The Nineteenth Byte

user202729

2:00 PM

@HyperNeutrino You can try the next one, it's very easy. oeis.org/A004903

HyperNeutrino

cool thanks!

wait is it finite or something

user202729

I forgot to remove the unused function print boolean matrix, otherwise it would be oeis.org/A004704

HyperNeutrino

nope doesn't have the fini tag so probably isn't

@user202729 I mean, you could change it right now and you probably wouldn't get murdered. Up to you :P

user202729

Already become the near-longest submission. Just shorter than this bash hardcode one.

HyperNeutrino

lol

Socratic Phoenix

2:12 PM

@HyperNeutrino I have been informed that you created an abstract interface , as a Java developer I am obliged to report you to the Java Police :)

HyperNeutrino

lol

I also made the method withing the abstract interface an abstract method. Sue me. :P

@SocraticPhoenix That comma should be a semicolon.

Socratic Phoenix

The thing is, Java is already really verbose... It doesn't need any more xD

HyperNeutrino

:P

user202729

It's only verbose in codegolf. -- Is there any problem with abstract interface except it's long?

Uriel

@user202729 and I thought the rust hardcode was bad 😱

HyperNeutrino

2:16 PM

I mean, even though interface is necessarily abstract by default and it's unnecessary to specify that, writing it explicitly shows that the interface is abstract. I was never taught to put the abstract there but I was also never taught to program Java. My method of learning was looking at my friend's code and reading Stack Overflow :P

(That's also how I learned Python)

(In fact the only language I've been mostly "taught" that I didn't learn by myself as much is Jelly lol)

Poke

@SocraticPhoenix pretty sure you never need to do this

Adám

@HyperNeutrino I've never used :Interface in APL… :-D

HyperNeutrino

:P

Poke

like, i think it's obsolete to put that modifier there

HyperNeutrino

It is

how do you return an array from a function in C++ without initializing it beforehand?

Poke

2:23 PM

I guess I can't fault you too hard for using it because I always add the public modifier to methods in interfaces

HyperNeutrino

So like currently I'm doing this:

user202729

Just do it.

Poke

but I'd almost certainly remove abstract if I saw it

HyperNeutrino

int value [2] = { a, b };
return value;

user202729

What is your function return value?

I'm afraid that the array self-destruct before you can possibly use it outside of the function.

Socratic Phoenix

2:25 PM

You guys must not use IntelliJ xD... (it points out useless/redundant modifiers)

HyperNeutrino

I use TIO as an IDE so maybe that's why :P

@user202729 int*

user202729

So it will self-destruct.

HyperNeutrino

it generates a warning

.code.tio.cpp:6:10: warning: address of stack memory associated with local variable 'singleton' returned [-Wreturn-stack-address]
                return singleton;
                       ^~~~~~~~~
1 warning generated.

user202729

You can't return array this way.

HyperNeutrino

I figured; how do I do it though?

user202729

2:26 PM

Either use proper stl tools std::vector, std::array

or to dynamically allocate memory. new int[2]

HyperNeutrino

oh there's a thingy for arrays?

user202729

But if you dynamically allocate memory you have to delete[] it, so I prefer the former one.

Also std::pair<int, int>

HyperNeutrino

I'm trying to do the next OEIS sequence in C++ for C++ practice :P

user202729

Yes, std::array it is a stl wrapper over C-style array.

HyperNeutrino

cool

Poke

2:31 PM

@SocraticPhoenix I'm in a transition phase between intellij and eclipse

HyperNeutrino

I just realized that I don't need arrays for this one

I realized that the sum of at most 8 non-zero powers of 10 is just the sum of 8 powers of 10 because the remaining ones you're not summing can just be zeroes :P

CMP: zeros/zeroes

Adám

77

A: What is the plural form of "zero"?

Both zeros and zeroes are acceptable, see e.g. Merriam-Webster, Wiktionary or TheFreeDictionary. The usage stats from the Corpus of Contemporary American English (COCA) and the British National Corpus (BNC) look as follows: COCA BNC zeros 312 132 zeroes.[n] 1...

HyperNeutrino

I figured both were valid but which one do you use :P

Adám

@HyperNeutrino I would say He zeroes in on the zeros.

HyperNeutrino

lol

I use them interchangeably; I can never remember which one I use as a standard :P

Adám

2:43 PM

@HyperNeutrino VTC as unclear. (Verb or noun?)

HyperNeutrino

oh

revised:

CMP: zeros.[n] / zeroes.[n]

user202729

@HyperNeutrino I've done coding that in C++ you may want to use as reference.

Adám

@HyperNeutrino Zeros.

@HyperNeutrino May we use 1 indexing?

HyperNeutrino

@user202729 I'll try to code it myself first then look at yours to see how horrible my coding style is :)

@Adám No, you must use (-1+i)-indexing.

lol the C iostream log message output stream is called clog which you use when your program is clogged :P

user202729

I never use it. Just cout and cerr.

HyperNeutrino

2:45 PM

I didn't know of anything other than cout and cin before I read the I/O tutorial :P

Anonymous

@HyperNeutrino The latter

HyperNeutrino

anyway gtg o/

Pavel

I'm still not quite sure how to use cin.

user202729

Is there comment in Jelly?

Pavel

Nope

user202729

2:53 PM

@Pavel cin >> {put your variable here};

Adám

@user202729 That would be a waste of a precious codepage point.

John Dvorak

"a comment is just an unused string expression";

Pavel

@user202729 Right, and this bothers me a lot. What do I do if I want to read an entire line? Or just a single word? Or read a number into an int?

Adám

@JohnDvorak Return; A comment is just an unused expression;

user202729

Line -> std::getline, word: std::cin >> {string lvalue reference},

int: std::cin >> {int lvalue reference}. Where variables are lvalue references.

John Dvorak

2:55 PM

please note that comments are very easy to make in INTERCAL

Pavel

I see. What does void *foo; cin >> foo do?

user202729

You read the value of a pointer and store it to the variable.

@Pavel Like this.

Pavel

I think I'll stick with fgets and scanf.

John Dvorak

[in Brainfuck you use a loop that never executes, which even lets you put commas and periods inside.]

user202729

Alternatively cin >> *a can't be compiled, because void pointer can't be deref.

You need cin >> *(int*)a, for example.

Pavel

2:59 PM

I also generally use char* instead of std::string

user202729

So you should use C instead.

Pavel

Eh, I'm mostly a C# person, classes are nice.

New Main Posts

2

Q: Triangular Numbers

(no, not those ones) The Challenge You'll be given two inputs. The first is a positive integer n > 0, which is used to output an n x n right triangle of the numbers 1, 2, 3, ... n. This triangle starts in a corner and increases horizontally and vertically by one and diagonally by two. See examp...

Pavel

My C++ is basically "C with classes"

AdmBorkBork

Huh. The chat search page still has the old top nav-bar design.

Anonymous

3:05 PM

@Pavel Gross

Pavel

I use a few C++ standard library features. Like cout.

feersum

But cout is one of those C++ things which is more gross than the C version.

Pavel

Nowhere did I say I write good C++

Anonymous

@feersum I disagree. Stream-based IO is much cleaner than format-string-based IO.

user202729

Jelly, 13 bytes: (can anyone saves some bytes?) (for A004903) Try it online!

feersum

3:08 PM

If you use actual printf then that's gross.

user202729

@Mego But also more verbose.

Pavel

@feersum What do you suggest?

feersum

The good way is a templated format string library that's type-safe.

Pavel

I basically chose cout because it looks pretty to use and I don't have to worry about having % signs in my strings.

Anonymous

std::cout << "The " << n << ordinal_suffix(n) << " fibonacci number is: " << fib(n) << std::endl; is way cleaner than printf("The %d%s fibonacci number is %d\n", n, ordinal_suffix(n), fib(n));

feersum

3:09 PM

cout is bad because as soon as you have to use a format specifier, it gets disgusting.

Because then you have to undo it.

Mutable state in the stream is what I don't like.

It's also very verbose if you just want to print a table with some aligned columns of numbers.

Anyways I've got to be going now.

Pavel

@feersum The one time I did that I combined cout and sprintf

Which is probably godawful

Anonymous

Hacktoberfest is live!

Pavel

ooh cool

I wonder if I'll be able to participate

I hope so, but it's probably too advanced for me

Anonymous

I've labeled the only outstanding reported bug in Actually for hacktoberfest, if anyone wants to contribute

Sherlock9

If I knew how to implement ECM, I might consider that

I think sympy.factorint() already uses that, if I'm not mistaken

Anonymous

3:15 PM

@Sherlock9 I'd rather only use numpy or sympy as a last resort

Pavel

@Mego What is temp=[1]*(n-max_tested)? It appears to be multiplying an array by a number.

Anonymous

@Sherlock9 Miller-Rabin should be sufficient

Sherlock9

I thought this was factoring not primality?

Anonymous

@Pavel list*int returns a list with int copies of the element(s).

Anonymous

@Sherlock9 Oops, brain signals crossed

Nathan Merrill

3:17 PM

@Mego congrats on the stickers

Pavel

@Sherlock9 The error is thrown in init_primes_up_to(n):

Anonymous

@NathanMerrill Thanks! Now if only I can get them...

Anonymous

I meant Pollard's rho algorithm for factoring. It's a very efficient algorithm, both in terms of space and time.

Sherlock9

I also don't know how that works at the moment

Though I can't imagine that the Internet doesn't have some resources for that

Also, I know that sympy.factorint() also uses Pollard's rho, so their source code would probably help as well

Pavel

@Mego so is factoring just getting all the factors of a number in a list?

Or is it a prime factorization

Anonymous

3:20 PM

@Pavel Yep

Anonymous

Well, prime factors

Pavel

Ok

Anonymous

Divisors is the other one

Anonymous

@Sherlock9 It's quite simple - if nobody claims it, I might implement it myself

Sherlock9

Well, I am in the middle of exams

Pavel

3:21 PM

@Mego I'm working on it

Sherlock9

Two more to go, and three assignments with them, but I'll see what I can do

Anonymous

A linearish-time prime factorization program isn't terribly hard, either - just iterate through the primes less than N and repeatedly divide when you find a factor

Anonymous

@Pavel Excellent!

Pavel

@Mego I'm not a great programmer, but I want that hacktoberfest shirt

I'm guessing "Get all divisors, check each one for primality" is too slow, because it's a pretty obvious solution

Nathan Merrill

figuring out if something is divisible is hard. That's why encryption works

it only works well with the average number because most numbers have lots of small divisors

Pavel

3:25 PM

@Mego According to that page, that algorithm occasionally randomly fails.

Sherlock9

@Mego Although implementing Miller-Rabin for your primality built-in may not be a bad idea, either

Anonymous

@Pavel Brent's improvement makes it deterministic

Anonymous

@Sherlock9 Yeah, I should add an issue for that, too

Anonymous

@Pavel Go ahead and add a test case for :1965593254291461501637330902918203684832716283082y in seriously/test/tests.py, under MathTests.test_arithmetic. The output should be [[2, 230936440213, 173258923096670681, 24562670344821643897]].

Pavel

@Mego I'm getting an error running the tests:

$ python tests.py
Traceback (most recent call last):
  File "tests.py", line 10, in <module>
    from lib.cp437 import CP437
ModuleNotFoundError: No module named 'lib'

Anonymous

3:39 PM

Run it from the root directory

Pavel

Ok

Anonymous

There's a tests.py in the root directory that handles properly importing stuff

user202729

@EriktheOutgolfer

In the triangular number challenge, how many times does the first ¡ repeat?

Erik the Outgolfer

floor(right argument/2)

Pavel

@Mego The issue that I've run into is that every version of the Brent algorithm I can find returns a divisor of the input, and I can't seem to get any of them to yield every divisor.

user202729

3:53 PM

@Pavel Divide the original number by that divisor, and then (recursively) factorize both that number and the divisor.

Anonymous

@Pavel What ^ said. Repeat until prime.

Pavel

I feel dumb now

Anonymous

And if it doesn't return a prime, recurse until you get all primes.

user202729

@EriktheOutgolfer I see. Jelly support binding link to ¡? I see only <link><repetitions>¡, nowhere <link><link>¡. Suggest updating documentation.

Stewie Griffin

4

Q: Invert a Quine!

We all know what a quine is. An inverted quine is a non-empty program that prints the inverse of its source code without reading its source code, and consists solely of printable-ascii characters. The inverse of the source code of a program is all the printable-ascii characters, duplicated for ...

code-golf restricted-source

^^ This isn't a duplicate, is it?

Erik the Outgolfer

3:55 PM

who said repetitions is a nilad

user202729

@EriktheOutgolfer Indeed. But that isn't very clear from the docs.
BTW can I join Jelly Hypertraining?

Stewie Griffin

It's way harder than the "Print every character your program doesn't have".

Erik the Outgolfer

@user202729 sure

I suggest reading the tutorial

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Make Three from One code-golf math arithmetic integer Write the shortest possible formula using only a given digit and certain mathematical operators that evaluates to equal a given three digit integer. Input: A one digit positive integer (1-9) and a three digit integer (100-999) Output: A ...

user202729

It is not mentioned there (tutorial) at all. All examples in the tutorial use nilad (in fact, number literal) before ¡. Even "The number of repetitions is the floor of the ..." is not mentioned.

AdmBorkBork

4:05 PM

@StewieGriffin I hammered it open. It's not in any way a dupe. It looks like a dupe on first glance, but reading into it the new challenge is much more difficult.

Pavel

@Mego I have to stop, I have class now.

I don't think I'll keep trying later, I'm not a python person.

Lynn

I wrote a script to generate vimgolf animations from code.

14

caird coinheringaahing

@AdmBorkBork /o_o I was just about to VTR. Now, I feel like my vote means nothing :(

H.PWiz

@Lynn Is it available online?

AdmBorkBork

#sorrynotsorry

Lynn

4:08 PM

it isn’t quite perfect yet, but I like the look of it!

AdmBorkBork

@Lynn That's pretty, and pretty dang cool.

caird coinheringaahing

Although, that's the reason I dislike hammers. If a question divides those with gold badges, it can be unilaterally opened and closed multiple times

Anonymous

@cairdcoinheringaahing In that case, a meta discussion is appropriate. The problem isn't the hammers, but the people wielding the hammers.

Anonymous

If I think something is a dupe and should be closed (or isn't a dupe and should be reopened), and hammers have already been involved, I start a meta discussion (or, more often, encourage those involved to start a meta discussion) before swinging my hammer.

AdmBorkBork

And, ideally, the people with hammers should know better and do that anyway before getting into a close/reopen war.

Anonymous

4:17 PM

@AdmBorkBork Yep. Though it's easy to forget to look at the history before swinging the hammer, so I give people the benefit of the doubt.

AdmBorkBork

True enough.

Anonymous

I also think that not having our rep requirements raised yet is causing problems, too. This site is very generous with rep relative to other SE network sites, so a new user can earn the close/reopen vote privilege without really digesting all of the relevant policies, leading to incorrect close/reopen votes. It's a pattern I've seen happen multiple times.

muddyfish

I know I've accidentally swung my hammer a couple of times, firstly when I'd just got it (there should be a notification that tells you actually) and in a case like that

Anonymous

@muddyfish The first few times should definitely come with a confirmation, because it's easy to miss the award notification for a gold badge

Stewie Griffin

I'm considering placing a substantial bounty on this question. Will there be any interest?

12

Q: Optimize sorting, using "Sub-vector reversals"

This is a fewest-operations challenge where the objective is to sort a vector into ascending order using the fewest reversals. Your algorithm can only sort the vector using "sub-vector reversals"1, but it can use other operations for arithmetic operations, loops, checking if it's sorted etc. The ...