2.格式啥的太烦
今天忽然就想用百度学术里面的“引用”来将PDF上面的参考文献写成标准的形式，然后直接copy去word里面 思路如下：
1.先把pdf参考文献直接copy到txt里，整个过程不需要花费超过五秒吧 2.从txt里用正则表达式匹配到论文名字，然后在百度学术里面搜索
3.点击引用（注意实际过程中可能会出现两个不一样的页面） 4.把引用内容爬下来放在另一个txt里
这个网络密集型可以用多线程实现，但是懒得搞了，后面再改吧，很简单 注意，使用webdriver必须要先搭配环境装一个exe 很简单 大家可以搜一下
最后效果如下：
pdf里面的：
爬虫写下来的：
代码贴下：import selenium.common.exceptions as sele_errorfrom selenium.webdriver.support.wait import WebDriverWaitfrom selenium.webdriver.support import expected_conditions as ECfrom selenium.webdriver.common.by import Byfrom bs4 import BeautifulSoupimport seleniumfrom selenium import webdriverimport urllibimport randomimport timeimport refile= "./未找到的参考文献.txt" #记录没有爬取到的参考文献名字file1= "./bbb.txt" #pdf里面的参考文献file_to_w= "./real.txt" #写入的txturl= "http://xueshu.baidu.com/"'''[32] A. Gretton, A. Smola, J. Huang, M. Schmittfull, K. Borgwardt, andB. Schölkopf, ‘‘Covariate shift by kernel mean matching,’’ Dataset shiftMach. Learn., vol. 3, no. 4, pp. 131–160, 2009.''''''<input name="wd" id="kw" maxlength="256" class="s_ipt" autocomplete="off" type="text">'''def get_name(string):names= re.findall(",.*?‘‘(.*?)’’",string,re.S)if len(names)== 0:return ""return names[-1].strip(",")# a= get_name('''# [32] A. Gretton, A. Smola, J. Huang, M. Schmittfull, K. Borgwardt, and# B. Schölkopf, ‘‘Covariate shift by kernel mean matching,’’ Dataset shift# Mach. Learn., vol. 3, no. 4, pp. 131–160, 2009.# ''')# print(a)def get_real_name(driver,name_to_search):# driver= webdriver.Firefox()# driver.set_page_load_timeout(20)try:try:input= driver.find_element_by_id("kw")except sele_error.NoSuchElementException:try:driver.switch_to.default_content()try:input= driver.find_element_by_id("kw")except sele_error.NoSuchElementException:driver.back()locator = [By.ID, "kw"]WebDriverWait(driver, 60, 1).until(EC.presence_of_element_located(locator))input= driver.find_element_by_id("kw")except Exception as e:with open(file,'a+') as f:f.write("%sn"%name_to_search)input.click()input.send_keys(name_to_search)key= driver.find_element_by_id("su")key.click()time.sleep(random.randint(0,3))'''<a class="sc_q c-icon-shape-hover" rel="external nofollow" href="javascript:;" data-link="http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6279353"data-sign="9f7be0c76448d79a7e0fe57d68b5b62e" data-click="{'button_tp':'cite'}"title="引用"><i class="iconfont icon-cite"></i>引用</a>'''try:locator= [By.CSS_SELECTOR,"a[class="sc_q c-icon-shape-hover"]"]WebDriverWait(driver,10,1).until(EC.presence_of_element_located(locator))except:locator = [By.CSS_SELECTOR, "a[class="sc_q"]"]WebDriverWait(driver, 10, 1).until(EC.presence_of_element_located(locator))get_way = driver.find_elements_by_css_selector("a[class="sc_q"]")[0]get_way.click()time.sleep(random.randint(0, 3))locator = [By.CSS_SELECTOR, "div[class="sc_quote_list_item_r"]"]WebDriverWait(driver, 30, 1).until(EC.presence_of_element_located(locator))soup = BeautifulSoup(driver.page_source)driver.back()div_ = str(soup.find_all(name="div", attrs={"class": "sc_quote_list_item_r"})[0])return re.findall(">(.*?)</div>", div_, re.S)[0]get_way= driver.find_elements_by_css_selector("a[class="sc_q c-icon-shape-hover"]")[0]get_way.click()time.sleep(random.randint(0, 3))'''<div class="sc_quote_list_item_r" data-type="GB/T" tabindex="0"id="sc_cit0">Quiñonero-Candela J, Sugiyama M, Schwaighofer A, et al. Covariate Shift by Kernel Mean Matching[C]// MIT Press, 2009:131-160.</div>'''locator= [By.CSS_SELECTOR, "div[class="sc_quote_list_item_r"]"]WebDriverWait(driver, 30, 1).until(EC.presence_of_element_located(locator))soup= BeautifulSoup(driver.page_source)driver.back()div_= str(soup.find_all(name="div",attrs={"class":"sc_quote_list_item_r"})[0])return re.findall(">(.*?)</div>",div_,re.S)[0]except Exception as e:driver.back()print(e)with open(file, 'a+') as f:f.write("%sn" % name_to_search)if __name__ == "__main__":need= str(open(file1,'r').read()).replace("n","").split("[")driver = webdriver.Firefox()driver.set_page_load_timeout(20)driver.get(url)locator = [By.ID, "kw"]WebDriverWait(driver, 60, 1).until(EC.presence_of_element_located(locator))# with open(file_to_w,'a+',encoding="utf-8") as f1:index= 1for value in need:name= get_name(value)if name== "":continuename_to_write= get_real_name(driver,name)if isinstance(name_to_write,str):with open(file_to_w, 'a+', encoding="utf-8") as f1:f1.write("[%d] %sn"%(index,name_to_write))index+= 1