage people to compete per language

rather than just seeing a Jelly answer and giving up

which means I will pose my numerical integration question

dzaima

3:00 PM

@Anush does it need to succeed once or always? e.g. the time of this has varied from 4.8s to 7s even though it's doing the same exact thing every time

@Anush many challenges do say that competition is per-language and it's often said to specifically not be discouraged by answers in golflangs

Anush

@dzaima it's a good point but how do you test always?

dzaima

@Anush i have no idea, you specify that ¯\_(ツ)_/¯

Anush

:)

what does that code do?

CPU share: 120.88 % Is it running in parallel?

dzaima

@Anush Java probably doing GC or something idk

Anush

what is it computing?

I can't read the code

dzaima

3:07 PM

@Anush generate a 1000 element array where each contains the numbers 2..(1+n) in the most horrible & inefficient way

Anush

:)

dzaima

(1+⍳¨⍳1000 being the more proper way to do that, running in 0.3s. i just needed something that didn't use much memory and calling a full function many times kills dzaima/APL)

DJMcMayhem

4:33 PM

Random question: Would it be acceptable to edit an old question to remove rules that are no longer considered good rules by meta?

For example, there's a question that explicitly says "Languages invented after this challenge was posted are not allowed" but we've agreed to allow new languages. This question has been racking up flags for answers that break that rule, and most of those answers are entirely harmless so I don't want to delete them

Simply editing out that rule seems like a much nicer solution

AdmBorkBork

I don't see why that'd be a problem. I vote go for it.

Business Cat

4:54 PM

Sounds fine to me

DJMcMayhem

On a completely unrelated note, does anyone know a SEDE query that shows you how much rep you have from questions vs answers?

New Sandboxed Posts

5:36 PM

0

A: Sandbox for Proposed Challenges

Fermat's Last Theorem, mod n It is a well known fact that for all integers \$p>2\$, there exist no integers \$x, y, z>0\$ such that \$x^p+y^p=z^p\$. However, this statement is not true in general if we consider the integers modulo \$n\$. You will be given \$n\$ and \$p\$, which are two positive...

randomdude999

6:27 PM

would submitting a solution to this problem: codegolf.stackexchange.com/q/58522/71803 that uses regexes and passes all of the test cases in the question, but not much else, be frowned upon?

AdmBorkBork

7:24 PM

@randomdude999 Optimizing for test cases is a standard loophole for code-challenge or fastest-code challenges. I'm not sure that it would apply to code-golf challenges, but I would think it should.

randomdude999

the scoring criteria seem to imply that detecting all the required test cases is good enough though... and besides, it's kinda very hard to do this completely accurately anyways, so where is the line between what must be recognized and what not?

AdmBorkBork

Very true. I think that challenge should have been a code-challenge and not a code-golf anyway, but /shrug

New Main Posts

9:23 PM

0

Q: IndexError: list assignment index out of range

I want write a program about the progressive sustitution algorithm, i wrote line for line, but Python 3.7 give me this error all of time: IndexError: list assignment index out of range I don't achieve find where i'm bad, can you help me please? Quiero escribir un programa sobre el algori...

python

Jono 2906

9:53 PM

Hey everyone! Merry Christmas today! :P

DJMcMayhem

Because Oct 31 == Dec 25? :P

Jono 2906

Exactly!

El'endia Starman

@NewMainPosts It's mildly amusing to me to see Spanish text and English code intermingled.

John Dvorak

@El'endiaStarman That's right. The correct thing would have been to use an i18n library and stuff all of the text into an external file :D

DJMcMayhem

I was very proud of myself being able to understand most of the text

El'endia Starman

10:29 PM

You should be! :)

New Main Posts

10:46 PM

0

Q: A Nice, Simple Web Scraping Code Golf

Introduction This is a web scraping Code Golf challenge that attempts to take stock Name and price columns from the 'Most Active' list on this New York Times URL. Why is this challenge interesting? Challenge: Scrape these figures from this NYT page and turn it into a pandas DataFrame using the...

code-golf python data-structures

New Main Posts

11:06 PM

0

Q: How many little boxes fit inside a big box

Rules: I have a random number of boxes... n (3 dimensions) Each box has a random sizing (integer), but that sizing is constant on start up. It may be rectangular or a cube. Challenge: Work out how many different combinations fit. You can have an unlimited number of reuses of the same box.

math

feersum

11:44 PM

@Anush You can checkmate the other player.

@DJMcMayhem I decided how I wanted it to work before decided to use python-chess in the implementation. The rules would anyway be much easier to implement if check was ignored.

caird coinheringaahing

Tetris GoL just hit 1000 upvotes :D

4

Transcript for