« first day (1990 days earlier)      last day (3237 days later) » 
00:00 - 20:0020:00 - 00:00

user54412
20:05
#!/usr/bin/python
import survive
@ChrisWhite You have to 'import petrified' first.
@ChrisWhite In the past week nothing of what I tried to build in C++ worked
Last time I got this frustrated someone had just explained to me the concept of dating
Carbon dating?
I'll pencil you in...
20:10
I was optimistic about this one then it failed compiling
Why does it fail?
If I recall something about incompatible types on line 29
But everything there is cpp_int so I dunno
@BernardMeurer Was this the same program that was compiling earlier but SEG faulting?
@barrycarter Nope, that was my other project that is still SEG faulting. I'll maybe try to transform recursion into iteration there
@BernardMeurer Well, you should be punished for using the ternary operator anyway. Try writing it as an if-else and see what happens?
20:16
@barrycarter I want to know how to use ternaries because I hate them and they are ugly, thus I'm using it in my practice :)
My vague guess is that the compiler sees 'return d;' somehow
user54412
Can you really recursively define functions without declaring them first?
@BernardMeurer Fine, but undo them for debugging. You're a functional programmer at heart, admit it!
I think he's declaring in one of the header files, but haven't checked.
@ChrisWhite Which function?
user54412
@barrycarter Definitely not. return takes the full expression following it.
20:19
@ChrisWhite In theory, yes...
I assume the spaces are allowed too, though I've always seen it as x?y:z
user54412
@BernardMeurer Like, cpp_int phi(const cpp_int x); somewhere before the definition of the function. Maybe it's not a problem -- I just so rarely see things defined without separate declarations.
user54412
@barrycarter 99.99999% of "compiler bugs" are programmer errors ;)
user54412
For that matter, what is primesieve.hpp?
@ChrisWhite A very, very fast implementation of the Sieve of Erastothenes
My theory of debugging is to randomly tweak things that should be working until they actually start working and then tweak back until they break.
20:22
berna@DESKTOP-7E2QL94 ~/euler-function
$ make
g++ -std=c++14 -Wall -O2 -lboost_system -lboost_thread -lgmp -lprimesieve euler.cpp -o euler.o
euler.cpp: In function ‘boost::multiprecision::cpp_int phi(boost::multiprecision::cpp_int)’:
euler.cpp:29:23: error: operands to ?: have different types ‘boost::multiprecision::detail::expression<boost::multiprecision::detail::multiply_immediates, boost::multiprecision::number<boost::multiprecision::backends::cpp_int_backend<> >, boost::multiprecision::number<boost::multiprecision::backends::cpp_int_backend<> >, void, void>’ and ‘boost::multiprecision:
@BernardMeurer I know the ternary thing SHOULD work, but, seriously, try it without. After you figure out the problem, you can retroactively say "oh, that's what I did wrong"
You are dividing by phi(d), but that's integer division, so you still should be OK, but still.
@barrycarter I'm removing the ternaries for testing
user54412
@BernardMeurer So you compiler thinks phi(k)*phi(o) and phi(k)*phi(o)*d/phi(d) have different types. How did boost define operator* and operator/ for cpp_int?
user54412
(Side note: this is why boost, like all external libraries, sucks.)
Yeah, that might not be integer division.
@ChrisWhite You use NO external libraries when coding?
20:31
@ChrisWhite I couldn't tell you, but maybe this can
user54412
@barrycarter MPI and HDF5, because I don't want to rewrite infiniband calls every time I go to another machine. Other than that, no.
@ChrisWhite And "-lm"? Or do you consider that part of the base.
user54412
That's part of the language, just awkwardly not part of the default linking.
I only use C when I need an external library.
Sigh, now compilation is failing because I'm on windows and idk how to use windows
or rather, I don't know how to develop on windows
20:43
Why would anyone want to develop for Windows?
user54412
@BernardMeurer Well, I can't make heads or tails of that file.
for Windows? Large userbase
on Windows? No clue
@ChrisWhite ;-;
Fair enough. Ages ago, I wrote some Java that does run on Windows.
user54412
Windows goes out of its way to make coding impossibly hard. "Our OS is so special it deserves its own unique API."
@ChrisWhite It's bizarre
I'll run an ArchLinux VM to develop on this
it'll be easier
20:47
Windows does a lot to make sure you can't use it as a true computer. It's more of a limited function quasi-entertainment device.
21:37
Y'all computer-snobs :3
@Danu At least that's the best type of snob :p
A bold claim!
@Danu Which claim? The snob one or the Windows one?
22:01
@vzn "Profile"?
What do you mean?
@ChrisWhite Oh give me a break. Windows is as "unique" as Unix.
:-)
Were they really going against the grain as much as people say?
@DanielSank Yeah, I agree, "apt-get install libsdl && g++ -lSDL2 mycode.cpp mycode" is such a terribly long time to set up a development environment and compile. Visual studio does it way better.
@DanielSank UNIX is much better than windows for development
@BernardMeurer That is very unfair.
Development for Windows on Windows is an very pleasant experience.
@NeuroFuzzy Can't tell if that's sarcasm.
@DanielSank I was wondering the same thing.
@Danu Internets aren't good for tone of voice.
22:15
I'm taking a general ed art course
so consider the ambiguity a form of abstract modern art.
a performance piece.
@DanielSank Oh, don't get me started ;)
@BernardMeurer False. Food snob masterrace.
Hear, hear
@DanielSank You just say that because you can actually cook :p
@BernardMeurer Well, at least unix exposes a much simpler basic internal structure than windows does. Which makes the minimum set of APIs you need to know a lot easier to comprehend.
22:19
You're so lucky I'm not going to UCSB, I'd bother you so much
@BernardMeurer Anyone can cook if you're interested in eating good food + willing to spend time.
@BernardMeurer Well... yeah, duh.
You should try it too :D
@Danu correct.
If you can physics you can cook.
(not to detract from DS's foodery)
22:20
That of course means that it has have the multitudes-of-different-third-party-standards-for [name you favorite advanced service]-problem in spades.
@dmckee My whole point regarding this is based on the fact that I am a beginner developer, one who never had a single programming class and had to figure things out by himself. I can code alright on Linux, I can't do anything on Windows (And I started on Windows before I felt forced to move)
But at least the run-time linker is flexible enough to deal with that.
@DanielSank I can't either :p
I can create Segmentation faults, is that good?
The thing is, though, that I know a number of professional developers who consider it easier to write for the windows ecosystem than for the unix ecosystem even though the APIs are harder.
Because the ecosystem is less fragmented.
@dmckee The tools are much better.
Visual Studio is incredible (so I'm told) and C# is rather nice.
22:23
Real programmers use a magnetic tape and a needle
@DanielSank That's one of the things they say.
@BernardMeurer Have fun saying that and living off of money you make playing guitar for bologna sandwiches on the beach.
@DanielSank Visual studio is amazing, but: 1. CLion also is 2. It's coming to linux soon
@BernardMeurer C-x M-c M-butterfly...
@dmckee Hahaha
22:23
@BernardMeurer "soon"
In computer land that means "maybe".
@DanielSank IT'S OKAY TO HAVE FAITH
ALRIGHT?
What can I say. I'm an emacs guy. Of course I like that cartoon.
@dmckee VIM masterrace (because I can't emacs)
@BernardMeurer I know enough vi or the most important task you can do on a system with vi as the main editor: installing emacs.
22:27
@dmckee So you use vi basically how everyone uses internet explorer?
Of course, my day-to-day machine in a MacBook Pro (because it's a unix laptop with all peripherals working out of the box). On which I write LaTeX documents in emacs.
Peripherals not working on Linux is a myth
I literally never had that problem
@BernardMeurer Well, I used to. When you had to edit config files to get emacs to build. These days the package manager takes care of it.
@BernardMeurer I had it many, many times. Now, I haven't put linux onto a laptop in about five years and even the last time it was getting better, but it was a real problem for a long time.
@dmckee Linux changed a lot, it's a genuinely usable system nowadays
honestly never had problems with it
apart from some issues with Ubuntu
I do think OSX is superior though, and would use it if I had a computer
I've never stopped using linux systems, they just haven't been my main system since I (or the grant I was working) could afford an apple laptop.
The installation process has fewer hiccups these days, but laptops fall into two categories: (a) those that cost as much as an apple machine and have really solid, highly standard hardware and last a long time and (b) those that cost a lot less than an apple machine but share pennies here and there and have the highest risk of trouble with linux and will die in only a few years of heavy use.
22:40
I've been dying to try and use OSX, I never have. It looks like Linux with a good UI and perhaps less 'fuckupability'
@BernardMeurer It's BSD underneath and the standard commandline toolset is closer to raw POSIX than the GNU tools you're used to.
I use brew and install the GNU commandline tools to feel a little bit more at home.
@dmckee I've been considering moving to FreeBSD because it's fully UNIX and completely POSIX compliant
These days apple is using clang rather than gcc for the system compiler, too.
clang is faster than gcc
If you really want gcc you have to install it.
22:45
and the intel compiler beats them all
I feel that Stallman's line of thought is being the bane of some OpenSource projects
@BernardMeurer The speed isn't as big a deal to me and the pickier and more detailed error reports, which I like a lot.
@BernardMeurer No, it's not.
by forcing them to be libre and by not accepting new features
But some of the particle physics software package I use are gcc specific
@DanielSank You don't count :p
22:47
@BernardMeurer >:(
You're peripherals are quantum devices you hack!
@dmckee I compile a lot of large stuff, so the speed ends up being a part of it for me (and my computer is very slow)
I just wish I could use the Intel compiler, it's so beautiful
@BernardMeurer Oh, the ones I work on are big, too, but they have fairly smart builds (cmake based) so unless I touch a lot of files I don't have long compilation waits anyway 'round.
cmake <3
@DanielSank Definitely sarcasm.
@FaheemMitha It's weird because the command line example given has nothing to do with having a nice IDE, so why compare to Visual Studio?
22:56
@DanielSank I think the point is you can get it done easily.
For the record, I use (or used to use) gcc with emacs.
Perfectly usable, imo. Though maybe a custom IDE would be better. Dunno.
The real problem is that C++ is a sucky language.
2
I dream of the day Rust will be a usable alternative to C++
Rust <3
23:11
hello
I want to start coding something new.. . . my last project was so hard I dont code it any more. Other more talented people are doing the heavy lifting in the backend amallu.com
Does anyone have any cool coding projects (not scripting stuff)
. . .
@kevinTahN. Which language?
@BernardMeurer C++ or java. . . let's do c++
@kevinTahN. I was just curious, I don't have any public projects currently
@BernardMeurer I have a small project going on quantconnect , I have a few projects, I am thinking they are kind of just alright for me to put in github now
lol, it is kind of useless though, the code works but the math is so silly even a donkey could have written it. It predicts some stuff though, but it is far from usable, actually, may be I should not put it up, someone might actually try to use it and loose a lot of money lol
00:00 - 20:0020:00 - 00:00

« first day (1990 days earlier)      last day (3237 days later) »