我是Python的新手,但需要修改别人创建的代码。我无法发布完整代码,但以下是我发布的大部分内容:
from bs4 import BeautifulSoup
import datetime
import getpass
from gmail import Gmail
from selenium import webdriver
from selenium.common.exceptions import NoSuchElementException
from selenium.common.exceptions import ElementNotVisibleException
from time import sleep
from selenium.common.exceptions import NoAlertPresentException
from selenium.webdriver.support import expected_conditions as EC
def soupify(session, url):
"""
Makes parse-able HTML from any given URL.
:param session: requests.Session()
:param url: str
:return: BeautifulSoup object
"""
while True:
try:
r = session.get(url)
break
except Exception as e:
print(e)
return BeautifulSoup(r.content, 'html.parser')
def create_http_session():
"""
Quick little function for returning a requests.Session() instance
with a properly set User-Agent header.
:return: requests.Session()
"""
session = requests.Session()
session.headers.update({'User-Agent':
'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36'
})
return session
def retrieve_hidden_input(soup, name):
return soup.find('input', attrs={
'type': 'hidden',
'name': name
}).get('value')
class AmericanHomesScraper:
def __init__(self, username, password, testing):
self.username = username
self.password = password
self.testing = testing
self.chrome_session = webdriver.PhantomJS()
self.chrome_session.maximize_window()
self.g = Gmail()
self.g.login(username, password)
self.index = 0
def check_for_new_emails_from_sender(self,
sender='info@mailer.netflix.com'):
self.index += 1
if self.index % 1000 == 0:
print('Checking emails - {0}.'.format(datetime.datetime.now()))
elif self.index == 1:
print('Checking emails - {0}.'.format(datetime.datetime.now()))
else:
pass
for s in sender:
messages = self.g.inbox().mail(sender=s,
unread=True)
for message in messages:
message.read()
print(
'Email from {0}: {1}.'.format(s, datetime.datetime.now()))
self.check_for_listings()
self.g.logout()
self.g = Gmail()
self.g.login(self.username, self.password)
def login(self, username, ahsPassword):
self.ahsPassword = ahsPassword
self.chrome_session.get('https://www.aaa.com/Login.aspx')
self.chrome_session.find_element_by_xpath(
'//*[@id="txtUsername"]'
).send_keys(username)
self.chrome_session.find_element_by_xpath(
'//*[@id="txtPassword"]'
).send_keys(ahsPassword)
self.chrome_session.find_element_by_xpath(
'//*[@id="btnSubmit"]'
).click()
def check_for_listings(self):
#code block
links = self.chrome_session.find_elements_by_class_name('link-record')
links = [(link.text, link.get_attribute('href').decode('utf-8'))
for link in links]
if len(links) == 0:
print("No work orders available at {0}".format(
datetime.datetime.now())
)
else:
for link_text, link_url in links:
print("Clicking work order {0} at {1}".format(link_text,datetime.datetime.now()))
self.chrome_session.get(link_url)
print("Attempting to accept at {0}".format(datetime.datetime.now()))
try:
self.chrome_session.find_element_by_xpath("//input[@value='Accept']").click()
try:
WebDriverWait(self.chrome_session, 1).until(EC.alert_is_present)
self.chrome_session.switch_to().alert().accept()
print("Accepted work order {0} at {1}.".format(link_text,datetime.datetime.now()))
except:
print "no alert"
except ElementNotVisibleException:
print("Accept input not found at {0}".format(datetime.datetime.now()))
self.chrome_session.back()
def main():
username = raw_input(
'Please enter the username of the GMail account you want to monitor.\n>')
password = getpass.getpass(
'Please enter the password of the GMail account you want to monitor.\n>')
ahsPassword = getpass.getpass('Please enter the password for ahs. \n>')
ahs = AmericanHomesScraper(username, password, testing)
ahs.login(username,ahsPassword)
print("Starting script")
while True:
ahs.check_for_new_emails_from_sender([
'crm@aaa.com',
'ds@aaa.com',
])
if __name__ == '__main__':
main()
这段代码可以正确地找到并点击“接受”输入按钮。点击后,会弹出一个JavaScript警告(确定/取消)以确认接受。但是,该脚本无法找到结果警报。或者至少不能接受它,因为会调用异常。
如您所见,我已经尝试了
switch_to().alert()
,但它没有起作用。我做错了什么?非常感谢您的帮助,我已经花费了数小时时间在这上面。
更新
此代码在虚拟服务器上运行,不使用用户界面。我刚意识到驱动程序是PhantomJS,而不是Chrome。所以,这可能是失败的原因。我已经更新了问题以反映这一点。
我尝试了类似问题的建议,但它没有起作用。第三方网站需要警报确认才能完成流程,但我的脚本无法看到它,因为它是无头的。