The APL Orchard

« first day (725 days earlier) ← previous day next day → last day (1922 days later) »

jordancurve

3:30 AM

@ngn How do you do IO in K4?

jordancurve

4:13 AM

@ngn Wow! |+\` is brilliant. That makes this the full solution: +/(~2!)#*:'{4000001>x@1}(|+)\1 1`

Is there any way to clean up the rest? I especially don't like putting the constant 4000001 instead of 4000000, but there seems to be no greater-equal (>=) operator in K.

My backslashes are getting eaten by the code formatter and I ran out of edits trying to fix it.

jordancurve

4:25 AM

The implementation of Euclid's greatest common factor algorithm at http://kparc.com/z/fun.k is great. Here's a tacit gcf routine based on it: gcf:*|*|(*)(|!\)\

Example usage: gcf 55 60. I don't get why I have to enclose the third star in parenthesis. Originally I had {*x} instead of (*). {*x} is clearer (maybe?) but (*) is shorter.

I seem to have fixed my code formatting issues here. I now use two backticks before code and two backticks after it.

The example code for Conway's Game of Life at kparc.com/z/fun.k doesn't work in oK. The error is: invalid arguments to ': [-1 +/ (valence 1), (0 1 1 0...

jordancurve

4:53 AM

The Collatz one is amazing too.

Each of them is so compact! The meaning doesn't become clear until I study it carefully. Like a Zen koan.

6 hours later…

ngn

10:29 AM

@jordancurve if it's for project euler, this could be the only verb you need: 0:"filename" (and/or 0:`:filename in some versions of k) reads a file and returns a list of lines

0: in this case is not "assignment to 0", but special syntax for a verb

more details on io - in the reference and in the "q for mortals" book (in q syntax)

ngn

10:49 AM

i tested - in k4 0:`:filename works and 0:"filename" doesn't

@jordancurve there's no greater-equal but you can use "not less" instead: ~x<y

@jordancurve we've all been bitten many times by the peculiarities of chat markdown...

there's a sandbox room where you can test if you like

@jordancurve that * acts as the left argument to (|!\)\ . if you use a naked *, it will parse as * applied to (!\)\ , applied to etc.; however the parentheses "nominalise" it, i.e. turn in into a noun

hm... i'm not sure oK implemented this correctly

something's not quite right with the parsing of that (*) in oK, but anyway...

you could simplify *|*|{*x}(|!\)\ to *|{*x}(|!\)/

ngn

11:46 AM

@jordancurve i've no clue what's going on there. that game of life implementation doesn't work in any version of k known to me. it's also wrong, unless the meanings of some primitive has changed

-1 f': x used to mean "each-prior with initial value -1". could it be that now it's "apply f to a sliding window of 1 before and one after", like dyalog's ⌺3 ("stencil")?

Adám

12:13 PM

Easy main challenge up for grabs using ⌺:

Q: Implement the game of life in 3D

The challenge is to find the shortest implementation of the game of life in 3D (example). These are the rules: Cells (in this case, cubes) with only 1 or less neighbours die, as if by lonliness. If exactly 5 cells surround an empty cell, they breed and fill it. If a cell has 8 or more neighbours...

ngn

@ngn oops, i missed a | there

4 hours later…

Feeds

3:56 PM

Q: In what way is APL made more useful by the distinction between scalars and vectors?

I'm an experienced software engineer beginning to study APL, and in order to get a better feel for the language early on, I would like to understand why the language exposes to the coder the difference between scalars and vectors. As far as I can tell this far into my studies, this only introduc...

Transcript for