|
楼主 |
发表于 2019-2-23 05:17:02
|
显示全部楼层
- import urllib.request
- import urllib.error
- import chardet
- import re
- import easygui as g
- def get_response_r(url):
- req = urllib.request.Request(url)
- req.add_header('User-Agent','Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36')
- try:
- response = urllib.request.urlopen(req)
- except HTTPError as e:
- print('The serve couldn\'t fulfill the request.')
- print('Error code: ',e.code)
- except URLError as e:
- print('We failed to reach a server')
- print('Reason: ',e.reason)
- response_r = response.read()
- return response_r
- def get_code(url):
- req = urllib.request.Request(url)
- req.add_header('User-Agent','Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36')
- try:
- response = urllib.request.urlopen(req)
- except HTTPError as e:
- print('The serve couldn\'t fulfill the request.')
- print('Error code: ',e.code)
- except URLError as e:
- print('We failed to reach a server')
- print('Reason: ',e.reason)
- response_r = response.read()
- encode = chardet.detect(response_r)['encoding']
- return encode
- def get_page(url):
- html = get_response_r(url).decode(get_code(url),'ignore')
- page = re.findall(r'page-numbers current\'>(.+)</span>',html)[0]
- return page
- def find_imgs(page):
- url = 'https://www.mzitu.com/zipai/comment-page-%s/#comments'%page
- imgs_list = []
- html = get_response_r(url).decode(get_code(url),'ignore')
- imgs_list = re.findall(r'data-original="([^"]+\.jpg)',html)
- return imgs_list
-
- def main():
- url = 'https://www.mzitu.com/zipai/'
- page = int(get_page(url))
- #print('该网页获得的页码为%s\n'%page)
- page = g.integerbox(msg='该网页有妹纸%s页,你想从第几页开始看!!!'%page, title=' ', default=None, lowerbound=0, upperbound=page, image=None, root=None)
- page_num = g.integerbox(msg='你目前在%s页,你要看几页!!!'%page, title=' ', default=None, lowerbound=0, upperbound=page, image=None, root=None)
- file_adress = g.diropenbox(msg='请选择存储路径', title=None, default=None)
- for i in range(page_num):
- if page > 0:
- print('\n当前正在获取第%s页的图片\n'%page)
- imgs_list = find_imgs(page)
- for each_imgs in imgs_list:
- imgs_name = '\\'.join([file_adress,each_imgs.split('/')[-1]])
- print(imgs_name)
- urllib.request.urlretrieve(each_imgs,imgs_name)
- else:
- print('没有更多的妹纸了!!!')
- break
- page -=1
-
- if __name__=='__main__':
- main()
-
复制代码 |
|