Python asyncio 爬蟲

TwinkleStar發表於2020-04-28
Python爬蟲

import aiohttp, aiofiles
from aiohttp.client_exceptions import ClientConnectionError
import asyncio
import os
import re


RE_IMG_PAGES = re.compile('''<li><a href=["'](https://www.mzitu.com/\d+)["']''')  # 每一個套圖的入口URL
RE_LIST_NEXT_PAGE = re.compile('''next page-numbers" href=["'](https://www.mzitu.com/page/\d+/)["']>''')  # 列表的下一頁
RE_IMG_INFO = re.compile('''<div class="main-image">.+?<img src=["']([^"']+?)["'] alt=["']([^"']+?)["']''')  # 圖片URL和名稱
RE_IMG_NEXT_PAGE = re.compile('''href=["']([^"']+?/\d+/\d+)["']><span>下一頁''')  # 圖片的下一頁
RE_SUB_DIRNAME = re.compile(r'[<>/\\|:*?]')  # 圖片名稱字元過濾


async def download(url, retries=0):
    headers = {'User-Agent': 'Mozilla', 'Referer':'https://www.mzitu.com/'}
    if retries < 3:
        async with aiohttp.request('GET', url, headers=headers, allow_redirects=False, expect100=True) as resp:
            if resp.status == 200:
                return await resp.read()
            else:
                await asyncio.sleep(10)
                return await download(url, retries+1)
    else:
        raise ClientConnectionError


async def save_image(img_url, save_dir=''):
    img = await download(img_url)
    save_dir = RE_SUB_DIRNAME.sub('_', save_dir)  # 過濾目錄名中的不規範字元
    save_path = os.path.join(save_dir, os.path.split(img_url)[-1])

    try:
        # 直接儲存，如果不成功再目錄，再儲存
        async with aiofiles.open(save_path, mode='wb') as img_fp:
            await img_fp.write(img)
    except FileNotFoundError:
        os.mkdir(save_dir)
        async with aiofiles.open(save_path, mode='wb') as img_fp:
            await img_fp.write(img)

    print(save_path)


async def process_list_page(list_page_url):
    list_page = await download(list_page_url)
    list_page = list_page.decode('utf-8')
    img_page_list = RE_IMG_PAGES.findall(list_page)

    # for img_page in img_page_list:
    # 爬取列表中每一個專案，謹慎開啟
    for img_page in img_page_list[:1]:
        await process_img_page(img_page)

    # list_next_page_list = RE_LIST_NEXT_PAGE.findall(list_page)
    # for list_next_page in list_next_page_list:
    #     await process_list_page(list_next_page)
    # 爬取列表中的下一頁，謹慎開啟


async def process_img_page(img_page_url):
    img_page = await download(img_page_url)
    img_page = img_page.decode('utf-8')
    img_info_list = RE_IMG_INFO.findall(img_page)
    for img_url, img_title in img_info_list:
        await save_image(img_url, img_title)

    img_next_page_list = RE_IMG_NEXT_PAGE.findall(img_page)
    for img_next_page in img_next_page_list:
        await process_img_page(img_next_page)


base_url = 'https://www.mzitu.com/'
loop = asyncio.get_event_loop()
loop.run_until_complete(process_list_page(base_url))
本作品採用《CC 協議》，轉載必須註明作者和本文連結

相關文章