#开始循环爬取
row = 1 #初始化写入元素的行
book = xlwt.Workbook() #打开一个excel
sheet1 = book.add_sheet('sheet1', cell_overwrite_ok=True)
heads = ['任务分类ID', '任务分类名称', '任务ID', '任务名称', '分发UID', '创建时间', '任务状态', '执行时间',
'完成时间', '责任人姓名', '责任人工号', '所属分组ID', '所属分组名称', '所属部门ID', '所属部门名称']
print(u'\n准备将数据存入表格...')
for x in range(int(page)):
# 爬取数据对应的html
soup = BeautifulSoup(driver.page_source, 'html.parser')
content = soup.find_all('td',
class_=['td-0', 'td-1', 'td-2', 'td-3', 'td-4', 'td-5', 'td-6', 'td-7', 'td-8', 'td-9',
'td-10', 'td-11', 'td-12', 'td-13', 'td-14'])
#将信息放入一个list中,创建new_list(方便后续存入excel)(语法糖)
data_list=[]
for data in content:
if data.text =='':
data_list.append(' ')
else:
data_list.append(data.text)
new_list=[data_list[i:i+12] for i in range(0,len(data_list),12)]
#数据存入excel表格
ii=0
#写入表头
for head in heads:
sheet1.write(0,ii,head)
ii+=1
#写入表元素
for list in new_list:
col=0
for data in list:
sheet1.write(row,col,data)
col+=1
row+=1
driver.find_element_by_class_name("next turn_page").click()
book.save('D:\研究生\实习\捷云\授权服务中心任务完成报表.xls')
print(u'\n录入成功!')