【並列処理】パスワードクラッキング

これをRustのthread poolで実装したい。。

import time
import math
import hashlib
import typing as T
import os
from multiprocessing import Pool

ChunkRange = T.Tuple[int, int]

def crack_chunk(crypto_hash: str, length: int, chunk_start: int, chunk_end: int)-> T.Union[str, None]:
    print(f"Processing {chunk_start} to {chunk_end}")
    combinations = get_combinations(length=length, min_number=chunk_start, max_number=chunk_end)
    for combination in combinations:
        if check_password(crypto_hash, combination):
            return combination
    return


def get_chunks(num_ranges: int, length: int) -> T.Iterator[ChunkRange]:
    max_number = int(math.pow(10, length) - 1)
    chunk_starts = [int(max_number / num_ranges * i) for i in range(num_ranges)]
    chunk_ends = [start_point - 1 for start_point in chunk_starts[1:]] + [max_number]
    return zip(chunk_starts, chunk_ends)

def crack_password_parallel(crypto_hash: str, legnth: int) -> None:
    num_cores = os.cpu_count()
    print("Processing number combinations correctly")
    start_time = time.perf_counter()

    with Pool() as pool:
        arguments = ((crypto_hash, length, chunk_start, chunk_end)
            for chunk_start, chunk_end in get_chunks(num_cores, length))
        results = pool.starmap(crack_chunk, arguments)
        print("waiting for chunks to finish")
        pool.close()
        pool.join()
    
    result = [res for res in results if res]
    print(f"PASSWORD CRACKED: {result[0]}")
    process_time = time.perf_counter() - start_time
    print(f"PROCESS TIME: {process_time}") 

def get_combinations(*, length: int,
    min_number: int = 0,
    max_number: T.Optional[int] = None) -> T.List[str]:

    combinations = []
    if not max_number:
        max_number = int(math.pow(10, length) - 1)
    for i in range(min_number, max_number + 1):
        str_num = str(i)
        zeros = "0" * (length - len(str_num))
        combinations.append("".join((zeros, str_num)))
    return combinations

def get_crypto_hash(password: str) -> str:
    return hashlib.sha256(password.encode()).hexdigest()

def check_password(expected_crypto_hash: str,
    possible_password: str) -> bool:
    actual_crypto_hash = get_crypto_hash(possible_password)
    return expected_crypto_hash == actual_crypto_hash

def crack_password(crypto_hash: str, length: int) -> None:
    print("Processing number combinations sequentially")
    start_time = time.perf_counter()
    combinations = get_combinations(length=length)
    for combination in combinations:
        if check_password(crypto_hash, combination):
            print(f"PASSWORD CRACKED: {combination}")
            break

    process_time = time.perf_counter() - start_time
    print(f"PROCESS TIME: {process_time}" )

if __name__ == "__main__":
    crypto_hash = "8bb0cf6eb9b17d0f7d22b456f121257dc1254e1f01665370476383ea776df414"
    length = 7
    crack_password_parallel(crypto_hash, length)

$ python3 password_cracking_parallel.py
Processing number combinations correctly
Processing 4999999 to 9999999
Processing 0 to 4999998
waiting for chunks to finish
PASSWORD CRACKED: 1234567
PROCESS TIME: 3.9475781959481537

【Rust】rustのthread Pool【並列処理】

use threadpool::ThreadPool;
use std::sync::mpsc;

pub struct Data { n: i32 }

impl Data {
    fn incr(&mut self) { self.n += 1; }
}

fn main() {
    
    let n_workers = 4;
    let pool = ThreadPool::new(n_workers);
    let (tx, rx) = mpsc::channel();
    let v = vec![Data{n:0}, Data{n:1}, Data{n:2}];
    let n_jobs = v.len();
    for mut data in v {
        let tx = tx.clone();
        pool.execute(move || {
            data.incr();
            tx.send(data).expect("channel will be there waiting for the pool");
        });
    }
    let sum: i32 = rx.iter().take(n_jobs).map(|data| data.n).sum();
    println!("sum= {}", sum);
}

sum= 6

thread poolはデフォルトの選択肢?

use threadpool::ThreadPool;
use std::sync::mpsc;
use std::thread;


fn main() {
    
    let n_workers = 4;
    let pool = ThreadPool::new(n_workers);
    let (tx, rx) = mpsc::channel();
    for i in 0..20 {
        let tx = tx.clone();
        pool.execute(move || {
            // println!("{}", thread::current().name().unwrap());
            tx.send(i).expect("channel will be there waiting for the pool");        
        });
    }
    for i in 0..20 {
        let j = rx.recv().unwrap();
        println!("{}", j);
    }
}

【Python】thread Pool【並列処理】

import time
import queue
import typing as T
from threading import Thread, current_thread

Callback = T.Callable[..., None]
Task = T.Tuple[Callback, T.Any, T.Any]
TaskQueue = queue.Queue

class Worker(Thread):
    def __init__(self, tasks: queue.Queue[Task]):
        super().__init__()
        self.tasks = tasks

    def run(self) -> None:
        while True:
            func, args, kargs = self.tasks.get()
            try:
                func(*args, **kargs)
            except Exception as e:
                print(e)
            self.tasks.task_done()

class ThreadPool:
    def __init__(self, num_threads: int):
        self.tasks: TaskQueue = queue.Queue(num_threads)
        self.num_threads = num_threads

        for _ in range(self.num_threads):
            worker = Worker(self.tasks)
            worker.setDaemon(True)
            worker.start()

    def submit(self, func: Callback, *args, **kargs) -> None:
        self.tasks.put((func, args, kargs))

    def wait_completion(self) -> None:
        self.tasks.join()

def cpu_waster(i: int) -> None:
    name = current_thread().getName()
    print(f"{name} doing {i} work")
    time.sleep(3)

def main() -> None:
    pool = ThreadPool(num_threads=5)
    for i in range(20):
        pool.submit(cpu_waster, i)

    print("All work requests sent")
    pool.wait_completion()
    print("All work completed")

if __name__ == "__main__":
    main()

name = current_thread().getName()
Thread-2 doing 0 work
Thread-5 doing 1 work
Thread-4 doing 2 work
Thread-1 doing 3 work
Thread-3 doing 4 work
Thread-2 doing 5 work
Thread-5 doing 6 work
Thread-1 doing 8 work
Thread-4 doing 7 work
Thread-3 doing 9 work
Thread-5 doing 10 work
Thread-1 doing 11 work
Thread-4 doing 12 work
Thread-2 doing 14 work
Thread-3 doing 13 work
All work requests sent
Thread-1 doing 15 work
Thread-4 doing 16 work
Thread-2 doing 17 work
Thread-3 doing 18 work
Thread-5 doing 19 work
All work completed

スレッドプールを実装してそれぞれのスレッドで並列処理を行なっているのはわかるんだが、
callbackとqueue, typingの使い方がイマイチよくわからん…

【Rust】rustでunix domain socket【並列処理】

use std::io::prelude::*;
use std::os::unix::net::{UnixListener, UnixStream};
use std::path::Path;

fn handle_client(mut stream: UnixStream) -> std::io::Result<()> {
    let mut buf = [0; 1024];

    let n = stream.read(&mut buf)?;
    let s = String::from_utf8_lossy(&buf[..n]);
    println!("{}", s);

    Ok(())
}

fn main()-> Result<(), Box<dyn std::error::Error>> {
    let sockfile = Path::new("/tmp/uds.sock");
    if sockfile.exists() {
        fs::remove_file(&sockfile)?;
    }

    let listner = UnixListener::bind(sockfile)?;
    for stream in listner.incoming() {
        let stream = stream?;
        thread::spawn(move || handle_client(stream).unwrap());
    }

    Ok(())
}

Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.28s
Running `target/debug/parallel`

これは基本的に、socketのincomingの待ち受け

use nix::unistd::{fork, getpid, getppid, ForkResult};
use std::thread;
use std::fs;
use std::io::prelude::*;
use std::os::unix::net::{UnixListener, UnixStream};
use std::path::Path;

use futures::StreamExt;
use tokio;
use tokio::io::AsyncReadExt;
use tokio::net::unix::{UnixListener, UnixStream};

async fn handle_client(mut stream: UnixStream) -> Result<(), Box<dyn std::error::Error>> {
    let mut buf = [0; 1024];

    let n = stream.read(&mut buf).await?;
    let s = String::from_utf8_lossy(&buf[..n]);
    println!("{}", s);

    Ok(())
}

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // let mut stm = UnixStream::connect("/tmp/uds.sock");
    // stm?.write_all(b"hello world")?;

    let sockfile = Path::new("/tmp/uds.sock");
    if sockfile.exists() {
        fs::remove_file(&sockfile)?;
    }

    let listner = UnixListener::bind(sockfile)?;
    let mut incoming = listner.incoming();
    while let Some(stream) = incoming.next().await {
        let stream = stream?;
        tokio::spawn(async move {
            handle_client(stream).await.unwrap();
        });
    }

    Ok(())
}

動かない….

やりたいことはこれなんだが、これも上手くいかん…

use std::io::prelude::*;
use std::os::unix::net::{UnixListener, UnixStream};
use std::thread;

pub static SOCKET_PATH: &'static str = "rst.sock";

fn sender() {
    let name = "Sender".to_string();
    let mut stream = UnixStream::connect(SOCKET_PATH).unwrap();
    let messages = vec!["Hello", " ", "world!"];
    for message in messages {
        stream.write_all(b"hello world").unwrap();
        let mut response = String::new();
        stream.read_to_string(&mut response).unwrap();
        println!("{response}");
    }
}

fn handle_client(mut stream: UnixStream) -> std::io::Result<()> {
    let mut buf = [0; 1024];

    let n = stream.read(&mut buf)?;
    let s = String::from_utf8_lossy(&buf[..n]);
    println!("{}", s);

    Ok(())
}

fn receiver() {
    let handle = thread::spawn(move || {
        let listner = UnixListener::bind(SOCKET_PATH).unwrap();
        for stream in listner.incoming() {
            let stream = stream.unwrap();
            thread::spawn(move || handle_client(stream).unwrap());
        }
    });
    handle.join().unwrap();
}

fn main() {
    receiver();
    sender();
}

【python】UNIXドメインソケットによるデータの受け渡し【並列処理】

ソケット通信は異なるサーバ間同士でのやりとりかと思っていたが、別々のスレッド同士でもできるのね。

import socket
import os.path
import time
from threading import Thread, current_thread

SOCK_FILE = "./mailbox"
BUFFER_SIZE = 1024

class Sender(Thread):
    def run(self) -> None:
        self.name = "Sender"
        client = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
        client.connect(SOCK_FILE)

        messages = ["Hello", " ", "world!"]
        for msg in messages:
            print(f"{current_thread().name}: Send: '{msg}'")
            client.sendall(str.encode(msg))
        
        client.close()

class Receiver(Thread):
    def run(self) -> None:
        self.name = "Receiver"
        server = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
        server.bind(SOCK_FILE)
        server.listen()
        print(
            f"{current_thread().name}: Listening for incoming messages...")
        conn, addr = server.accept()

        while True:
            data = conn.recv(BUFFER_SIZE)
            if not data:
                break
            message = data.decode()
            print(f"{current_thread().name}: Received: '{message}'")
        server.close()

def main() -> None:
    if os.path.exists(SOCK_FILE):
        os.remove(SOCK_FILE)

    receiver = Receiver()
    receiver.start()
    time.sleep(1)
    sender = Sender()
    sender.start()

    for thread in [receiver, sender]:
        thread.join()

    os.remove(SOCK_FILE)

if __name__ == "__main__":
    main()

$ python3 sockets.py
Receiver: Listening for incoming messages…
Sender: Send: ‘Hello’
Receiver: Received: ‘Hello’
Sender: Send: ‘ ‘
Receiver: Received: ‘ ‘
Sender: Send: ‘world!’
Receiver: Received: ‘world!’

【python】メッセージキューによるデータの受け渡し【並列処理】

q = Queue() で初期化した後は、qでやり取りしていますね。

import time
from queue import Queue
from threading import Thread, current_thread

class Worker(Thread):
    def __init__(self, queue: Queue, id: int):
        super().__init__(name=str(id))
        self.queue = queue

    def run(self) -> None:
        while not self.queue.empty():
            item = self.queue.get()
            print(f"Thread {current_thread().name}: "
                f"processing item {item} from the queue")
            time.sleep(2)

def main(thread_num: int) -> None:
    q = Queue()
    for i in range(10):
        q.put(i)

    threads = []
    for i in range(thread_num):
        thread = Worker(q, i + 1)
        thread.start()
        threads.append(thread)
    
    for thread in threads:
        thread.join()

if __name__ == "__main__":
    thread_num = 4
    main(thread_num)

$ python3 message_queue.py
Thread 1: processing item 0 from the queue
Thread 2: processing item 1 from the queue
Thread 3: processing item 2 from the queue
Thread 4: processing item 3 from the queue
Thread 2: processing item 4 from the queue
Thread 3: processing item 5 from the queue
Thread 1: processing item 6 from the queue
Thread 4: processing item 7 from the queue
Thread 3: processing item 8 from the queue
Thread 2: processing item 9 from the queue

【Rust】rustでパイプを使ったデータの受け渡し【並列処理】

let (mut tx, mut rx) = channel();でチャンネルを開設したとき、tx, rxの型が何かに苦戦しました。結局、pipe_channel::Sender、pipe_channel::Receiverになるのね。相互にやり取りするのではなく、一方的にデータを送るときなどに有効活用できそうです。

use std::thread;
use pipe_channel::*;

fn writer(mut tx: pipe_channel::Sender<String>) {
    let name = "Writer".to_string();
    let handle = thread::spawn(move || {
        println!("{}: Sending rubber duck...", name);
        tx.send("Rubber duck".to_string()).unwrap();
    });
    handle.join().unwrap();
}

fn reader(mut rx: pipe_channel::Receiver<String>) {
    let name = "Reader".to_string();
    let handle = thread::spawn(move || {
        println!("{}: Reading...", name);
        println!("{}: received ({})", name, rx.recv().unwrap());
    });
    handle.join().unwrap();
}

fn main() {
    let (mut tx, mut rx) = channel();
    writer(tx);
    reader(rx);
}

fn print_typename<T>(_: T) {
    println!("{}", std::any::type_name::<T>());
}

Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.29s
Running `target/debug/parallel`
Writer: Sending rubber duck…
Reader: Reading…
Reader: received (Rubber duck)

【Python】パイプによるIPC【並列処理】

reader_conn, writer_conn = Pipe() でpipeを接続してるのはわかります。
共有メモリの場合は、グローバルな変数で値をやり取りしたが、Pipeの場合はその名の通り、一方のスレッドからもう一方のスレッドへ値を渡すときに使われるってことかな。。

from threading import Thread, current_thread
from multiprocessing import Pipe
from multiprocessing.connection import Connection

class Writer(Thread):
    def __init__(self, conn: Connection):
        super().__init__()
        self.conn = conn
        self.name = "Writer"

    def run(self) -> None:
        print(f"{current_thread().name}: Sending rubber duck...")
        self.conn.send("Rubber duck")

class Reader(Thread):
    def __init__(self, conn: Connection):
        super().__init__()
        self.conn = conn
        self.name = "Reader"

    def run(self) -> None:
        print(f"{current_thread().name}: Reading...")
        msg = self.conn.recv()
        print(f"{current_thread().name}: Received: {msg}")

def main() -> None:
    reader_conn, writer_conn = Pipe()
    reader = Reader(reader_conn)
    writer = Writer(writer_conn)
    threads = [
        writer,
        reader
    ]
    for thread in threads:
        thread.start()
    for thread in threads:
        thread.join()

if __name__ == "__main__":
    main()

$ python3 pipe.py
Writer: Sending rubber duck…
Reader: Reading…
Reader: Received: Rubber duck

【Python】共有メモリによるIPC【並列処理】

shared_memory = [-1] * SIZE は、[-1, -1, -1, -1, -1] になり、producerで[0, 1, 2, 3, 4]に変わる。ここがややこしい。

import time
from threading import Thread, current_thread

SIZE = 5
shared_memory = [-1] * SIZE

class Producer(Thread):
    def run(self) -> None:
        self.name = "Producer"
        global shared_memory
        for i in range(SIZE):
            print(f"{current_thread().name}: Writing {int(i)}")
            shared_memory[i - 1] = i

class Consumer(Thread):
    def run(self) -> None:
        self.name = "Consumer"
        global shared_memory
        for i in range(SIZE):
            while True:
                line = shared_memory[i]
                if line == -1:
                    print(f"{current_thread().name}: Data not available\n"
                            f"Sleeping for 1 second before retrying")
                    time.sleep(1)
                    continue
                print(f"{current_thread().name}: Read: {int(line)}")
                break

def main() -> None:
    threads = [
        Consumer(),
        Producer(),
    ]

    for thread in threads:
        thread.start()

    for thread in threads:
        thread.join()

if __name__ == "__main__":
    main()
    

【Rust】Rustで新規のthreadを生成する【並列処理】

use nix::unistd::{fork, getpid, getppid, ForkResult};
use std::thread;
use std::time::Duration;

fn main() {
    thread::spawn(|| {
        for i in 1..10 {
            println!("hi number {} from the spawned thread!", i);
            thread::sleep(Duration::from_millis(1));
        }
    });

    for i in 1..5 {
        println!("hi number {} from the main thread!", i);
        thread::sleep(Duration::from_millis(1));
    }
}

Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.75s
Running `target/debug/parallel`
hi number 1 from the main thread!
hi number 1 from the spawned thread!
hi number 2 from the main thread!
hi number 2 from the spawned thread!
hi number 3 from the spawned thread!
hi number 3 from the main thread!
hi number 4 from the spawned thread!
hi number 4 from the main thread!
hi number 5 from the spawned thread!

joinを呼び出すことで、実行されたことを保証する

fn main() {
    let handle = thread::spawn(|| {
        for i in 1..5 {
            println!("hi number {} from the spawned thread!", i);
            thread::sleep(Duration::from_millis(1));
        }
    });

    for i in 1..5 {
        println!("hi number {} from the main thread!", i);
        thread::sleep(Duration::from_millis(1));
    }

    handle.join().unwrap();
}

所有権の移転

fn main() {
    let v = vec![1, 2, 3]; 
    let handle = thread::spawn(move|| {
        for i in 0..3 {
            println!("hi number {} from the spawned thread!", v[i]);
            thread::sleep(Duration::from_millis(1));
        }
    });

    for i in 1..3 {
        println!("hi number {} from the main thread!", i);
        thread::sleep(Duration::from_millis(1));
    }

    handle.join().unwrap();
}

なんか違うような気がするが…

use nix::unistd::{fork, getpid, getppid, ForkResult};
use std::thread;
use std::time::Duration;
use rayon::current_num_threads;

fn cpu_waster(i: u32) {
    let name = gettid::gettid().to_string();
    println!("{} doing {} work", name, i);
    thread::sleep(Duration::from_millis(5));
}

fn display_threads() {
    let str: String = "-".to_string();
    println!("{}", str.repeat(10));
    println!("Current process PID: {}", getpid());
    println!("Thread Count: {}", current_num_threads());
}

fn main() {
    let num_threads = 5;
    display_threads();

    println!("Starting {} CPU waters...", num_threads);
    let handle = thread::spawn(move|| {
        for i in 0..num_threads {
            cpu_waster(i);
        }
    });
    display_threads();

    handle.join().unwrap();
}

Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.23s
Running `target/debug/parallel`
———-
Current process PID: 582557
Thread Count: 2
Starting 5 CPU waters…
———-
Current process PID: 582557
Thread Count: 2
582600 doing 0 work
582600 doing 1 work
582600 doing 2 work
582600 doing 3 work
582600 doing 4 work