AI摘要
文章介绍Wallhaven壁纸站特色，并给出完整Python爬虫：用tkinter构建GUI，支持关键词、页码、线程数、保存路径配置，多线程下载高清壁纸，可暂停，含异常重试。
导读
Wallhaven是一个令人感到惊艳的壁纸网站，汇集海量精美壁纸，还为用户提供了榜单、精选、专题等，更有分类、标签、搜索、收藏、同步等贴心服务，旨为让用户能够更轻松的发现优质壁纸。
代码
import requests
from lxml import etree
import tkinter as tk
from tkinter import filedialog
import os
import threading
import time
import tkinter.messagebox as messagebox
def select_save_directory():
save_directory = filedialog.askdirectory()
if save_directory:
save_entry.delete(0, tk.END)
save_entry.insert(0, save_directory)
stop_event = threading.Event() # 停止事件，默认为未设置
max_threads = 5 # 最大线程数限制
semaphore = threading.Semaphore(max_threads) # 信号量，用于控制并发线程数
def perform_download(series_name, save_option, page_start, page_end, save_directory):
headers = {
'User-Agent': 'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.107 Mobile Safari/537.36'
}
url_template = 'https://wallhaven.cc/search?q={}&page=%d'
try:
for page in range(page_start, page_end + 1):
if stop_event.is_set(): # 如果停止事件被设置，则终止下载
return
print('正在下载第%d页图片' % page)
url = url_template.format(series_name)
new_url = url % page
page_text = requests.get(url=new_url, headers=headers).text
tree = etree.HTML(page_text)
li_list = tree.xpath('//*[@id="thumbs"]/section/ul/li')
for li in li_list:
detail_url = li.xpath('./figure/a/@href')[0]
detail_page_text = requests.get(url=detail_url, headers=headers).text
tree2 = etree.HTML(detail_page_text)
img_src = tree2.xpath('//*[@id="wallpaper"]/@src')[0]
if save_option == 1:
img_name = img_src.split('/')[-1]
img_path = os.path.join(save_directory, img_name)
img_data = requests.get(url=img_src, headers=headers).content
with open(img_path, 'wb') as fp:
fp.write(img_data)
print(img_name, '下载成功！')
messagebox.showinfo("提示", "图片下载完成！")
except Exception as e:
print('下载过程中出现错误:', str(e))
print('等待5秒后继续尝试...')
time.sleep(5) # 等待5秒后继续尝试爬取
perform_download(series_name, save_option, page_start, page_end, save_directory)
finally:
semaphore.release() # 释放信号量，表示线程结束
def update_thread_count_label():
current_thread_count = max_threads - semaphore._value
thread_count_label.config(text="当前线程数: " + str(current_thread_count))
def download_images(max_threads):
global semaphore
semaphore = threading.Semaphore(max_threads)
stop_event.clear() # 清除停止事件，表示开始下载
# 获取用户输入的系列名称、保存选项和下载页码范围
series_name = series_entry.get()
save_option = save_var.get()
page_start = int(start_entry.get())
page_end = int(end_entry.get())
save_directory = save_entry.get() # 获取保存路径
# 创建并启动线程
for page in range(page_start, page_end + 1):
if stop_event.is_set(): # 如果停止事件被设置，则终止下载
break
semaphore.acquire() # 获取信号量，控制线程数量
t = threading.Thread(target=perform_download, args=(series_name, save_option, page, page, save_directory))
t.start()
update_thread_count_label()
def stop_download():
stop_event.set() # 设置停止事件，表示停止下载
# 创建主窗口
window = tk.Tk()
window.title("图片下载应用程序")
window.geometry("800x450") # 设置窗口大小
#布局容器
frame = tk.Frame(window)
frame.pack(pady=20)
#输入系列名称的窗口
series_label = tk.Label(frame, text="系列名称:")
series_label.grid(row=0, column=0, padx=10, pady=5)
series_entry = tk.Entry(frame)
series_entry.grid(row=0, column=1, padx=10, pady=5)
#保存选项的窗口
save_option_label = tk.Label(frame, text="保存选项:")
save_option_label.grid(row=1, column=0, padx=10, pady=5)
save_var = tk.IntVar()
save_radio1 = tk.Radiobutton(frame, text="下载照片", variable=save_var, value=1)
save_radio1.grid(row=1, column=1, padx=10, pady=5)
#输入页码范围的窗口
page_range_label = tk.Label(frame, text="页码范围:")
page_range_label.grid(row=2, column=0, padx=10, pady=5)
start_label = tk.Label(frame, text="起始页码:")
start_label.grid(row=2, column=1, padx=10, pady=5)
end_label = tk.Label(frame, text="结束页码:")
end_label.grid(row=3, column=1, padx=10, pady=5)
start_entry = tk.Entry(frame)
start_entry.grid(row=2, column=2, padx=10, pady=5)
end_entry = tk.Entry(frame)
end_entry.grid(row=3, column=2, padx=10, pady=5)
#选择保存路径的窗口
save_directory_label = tk.Label(frame, text="保存路径:")
save_directory_label.grid(row=4, column=0, padx=10, pady=5)
save_entry = tk.Entry(frame)
save_entry.grid(row=4, column=1, padx=10, pady=5)
select_button = tk.Button(frame, text="选择路径", command=select_save_directory)
select_button.grid(row=4, column=2, padx=10, pady=5)
#最大线程数选择
thread_count_label = tk.Label(frame, text="最大线程数:")
thread_count_label.grid(row=5, column=0, padx=10, pady=5)
thread_count_var = tk.StringVar()
thread_count_dropdown = tk.OptionMenu(frame, thread_count_var, "1", "2", "3", "4", "5")
thread_count_dropdown.grid(row=5, column=1, padx=10, pady=5)
thread_count_var.set("5") # 默认值设置为5
#启动下载和停止下载按钮
download_button = tk.Button(frame, text="启动下载", width=15, command=lambda: download_images(max_threads))
stop_button = tk.Button(frame, text="停止下载", width=15, command=stop_download)
#设置按钮居中对齐
download_button.grid(row=6, column=1, padx=10, pady=5, sticky="nsew")
stop_button.grid(row=6, column=2, padx=10, pady=5, sticky="nsew")
#显示当前线程数的标签
thread_count_label = tk.Label(frame, text="当前线程数: " + str(max_threads - semaphore._value))
thread_count_label.grid(row=5, column=1, columnspan=3, padx=10, pady=5)
#运行主循环
window.mainloop()