« first day (2600 days earlier)      last day (2244 days later) » 
00:00 - 17:0017:00 - 00:00

5:10 PM
I can't find much information on SO, answers on this topic aren't really helpful. One says much different things than the other. Whatever, can I use extended ASCII characters in the after-hashtag part of URI?
 
5:38 PM
I think I'm finally gonna use APL for actual programming again ._.
 
@Soaku I have an idea for compressing better. I will implement a proof of concept to demonstrate.
 
@Zacharý Cool. What are you working on?
 
*Mentions APL .... summons Adám
@Adám pic-cross solver.
 
@Zacharý What is a pic-cross?
 
It goes by many names, I'll find an example.
 
5:45 PM
I don't know if you folks saw insights.stackoverflow.com/survey/2018 but I imagine it would be starred if you did
 
@Poke Nah, I guess at PPCG we don't care about trends.
 
surely not
 
@Poke I think I got an email about something to do with that (but I don't read emails from anyone I don't know so ¯\_(ツ)_/¯)
 
"Also for the third year in a row, Visual Basic 6 ranks as the most dreaded programming language"
i laughed
 
And if anyone can figure out what is wrong with bitbucket.org/zacharyjtaylor/jagony/src/f7af5c5659e9/Jagony/…, I will be forever in your debt.
 
5:48 PM
What's the error? That link just goes to the repo
 
When running test.py, I get this:
Traceback (most recent call last):
  File "test.py", line 11, in <module>
    print U._eval()
  File "/home/zachary/Jagony/jagonyObject.py", line 12, in _eval
    return self.__dict__["_fn"]()
  File "/home/zachary/Jagony/jagonyObject.py", line 16, in <lambda>
    return Jagony_Object("$CHAIN", lambda: self._eval() + "\n" + obj._eval())
  File "/home/zachary/Jagony/jagonyObject.py", line 12, in _eval
    return self.__dict__["_fn"]()
  File "/home/zachary/Jagony/jagonyObject.py", line 14, in <lambda>
 
@Poke I think we all got both emails and navbar notifications.
 
@Zacharý lambda: self.__dict__["_fn"](p) The (p) is calling the lambda
 
I'm not sure which lambda I'm passing in though...
 
oh
i don't know if i got either of those
or maybe i ignored it
 
5:51 PM
@Zacharý Sounds sudoku-like.
 
It's not the lambda you're passing in, you're calling a defined lambda, that takes no arguments
Wait. I misread the error ಠ_ಠ
 
@cairdcoinheringaahing I thought I overloaded that everytime I created a Jagony_Object or subclass that has __call__ called.
@Adám Yep. But it should be easier.
 
@Zacharý I suppose you want the fun of solving it from scratch, but for the benefit of other readers: dfns has tools that can probably help.
 
There's one thing I might need help on... but I'll wait until I give up on it (additive partitioning that doesn't eat memory like candy)
Okay, when I call inspect.getsource(p._fn) above the error-causing line in my code, it prints the source for .__init__... IDK what's going on. I had a similar error last time I tried to write some code for Jagony.
 
Does anyone know of something like strace but with human-readable output
 
6:04 PM
@Zacharý What is additive partitioning? +/¨cutoffs⊂data?
 
Spliting a number into N numbers that sum to X.
I can do it easily, it's just my solution is golf-like and will probably run out of memory fast
I might have used the wrong term
 
@Zacharý And I suppose you want the smallest +/×⍨nums-X÷N, right?
 
No, I want all of them (that satisfy the row/column clues)
I haven't given up yet, though
 
Um, is it a bug that I can VTD this post? Doesn't it have to have a negative score?
 
@Zacharý You said you might need help on one thing, but that one thing is the entire problem ⍨
 
6:14 PM
Pretty much... except maybe finding the common ground between the partitions, then applying that to each row/column, then doing it until every square is filled out.
 
@Zacharý OK. I hope you'll have fun with it.
 
Yep.
 
[Try it online!](https://tio.run/##dVTbctowEH33V@y4kxkbEoZA6cUtDymlaXpJUpL05mE6sr0EJbbsyjYJ/Xm6kvCFZKJhNLB7zq50zqJsXSxTMdxseJKlsgCJ1kKmCSxKERZpGudQJ6IyRMvK10mgwmOwj95N3k8/HH88@fT5y9fTs/Nvs4vLq@8/fv76zYIwwsX1kt/cxolIs78yL8rV3f36X/9wMHw@evHy1euDP7ZlEQrCNMkk5vk5ZuhIRNezgFYVxkj1snUsYUW4RNVcYm/BRcTi2LEldux9iqDaeifHp2ez6eToYupqziKVhgdcVAVMB7UCXqhy/uEcurQDp/P0eF5mGUrHBYxzhL6uEdb8@UP2QYzCUd9d2IMRdMDvq3IqUkPvljxGHWqaq7VicYn6QkpfJ2ZJEDFg@4EHgw7rBvua43ujubvD27bWyZE338nRJXSjbfHuGIaDHUBLW0puLfU12lSSWJRStHG0jF00GG3DGsjWN/KhMYyyARfaQtLEqi@sNd/27ZGPeO@EbiNzU9RwjHiG2chXSWfivSzNnH4jUtXaPyBvx7s/OyQIGWQK7LXVIe
 
No markdown in multiline messages
 
@recursive yeah, link works but markdown dun goof'd
 
6:25 PM
Tried to fix it, but the tio link is too long now (?)
@Soaku Here's a link to a link
It compresses "REeE REeE RE REEEEE Reee" to "ddHjfY" using 64 safe-ish symbols.
 
@recursive wow, thanks. o_o
it's a 75% compression, that's nice
 
It helps when the domain is so sparse.
 
0
A: Sandbox for Proposed Challenges

J. SalléHelp me play my trumpet code-golf The trumpet is a valved aerophone instrument, usually pitched in B♭. The sound is made when the player vibrates their lips to displace air inside the instrument. That vibration is acquired by setting one's mouth in a specific way, called the embouchure. Differe...

 
@Adám, is there a way to apply ¨ a certain number of times?
 
@Zacharý dfns.
 
6:59 PM
@recursive I appreciate the work, but you know what is the problem? Pepe and the website is written in JS, not Python ._.
I've been wondering; why are TIO links so long?
 
@Adám Thanks. I got the additive partition function done now. Now there's still a bit more, but I think I've got it now.
 
@Soaku Because they have to encode a lot of data. Check the byte lengths of really long programs in TIO
 
@cairdcoinheringaahing So the whole code is encoded in the url?
 
How else would you keep the code the same when you follow the link?
 
@cairdcoinheringaahing Well, a lot of services of this type save the code locally...
and when I open tio links, I often see it loading some kind of output cache
 
7:14 PM
@Soaku it caches the results of the program, not the program itself IIRC
 
@Οurous I understand... but the output is often bigger than the program...
so why isn't it cached too...?
 
@Soaku that's what I meant. The output is cached.
 
@Οurous I know! I get it! But why isn't the code cached too?
@Οurous Also, quoting tio.run home: "server-side permalinks will eventually require a separate storage"
Which sounds like if the code was stored on the server...?
even if it isn't because url length does depend on the code length
 
7:31 PM
@Soaku A cache is something you can drop at any time; the output cache holds results for more or less one week (500 MB) before forgetting them. Code would have to be stored permanently on the server side.
 
that could make sense
 
7:46 PM
CMQ for those knowing at least a little APL: What other languages maintain something akin to APL's workspace, i.e. a state (collection of stuff) which doesn't necessarily exist as source files on disk?
 
@Adám Mathematica
Called a Notebook
 
@Soaku The code isn't too complex. It shouldn't be a problem to rewrite in js. Let me know if there's a part you don't understand.
 
Doesn't JavaScript have it, as I can define functions in my browser console?
Anyone know about Matlab in that regard?
 
@Adám If you count that, then you count anything with a repl
 
I mean, a lot of repls have something similar. But it's not like you can load a workspace
 
7:51 PM
@Pavel Yeah, that's what I though too.
@H.PWiz Ah, maybe the distinction should be whether it can be saved and loaded.
The original question (from another APLer) was:
"Of course, most other languages don't have an active workspace containing execution state, which might complicate things a bit."
What are the ones, besides APL, that do have an active workspace containing execution state? Anybody know?
 
@Adám What is execution state?
 
@H.PWiz Unclear, except that in traditional APL, the workspace contains EVERYTHING.
 
Python, you can pickle globals() to save it.
 
@H.PWiz A saved APL workspace (i.e. a file) is basically a memory dump, which can be loaded back into the computer's RAM and you can resume wherever you left off. In a sense, APL implemented hibernation half a century ago.
You can hit an error which causes execution to be suspended with a huge stack, save your workspace, shut down the computer or even move the workspace file to a different computer with different hardware, memory architecture (32 vs 64 bit), and OS, load your workspace (in a different version of the same APL flavour) and continue execution as if nothing happened.
 
Mathematica notebooks are a bit different
 
8:03 PM
@Adám I believe SQF (Arma 3 scripting language) has a very similar mechanism for pausing and resuming execution, and transferring partially executed scripts from / to server / client.
Not sure if it counts though.
 
They aren't immediatly evaluated when loaded, instead, it saves all your code and lets you reavluate it.
You can have "initialization cells" and evaluate them all at once
 
@Pavel Right, in normal cases APL workspaces don't do anything when you load them either, but you can tell APL to do something upon workspace load, and if the workspace was in a suspended state when saved, you get that stack back. The most extreme is saving the workspace under program control, which will then cause the system to continue from that point when loading.
 
Mathematica Notebooks are more a way of organizing your code
 
I don't know of another general-purpose language that would come as close as SQF (which is very not general-purpose) though.
 
Not just code though, you can also use formatting and define normal text cells.
So you can have like a research paper with interactive examples built right in
 
8:10 PM
Is anyone else getting a really weird padding error here?
 
like what
no I don't think so
 
 
refresh?
 
It's been like this for several days each time I've popped in
Refreshing doesn't help
 
Hard refresh?
 
8:13 PM
ctrl + f5? what browser are you using?
 
Nor that apparently
 
hmmmmmmmm...
 
Perhaps it's due to the recent CSS changes
 
but it doesn't happen for me
 
Chrome Version 65.0.3325.146 (Official Build) (64-bit)
Tis like this in every chatroom
 
8:14 PM
hm, I'll update and see
 
Maybe I should report something to meta
 
@Blue it works fine for me. Same browser.
 
huh
 
Plugin or userscript?
 
Works on my phone's chrome as well.
 
8:15 PM
No plugins I don't think
oh
 
oh?
 
Apparently the colourblind simulator plugin I'm using is killing it
 
the what
 
Why
 
No idea why
Turning it off fixes it though
 
8:17 PM
No I mean why are you simulating colour blindness
 
My uni coursework is marked on how accessible it is
Usability for colourblind people was very much mentioned
 
0
A: Sandbox for Proposed Challenges

Magic Octopus UrnName... That... N-gon! Math prefixes have been around for a very long time, and there is actually two main-stream sets that are used today (Greek and Latin). These prefixes most commonly align to shapes, and expressing how many sides are a part of a polygon. Usually, for anything above 20 we c...

 
@Blue make it a nice tasteful grey scale :P
 
:P
 
Or grey with a single tone highlight.
 
8:19 PM
God it's nice knowing artists now
 
"pay sace"?
 
sage :P
How am I the only one who could read that font first time :P
 
I think it's pa(upside down lambda) sage
 
xD
 
what a secret site
 
8:32 PM
(anyone up for a game of spyfall? we need just one person more .-.)
 
tio.run/##S8ksKqn8/9/QwMBA71FHv@L////yC0oy8/OK/… the feeling when you wanted to see how fast it was but instead it reminded you that you haven't finished implementing math.
 
 
9:02 PM
I just downloaded the Python plugin for IntelliJ. I'm kind of a big deal
 
@Poke So, not PyCharm?
 
@Pavel that's right
 
But... why?
And how is this special?
 
^
 
there's a directory in a java project i'm working on that has various scripts
and one of them is written in python
and i wanted syntax highlighting
problem solved
It is in no way special. That's why it's a big deal :]
 
9:16 PM
@Poke this is a new one to me thank you
I'm adding this to my collection
 
breakfast club, man
 
tfw you find a rather active IRC chat.. for minecraft? not what I expected
 
Well, SE doesn't allow 9 year olds to join :P /s
 
The sandbox hit 2,000 answers yesterday, and 2018 answers today haha.
 
0
A: Sandbox for Proposed Challenges

Magic Octopus UrnImage to HTML ASCII-Art Given an input image n, output each pixel as a 0 using HTML for coloring and <br/> as a delimiter for the rows. Example: Input Image: Example HTML Output: https://pastebin.com/jDHZwb4P Rendered Output: Rules: You may color the 0's using any HTML trick you...

 
9:43 PM
0
Q: Calculate the Peirce quincuncial projection of the Earth

PyRulezThe Peirce quincuncial projection is pretty neat. It actually tessellates the whole plane: Your job is to generate this map projection. To simplify things, you will only need to generate the northern hemisphere. Also, you will be given the stereographic projection of the northern hemisphere ...

 
9:53 PM
@Pavel both congrats on hnq and congrats on coming to the nethack side
 
@Riker I've been playing nethack for months (still crap at it tho)
Thanks though
 
ಠ_ಠ For once I wish Python did a JS and didn't care if extra function arguments were passed
 
@cairdcoinheringaahing why.
@cairdcoinheringaahing also it's called *ignored...
 
I'm adding version support to Add++ and all but the latest version have the main function take 2 arguments, but the latest version requires 4, so I've been adding _, __ to about 20 different files
 
@cairdcoinheringaahing you should really have used *ignored...
@cairdcoinheringaahing why do you even have 20 different files...
use git -> magic
 
10:04 PM
There are many better design choices I could've done :P
@ASCII-only I can't use git, and can't be bothered to learn
 
@cairdcoinheringaahing ...
:/ then don't complain when you're making it harder for yourself
 
I never complained, just remarked on a feature I wish Python had
 
@cairdcoinheringaahing ... if it was there it wouldn't be a feature it would be a bug
 
Happy Pi Day from Europe!
with an inferior date format
 
Neat, LetsEncrypt wildcard certs are live.
 
10:13 PM
CMP: Nethack display mode of choice? TTY, Curses, Tiles?
I use curses
 
Is there an extension that adds a side of the force for users, where the side is defined by code-gold rep vs code-review rep?
 
10:29 PM
@jrtapsell not yet
@jrtapsell also an extension is overkill, a userscript is enough
 
10:42 PM
@ASCII-only it's treason then
 
@cairdcoinheringaahing **kwargs?
 
11:05 PM
o/ Hola amigos.
 
I'll wait though
 
well, if it's uncursed then go for it
 
o/ taco
 

 CG&CC-gaming

For CG&CC users to discuss and play video games and board game...
maybe?
 
11:06 PM
65 messages moved to Nethack
 
(I've seen some other discussion is going on there)
oh we have a nethack room apparently
 
eh, already moved to nethack
there's 2: PPCG and arqade
 
Oh yeah, It's backwards pi day in Australia.
 
Anyone know how to make \ go LTR in APL?
 
@ATaco .... grats?
 
11:15 PM
What do you mean by "backwards"?
 
oh, I forgot it was the 14th there
 
It's 14/3
 
just fyi: tha'ts forwards pi day not among heathens
 
It's not backwards pi...
I think you mean if the order of the dates are backwards
 
yes it is
 
11:16 PM
3.41 ???
 
everything in AUS is backwards
did you not know that?
they put their toast on their vegemite
 
Anonymous
@Zacharý \⌽?
 
no way that works
 
no... that doesn't work.
Or, equivalently, how to make Jelly's cumulative reduction (whatever that was) go RTL.
 
Anonymous
So you have a function A and an array B, and you want to reduce A over B, starting from the last element in B?
 
11:18 PM
but if {⍺⍺/⍵} goes RTL, then {⍺⍺⍨/⌽⍵} goes LTR
 
@Mego RTL refers to the associativity, so RTL: f\a b c d = a (a f b) (a f (b f c)) (a f (b f (c f d)))
LTR has the parenthesis going the other way
 
that's cumulative reduction though
 
Which is what was initially asked about ...
 
Anonymous
@H.PWiz Ah, I thought you meant reducing RTL, not switching associativity
 
Yeah RTL refers to evaluation order (loosely)
 
11:27 PM
for cumulative reduce, it's actually much harder
actually, it's the reversed reduce as above on each of the prefixes
 
Well, it really depends on how you define LTR cumulative reduce. I mean, APL's isn't exactly cumulative
 
it's actually called scan
 
Thought scan was just an APL term
So it's {⌽⍺⍺⍨\⌽⍵}?
 
Haskell uses scan
 
@Zacharý unfortunately no
you want the scan to go over a, a b, a b c, a b c d
instead, it will go over a b c d, b c d, c d, d
it's not just the ordering, but the values
 
11:41 PM
Wait ... let me look at MY's code
I'm pretty confident it used something equivalent to ,\
 
no, ,\ won't always work
 
def scanr(f,a):
    r = []
    for i in range(len(a)):
        r.append(reducer(f,a[:i+1]))
    return r
 
, is catenate/laminate, and is therefore not a safe way to get prefixes
 
{f/⍵↑a}¨⍳≢a, where a is the argument, and f is the function ?
 
first of all, (⍳∘≢↑¨⊂)⍉ is one function you can use that safely takes prefixes
(or (⍳∘≢↑¨⊂)∘⍉ if not within parentheses)
gtg unfortunately, already 01:50 here lol
 
11:55 PM
Huh, sometimes I don't known if I'm in TNB or the APL Orchard…
 
lol
 
@Zacharý {⍺⍺⍨/∘⌽¨,\⍵} or {⍺⍺⍨⌿∘⊖¨⍪⍀⍵} ?
 
00:00 - 17:0017:00 - 00:00

« first day (2600 days earlier)      last day (2244 days later) »