Python 自动下载论文

Posted 2021-07-31 yangbocsu

目录

0  导入相关的库

1  获取URL信息

​2 设置请求头  

3  获取相关论文的DOI列表

4 构建sci-hub下载链接

5 检索及下载

6 完整代码

---

0  导入相关的库

# 导入所需模块
import requests
import re
import os
from urllib.request import urlretrieve

1  获取URL信息

# 获取URL信息
def get_url(key):

    url = 'https://xueshu.baidu.com/s?wd=' + key + '&ie=utf-8&tn=SE_baiduxueshu_c1gjeupa&sc_from=&sc_as_para=sc_lib%3A&rsv_sug2=0'
    return url

https://xueshu.baidu.com/s?wd=python&rsv_bp=0&tn=SE_baiduxueshu_c1gjeupa&rsv_spt=3&ie=utf-8&f=8&rsv_sug2=0&sc_f_para=sc_tasktype%3D%7BfirstSimpleSearch%7D&rsv_n=1

2 设置请求头  

# 设置请求头   用Python 模拟浏览器的行为 
headers = {
    'user-agent':
    'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (Khtml, like Gecko) Chrome/86.0.4240.198 Safari/537.36',
    'Referer': 'https://googleads.g.doubleclick.net/'
}

 

3  获取相关论文的DOI列表

# 获取相关论文的DOI列表
def get_paper_link(headers, key):
    response = requests.get(url=get_url(key), headers=headers)
    res1_data = response.text
    #找论文链接
    paper_link = re.findall(r'<h3 class=\\"t c_font\\">\\n +\\n +<a href=\\"(.*)\\"',
                            res1_data)

    doi_list = []  #用一个列表接收论文的DOI
    for link in paper_link:
        paper_link = 'http:' + link
        response2 = requests.get(url=paper_link, headers=headers)
        res2_data = response2.text
        #提取论文的DOI
        try:
            paper_doi = re.findall(r'\\'doi\\'}\\">\\n +(.*?)\\n ', res2_data)
            if str(10) in paper_doi[0]:
                doi_list.append(paper_doi)

        except:
            pass
    return doi_list

4 构建sci-hub下载链接

#构建sci-hub下载链接
def doi_download(headers, key):
    doi_list = get_paper_link(headers, key)
    lst = []
    for i in doi_list:
        lst.append(list(i[0]))
    for i in lst:
        for j in range(8, len(i)):
            if i[j] == '/':
                i[j] = '%252F'
            elif i[j] == '(':
                i[j] = '%2528'
            elif i[j] == ')':
                i[j] = '%2529'
            else:
                i[j] = i[j].lower()
    for i in range(len(lst)):
        lst[i] = ''.join(lst[i])
    for doi in lst:
        down_link = 'https://sci.bban.top/pdf/' + doi + '.pdf'
        print(down_link)
        file_name = doi.split('/')[-1] + '.pdf'
        try:
            with open(file_name, 'wb') as f:
                r = requests.get(url=down_link, headers=headers)
                f.write(r.content)
                print('下载完毕：' + file_name)
        except:
            print("该文章为空")
            pass

5 检索及下载

# 检索及下载
key = input("请输入您想要下载论文的关键词（英文）：")
doi_download(headers, key)

 

6 完整代码

# 导入所需模块
import requests
import re
import os
from urllib.request import urlretrieve

# 获取URL信息
def get_url(key):

    url = 'https://xueshu.baidu.com/s?wd=' + key + '&ie=utf-8&tn=SE_baiduxueshu_c1gjeupa&sc_from=&sc_as_para=sc_lib%3A&rsv_sug2=0'
    return url

# 设置请求头
headers = {
    'user-agent':
    'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36',
    'Referer': 'https://googleads.g.doubleclick.net/'
}

# 获取相关论文的DOI列表
def get_paper_link(headers, key):
    response = requests.get(url=get_url(key), headers=headers)
    res1_data = response.text
    #找论文链接
    paper_link = re.findall(r'<h3 class=\\"t c_font\\">\\n +\\n +<a href=\\"(.*)\\"',
                            res1_data)

    doi_list = []  #用一个列表接收论文的DOI
    for link in paper_link:
        paper_link = 'http:' + link
        response2 = requests.get(url=paper_link, headers=headers)
        res2_data = response2.text
        #提取论文的DOI
        try:
            paper_doi = re.findall(r'\\'doi\\'}\\">\\n +(.*?)\\n ', res2_data)
            if str(10) in paper_doi[0]:
                doi_list.append(paper_doi)

        except:
            pass
    return doi_list

#构建sci-hub下载链接
def doi_download(headers, key):
    doi_list = get_paper_link(headers, key)
    lst = []
    for i in doi_list:
        lst.append(list(i[0]))
    for i in lst:
        for j in range(8, len(i)):
            if i[j] == '/':
                i[j] = '%252F'
            elif i[j] == '(':
                i[j] = '%2528'
            elif i[j] == ')':
                i[j] = '%2529'
            else:
                i[j] = i[j].lower()
    for i in range(len(lst)):
        lst[i] = ''.join(lst[i])
    for doi in lst:
        down_link = 'https://sci.bban.top/pdf/' + doi + '.pdf'
        print(down_link)
        file_name = doi.split('/')[-1] + '.pdf'
        try:
            with open(file_name, 'wb') as f:
                r = requests.get(url=down_link, headers=headers)
                f.write(r.content)
                print('下载完毕：' + file_name)
        except:
            print("该文章为空")
            pass
# 检索及下载
key = input("请输入您想要下载论文的关键词（英文）：")
doi_download(headers, key)