\ctexset{
    section = {
        format = \Large\bfseries
    },
    today = old     % 使日期显示为英文格式
}

root@ub22:/# apt-get update
...
获取:4 http://mirrors.aliyun.com/kali kali-rolling InRelease [41.5 kB]
...
错误:4 http://mirrors.aliyun.com/kali kali-rolling InRelease
  由于没有公钥, 无法验证下列签名: NO_PUBKEY xxxxxxxx
...
正在读取软件包列表... 完成
W: GPG 错误: http://mirrors.aliyun.com/kali kali-rolling InRelease: 由于没有公钥, 无法验证下列签名:  NO_PUBKEY xxxxxxxx
E: 仓库 "http://mirrors.aliyun.com/kali kali-rolling InRelease" 没有数字签名.
N: 无法安全地用该源进行更新, 所以默认禁用该源.
N: 参见 apt-secure(8) 手册以了解仓库创建和用户配置方面的细节.

W: http://mirrors.aliyun.com/kali/dists/kali-rolling/InRelease: 密钥存储在过时的 trusted.gpg 密钥环中(/etc/apt/trusted.gpg), 请参见 apt-key(8) 的 DEPRECATION 一节以了解详情.

Warning: apt-key is deprecated. Manage keyring files in trusted.gpg.d instead (see apt-key(8)).

import matplotlib.pyplot as plt
from matplotlib.patches import Polygon
import numpy as np
import random
from scipy.spatial import ConvexHull

# 回溯法计算是否是有效的听牌形式或胡牌形式(不含国士)
def is_jinzhang_OK(arr):
    arr = arr.copy()
    arr.sort()
    if not arr:
        return True
    # 刻子
    if arr.count(arr[0]) >= 3:
        new_arr = arr.copy()
        new_arr.remove(arr[0])
        new_arr.remove(arr[0])
        new_arr.remove(arr[0])
        if is_jinzhang_OK(new_arr):
            return True
    # 顺子
    if arr[0] + 1 in arr and arr[0] + 2 in arr:
        new_arr = arr.copy()
        new_arr.remove(arr[0])
        new_arr.remove(arr[0] + 1)
        new_arr.remove(arr[0] + 2)
        if is_jinzhang_OK(new_arr):
            return True
    # 对子, 两面, 坎张的情况的处理是类似的, 此处略去
    return False

# 列表中对应位置处的元素代表对应数牌是否是有效进张
def jinzhang(arr):
    return [0] + [is_jinzhang_OK(arr + [i]) for i in range(1, 10)]

# 进张种数
def jinzhang_zhongshu(arr):
    return jinzhang(arr).count(True)

# 计算除了这个搭子外, 所有数牌的残余数量
def shengyu_meishu(arr):
    return [0] + [(4 - arr.count(i)) for i in range(1, 10)]

# 进张枚数
def jinzhang_meishu(arr):
    remaining = shengyu_meishu(arr)
    jinzhang_result = jinzhang(arr)
    return sum([(remaining[i] * jinzhang_result[i]) for i in range(1, 10)])

# 生成所有的单张数牌
def generate_all_A():
    return [[i] for i in range(1, 10)]

# 生成所有的 ABCD 型搭子
def generate_all_ABCD():
    arrays = []
    for i in range(6):
        arrays.append([1 + i, 2 + i, 3 + i, 4 + i])
    return arrays

# ...

# 单张数牌 A 的情况
arrays0 = generate_all_A()
x0 = [jinzhang_zhongshu(arr) for arr in arrays0]
y0 = [jinzhang_meishu(arr) for arr in arrays0]
points0 = [(x + random.uniform(0, 0.5), y + random.uniform(0, 0.5)) for x, y in zip(x0, y0)]
points0 += [(x - random.uniform(0, 0.5), y + random.uniform(0, 0.5)) for x, y in zip(x0, y0)]
points0 += [(x - random.uniform(0, 0.5), y - random.uniform(0, 0.5)) for x, y in zip(x0, y0)]
points0 += [(x + random.uniform(0, 0.5), y - random.uniform(0, 0.5)) for x, y in zip(x0, y0)]

# 生成凸包并绘制
hull0 = ConvexHull(points0)
points0_np = np.array(points0)
plt.scatter(x0, y0, color='#FF4B00', alpha=0.2)
poly0 = Polygon(points0_np[hull0.vertices], fill=True, color='#FF4B00', alpha=0.2, linestyle='dashed', label='A')
plt.gca().add_patch(poly0)

plt.xlabel('zhongshu')
plt.ylabel('meishu')
plt.legend()
plt.show()

import os
import time
import argparse
from PIL import Image, ImageOps

# 创建命令行参数解析器, 并解析命令行参数
parser = argparse.ArgumentParser(description='Combine images into a grid.')
parser.add_argument('image_folder', help='The folder containing the images.')
parser.add_argument('image_format', help='The format of the images (e.g., .png, .jpg).')
parser.add_argument('num_rows', type=int, help='The number of rows in the grid.')
parser.add_argument('num_cols', type=int, help='The number of columns in the grid.')
parser.add_argument('--rotate', type=int, choices=[0, 90, 180, 270], help='The angle to rotate each image.')
parser.add_argument('--log', action='store_true', help='Log the process.')
args = parser.parse_args()

timestamp = int(time.time())

# 获取图片文件夹中的所有图片文件, 按文件名排序, 只保留前 num_rows * num_cols 张图片
image_files = sorted([os.path.join(args.image_folder, file) for file in os.listdir(args.image_folder) if file.endswith(args.image_format)])
image_files = image_files[:args.num_rows * args.num_cols]
images = [Image.open(image_file) for image_file in image_files]

# 如果指定了旋转角度, 就旋转每一张图片, 并获取旋转后的宽度和高度
if args.rotate:
    images = [image.rotate(args.rotate, expand = True) for image in images]
    image_widths_and_heights = [(image.getbbox()[2] - image.getbbox()[0], image.getbbox()[3] - image.getbbox()[1]) for image in images]
    image_width = max(width for (width, height) in image_widths_and_heights)
    image_height = max(height for (width, height) in image_widths_and_heights)
else:
    image_width = max(image.size[0] for image in images)
    image_height = max(image.size[1] for image in images)

total_width = image_width * args.num_cols
total_height = image_height * args.num_rows

if args.log:
    print(f'Max image width: {image_width}, max image height: {image_height}\n')
    print(f'Total width: {total_width}, total height: {total_height}\n')

# 遍历图片列表, 将每一张图片粘贴到新的图片上的正确位置
new_image = Image.new('RGB', (total_width, total_height), (255, 255, 255))
for index, image in enumerate(images):
    row = index // args.num_cols
    col = index % args.num_cols

    # 创建一个新的空白图片 background , 大小为最大宽度和高度, 将图片居中粘贴到背景图片上
    background = Image.new('RGB', (image_width, image_height))
    offset = ((image_width - image.size[0]) // 2, (image_height - image.size[1]) // 2)
    background.paste(image, offset)

    # 将 background 粘贴到新的图片上的正确位置
    new_image.paste(background, (col * image_width, row * image_height))

    if args.log:
        print(f'Pasted image {index} at offset: {offset}, at position: {(col * image_width, row * image_height)}\n')

new_image.save(f'output-{timestamp}{args.image_format}')

import os
import time
import argparse
from PIL import Image
import imagehash

# 创建命令行参数解析器, 并解析命令行参数
parser = argparse.ArgumentParser(description='Output images as a PDF file, each image on a separate page, removing duplicates.')
parser.add_argument('image_folder', help='The folder containing the images.')
parser.add_argument('image_format', help='The format of the images (e.g., .png, .jpg).')
parser.add_argument('--rotate', type=int, choices=[0, 90, 180, 270], help='The angle to rotate each image.')
parser.add_argument('--log', action='store_true', help='Log the process.')
args = parser.parse_args()

timestamp = int(time.time())

# 获取图片文件夹中的所有图片文件, 按文件名排序
image_files = sorted([os.path.join(args.image_folder, file) for file in os.listdir(args.image_folder) if file.endswith(args.image_format)])
unique_images = []
hashes = []

for image_file in image_files:
    current_image = Image.open(image_file)
    current_hash = imagehash.average_hash(current_image)
    
    # 检查当前图片与已有图片的相似度
    if not any(current_hash - h < 5 for h in hashes):
        unique_images.append(current_image)
        hashes.append(current_hash)
    elif args.log:
        print(f'Skipped duplicate image: {image_file}')

# 如果指定了旋转角度, 就旋转每一张图片
if args.rotate:
    unique_images = [image.rotate(args.rotate, expand=True) for image in unique_images]

# 使用第一张图片作为封面, 其余图片追加到 PDF 中
if unique_images:
    last_folder_name = os.path.basename(os.path.normpath(args.image_folder))
    pdf_path = f'{last_folder_name}-{timestamp}.pdf'
    unique_images[0].save(pdf_path, "PDF", resolution=100.0, save_all=True, append_images=unique_images[1:])
    
    if args.log:
        print(f'Saved {len(unique_images)} images to {pdf_path}')

import os
import datetime

directory = '.'

for file in os.listdir(directory):
    if file.endswith('.md'):
        full_path = os.path.join(directory, file)
        mod_time = os.path.getmtime(full_path)
        readable_time = datetime.datetime.fromtimestamp(mod_time).strftime('%Y-%m-%d %H:%M:%S')
        update_string = f"updated: {readable_time}\n"

        with open(full_path, 'r', encoding='utf-8') as file:
            lines = file.readlines()

        if len(lines) >= 3:
            lines.insert(3, update_string)
        else:
            lines.append(update_string)

        with open(full_path, 'w', encoding='utf-8') as file:
            file.writelines(lines)

wsl: 检测到 localhost 代理配置, 但未镜像到 WSL. NAT 模式下的 WSL 不支持 localhost 代理.

[experimental]
autoMemoryReclaim=gradual  # gradual  | dropcache | disabled
networkingMode=mirrored
dnsTunneling=true
firewall=true
autoProxy=true