1

使用 PhantomJS，下面显示的 ActionChain 永远不会执行：

import sys

from selenium import webdriver
from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
from selenium.webdriver.common.action_chains import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC

# have PhantomJS pretend to be Firefox to make things more equal
desired_capabilities = dict(DesiredCapabilities.PHANTOMJS)
desired_capabilities["phantomjs.page.settings.userAgent"] = (
    'Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:21.0) '
    'Gecko/20100101 Firefox/21.0'
)
wd = webdriver.PhantomJS(desired_capabilities=desired_capabilities)
# wd = webdriver.Firefox()

# get a page with some AJAXy comments
wd.get('http://www.cartoonbrew.com/disney/disneys-princess-makeover-of-merida-leads-to-uproar-and-petition-82636.html')

# switch to iframe with comments
wd.switch_to_frame('dsq1')

# these selectors are used throughout
load_more_div_selector = "//div[@class='load-more']"
load_more_button_selector = "//div[@class='load-more']/a"

# wait until the comments are loaded
WebDriverWait(wd, 10, 0.1).until(
    EC.presence_of_element_located((By.XPATH, load_more_div_selector))
)

# print out the status of the load more button (the display is 'block'
# if there are more comments to be loaded), and the button should have
# the class 'btn'
load_more_div = wd.find_element_by_xpath(load_more_div_selector)
load_more_button = wd.find_element_by_xpath(load_more_button_selector)
print >> sys.stderr, '(%s, %s)' % (load_more_div.value_of_css_property('display'), load_more_button.get_attribute('class'))

# click on the 'load more comments' button
action_chain = ActionChains(wd).click(on_element=load_more_button)
action_chain.perform()

# wait until the comments are loaded
def comments_no_longer_loading(wd):
    load_more_div = wd.find_element_by_xpath(load_more_div_selector)
    load_more_button = wd.find_element_by_xpath(load_more_button_selector)
    is_loading = 'busy' in load_more_button.get_attribute('class')
    if is_loading:
        print >> sys.stderr, 'comments are loading, (%s, %s)' % \
            (load_more_div.value_of_css_property('display'),
             load_more_button.get_attribute('class'))
    else:
        print >> sys.stderr, 'comments done loading (%s, %s)' % \
            (load_more_div.value_of_css_property('display'),
             load_more_button.get_attribute('class'))
    return not is_loading

# wait until comments are loaded
WebDriverWait(wd, 10, 0.1).until(comments_no_longer_loading)

# print out the status of the button
load_more_div = wd.find_element_by_xpath(load_more_div_selector)
load_more_button = wd.find_element_by_xpath(load_more_button_selector)
print >> sys.stderr, '(%s, %s)' % (load_more_div.value_of_css_property('display'), load_more_button.get_attribute('class'))

print wd.page_source

wd.quit()

使用 Firefox 运行会将此输出打印到 stderr，这是正确的：

(block, btn)
comments are loading, (block, btn busy)
comments are loading, (block, btn busy)
comments done loading (none, btn)
(none, btn)

但是使用 PhantomJS 运行会将此输出打印到 stderr，这是不正确的：

(block, btn)
comments done loading (block, btn)
(block, btn)

我希望能够为我正在编写的应用程序使用 ActionChain 版本，但也需要使用 PhantomJS。为什么 ActionChain 不使用 PhantomJS webdriver 执行的任何想法？

4

1 回答 1

1

我无法判断您的真实网站有什么问题，但http://example.com我相信这只是时间问题。time.sleep(5)之前添加print wd.current_url，应该可以得到正确的url。

import time

# your code
# ...

ActionChains(wd).click(on_element=button).perform() 
time.sleep(5)
print wd.current_url

（如果您想做一些真正的事情，请删除睡眠并使用 WebDriverWait。）

于 2013-05-29T01:18:13.890 回答