Python实现批量下载文件的代码详解

pip install requests tqdm

[file1]
url = http://example.com/file1.jpg
path = downloads/file1.jpg

[file2]
url = http://example.com/file2.jpg
path = downloads/file2.jpg

[file3]
url = http://example.com/file3.jpg
path = downloads/file3.jpg

import os
import requests
from concurrent.futures import ThreadPoolExecutor
from tqdm import tqdm
import configparser
import argparse

def download_file(task):
    """
    下载单个文件并保存到指定路径，并显示下载进度
    """
    url = task['url']
    path = task['path']
    
    try:
        os.makedirs(os.path.dirname(path), exist_ok=True)
        
        with requests.get(url, stream=True) as r:
            r.raise_for_status()
            total_size = int(r.headers.get('content-length', 0))
            with open(path, 'wb') as f, tqdm(
                total=total_size, unit='B', unit_scale=True, desc=os.path.basename(path)
            ) as pbar:
                for chunk in r.iter_content(chunk_size=8192):
                    f.write(chunk)
                    pbar.update(len(chunk))
        
        print(f'{path} 下载完成')
    except Exception as e:
        print(f'下载 {url} 失败: {e}')

def read_config(config_path):
    """
    从配置文件读取下载任务列表
    """
    config = configparser.ConfigParser()
    config.read(config_path)
    tasks = []
    for section in config.sections():
        url = config[section]['url']
        path = config[section]['path']
        tasks.append({'url': url, 'path': path})
    return tasks

def main(config_path):
    """
    主函数，读取配置文件并并行下载文件
    """
    tasks = read_config(config_path)
    
    with ThreadPoolExecutor(max_workers=5) as executor:
        executor.map(download_file, tasks)

if __name__ == '__main__':
    parser = argparse.ArgumentParser(description='批量下载文件工具')
    parser.add_argument('config', type=str, help='配置文件路径')
    args = parser.parse_args()
    
    main(args.config)

python batch_download.py downloads.ini

import os
import requests
from concurrent.futures import ThreadPoolExecutor
from tqdm import tqdm
import configparser
import argparse

def download_file(task):
    url = task['url']
    path = task['path']
    
    try:
        os.makedirs(os.path.dirname(path), exist_ok=True)
        
        with requests.get(url, stream=True) as r:
            r.raise_for_status()
            total_size = int(r.headers.get('content-length', 0))
            with open(path, 'wb') as f, tqdm(
                total=total_size, unit='B', unit_scale=True, desc=os.path.basename(path)
            ) as pbar:
                for chunk in r.iter_content(chunk_size=8192):
                    f.write(chunk)
                    pbar.update(len(chunk))
        
        print(f'{path} 下载完成')
    except Exception as e:
        print(f'下载 {url} 失败: {e}')

def read_config(config_path):
    config = configparser.ConfigParser()
    config.read(config_path)
    tasks = []
    for section in config.sections():
        url = config[section]['url']
        path = config[section]['path']
        tasks.append({'url': url, 'path': path})
    return tasks

def main(config_path):
    tasks = read_config(config_path)
    
    with ThreadPoolExecutor(max_workers=5) as executor:
        executor.map(download_file, tasks)

if __name__ == '__main__':
    parser = argparse.ArgumentParser(description='批量下载文件工具')
    parser.add_argument('config', type=str, help='配置文件路径')
    args = parser.parse_args()
    
    main(args.config)