So today I was parsing a new file in the VSL standard library (that connects to DOM apis) and apparently nodejs needs more than 1.7GB to parse this 2K+ line file so it is now crashing from out of memory ;_;
probably memory leak but wow
@ASCII-only do you have any potential idea on how to address
Unless nearley creates megabytes worths of objects I don't think a memory leak there could be the issue
and additionally they are no reference cycles until the transformation passes so given the simple tree-like data structure it should be pretty easy on the GC
Loop with no loop
code-challenge restricted-source
Challenge: Create a loop without your language's loops features.
The point is to print to stdout this text:
Hello World #0
Hello World #1
Hello World #2
Hello World #3
...
Hello World #97
Hello World #98
Hello World #99
...but without lo...
hmm. if the goal of the builtins is to be as minimal as possible, what kind of conditionals/branching should it have?
just branch if? (i.e. goto_if?)
@DJMcMayhem which one is which though? (pls answer seriously) (wait, don't actually answer seriously - i'm just saying i'm actualyl confused which is which)
also, how bad of an idea is it to make 1. boolean literals, 2. strings and 3. integers behave like identifiers (i.e. you can assign to them) (returning their value by default of course)
hmm. what should i do if generics are required for core features (lists)
because e.g. .NET has the System.Collections.Generic namespace (and possibly others), but for them generics aren't core features
so... i guess the real question is: should generics be handled specially? or... if it's a core feature, it might be even better to treat all normal types as essentially zero-parameter generics??? (i.e. all types are treated as (well, kinda) generic types - this would reduce eliminate the need to specialcase generics from normal types)
I have this challenge idea: Given a Boolean matrix compute the next generation which has each row be the XOR of the row above it and the row below it. How does that sound? Does that have a name?
You don't need to code golf this puzzle.
The goal of this is to input a string (containing +, -, * , /, %, ^, ==, !=, >=, <=, >, <, !, &, |, parenthases, and integers). You must implement a function that returns the integer result without using a function that evaluates an expression(like eval()...
Ah, even more fun. A 3D Boolean array. Process layers. Rows of layer above indictate per-row how to rotate. Columns of layer below indicate per-column how to rotate.
@Adám it can expand (I'm pretty sure), i was just happy that it didn't expand constantly infinitely
@Adám also thanks for giving me the idea to try to find non-normal cellular-automaton-ish thing, now i have something do today that doubles as a backup plan for a homework :D
hmmm for a fancier editor/display do i try to implement the thing in straight up processing java or do i try to do it in my APLP5 and deal with the slow speed
@El'endiaStarman take the 5x5 neighborhood, sum the on cell relative positions to the current cell, and take the value of the cell at that relative position, cycling as a torus in the 3x3 neighborhood
@El'endiaStarman exactly. though i wonder if it's possible to take this idea further and make that picking part not only in the 3x3 neighborhood but the whole board, without making almost nothing possible
@El'endiaStarman hmm really any cellular automaton which evaluates everything at once can be called an elementary cellular automaton with an ∞×∞ neighborhood :p
The bank has been broken into, and all the local mafia thugs have an unusual alibi: they were at home playing Connect 4! In order to assist with the investigation, you are asked to write a program to validate all the Connect 4 boards that have been seized in order to check that the positions are...
@El'endiaStarman um, yeah, but my point is that the backstory doesn't really stand like that, detectives can't know who started or if they switched the starting color between different games
That Android Studio one is too relatable. I've just spent 36 hours at a hackathon (and accomplished very little :(). Struggles with Andriod Studio and with Qt Creator, as well as with trying to write C++ like Haskell.
@ngn Thanks, it was posted on my website (the clues). I'm currently decoding a ceaser cipher
oh, maybe we should make a PPCG challenge for a ceaser cipher
I'm worried it may be closed as too broad
Maybe the user would give the substitution, and then the program would take another input as the input to the substitution input, and output the ceaser cipher output
Cicada 3301 is a nickname given to an organization that on three occasions has posted a set of puzzles to recruit codebreakers/linguists from the public. The first internet puzzle started on January 4, 2012, and ran for approximately one month. A second round began one year later on January 4, 2013, and a third round following the confirmation of a fresh clue posted on Twitter on January 4, 2014. The stated intent was to recruit "intelligent individuals" by presenting a series of puzzles which were to be solved. No new puzzles were published on January 4, 2015. However, a new clue was posted on...
i'm not willing to spend too much effort on this, as at some point there's likely to be a US location-based task, but i was thinking, maybe a combined effort among ppcg, the side channel, and puzzling.SE can go a long way :)
@MilkyWay90 the real challenge is, what could be a more motivating reward for people to publicly share their progress than a potential job at whatever super-awesome company or intelligence agency is behind cicada? is stackoverflow reputation worth more? :)
btw, we haven't verified yet it's truly cicada. at some point there should be something signed with the key they published a couple of years ago
@ngn I really doubt it's Cicada, although I could post this on Puzzling.SE
And reputation
would be the payoff, and it could also be a collaborative effort, like the Quest for Tetris
But if it is Cicada 3301, people could also be blocked from entry because it was a group effort, right? This is also why people were blocked from Cicada in 2014
@MilkyWay90 Maybe. I wonder how Jelly would do in front-end dev though
I suppose it could be measured by, say, ratio of longest 5 outliers to mean code size on a collection of tasks meant to represent the universe of problems. Maybe drawn from PPCG problems, real-life code examples in a variety of areas, and reimplementing its own builtins.
The other question is, what causes a language to have or not have that?
Anonymous
@lirtosiast I think that some basic stuff like number theory (basic arithmetic, primality, fibonacci, factoring), string manipulation (reversing, contains, substring, replacement, Levenshtein distance), and fundamental data structures (array, linked list, dictionary, stack, queue) would be a pretty good corpus
I'm working on a Python project with like 5 other people and magically everyone had Python 3.7.2 so I get to use Dataclasses and I'm pretty happy about that