Sandbox for Proposed Challenges

Question

This "sandbox" is a place where Code Golf users can get feedback on prospective challenges they wish to post to main. This is useful because writing a clear and fully specified challenge on your first try can be difficult, and there is a much better chance of your challenge being well received if you post it in the sandbox first.

Sandbox FAQ

Posting

To post to the sandbox, scroll to the bottom of this page and click "Answer This Question". Click "OK" when it asks if you really want to add another answer.

Write your challenge just as you would when actually posting it, though you can optionally add a title at the top. You may also add some notes about specific things you would like to clarify before posting it. Other users will help you improve your challenge by rating and discussing it.

When you think your challenge is ready for the public, go ahead and post it, and replace the post here with a link to the challenge and delete the sandbox post.

Discussion

The purpose of the sandbox is to give and receive feedback on posts. If you want to, feel free to give feedback to any posts you see here. Important things to comment about can include:

• Parts of the challenge you found unclear
• Comments addressing specific points mentioned in the proposal
• Problems that could make the challenge uninteresting or unfit for the site

You don't need any qualifications to review sandbox posts. The target audience of most of these challenges is code golfers like you, so anything you find unclear will probably be unclear to others.

If you think one of your posts requires more feedback, but it's been ignored, you can ask for feedback in The Nineteenth Byte. It's not only allowed, but highly recommended! Be patient and try not to nag people though, you might have to ask multiple times.

It is recommended to leave your posts in the sandbox for at least several days, and until it receives upvotes and any feedback has been addressed.

Other

Search the sandbox / Browse your pending proposals

The sandbox works best if you sort posts by active.

To add an inline tag to a proposal, use shortcut link syntax with a prefix: [tag:king-of-the-hill]. To search for posts with a certain tag, include the name in quotes: "king-of-the-hill".

Answer 1

Smallest integer divisible by 2..n

Given an integer n, output the smallest integer divisible by 2,3,4,...,n inclusive.

Example

2520 is divisible by every integer from 2 to 10.

Scoring

Shortest code in bytes wins.

Sandbox

• Dup?
• Better Wording?
• Restrictions/Rules?

Answer 2

Pseudoku Cops and Robbers King of the Hill

(I know that another user, @NathanMerrill, is proposing a similar contest. I started playing with the idea for this type of contest independently yesterday, but have since chatted with in The Nineteenth Byte. He is currently undecided on the type of puzzle to use and has some different ideas on how to evaluate participants' performance, so I feel comfortable proposing my idea as a separate challenge.)

Sudoku is a well-known logic puzzle. It is a puzzle of four nines: nine rows of cells, nine columns of cells, nine 3x3 adjacent and distinct blocks of cells, and nine values that any cell can have. A valid Sudoku arrangement or solution is one in which every row, cell, and block has all nine values exactly one time. For example, consider the following valid solution:

+-----+-----+-----+
|4 6 1|5 7 3|2 8 9|
|5 7 8|2 1 9|4 6 3|
|3 2 9|6 8 4|1 7 5|
+-----+-----+-----+
|9 8 4|7 6 2|3 5 1|
|7 5 6|3 4 1|9 2 8|
|2 1 3|9 5 8|7 4 6|
+-----+-----+-----+
|8 3 5|1 2 7|6 9 4|
|6 9 7|4 3 5|8 1 2|
|1 4 2|8 9 6|5 3 7|
+-----+-----+-----+

These are turned into puzzles by removing many of the values in the arrangement in such a way that all blanks are mirrored horizontally and vertically across the puzzle and so there is only one valid way to fill in the blanks to get a valid Sudoku solution. For the above puzzle, this might look like this:

+-----+-----+-----+
|4    |     |2 8  |
|  7  |  1  |    3|
|    9|    4|1    |
+-----+-----+-----+
|     |7 6  |3 5 1|
|     |     |     |
|2 1 3|  5 8|     |
+-----+-----+-----+
|    5|1    |6    |
|6    |  3  |  1  |
|  4 2|     |    7|
+-----+-----+-----+

Someone who wished to play this Sudoku puzzle would then use the information provided to find the original solution.

Sudoku has some interesting properties that allow it to be generalized to similar puzzles with different rules that are sometimes called "Pseudoku" (which is pronounced the same way as the actual puzzle, SOO-DOE-KOO, so please stop saying SOO-DOO-KOO). For our purposes, we will make two differences. First, it may be possible to generate harder puzzles by removing the restriction for symmetric removals. The following is a valid puzzle by Sudoku rules, so why not allow it?

+-----+-----+-----+
|4   1|  7  |2    |
|5    |2   9|     |
|3    |  8  |     |
+-----+-----+-----+
|  8 4|7   2|    1|
|     |3   1|  2  |
|     |     |     |
+-----+-----+-----+
|     |     |6    |
|  9 7|     |     |
|1    |     |5 3  |
+-----+-----+-----+

Second, Sudoku's properties allow us to define games with different sizes. You can define a Pseudoku game with a parameter N where the resulting board has N^2 rows, N^2 columns, N^2 blocks of size NxN, and N^2 values for each cell. Standard Sudoku would be a Pseudoku variant with N=3. So the following would be a valid Pseudoku(2) game:

+---+---+
|1  |   |
|   |  3|
+---+---+
|  1|4  |
|4  |   |
+---+---+

and an example Pseudoku(4) game:

+-----------+-----------+-----------+-----------+
|11         |      10  4| 1     9   | 2       16|
| 6     5   |   15  1   |    3     2|12  8      |
|10    13 14|         12| 5       15| 4     7   |
| 2  3      |    6    13|       8 11|    5    10|
+-----------+-----------+-----------+-----------+
| 7 11 12  9|14        2|16  1  4   |           |
|           |   10  4   |          3|13    16 11|
| 4       10|   16    15|   12      |          6|
| 1 16  2   |11  3      |   10     8|           |
+-----------+-----------+-----------+-----------+
| 3        2|       6   |13         | 5 14     1|
|    7      |          5|           |           |
|   13 14  4|12    16   |           | 8  9      |
|    5 16   |13     9   | 4     2  1|           |
+-----------+-----------+-----------+-----------+
|14 12     7|           |   15     4|11     6   |
|    9      | 6         |11 16      |       3   |
|      11   | 8 13     1| 3 14      |    7      |
|13     8   | 7    11   | 2         |      15   |
+-----------+-----------+-----------+-----------+

Since Sudoku is NP-complete, so is Pseudoku. That means that it gets more difficult to solve a Pseudoku puzzle the larger N gets. However, it can take more time to generate Pseudoku puzzles than it does to solve them, since the naive algorithm for generating a puzzle requires solving the puzzle each time a value is removed! Solving Pseudoku puzzles is fun, but if it takes longer to generate them than it does to solve them, it becomes more work than play.

So help me out! I propose a Cops and Robbers style King of the Hill. The Cops will compete by writing programs to generate lots of Pseudoku puzzles to consume as much time as possible for solving, while the Robbers will compete by writing programs to solve Pseudoku puzzles to consume as little as time possible solving these puzzles.

I need some help ironing out the format, but here is what I have so far:

1. I will provide a Java framework for running the contest. This framework will connect to clients by TCP/IP so contestants can choose whatever language they want to write their Cops and Robbers (so long as I can run them on my system). I will also provide a basic Cop and Robber for these users to try out to see what sorts of times they take. I will publish the times they generate on my system so contestants can estimate how their entries will run on my system.

2. I will give each Cop ten minutes to generate as many Pseudoku(N>= 3) puzzles as they can, but they should be able to generate at least Pseudoku(N=4) puzzles. They can choose what sizes they want the puzzles to be, but they have to be valid with exactly one solution. My server will naively check each one to guarantee their validity; any Cop that generates an invalid puzzle is disqualified. I recommend configuring the Cop programs to be parameterized externally so that Robbers can test their code against basic Cop configurations, but then the Cops can send me secret, optimized configurations before the contest completes for their actual execution. I will provide a couple days after the deadline ends for conferring with the Cop programmers if their settings do not work as expected on my system. Cops should generate different puzzles every time with reasonable expectations; that means no spamming with the same puzzle repeatedly or reading pregenerated puzzles from a file system, Internet source, or internal cache. In addition, I don't want to see a Cop that uses the same removal pattern for every puzzle (that may not guarantee valid puzzles, anyway).

3. Each Robber will be tested against each puzzle generated by the Cops. The Robber will have to generate the correct solution for each puzzle as quickly as possible. I will probably need to see some timings before I make a final decision, but each Robber will be capped with some amount of time to solve a puzzle (maybe an hour?) before the time-to-completion defaults to twice that cap. These Robbers will be permitted to use any technique for solving the puzzles that my system supports except for packet sniffing. I am on the fence as to whether the Robbers will be on an honor code to not study Cop code since I plan to have secret parameterizations anyway.

4. All the times for all the puzzles will be sorted from least to greatest and then assigned an index as one would in a Mann-Whitney U test. Each Cop and Robber will be scored using the sum of the indices of their contributions: Cops for the times the Robbers spent solving their puzzzles, and Robbers for the times they spent solving puzzles. The winning Cop will have the highest sum and the winning Robber the lowest. Cop ties will be broken first by the average time required to solve one of its problems (more is better), then by the number of puzzles generated (more is better), then by the name I deem cooler (here's hoping that doesn't happen). Robber ties will be broken first by the average time spent solving puzzles, then by the sum of the time, then by the standard deviation, then by the cooler name.

5. This scoring scheme poses an interesting challenge to Cops: balancing the size of the problems (and the likely amount of time needed to solve them) against the number of problems generated. A Cop that generates only one puzzle that no Robber can solve in the time limit is likely to lose to another Cop that generates many moderate problems. Similarly, a Cop that spams many small problems is likely to be beaten by another Cop that generates fewer problems of larger sizes. Since the official contest configurations should be kept secret until the contest starts, other Cops can study the other programs to try to determine what their opponents are likely to do and plan accordingly.

I am interested in any and all feedback that the community might have about this challenge.

Answer 3

broken keyboard workaround

|nspired by BASTA´s song and memories from earlier work:

Your keyboard is broken but there is some urgent work you have to complete; you have no back^up hardware - and the shops are closed so you can´t buy a new keyboard!

All you have left to work with is your mouse.

6iven two texts as input (the one you have and the one you want to have), create a program or function that tells you the cut, copy&paste actions that will turn the one text into the other.

Using the mouse is strenuous, so you don´t want too many cut/copy/paste actions. Keep your output as short as possible.

Remember: Your keyboard is broken = you can´t use any characters in your code that you don't have ~ you must get alon9 with those that are provided in th1s {["te%t"]}. For7unately your keyboard 7ook qu17e a wh1le 7o bre4k down c0mpletely 4nd y0u used numb3rs 4nd sp3c*4l ch@r$ t0 r3pl@c3 br0k3n l3773r$; $0 y0u $h0uld h@v3 m0$7 0f 7h3m @v@*l@bl3.

Also, you don´t want to do too much C&P to cre8 your code, so keep that as short as possible, too.

---

• You can assume that the second text contains no characters that are not present in the first text.
• You can pick any input format and method that is convenient for you; but the output format should match that. (e.g. if you take input from files, output should also go to a file).

code-golf restricted-source

NOTES

Note that the challenge description contains all letters and digits except j and z. If you absolutely need them: they are hidden in the YouTube link. (I didn´t check for upper/lower case though.)

Curlys, brackets, braces, single and double quotes and all operators I could think of are there, so the challenge should be fine for most languages that use printable ASCII.

Still trying to find a more fluent way to include curlys, brackets, double quotes and circumflex, though.

I thought there was a tag [string-manipulation]; but couldn´t find it in the list.

I think about dropping the "output method should match input method" restriction.

Answer 4

Find B1nar0 Solutions

B1nar0 puzzle is a paper and pencil game with 0 and 1. The goal is to fill the grid accoring to 3 rules :

1. No more than 2 consecutive 0s or 1s
2. Each row/column has half 0s and half 1s
3. No identical row/columns

Example :

[

• A is 0 according to rule 1
• B is 1 according to rule 1
• C is 0 according to rule 2
• D is 1 according to rule 2
• etc.

Edit : Grids are square grids of even size (4, 6, 8, 10, 12 or 16 are usual sizes).

Input : Any binary grid (array or string) with 0,1 and any other character you want for empty cells.

Output : Same format as input but filled with a correct grid.

Test case (see GIF)

0 11
  0 
 0  
1 1 

Answer 5

Enthusiastically Russianify a String

Greetings Comrades,

Many of you may have interacted with people from Russia on the internet at some point, and a subset of you may have noticed the slightly odd method they have of expressing themselves.

e.g. деинсталляция игра нуб))) - (forgive the google translate)

where the ))) are added for emphasis on the previous statement, I have been working on a theory that the ratio of )'s to the rest of the string is directly proportional to the amount of implied emphasis, however I oftentimes find it difficult to compute the ratio on the fly, as I am also trying to cope with a slew of abuse, so I would like the shortest possible code to help me calculate what the resulting string should be, for a value of enthusiasm between 0 and 500%, given the original, unenthusiastic string, this will aid my research greatly as i will not have to type out bulky scripts every time I wish to test my hypothesis.

so, the challenge:

write a full program or function, which, provided two arguments, a string of unknown length, and a number, in either integer format (between 0 and 500) or in decimal format (between 0 and 5, with 2 points of accuracy) will

• return the original string, suffixed with a number of )'s
• the number will be the calculated as a ratio of the input number to the string length.
• so if the number 200, or 2.00 was provided, 200% of the string must be suffixed as )'s
• the number of brackets rounded to in decimal situations does not matter.
• script is required to support Printable ASCII characters.
• only has to support one input number format, of your choice.

examples:

"codegolf" 125      = codegolf))))))))
"codegolf" 75       = codegolf))))))
"noob team omg" 0.5 = noob team omg))))))
"hi" 4.99           = hi!)))))))))))))))

example code (powershell) (with decimal input):

Function Get-RussianString ([string]$InputStr,[decimal]$Ratio){
    $StrLen = $InputStr.Length
    $SuffixCount = $StrLen * $Ratio
    $Suffix = [string]::New(")",$SuffixCount)
    return $InputStr + $Suffix
}

Get-RussianString "codegolf" 0.5
codegolf))))

this is code-golf so shortest code wins!

---

This is my first challenge, any feedback is greatly appreciated.

Answer 6

Animate the text in your terminal

The goal

The goal is to "animate" the string "Hello world" in your output so that each character gets capitalised after each other.

Your program can exit after each letter has been capitalised.

For example;

# Iteration 1
Hello world

# Iteration 2
hEllo world

# Iteration 3
heLlo world

# Iteration 4
helLo world

# Iteration 5
hellO world

# Iteration 5
hello world

# Iteration 6
hello World

# Iteration 7
hello wOrld

# Iteration 8
hello woRld

# Iteration 9
hello worLd

# Iteration 10
hello worlD

Input

No input is required, but "Hello world" must be the string that is "animated".

Output

The string "Hello world" must be animated. The output must be as 1 line to create a sort of wave animation. Example gif;

^{I saw this on a metasploit youtube video and thought the effect was pretty cool, which is where I recorded the gif from, so it's a little laggy, but I hope it illustrates the output fine}

This is code-golf, lowest byte-count will be deemed the winner.

Answer 7

Packing Primes for Posterity

Introduction

You've calculated which of the first n numbers are prime, and want to save your achievement for all future generations. Unfortunately, you're broke, and want to minimize storage costs (you'll be paying them forever, after all.)

You need to determine the best way to pack all of the primes <=n and still be able to answer the question "is p prime?" in O(1) time.

Challenge

A submission to this challenge must include both a compress algorithm and an isPrime algorithm.

compress

Input: n -- the number that you have checked prime-hood through.

Output: Bytes to feed into your isPrime algorithm.

isPrime

Input: The output of your compression algorithm, and an integer i >= 0. i is guaranteed to be <= n.

Output: True if i is prime, otherwise False.

This algorithm must run in O(1).

The winner of this challenge is the (compression, isPrime) pair that is

• Correct
• Has the best compression ratio, as determined by the average compression ratio for

  n in {10^3, 10^4, 10^5, 10^6, 10^7, 10^8, 10^9}

as compared to the naive solution below.

Consider the following solution in Python:

def compress(n):
    # simple sieve of Eratosthenes. Note: this is not a 
    # prime generation challenge; a list of the first 
    # billion numbers will be provided in this format.
    primes = [1] * (n + 1)
    primes[0] = 0
    primes[1] = 0
    upper_bound = int(math.sqrt(n)) + 1
    for i in range(2, upper_bound):
        factor = i
        if not primes[factor]:
            continue
        factor += i
        while factor <= n:
            primes[factor] = 0
            factor += i
    primePackStr = ''.join(str(i) for i in primes)
    return primePackStr

def isPrime(compressed, i):
    return compressed[i] == '1'

Example Input and Output

Input to compress:

20

Output:

"001101010001010001010"

Input to isPrime:

("001101010001010001010", 13)

Output:

True

Notes

• This is not a prime generation challenge. The compress executable can assume that there is a file called primes.txt in the same directory that contains the first billion numbers in the format s[i] = 1 if i is prime, 0 otherwise. (Zero-indexed)
• Naturally, the isPrime executable cannot make use of this file.
• The isPrime executable must not hardcode any primes.
• Please provide instructions on how to compile/run your code on either OSX 10.12 or Ubuntu 16.04, if it's not obvious.
• This is not a code golf challenge. Any length of code is fine, as long as isPrime doesn't attempt to cheat.

Notes for sandbox

• I'll include a link to a downloadable primes.txt
• Any thoughts on a better restriction than "The isPrime executable must not hardcode any primes?"
• Should I test on random values of n instead?
• thanks!

Answer 8

Challenge about loudly interjecting in a courtroom

code-golf

One of the most important things about being a courtroom lawyer is loudly interjecting before you make your point. In this challenge, we're going to edit a typical courtroom transcript to include these interjections.

Any lawyer (and in fact, any character at all in the transcript), uses these rules to interject:

1. Use an interjection when the character who is speaking changes to you.

Take the following example:

SAHWIT: I remember the time I found the body exactly.
SAHWIT: It was 1 P.M.
PHOENIX: Frankly, I find that hard to believe!
PHOENIX: Your statement directly contradicts the autopsy report.

There is one change in speaker, so an interjection will be added in at that point like this:

SAHWIT: I remember the time I found the body exactly.
SAHWIT: It was 1 P.M.
PHOENIX: Hold it! Frankly, I find that hard to believe!
PHOENIX: Your statement directly contradicts the autopsy report.

2. Use Hold it! if the previous statement ends with a single full-stop or exclamation mark, Take that! if the previous statement ends with an elipses (...), and Objection! if the previous statement ends with a question mark.

For instance:

JUDGE: What evidence proves the clock is running slow?
PHOENIX: The victim had just returned from abroad the day before the murder.
PHOENIX: The time difference between here and Paris is 9 hours!
PAYNE: But modern day clocks automatically adjust for time zones...
PHOENIX: This is an antique!

Becomes:

JUDGE: What evidence proves the clock is running slow?
PHOENIX: Objection! The victim had just returned from abroad the day before the murder.
PHOENIX: The time difference between here and Paris is 9 hours!
PAYNE: Hold it! But modern day clocks automatically adjust for time zones...
PHOENIX: Take that! This is an antique!

That's about it. I'll write some longer test cases a bit later. This challenge is probably Retina-bait to be honest.

Answer 9

Normal quine, weird quine

Note: This Sandbox entry has a fairly long history, and is basically an attempt to produce a challenge inspired by this comment, but that's immune to wilful misinterpretation (or misunderstanding) of what counts as an error in order to trivialise the question.

Background

In the world of programming languages, there are lots of different ways to produce output on the usual output streams. Most languages have a way to print a string intentionally, called print, write, or something like that. Sometimes you can even just leave a value to be printed implicitly. Most languages also have situations in which the implementation interjects with its own output, e.g. warnings produced during the compile. We'll call this weird output.

Task

For each method of output to standard output or standard error in a programming language, consider how much of that output is under the programmer's control and thus can contain arbitrary text (e.g. specified as a parameter, part of the program's filename, taken from a variable that can be assigned to, or the like), as opposed to being a single possibility (or a finite set of possibilities) hardcoded into the interpreter. We'll call this output method normal if no more than 3 bytes are outside the programmer's control; and weird if there are 4 or more hardcoded bytes that the programmer cannot control.

In this challenge, you need to write a full program that's a quine variant, obeying the proper quine rules. Specifically, after performing the entire process of building and running the program (i.e. if there's a separate compile step required, its output counts too):

• All the output produced on standard output and standard error via normal output methods must be identical to the program's source code;
• All the output produced on standard output and standard error via weird output methods must also, separately, be identical to the program's source code.

In other words, the program is a quine in two different ways. You can think of this as being a quine that's also an error quine (also known as a "Kimian quine"), except that the notion of "error" is restricted in order to avoid abuse (mechanisms which would let the program provide an arbitrary "error message" count as normal output, not weird output, on the above definition), but generalized to allow things like warnings, banners that the implementation prints as it loads, and other weird ways to produce output.

Clarifications

• For the purpose of the proper quine definition, the fixed part of the output that's inherent in a weird output method is considered to not be encoded by the corresponding part of the program (even if that part of the program causes an error). As such, only the normal part of the quine can fail to be a proper quine.
• PPCG doesn't normally count output that's inherent to an implementation (such as compiler progress messages and fixed banners). This challenge is about handling that sort of thing, though, so such output is definitely relevant here (in addition to everything else on the standard output and error streams).
• Unlike in many challenges, the switches given to the compiler, and the program filename, are likely to be highly relevant in this challenge. Using an unusual build configuration may well be required to make the challenge possible, and as such is legal here; however, if you run the implementation in an unusual way, remember that PPCG rules charge a byte penalty for doing so (equal to the number of additional characters that you'd need to add on the command line over the shortest "normal" way to run a program), and thus you'll need to specify the size of the penalty in your post. (For example, if the interpreter you're using reads the program from a file, and has no particular restrictions on the filename, the shortest normal way to run the program would be from a file with a 1-character filename; thus, if you need a 100-character filename to make your program work, you'd incur a byte penalty of +99.)
• The compiler/interpreter version you use may well be relevant, so as part of your submission, please state a specific compiler or interpreter on which your program works, and which version is required. (For example, a C submission might state "C (gcc 6.2.0)" in the header.)
• Note that this task may not be possible in all languages. In the languages where it is, the easiest method will likely be to find an error or warning message for which it's possible to customize some subset of the text (via changing the name of something that gets quoted in the message; filenames are a common choice here, but not the only one). Obviously, if you could customize the entire thing, it wouldn't be weird output and thus wouldn't work. I'll be particularly impressed (and surprised) if someone finds a way to do this using only error and warning messages whose text is entirely fixed.

Victory condition

This is a code-golf challenge, so an entry is considered to be better if it has a smaller byte count. As such, once you've got your program working at all, you want to optimize it to bring the number of bytes down as far as possible. (However, don't be discouraged if there's already a shorter entry, especially if it's in a different language; what we're really looking for here is to shorten a particular algorithm or idea behind a program as much as possible, but seeing multiple solutions in different languages or that rely on different principles is always worthwhile.)

code-golfquine

Sandbox questions

This was moved here from main because many answerers seemed to disagree with everyone else as to what an error message was.

I've aimed to avoid the problem in this rewrite by focusing not on what is and isn't an error message, but rather on the amount of hardcoded content in the message. Is this likely to be interpreted the same way by everyone? Is it objective?

Also, should I edit the original challenge, or should I post it as a new challenge? Out of the two non-deleted answers, one will stay valid (although the explanation will end up somewhat out of context), the other will need to be deleted (although I consider it to be invalid under the original specification too, and thus arguably no changes are being made to which answers are correct).

Answer 10

Garbled Phone Numbers

code-golf (de)cipher

You know how you get a voicemail message and the person's connection wasn't great, and you're trying to figure out how to call them back, but you're not sure if that was a "5" or an "8" they said?

That's this challenge.

The good news is that the caller read off their number twice, but it's garbled in both places.

Your program should take input like this:

5551231234 / 5551231234

Where the first seven digits are the first time the phone number is said in the voice mail and the second set are the second time it's said. Only...it'll look more like this:

555?AAA1_36? / 55?522_1?234

A digit followed by a question mark means that that's the best-guess for that digit (e.g. "5?" means "probably a 5, compare with repeat"). An underscore indicates a known missing digit, something too fuzzed by static to be deciphered at all. Letters are just that: letters. Treat them as their respective digits (ABC -> 2, DEF ->3, HIJ -> 4, etc).

You can safely assume the following judgement calls:

5? / _     -> 5  //5 is the best guess we have, use it
5? / 4?    -> ?  //conflict
 5 / 4     -> ?  //conflict
5? / 4     -> 4  //solid information overrides possible value
 5 / 4?    -> 5  //solid information overrides possible value
 _ / _     -> ?  //no information available

Additionally you can assume that all inputs will contain ten-digit phone numbers, not including the question marks. Inputs that aren't ten digits (e.g. 1234567 / 1234567) can either be treated as unsolvable (falsey output) or throw an error.

Output option A: Output a truthy value indicating whether or not a given input can be resolved to a single valid ten-digit phone number.

Output option B: If it can be parsed to a single valid ten-digit phone number, output the phone number. Otherwise output some form of error indication (e.g. -1, false, empty line).

Shortest wins, as per usual.

[Sample inputs]

Answer 11

Write a "21" game in exactly 21 characters code-bowling

Challenge

You must write a program which implements the following algorithm:

Let x = 0
Let y = truthy value
while (y is not falsy AND x <= 21) do:
  Let x = x + a uniform random number from {1,2,3,...,11}
  Output the value of x
  Input a value of y from the user (you may assume input is valid)
Output the value of x

(You do not have to follow the pseudocode exactly. For example, if your language happens to initialise variables to a truthy value automatically, you don't have to include the y:=TRUE line. Similarly, you don't have to use a while loop. The important thing is that it repeatedly takes user input until either x exceeds 21 or the user chooses to stop, and it outputs the current value of x after each user input.)

Score

Let n be the length of the shortest program which meets the spec which can be obtained by deleting 0 or more characters from your code. Then your score is:

- 500            if n > 21
- 1 + (n-21)^2   if n < 21

The winner in each language is the program with the lowest score.

Questions

• Is this a resonable idea? I can't find similar challenges, so maybe there is a problem with ones like this? (Trivial solutions etc.)
• Is the specification too complicated (maybe more languages could enter if it was a simpler algorithm, for example just taking user input once?)

Answer 12

Tetris Programming

code-challengerestricted-sourcetetris

The Task

Your program or function should take as input one character from the set IJLOSTZ, which represents one of the seven tetrominos as usual:

I  J   L   O   S    T    Z

#   #  #   ##   ##  ###  ##
#   #  #   ##  ##    #    ## 
#  ##  ##
#

You should output the number of shapes which are equivalent to this tetromino up to rotation. For example, the I only has two arrangements, horizontal and vertical, whereas the J has four different orientations. The O looks the same no matter how you rotate it, so there's only one shape. Here all possible input/output pairs:

I  2
J  4
L  4
O  1
S  2
T  4
Z  2

The Source Code

The main part of this challenge is the source code restriction:

• You may choose either linefeeds (LF, 0x0A), carriage returns (CR, 0x0D) or LR-CR pairs to represent newlines in your code (consistently). These split the source code into a 2D grid of lines (which aren't necessarily the same length).

• This 2D grid must be completely made up of tetrominos where all 4 characters in each tetromino must be the same. For example, this would be a valid program:

  aaa
  bba
  cbddd
  cbd
  c
  c
  
  eeee
  

  Note that the individual characters don't necessarily need to be distinct, so there may be larger connected regions of the same character, as long as this region can be segmented into non-overlapping tetrominos. Also note that this restriction also applies to spaces, so the following is not a valid program, because the two spaces don't form a tetromino.:

    x
  xxx
  

The Score

For each of the seven tetrominos count how often it appears in your source code. Your score is the maximum of these seven values.

That means you don't want to make up your code entirely of Is but instead try to use about the same number of each of them to keep the maximum of the seven values down.

The Small Print

You may either write a programs or a functions and use any of the standard methods of receiving input and providing output, as usual. Note that these loopholes are forbidden by default.

Answer 13

So I've been puzzling over the best way to present this idea I had, so this will probably need a lot of help. I am open to completely reworking the challenge, but this is the best polish I've managed to figure out so far.

At work I have to secure my laptop with one of those 4-digit cable locks and it occurred to me that there was a puzzle here: figuring out the combination by looking at the typical behavior of setting the lock: never allowing any given wheel to "rest" on its unlocked value. e.g. if the combination is 1234 then never walking away with a 1 in the first position, a 2 in the second, and so on (e.g. 1111 would not be considered locked, but 2111 would be). Or possibly by not letting any digit of the unlocked combination be visible (so even 2111 would be "bad" but 6789 would be ok...unless a transpose was also considered to be insufficiently random, however such choices are often up to the user of the lock). I also subsequently changed my behavior (not that I have any real risk of my laptop being stolen).

A standard challenge of "write some code that examines a series of locked values to determine the unlocked value, scoring by number of entries needed" is non-viable, as the sequence list would need to be carefully chosen such that there is a strictly known optimal solution (i.e. a minimum number of locked values), as finding a shortcut in that specific sequence might be possible, but invalid on another sequence.

Then the other night it occurred to me that it might be possible to do this as cops-and-robbers: one side has to randomize their locks (albeit following a set of rules that allows exploitation), the other side has to break them open. The downside being that it will be a nightmare to validate scores as there will be no easy way to pipe input and output back and forth between two programs running arbitrary languages.

I'm also not sure if there's enough room for freedom in designing the lock randomization code (i.e. interesting for the cop) for it to be plausibly crackable without resorting to brute force (an uninteresting challenge for the robber). Ostensibly the robber half is brute force, but it's guided in some manner towards a determinable value ("ah, I see, the first spindle is never set to 1 when locked, ergo the first digit in the code must be a 1) rather than indeterminate ("ah, I see, no spindle is ever set to 1, ergo there is a 1 somewhere in the code" -> 4 digits ^ 4 spindles -> 256 plausible values with further attempts gaining no new information).

---

Combination Locks (Cops)

Your goal is to write a program that produces a 4 digit random number as a combination lock entry code. Your program needs to keep this value a secret, but must produce output that is the result of the lock being locked and its tumblers spun, the value printed being the digits shown along the set row (8585 in the above image).

Your program will then take input of a 4 digit code that is an attempt to unlock the lock. If it is the correct value, output the number of attempts made and the seed value, otherwise print another randomized lock value. Repeat until successfully unlocked.

Rules:

• Your program must have some way of setting the combination (for scoring), eg. providing a seed value for the random number generator (inputting the correct combination is allowed).
• All locked combinations must be considered random. However:
• The nature of "random" is what is to be exploited here. Obviously you wouldn't want your random lock to actually remain unlocked after shuffling the dials!
• You may chose any rules by which to keep the lock locked, provided that it can be exploited. No outputting 0000 every iteration or cycling between predetermined sets (1234,4567,7890,1234). You're trying to emulate what appears to be smart behavior of a human being, not create an unbreakable lock.
• Every digit from 0-9 should be possible with some degree of uniformity. That is, if the correct combination is 1234 you are allowed to prevent 1 from showing up as the fist digit, but you may not prevent 1 from showing up in other positions.
  • Blanket removal of all four digits of the combination from all four columns reduces the problem to brute forcing 256 possible combinations.
  • Similarly, allowing a ban on a digit for up to three columns reduces it to brute force against 3136 possible combinations (banning only the combination digits from 3 columns is 81 possible combinations). None of these are interesting challenges.
  • Entries shown to devolve to a brute force guessing will score based on the worst-case lucky guess (i.e. the number of attempts needed to identify the brute-force point, +1).
• If your language does not have a way to "wait for input" then....??? (requirements for fixed-seed randomness across multiple attempts, e.g. for a given combination and the same number of attempts made, the next output should be the same)
• Your program should store no data about attempts to break the lock or prior output values, the only data that may be stored are the Random instance (if needed), the correct combination, and number of attempts made. Outputting an attempt value back out (intentionally) would be underhanded.

Scoring

The ratio of your code's byte-length to the best (lowest) number of attempts needed by any robber against your lock.

---

Combination Locks (Robbers)

Your goal is to exploit the non-pure-random nature of locked 4 digit combination lock. After all, no one leaves a portion of the correct code in the lock after they shuffle the wheels!

You are to write a program that attempts to deduce the correct combination for a given lock, given only a series of locked (incorrect) combinations. Your program will read a single 4 digit number as the current state of the lock and produce a 4 digit number output as an attempt to unlock the lock. If additional input is given, the attempt was unsuccessful. Your program need not self-terminate (i.e. there is no requirement to take input telling your program that it was successful; ctrl-C interrupt is succificient).

As you are an accomplished thief, you know exactly how each lock gets randomized. You are to exploit the built-in rules to bypass the lock in the fewest number of attempts by looking for patterns in the lock's "output" and narrowing down the list of possible correct combinations.

• Locks will have a way to predetermine their combination (e.g. random seed or specific 4 digit combination). Your program may not know these values, they are used for scoring only. Remember the standard loopholes: hardcoding the output is disallowed.
• If a human is unable to find the solution with the data known at that the point of solving, the number of attempts will not count for scoring as it can be considered a lucky guess (arbitrary threshold: 10 or fewer attempts will be automatically assumed as such). This should be treated like a logic puzzle, not a slot machine.
• If your language of choice doesn't do "programs" it is acceptable to write a function taking in an array of inputs [XXXX, AAAA, XXXX, BBBB, XXXX] (where XXXX represents the combinations displayed on the lock, and AAAA/BBBB represents the prior attempts made) or similar. Note that there will be one more value from the lock than values from attempts, as your function would be producing the paired half as its output.
  • Supplementary output to support ease of alternative input methods acceptable (e.g. a newline followed by the input array for the next iteration to be copy-pasted).

Scoring

The ratio of your code's byte-length to the best (lowest) number of attempts needed top open any lock.

Answer 14

Wasn't able to find anything in my searches, but please let me know if this or something very similar has been done before. Appreciate any feedback, first post in sandbox.

Can I leave yet?

I'm bored at work, and want to know how close I am to being able to go home. To represent this, I wish to know what percentage of work I have completed for the day.

Inputs

None - Current local/computer time shall be used

Outputs

Percentage of work completed for this day

• Formatted as either a percentage value or a decimal value: 0.57, .57, 57%, 57.0%
• Output should be accurate to at least +/-0.5%. Additional accuracy/digits are allowed.
• The work day is a total of 8 hours.
• Work starts at 08:00, and ends at 17:00.
• Lunch is between 12:00 and 13:00. Working during lunch is forbidden, and thus should not count towards the percentage of work completed.
• Output should be correct during any time of day, including before work starts (0%) and after the work day ends (100%).

The response for a full 24 hour day is shown below:

Valid Test Cases

Time Ran     Output
03:55        0%
04:31        0.000
08:00        0
09:00        12.5%
11:31        44%
12:37        .5
16:30        0.94
21:08        100%

Incorrect Test Cases

09:00        12.5     (Interpreted as 1250%
11:31        43%      (Error of 0.9%)
12:37        .58      (Did not account for lunch)
16:30        0.9      (Error of 3.8%)

Notes

• I work 7 days a week 365 days a year; you do not need to check if it's a weekend, holiday, etc.
• I live in an area with no Daylight Savings Time, Leap Seconds, or any other confusing time-changing events.

This is code-golf, so lowest byte-count score wins.

Answer 15

Bitcoin Trading

code-challenge?

This is mainly an idea for something I could potentially host on my KOTH server.

Everybody knows that bitcoins are the next big thing. It's just a question of when they are going to take off. Right now, they are worth $250 each, but who knows, maybe someday they will be worth over $1000! The growth trend is phenomenal.

You are a tech-savvy investor who wants to get in on this action.

The Challenge

Your goal is to write a bot that can predict the market and tell you how you should invest your money, given hourly updates of the Bitcoin price.

Keeping Balance

To overcome the fastest-gun-in-the-west effect, wherein early answers have more time to make more money, this challenge will not keep track of any absolute balances. Instead, the assets of each entrant will be scaled up/down between each round.

Each entrant will be given a single float in the range 0 to 1 representing the percent of total assets are currently invested in Bitcoin. This is calculated by (BTC_cur_rate*BTC_owned)/(USD_owned + BTC_cur_rate*BTC_owned).

A value of 0 means that you currently have nothing invested in Bitcoin, while a value of 1 means that you have everything invested in Bitcoin. An input of 0.3 means that 30% of your total value is in Bitcoin, while the other 70% of your value is in dollars.

Examples

input    ->    assets as portion of your total value
                 BTC     %      USD     %
 0.0     ->      0.0    0%      1.0  100%
 0.3     ->      0.3   30%      0.7   70%
 0.6     ->      0.6   60%      0.4   40%

Price Data

Players will also have access to a file history.txt which will contain the BTC price history, measured in cents, over the duration of the competition. Each time a player is called, they are presented with a fresh copy of history.txt, with one line appended each turn. Do not attempt to modify this file.

Example File

This could be the history.txt file after 3 hours of competition. The most recent price is $247.49.

24694
24724
24749

There will be a trailing newline at the end of the file.

Making a Trade (ouput)

The output of your program should be another float in the range of 0 to 1, representing the new portion of your assets that you want invested in Bitcoin. The difference between your input number and output number represent the amount of value being exchanged.

Examples

input -> BTC USD  |  output -> BTC USD  |  trade being made
 0.3  -> 0.3 0.7  |   0.2   -> 0.2 0.8  |  0.1 in BTC -> 0.1 in USD
 0.3  -> 0.3 0.7  |   0.6   -> 0.6 0.4  |  0.3 in USD -> 0.3 in BTC

Calculating Score for a Round

Your score for a round is based on your change of value for that round. You start every round with a total value of 1, but your ending value is influenced by two things:

• A 0.2% commission on your trade taken by the controller
• The change in Bitcoin value over the next hour

Taking Commission

Commission is taken whenever you buy or sell bitcoins. Whenever you convert a certain amount of value from one currency to the other, you will receive 0.2% less of the new currency than what you actually ordered.

Examples

input  |  output  |    trade     |  commission  |  result after commission
 0.3   |   0.6    |  0.3 -> BTC  |  0.0006 BTC  |  0.5994 BTC & 0.4 USD
 0.75  |   0.05   |  0.7 -> USD  |  0.0014 USD  |  0.05 BTC & 0.9486 USD

Adjusting BTC Value

After taking commission, your value of BTC is multiplied by the price percent change in BTC over the next hour. The amount of value you have in USD will stay constant.

Examples

BTC after commission  |  prices in cents  |  % change  |  new BTC value
         0.5          |  30000 -> 29850   |   -0.5%    |    0.4975
         0.236        |  20000 -> 30447   |   +2.0%    |    0.24072

Overall Process of a Round

Below is an example showing all of the steps in a single round.

BTC    USD
.3     .7  = 1.0      input to entrant is 0.3
.6     .4  = 1.0      output of entrant is 0.6
.0006  .0             0.2% commission of the trade
.5994  .4  = 0.9994   result after commission
+0.3%                 percent change in bitcoin price over 1 hour
.6012  .4  = 1.0012   result after the flow of time = score for that round
.60048 .39952 = 1     input for the next round is 0.60048 after scaling

Determining the Winner

For a given round, your score is your new total value. This is after taking the 0.2% commission and calculating the change in Bitcoin value. For the above example, the score was about 1.0012.

At any given time, the aggregate score for an entrant shall be the product of the scores for its most recent (up to) 50 rounds. At any given time, the current winner is the player with the highest aggregate score.

For example, a bot could get these scores for its first 5 rounds: 1.001 1.002 0.998 0.999 1.003. The total score of the bot is about 1.00299.

The Controller

I haven't written the controller yet, but I think it's going to be written in Perl with support for entrants in a variety of other languages (Java/Python/Ruby/C++).

I plan to use this API for bitcoin price data.

The controller will probably run all of the entrants in parallel, each with their own thread. This simply allows it to put a stop to any infinite looping that may occur. I hope it will work if all of the programs are reading the same history file at once.

Additional Rules

Since this is a PvE competition and not a PvP competition, and takes place on a server, there are some slight differences in rules.

• There's no set restriction on submitting multiple bots, since you can't make a team.
• Similar to always, you can't call other programs, like the controller or other bots, during your turn.
• The time limit is loose. A long as a single round with all of the bots doesn't take up most of an hour to perform, it'll be fine. It really shouldn't take more than a couple minutes for each bot to make a move.
• You may create a single file, with the filename [botname]-data.txt, in the current directory. This file will persist, even across updates of your bot or the controller.

Answer 16

Raindrops are falling on my... glasses?

Answer 17

The alphabet - my way

Your pointy-haired boss gives you a list of words and tells you to sort them. So you give him back a sorted list. "Wrong!" he says. "I want them sorted according to MY alphabet..."

Challenge

Given a new ordering of the alphabet and a list of words, sort the words according to that new alphabet ordering.

The new ordering is given as a 26-character string, guaranteed to contain all letter of the alphabet exactly once, in lower case.

All words in the list of words will be made up of lower case letters only -- no capital letters or punctuation. There will be no repeat words.

If there is a word in the list that is the prefix of another word, then the shorter (prefix) word should appear first in the sorting. For example, "golf" should appear before "golfing".

Examples

Example 1

Input:

qwertyuiopasdfghjklzxcvbnm
apple
banana
currant
dragonfruit
elderberry
fejoia

Output:

elderberry
apple
dragonfruit
fejoia
currant
banana

Example 2

Input:

qwertyuiopasdfghjklzxcvbnm
uranium
plutonium
uranus
pluto
polonium

Output:

uranus
uranium
polonium
pluto
plutonium

Example 3

Input:

abcdefghijklmnopqrstuvwxyz
four
score
and
seven
years
ago

Output:

ago
and
four
score
seven
years

Answer 18

Print a shuffled deck of cards

Answer 19

Title

code-challenge number

---

Challenge

In your language of choice, write 25 programs, functions, or snippets that output or return the integers 1 through 25, inclusive. However, the goal is to simultaneously minimize the number of distinct chars used and the length of the code.

Scoring

This is a variation on code-golf: If your 25 entries have N distinct characters and a total length of L, your score is N × (L + N). The submission with the lowest score wins.

Example

Say the challenge only went up to 10, and your ten snippets were:

1
2
3
4
5
6
7
8
9
10

That's 11 bytes total, and 10 distinct chars; therefore, your final score would be 10 × (11 + 10) = 210.

Now, if your snippets were:

1
1+1
1+1+1
1+1+1+1
1+1+1+1+1
1+1+1+1+1+1
1+1+1+1+1+1+1
1+1+1+1+1+1+1+1
1+1+1+1+1+1+1+1+1
1+1+1+1+1+1+1+1+1+1

That's 100 bytes total, and 2 distinct characters (+ and 1); thus, the final score is 2 × (100 + 2) = 204, a small improvement over the literal numbers.

One last example. If your snippets were:

1
-~1
1-~1
-~1-~1
1-~1-~1
-~1-~1-~1
11-1-1-1-1
11-1-1-1
11-1-1
11-1

That's 58 bytes total, and 3 distinct chars (-, ~, 1); therefore, your final score would be 3 × (58 + 3) = 183, an improvement over both.

Rules

• Each output may be a string of digits rather than a literal number.
• Each output may have trailing decimals, as long as they are all 0s (e.g. 1.000 is allowed, but 1.000001 is not).

---

Meta

• Is 25 a good number? I originally had it at 100, but that seems a little tedious.
• Will the scoring system work well enough?
• Is this even a good idea?
• Title and tag suggestions?

Answer 20

Compare string repetitiveness code-challenge decision-problem

Given two strings, decide which one is more repetitive. This is the string whose most common character appears more often in it. If these are equal, then tiebreak by counting their respective second most common characters, and so on. Once a string's distinct characters have been exhausted, all further counts are zero.

Give one consistent output if the first string is more repetitive, and a different consistent output if the second one is. You will never be given a complete tie.

You may assume input strings will be non-empty and use only ASCII characters.

Scoring: Your code's score is its repetitiveness, with comparing lower being better. Put in your header the counts of the top 3 most common characters, and the full frequency list in your body.

Test cases:

TODO

Answer 21

Range without Range Builtin

Challenge

Given an integer y where y > 0, output a list of in some reasonable format that contains every integer in increasing order up to but not including y, without using any builtin that generates a range of any sort.

If your language has a feature whose specification uses a range, that is not allowed (for example, you cannot use the map quick in Jelly on a single integer because that maps over the range). You can assume that y will not exceed your program's capacity for integers, but it must be able to theoretically work on any integer given no memory, time, or otherwise language constraints.

A format is reasonable for a list a = [a0, a1, ..., an] if and only if there exists a string x, a string y, and a non-empty string z such that the output is x + z.join(a) + y.

code-golf

Answer 22

Fence Matrix

code-golf

Given a positive integer n, output the 2n+1 x 2n+1 "fence"-matrix

0  1  0  1  ...  0 
1  2  1  2  ...  1
0  1  0  1  ...  0 
1  2  1  2  ...  1 
⋮   ⋮  ⋮   ⋮       ⋮  
0  1  0  1  ...  0 

Alternatively you can also return a nested array or print a string (even with other entry delimiters than spaces or none at all) or output a raster image where each entry is represented by one pixel.

Examples

n = 1
0  1  0
1  2  1
0  1  0

n = 2
0  1  0  1  0
1  2  1  2  1
0  1  0  1  0
1  2  1  2  1
0  1  0  1  0

Answer 23

Anyone who would like may post this challenge to main. Just give credit to @Lordofdark.

How long will I sleep

You need to go to bed, but what you need more is to know how long you will sleep until your alarm rings.

Your task

Write a program or function that takes a time (hours and minutes) as input, and outputs the number of hours and minutes until the next occurence of this time.

Rules

• In this challenge every clock in 24h format.
• You must always get the current time for the same timezone; you can assume the input is in this timezone

Input

The input time must be in hours and minutes in any convenient 24h format for your language.
Hours and minutes must always be separated by at least one character

Valid inputs for 8h30:
"8h30"
8H30M
8,30
8 30
[8,30]
...

Invalid inputs for 8h30:
8.5
830
510min

Ouptut

The output is the difference between current time and the next occurrence of the input time (it can only be today or tomorrow).

The same formatting rules apply : hours and minutes separated by at least one character an in 24h format.

Note that the output will always be between 0h00 and 23h59

Examples :

If it is currently 20h10 :

7h30 -> 11h20m
20h -> 23h50m
21h -> 0h50m

Challenge

This is code-golf, so the shortest answer in bytes wins. Standard loopholes are prohibited

Answer 24

This message is open for anyone to adopt and post to main. For more details, see the chat room or meta post.

Visualized Tree of 3n+1 Conjecture

ascii-art math code-golf

Originally by @KeyuGan. Thanks for letting me use this!

Introduction

Probably you are already familiar with 3n+1 conjecture (aka Collatz conjecture). As is stated in this golfing problem:

• Start with an integer n > 1.
• Repeat the following steps:
  • If n is even, divide it by 2.
  • If n is odd, multiply it by 3 and add 1.

And it is proven that for all positive integers up to 5 * 2⁶⁰, or about 5764000000000000000, n will eventually reach 1.

It is easy to draw a chain of the whole process for an integer (e.g. for 5, the chain is 1<-2<-4<-8<-16<-5).

Task Description

You are asked to print a string of a visualized 3n+1 tree of all chains resulting from positive integers from 2 to n, containing new lines if necessary.

Input and Output

There is only one input n, which can be read from stdin, be a function parameter or from any external sources.

You can safely assume input is valid and does not exceed your language's processing ability. However, your code should be able to deal with inputs of 2 - 446. Under such circumstances, the biggest number involved is 13120.

[Sandbox note: Is 446 a proper minimum ? It turns out to be the largest number in which the biggest number involves is less than 32767]

The output is flexible, as long as:

• It is a textfile, or a string, or an array of characters, or an array of lines.
• It contains only 0-9, -, |, <, >, ^, v, spaces or new lines, where <, >, ^, v are for arrows, -, | are for lines.
• Not hardcoded
• Correctly visualized and in proper directions (for instance, 1->2->4->8->16->5 and 1-2-4-8-16-5 are not accepted.)
• All numbers included in the output occur exactly once.
• All leaves of the visualized tree should lie in the range of 2 ~ n, that is, all numbers in the output must be necessary for the result.
• the destination of every chain is 1

Besides, the output should meet the following formatting criterion:

• A number should be arranged horizontally and connected.
• There should not be horizontally-adjancent digits from different numbers. For instance, in the following example, 17<34 23<46 is OK, while 17<3423<46 is not. Space(s) should be put between the two numbers under this circumstance.
• There should not be vertically-adjancent digits as well.
• There should not be zero(es) before a number (such as 0016).
• There should be only one arrow for a line.
• The line between two numbers must be straight.
• As is demonstrated in output, - and | can be omitted if not neccessary.
• Lines should not be crossed. A solution without crossed lines is proved to be available. A simple explaination is: Thinking in reverse, you can start from integer x, and draw 2x and (x-1)/3 (if result is odd) following x, and repeat the process for every new number. Stop when you have all required integers from 2 ~ n in the graph and erase all unneccessary numbers.
• You can only draw a line onto and from a number directly, that is, the arrow of the line must be pointing at a digit. e.g. |<--, ^<--, |-- and ^-- are not accepted.
• The direction of arrows and lines must be correct. e.g. ^- and <| are not accepted.
• There should not be spaces between arrow and number.
• There should not be spaces between arrows and lines, neither in lines.

[Sandbox note: Tell me plz if you come up with other loopholes.]

Output is assumed to be printed in a monospaced font (all characters have same width).

Sample

Input

15

Possible Output 1

1<2<4<8<16
         ^
         5<10<20<40<80<160
           ^     ^     ^
           3<6<12|     53<106
                 |        ^
                 13<26<52 35<70
                       ^     ^
                       17<34 23<46
                          ^     ^
                          11<22 15
                             ^
                             7<14<28
                                  ^
                                  9

Possible Output 2

                   15
                   v
                   46
                   v
                   23
                   v
                   70
                   v
                   35
                   v
                   106
                   v
                   53
                   v
                   160
                   v
                   80
                   v
1<2<4<8<16<5<10<20<40<13<26<52<17<34<11<22<7<14<28<9
             ^
             3
             ^
             6
             ^
             12

Possible Output 3

1<2<4<8<16<5     80<160<53<106<35<70<23<46<15
           ^     v
    12>6>3>10<20<40<13<26<52<17<34<11<22<7<14<28<9

Scoring

Your answer should include verifiable output of input 42, without a violation to output requirements. And you should verify your answers with different answers on this page: TBD

[Sandbox note: I will provide a js checker on my site to validate an output.]

Among all accepted codes, shortest code wins.

Answer 25

Perfect Hash Generator

Given N words you are to generate a perfect hash function (ala gperf). A perfect hash function for a set of strings is a hash function with no collisions. An additional condition is that the range of the generated hash function must be [0...O(N)] (i.e. at most a constant times larger than N). You can use any language for the generated function.

Can we get some feedback on this old post? I'm wondering if it is possible to avoid the obvious loophole of a cat program.

Answer 26

This message is open for anyone to adopt and post to main. For more details, see the chat room or meta post.

Find the mines!

Tags: code-challenge algorithm

You are a mine remover. Your job is to find all mines on a field, without a mine explodes. So, you write an application that can find the mines carefully.

The input

The input can either be provided through command line arguments or through STDIN (tell what you use in your submission). The input items are separated by commas.

The input looks like this:

<current step (zero-based)>,<mine count>,<field width>,<field height>,<field data>

The field data is like a Minesweeper field. Rows in the field data are separated by semicolons, columns are separated by nothing, as each column is just one character. Here are the characters you can get:

• X This means that you don't yet know what's there, the real data is still hidden. At the start, you get a field full of Xs.
• / This means that there is nothing on that location.
• <number> Specifies the count of mines around the location of the number.
• F This is marked by a flag by you.
• ? This got a question mark from you. There might be a mine on it, but you are not sure. This is just used as a reminder for you, it doesn't mean something specific to the controller.

Example input:

2,1,3,3,XX1;X1X;XXX

That input means that it's currently your third step, there is one mine, the field is 3x3, and the field looks like:

X X 1
X 1 X
X X X

The output

The output consists of 4 parts: the X of which you want to see the data (like a click in Minesweeper), the location on which you want to put a flag mark, the location on which you want to put a question mark and a sign, used to let the controller know whether you are finished or not (0 for not finished, 1 for finished). Locations are written as X;Y, zero-based. If there is something you don't want to do, output -1. You can also remove flags/question marks using the same way.

Example output:

4;3,-1,3;3

Specifications

• If your first output data is the location of a mine, you hit the mine and you die, but you'll still get a score.
• If you select a X which hides an empty location (/), all adjacent empty fields (and their borders, which are numbers) will be revealed.
• For every step, your program is executed again with the updated arguments.
• When looking for mines, you are allowed to have more flags than the amount of mines. Only if you finish, the amount of flags must not be more than the amount of mines. If the amount of flags is greater than the amount of mines, your submission is disqualified (for every test!) and excluded from the scoreboard.

Testing

When I test your submission, I'll run 100 tests on every submission, with randomly generated fields, which I created using a program that I'll write after I got some feedback. Every submission gets the same test fields, so it's fair. Test fields look like this:

• 10 tests with a 10x10 field and 10 mines.
• 10 tests with a 10x10 field and 12 mines.
• 10 tests with a 12x12 field and 14 mines.
• 10 tests with a 15x10 field and 16 mines.
• 10 tests with a 15x15 field and 35 mines.
• 10 tests with a 20x20 field and 40 mines.
• 10 tests with a 25x25 field and 50 mines.
• 10 tests with a 25x25 field and 60 mines.
• 10 tests with a 50x50 field and 100 mines.
• 10 tests with a 50x50 field and 125 mines.

Scoring

You get 10 points for every mine you find, you lose 5 points if you think there is a mine somewhere when there is none and you lose 2 points for every mine you missed. The scoring is always the same, it doesn't matter whether you finish or die. The highest score wins.

In case of a tie, the count of steps is a tie breaker.

Controller

I'll start working on this after I got some feedback.

Answer 27

Inspired by lifecompetes.com

Multiplayer Game of Life (GOL)

There are n players that play the Game of Life (standard rules) on an 50 x 50 grid. (Size, border conditions? Toroidal, Absorbing, Mirroring?)

Before first GOL-Step

When the game starts, each player has 12 cells that he can place anywhere he want as long as they do not overlap. Before the first GOL-step occures every player has to place 6 cells. If two players place their cell on the same spot no cell wil be placed an thei cells will remain in each players bank.

During game

In each GOL-Step, each player can place as many new cell on the grid as he has in his bank. If two or more players want to place a cell on the same spot, no cell will be placed there (the cells will remain in each players bank). Every six GOL-steps all the players who have less that 12 cells in their bank will get a new cell in their bank.

Goal

The goal is achieving the maximum number of cells on the grid during 1000 GOL-steps.

How to participate

Each participant has to write a javascript function of the following form (multiple return statements allowed.)

function my_bots_name(field, bank, golstep){
  /* your code */
  return p;
}

Where p is a 2d array of points [[x1,y1],[x2,y2],[x3,y3],...] that the player wants to set. field contains a 2d array of the GOL grid, bank indicates the number of cells in the player's bank, and golstep is the index of the current GOL step (golstep == 0 before the first GOL-step occurs). The function may not the global variables and cannot access Math.random() or Date(). In field the empty cells will be set to 0, the own cells will be set to 2 and the other player's cells will be set to 1. (You will not be able to distinguish between various other players.)

The winner will be determined with a game that contains all valid submissions after one week after the first submission.

TODO

The exact environment will be provided so everyone can test the own function before the official runs.

• What size of the grid is appropriate? (dependent on number of players?)
• How many steps should be computed?
• What border conditions should be chosen?
• Is the restriction of Math.random() and Date() apropriate? (The idea was that the games will be the same no matter of who/when they will be run. (Deterministic) )

Answer 28

This message is open for anyone to adopt and post to main. For more details, see the chat room or meta post.

Edit: I'll have to clarify the optional features and there may be more. Essential I want to have the basic features covered, and if anything beyond gets implemented because it saves characters that is okay, but not required.

Edit: Updated but still working on it. Will definitely include edge cases and more examples as test cases.

Edit: It might be cool to have this implemented as a function, and then have some follow up questions where you are allowed to call the function created here while only counting the function call as characters, and not the contents.

Still working on the specifics, but getting it the idea out there.

---

2-Dimensional Regex

Given a 2-Dimensional regex and a block of text, do a match, a single search and replace or a global search and replace depending on the input.

Implement this as a function. The input should be two arguments to the function. The first is the regex, and the second argument is the string to match. The output should be a truthy or falsey value when doing a match, or the string when doing a search and replace.

---

Match

Input:

/aaa/
/bbb/
/ccc/

and

aaaab
dbbba
ecccc

Output:

A truthy value

---

Single Search and Replace

Replace the first instance of the match. The search order is the match that includes the top most character. If multiple matches end up with the top-most character on the same line, then include the left-most match.

Input:

s/bb/cc/
 /bb/cc/

and

abbaabba
bbbaabba
bbbaaaaa

Output:

accaabba
bccaabba
bbbaaaaa

---

Global Search and Replace

Replace all occurrences of the match. Matches do not overlap, and you use the same search order as the single search and replace.

Input:

s/bb/cc/
 /bb/cc/

and

abbaabba
bbbaabba
bbbaaaaa

Output:

accaacca
bccaacca
bbbaaaaa

Another example:

Input:

s/aba/bbb/
 /aba/aba/
 /aba/bbb/g

and

aaaabaaaa
  aabaababa
aababaababa
aabaaabbaba
 aba

Output:

aaabbbaaa
  aabaabbbb
aabbbbababa
aabaaabbbbb
 aba

---

Rules

• Only ASCII characters 32 through 126 are valid.
• Character classes are valid, [a-c4-6] would match an a,b, c, 4, 5, or 6. And [b?7] matches one b, one ?, or one 7.
• . matches any character.
• Use \ to escape there special characters to match their literal character instead of their special meaning: {}[]/\.?*.
• a? matches 0 or 1 a's.
• b+ matches 1 or more b's.
• c* matches 0 or more c's.
• b{2} matches 2 b's. Ranges such as {2,5} to match 2 to 5 b's is optional.
• g flag replaces all of the occurrences, without it only the first occurrence would be replaced. This flag is optional.

Optional features:

• Capture groups are optional. Please specify whether to use () or \(\) to match literal parenthesis.
• Grouping such as (ab)+ matching all of abababab.

Answer 29

Sandbox

This will be my first kolmogorov-complexity submission, does this question fall under that category?

Is the question clear enough?

Is it too trivial?

Problem

Given no input write a program or a function that outputs or returns the following string:

(<(<>(<>.(<>.<(<>.<>(<>.<>)<>.<>)>.<>).<>)<>)>)

Rules

• Shortest program wins.
• Trailing whitespace allowed.
• Trailing newlines allowed.
• Unused parameters for functions allowed.

Answer 30

Implement Nopfunge

Background

Nopfunge is a very simple Befunge derivative, with no stack and no real form of data storage; all it has is an instruction pointer and five commands ("turn north", "turn south", "turn east", "turn west", "continue in same direction"). However, what it does have is a program that repeats forever in two dimensions, which turns out to be enough to make it Turing complete.

The task

Your program must take, as input, four 2D arrays of characters, a, b, c, and d, via any reasonable means. Each of these will have the same dimensions, and they will be rectangular (although not necessarily square). (Note: This is a simplification from the actual syntax of Nopfunge, but is still just as Turing-complete.)

The program will then conceptually work on an infinite array formed out of sub-arrays identical to the inputs, formed by placing a in the top left corner, b along the top edge, c along the left edge, and d everywhere else, like this:

abbbbbbb …
cddddddd …
cddddddd …
cddddddd …

⫶⫶⫶⫶⫶⫶⫶

(Note that because you can't assume your language can do infinite work in finite time, you won't be able to actually represent the compressed array in memory, so you won't be able to calculate the value at a particular coordinate pair in advance; rather, you'll have to calculate it lazily and/or every time it's needed.)

Then simulate the progress of a Nopfunge program on this array. Specifically, there's an instruction pointer that has a position on the array, and a direction; it starts at the top-left corner, going to the right. Repeatedly, run the command specified by the character at the instruction pointer:

• >: Move the instruction pointer one position to the right; it now points right
• <: Move the instruction pointer one position to the left; it now points left
• ^: Move the instruction pointer one position upwards; it now points up
• v: Move the instruction pointer one position downwards; it now points down
• (space): Move the instruction pointer one position in its current direction

Should the instruction pointer ever attempt to move outside its infinite array (by moving above the top edge or to the left of the left edge), your program should exit. If the instruction pointer never attempts to move outside its infinite array, the program should never exit. (Or in other words, the task is to halt if and only if the Nopfunge program does.)

Clarifications

• You can input the arrays as arrays of character codes (i.e. integers) rather than characters if you wish, but if you do, they must use the ASCII/Unicode encodings, 62 60 94 118 32 for > < ^ v .
• Your program can do anything if the input arrays aren't all the same size or aren't rectangular.
• You may take additional inputs for the dimensions (width and height) of the input arrays, if you wish.
• It doesn't matter what (if anything) your program outputs; all that matters is whether or not it halts.

Victory condition

This is code-golf, so shorter programs (in bytes) are considered to have a better score.

Sandbox questions

• Is the specification easy to understand and unambiguous?

• Is this the best possible input format? I wanted to avoid the secondary task of having to parse Nopfunge's input format (which uses = and ; to split one array into the four given here), but am not sure whether this format or that one is more convenient for solving the task.