用phantomjs进行web界面自动化测试的几个注意点

貌似我以前说过不少界面自动化测试的坏话，哈哈。最近接触了phantomjs，发现用它进行web界面测试也挺有意思的，下面举几个我使用过程中发现的注意点。

1、需要指定phantomjs位置，否则使用时会报错

driver = webdriver.PhantomJS(executable_path=r'D:\phantomjs-2.1.1-windows\bin\phantomjs.exe')#在windows中使用phatomjs，需要在这里指定可执行文件的位置，或者修改系统宏也行

2、需要指定虚拟窗口的分辨率，否则截图时可能会报错：Element is not currently visible and may not be manipulated exception。详细讨论看：https://github.com/ariya/phantomjs/issues/11637

driver.set_window_size(1024, 768)

3、需要修改源码service.py的send_remote_shutdown_command，否则driver.quit()　时会报错。详细讨论看：http://stackoverflow.com/questions/36153007/permission-error-if-to-use-phantomjs

def send_remote_shutdown_command(self):
        try:
            if self._cookie_temp_file:
                os.remove(self._cookie_temp_file)
        except Exception,info:
            pass

4、有时候定位失败可能是因为网页还未加载完，可以加个sleep或者WebDriverWait试试

无论是web还是gui，界面测试最重要的就是定位控件，再进行操作。phantomjs提供的定位控件方法如下：

def find_element_by_id(self, id_):
def find_element_by_xpath(self, xpath):
def find_element_by_link_text(self, link_text):
def find_element_by_partial_link_text(self, link_text):
def find_element_by_name(self, name):
def find_element_by_tag_name(self, name):
def find_element_by_class_name(self, name):
def find_element_by_css_selector(self, css_selector):

还有一类是find_elements的，含义和上面的相同，只不过是寻找出目标网页中具有相同条件的一批控件罢了，这样就可以同时对多个控件进行同样的操作了。

说到操作，定位控件之后，接着就是对控件进行操作，常见的控件操作如下：

.click 点击
.send_keys 按键输入
.text  获取控件的文本
.get_attribute  获得控件指定的属性值

 

使用phantomjs时最好结合chrome的f12（dev_tool），这样对于不懂html的同学（如我）来说事半功倍。

最后，再举几个phantomjs的简单使用例子：

url = r'https://www.so.com/'
driver.get(url)
input_by_id = driver.find_element_by_id('input')
input_by_id.send_keys(u'测试find_element_by_id')
button_by_id = driver.find_element_by_id('search-button')
button_by_id.click()
time.sleep(1)
driver.save_screenshot('byid.png')


url = r'https://www.so.com/'
driver.get(url)
input_by_name = driver.find_element_by_name('q')
input_by_name.send_keys(u'测试find_element_by_name')
button_by_class_name = driver.find_element_by_class_name('skin-search-button')
button_by_class_name.click()
time.sleep(1)
driver.save_screenshot('byname.png')


url = r'https://www.so.com/'
driver.get(url)
input_by_class_name= driver.find_element_by_class_name('placeholder')
input_by_class_name.send_keys(u'测试find_element_by_class_name')
button_by_class_name = driver.find_element_by_class_name('skin-search-button')
button_by_class_name.click()
time.sleep(1)
driver.save_screenshot('byclassname.png')


url = r'https://www.so.com/'
driver.get(url)
button_by_link_text = driver.find_element_by_link_text('问答')
button_by_link_text.click()
time.sleep(1)
driver.save_screenshot('bylinktext.png')


url = r'https://www.so.com/'
driver.get(url)
button_by_partial_link_text = driver.find_element_by_partial_link_text('问')
button_by_partial_link_text.click()
time.sleep(1)
driver.save_screenshot('bypartiallinktext.png')


url = r'https://www.so.com/'
driver.get(url)
input_by_xpath = driver.find_element_by_xpath('//*[@id="input"]')
input_by_xpath.send_keys(u'测试find_element_by_xpath')
button_by_id = driver.find_element_by_id('search-button')
button_by_id.click()
time.sleep(1)
driver.save_screenshot('byxpath.png')


url = r'https://www.so.com/'
driver.get(url)
input_by_css_selector = driver.find_element_by_css_selector('#input')
input_by_css_selector.send_keys(u'测试find_element_by_css_selector')
button_by_id = driver.find_element_by_id('search-button')
button_by_id.click()
time.sleep(1)
driver.save_screenshot('bycssselector.png')


url = r'https://www.so.com/'
driver.get(url)
input_by_tag_name = driver.find_elements_by_tag_name('input')#tag表示标签，当网页存在同标签名的控件时，一般用for进行定位比较好
for a in input_by_tag_name:
    if a.get_attribute('type') == 'text':
        a.send_keys(u'测试find_element_by_tag_name')
button_by_tag_name= driver.find_element_by_id('search-button')
button_by_tag_name.click()
time.sleep(1)
driver.save_screenshot('bytagname.png')