/dpt/ - Daily Programming Thread

Rust >>>>> C++ >>>>>> Power gap >>>>>>> C > Ada > Fortran > Forth

>Forth at the end
can't even fit this bait into my mouth to bite it

Why do C trannies get mad when people call C a subset of C++?
mcla.ug/blog/cpp-is-not-a-superset-of-c.html

Attached: 8yvcqurvzun11.jpg (490x586, 19K)

I'm a faggot that only likes python
I hate JS
I hate Go
functional autism isn't for me
I want to learn something compiled but nim and crystal are both immature
I plan to learn C/C++ eventually, but for now, I'm going through the rust book

Because it's literally not.

QuickCheck doesn't do formal verification. It tries random inputs until one fails, then it attempts to shrink the failing input to a minimal size (so you don't have to).

I see. Still sounds useful.

>lisp
>failed
wat

because it's not and you shouldn't treat it like it is
recommended idioms, general style, pretty much everything differs between C and C++
they just happen to have some level of compatibility
it's only retards who know neither who insist on the phrase "C/C++" - that or it's one of the rare times when someone is talking about behavior common to both
you aren't a C++ programmer if you don't understand this
you may think you are, but you're not, you're not writing idiomatic modern C++, you're writing shitty c++98 tier C++

It's a false statement. You're just retarded or ignorant for defending it.

Because it isn't. Originally, it was, but that's no longer true since the later changes to C and C++.

It's incredibly useful! I use it regularly in Clojure via the test.check library. It invents test cases I would never think up.

Why put Forth last?
Kek, Ritchie even used a C++ compiler when making r&k.

>when making r&k.
>r&k
> C++
>1970s

Second edition.

>Jow Forums -> Fortran
>Jow Forums -> r&k
Why are there so many boomer programming references on this site?

forchan
fortran

The similarity is obvious to anyone who's not autistic.

Because C++ is not a superset of C. There is perfectly valid C code that fails to compile as C++. Obviously, this falsifies the claim that C is a subset of C++ and you're retarded

>ctrl + f "lua"
>0 results
holy shit you guys are all faggots

nobody talks about Lua because it's boring and Portuguese. It's the very best of scripting languages except for being unusable (by default) for suid scripts since it has an environment variable backdoor.

does getting into anime make me a better programmer? i'm tired of chasing 3d every weekend, it wastes a whole weekend. women are way too needy and securing a regular clam supply is not worth the time and effort, and chasing around birds takes friday and saturday night away from me every week. how do i get into 2d? i've tried but i couldn't find any anime that really interested me. i need suggestions, i like paranormal stuff if that helps.

anime is programming related in this case, because i think it might help me become a better programmer.

struct girl *girl = (struct girl *) malloc(sizeof(struct boy));

Attached: chihiro 6.jpg (1280x720, 467K)

>Casting malloc
Literal fags confirmed for retarded.

you will be sorely disappointed

lua's fucking great
more people should use it

i wish there were COM bindings to the CLR so lua could easily interop with .NET without using a compiler written in a CLR language

You're stretching.

>casting malloc
also
sizeof(struct boy) == 6
sizeof(struct girl) == 8
woops! segfault. it looks like your trying to access data that doesn't exist. i guess reality disagrees with your expectations.

Why is that? I see weebs when I'm at the mall and they always seem to be having fun, laughing about stuff, and generally enjoying themselves.

>fuck seven bitches (ever in your life)
>get casted into a girl
>suddenly have seven children

Why is the girl bigger?

eventually you'll come to your senses and realize you're a degenerate.

It holds isPregnant

kek

Tag based image management (fap enhancement) software.

you mean a void pointer to another struct girl??

I'm already a degenerate. I chase around women 10 years younger than myself trying to dip my dirty old beak in their sweet young nectar.

I try to keep it related to it's function, though that depends on how frustrated I am. Some times my code does look like though.

naming and cache invalidation are the two hardest problems in computer science.
This is true and not a joke or a meme.
You don't solve it. You just keep doing it.

It's brazilian actually

>Ritchie even used a C++ compiler when making r&k
[citation needed]

"hos are useless" - simon peyton jones

Attached: useless.jpg (480x360, 16K)

he's just too lazy to fuck them.

don't want to cause side effects

sex is a function that returns a maybe(list(child)). The pleasure is the side effect, degenerate.

>I do say madam your crumpets are looking exquisite this afternoon
>Care for a tea and shag at my flat?

hahaha hilarious joke definitely not cringe

Attached: 1565361823321.png (1122x852, 466K)

What's stoppiing me from writing an x86 emulator in CUDA and having access to thousands of threads?

Attached: 1567115198618.jpg (598x522, 60K)

Here's a reasonably sized, reasonably understandable, reasonably comparable Rust program: rosettacode.org/wiki/Hunt_The_Wumpus/Rust
how does /dpt/ rate it?
the slowness of each thread.

branching

>println!("There are no tunnels from here to that room. Where do you wanto do go?");
nevermind it's shit.

What if I use a single instruction ISA?

en.wikipedia.org/wiki/One_instruction_set_computer

JEEEESUS FUCKING CHRIST WHY IS INXI SO FUCKING SLOW? FUCK FUCK FUCK FUCK.
[user0@primarypc ~]$ time inxi -F -xxx -v 8 > /dev/null

real 0m1.273s
user 0m0.466s
sys 0m0.171s
[user0@primarypc ~]$ inxi -C
CPU:
Topology: 8-Core model: AMD Ryzen 7 1700X bits: 64 type: MT MCP
L2 cache: 4096 KiB
Speed: 3266 MHz min/max: 2200/3400 MHz Core speeds (MHz): 1: 2872 2: 3123
3: 1714 4: 1685 5: 2406 6: 2409 7: 2404 8: 2410 9: 1746 10: 1750 11: 1749
12: 1740 13: 2334 14: 2342 15: 2401 16: 2410

>user0
doxxed

what the fuck is inxi? github.com/smxi/inxi
>FEATURES AND FUNCTIONALITY
>---------------------------------------------------------------->---------------
>
>inxi's functionality continues to grow over time, but it's also important to
understand that each core new feature usually requires about 30 days work to
get it stable. So new features are not trivial thing
...
>The core mission of inxi is to always work on all systems all the time
...
>README for inxi - a command line system information tool
I haven't been this baffled since I couldn't tell from Eclipse's website that it was any of
1. related to code
2. related to java
3. an editor or an IDE

also it's Perl so it's probably slow because it's looking for and loading modules. You need a fast disk, not a fast CPU.

inxi -D
Drives:
Local Storage: total: 2.84 TiB used: 55.11 GiB (1.9%)
ID-1: /dev/sda vendor: Seagate model: ST1000DM010-2EP102 size: 931.51 GiB
ID-2: /dev/sdb vendor: Kingston model: SUV400S37120G size: 111.79 GiB
ID-3: /dev/sdc type: USB vendor: Western Digital model: WD20EZRZ-00Z5HB0
size: 1.82 TiB

I webcrawled infowars.

Attached: 63DEFFC5-8F5F-4C56-A98A-0BA732985C9B.png (533x672, 380K)

wtf
I meant to show that I'm using an SSD

>anime
>infowars
How does a sane person sit through these without visibly cringing?

ok. well, actually looking at it, it's slow because of all the shit it calls.
Take a look at this:
strace -vtfff -s100 time perl inxi -F -xxx -v 8 2>&1 |grep -Po 'execve\([^,]+, \[[^\]]+'

inline void safe_exit()
{
asm volatile (
"jmp label_after\n"
);
}

0x56556d3e push ebp
0x56556d3f mov ebp, esp
0x56556d41 call 0x56556d53
0x56556d46 add eax, 0x22ba
0x56556d4b jmp 0x56556ab1
0x56556d50 nop
0x56556d51 pop ebp
0x56556d52 ret

gcc, what the heck man

godbolt.org/z/f4DWK4

-O3 in that example will just return 0 since nothing is done in any of the functions, though.

the point is, there's nothing weird about how safe_exit() gets handled. The assembler is inlined into the 'caller' with no extra baggage as you wanted.
godbolt.org/z/JLoOrL

(C label is from me trying that in a dosomething() function)

#include

auto main() -> int{
return EXIT_SUCCESS;
}

Fuck off

bfilipek.com/2019/08/newnew-align.html

#include
#include

inline static constexpr const auto no_string = "no";
constexpr const auto no() -> decltype(no_string) {
return no_string;
}

auto main() -> int{
std::cout

*inline static constexpr const auto no()

can you do something with some constexpr friends?

aside from this one trick you've seen before with the compile time state?
no

Attached: 1564785644049.png (1541x1208, 125K)

You need to throw some more namespaces or some shit in there.

this takes 7.625x as long (and 2.8x as much memory) to compile as
#include

int main() {
puts("no");
return 0;
}
... and then Zig takes 2.6x as long as even C++.

good idea

#include
#include
namespace nolib::no_string_space{
namespace {
inline static constexpr const auto no_string = "no";
}
}

namespace nolib::no_function_space{
namespace {
inline static constexpr const auto no() -> decltype(::nolib::no_string_space::no_string) {
return ::nolib::no_string_space::no_string;
}
}
}

auto main() -> int{
std::cout

>join leetcode
>get stuck at "reverse the digits of an integer"
life is over

look up the 'modulo' operator in your language. It's probably spelled %

As an intern I've been given admin level privileges to the production database. Instead of giving me a test database they had me do development on the production database. Now my boss is angry about some rows in the table getting edited. Why wasn't I given a test database for testing my code???

What are some of the joys of programming?

Working on a socket application. Trying to add a high-level read function that can receive multiple transport-layer packets corresponding to a single application-layer packet. My question is, if I have multiple different sockets waiting to be read from, and say, the one currently being read from times out (i.e a slow connection, or a connection that gets dropped), could that potentially mean the program gets bogged down and can't do any other work?

Maybe each tick of the program I could make threads for each socket that's ready to be read from to make use of thread-level parallelism. However, say I want to do an application-level broadcast to all the connected sockets by writing to each one. Would this break the program if a socket that's currently being read from is written to? Is there a way to make sockets full-duplex?

That's the dumbest question. I don't get the point of that one. Maybe if you're doing some embedded coding it would be nice to show off your ability to do some bit manipulation, but no one is doing that in user space code.

>Why wasn't I given a test database for testing my code???
Pursuit of this inquiry leads to eventual insanity.

religious wars
runtime introspection
editor ricing
unemployment

Why should I use lua when I basically want to call C functions based on strings from a test file. I can make a hash table of and compare strings in the file to the ones in the table and call the function I want.

If you're using asynchronous I/O it doesn't really matter as long as you handle the socket teardown gracefully (i.e. if API throws an exception you should handle it).

>if a read blocks for a long time, could that block my program
yeah, duh. so either don't use blocking read, or use it in a thread by itself, or use it only after something like epoll() has told you a read definitely won't block.
>say I want to do an application-level broadcast to all the connected sockets by writing to each one. Would this break the program
writes can also block, but I'm not as sure how bad this is in the real world.

>trying to write a macro in scheme after knowing CLs defmacro
Yikes

>big joy
working on some complex really intricate bit of code for hours and it compiles and runs perfectly the first time
>medium joy
finding out about a neat new thing/feature/compiler setting
>little joy
finding a library that has followed proper conventions and not created a bunch of pointless typedefs of standard types

Mate, your code is not even safe.

mod no {
pub mod traits {
pub trait No {
type No;
fn no() -> Self::No;
}
}
pub mod types {
pub struct No {
phantom: std::marker::PhantomData,
}

impl

Yeah, I poll to get which sockets are ready to read from, then make threads for each of them to actually do their reading, and logic for what to do based on what was read. After that, the thread would exit.