Automated chrome driver tests for about:plugins functionality tests

functional/aboutplugins_ui.py

Index: functional/aboutplugins_ui.py
===================================================================
--- functional/aboutplugins_ui.py	(revision 0)
+++ functional/aboutplugins_ui.py	(revision 0)
@@ -0,0 +1,145 @@
+import os

[kkania, 2012/01/20 17:16:28]

rename this file to about_plugins_ui.py

[vivianz, 2012/01/27 23:24:36]

Done.

+import re
+
+import pyauto_functional  # Must be imported before pyauto
+import pyauto
+import test_utils

[kkania, 2012/01/20 17:16:28]

you don't use this; remove?

[vivianz, 2012/01/27 23:24:36]

Done.

+
+
+class AboutPluginsTest(pyauto.PyUITest):
+  """TestCase for chrome://plugins."""
+
+  def Debug(self):
+    """Test method for experimentation.
+
+    This method will not run automatically.
+    """
+    while True:
+      raw_input('Interact with the browser and hit <enter> to list plugins...')

[kkania, 2012/01/20 17:16:28]

this method is more useful for the plugins test; how about change it to be more
useful for us:

instead of
while True:
  ...

just do
self.NavigateToURL('chrome://plugins/')
driver = self.NewWebDriver()
import pdb
pdb.set_trace()

[vivianz, 2012/01/27 23:24:36]

Done.

+      self.pprint(self.GetPluginsInfo().Plugins())
+
+  def _ClearLocalDownloadState(self, path):
+    """Prepare for downloading the given path.
+
+    Clears the given path and the corresponding .crdownload, to prepare it to
+    be downloaded.
+    """
+    if os.path.exists(path): os.remove(path)
+    crdownload = path + '.crdownload'
+    if os.path.exists(crdownload): os.remove(crdownload)
+
+  def _IsEnabled(self, plugin_name):
+    """Checks if plugin is enabled."""
+    for plugin in self.GetPluginsInfo().Plugins():
+      if re.search(plugin_name, plugin['name']):
+         return plugin['enabled']    
+
+  def _GoPDFSiteAndVerify(self, pdf_name, tab_index=0, windex=0,
+                          is_pdf_enabled=True):
+    """Navigate to PDF file site and verify PDF viewer plugin.
+
+    When PDF Viewer plugin is enabled, navigate the site should not trigger

[kkania, 2012/01/20 17:16:28]

navigating to the site

[vivianz, 2012/01/27 23:24:36]

Done.

+    PDF file downloade and if PDF viewer plugin is disabled, PDF file should

[kkania, 2012/01/20 17:16:28]

downloade?

[vivianz, 2012/01/27 23:24:36]

fixed

+    be viewable in browser.

[kkania, 2012/01/20 17:16:28]

if plugin is disable, the pdf should be viewable in the browser? i think you
wrote it backwards

[vivianz, 2012/01/27 23:24:36]

my bad :) fixed

+
+    Args:
+      pdf_name: the name of the PDF file.
+                This should exist in the 'dangerous' directory.
+      tab_index: tab index. Default 0.
+      windex: window index. Default 0.
+      is_pdf_enabled: the boolean flag check if PDF viewer plugin is enabled.
+                      Default True.
+    """
+    file_url = self.GetHttpURLForDataPath('downloads', 'dangerous',
+        'download-dangerous.html' + '?' + pdf_name)
+    num_downloads = len(self.GetDownloadsInfo().Downloads())
+    self.NavigateToURL(file_url, windex, tab_index)
+    # It might take a while for the download to kick in, hold on until then.
+    if is_pdf_enabled == False:
+      self.assertTrue(self.WaitUntil(lambda:
+          len(self.GetDownloadsInfo().Downloads()) == num_downloads + 1))
+    else:
+      driver = self.NewWebDriver()

[kkania, 2012/01/20 17:16:28]

don't create a new webdriver here; pass it in as a argument to this func

[vivianz, 2012/01/27 23:24:36]

Done.

+      self.assertTrue(self.WaitUntil(lambda: len(
+          driver.find_elements_by_name('plugin')) == 1))
+
+  def testAboutPluginDetailInfo(self):
+    """Verify about:plugin page shows plugin details."""
+    self.NavigateToURL('chrome://plugins/')
+    driver = self.NewWebDriver()
+    detail_link = driver.find_element_by_id('details-link')
+    while not detail_link.is_displayed():
+      pass
+    detail_link.click()
+    # Verify the detail info for Remote Viewer plugin show up.
+    self.assertTrue(self.WaitUntil(lambda: len(driver.find_elements_by_xpath(
+        '//*[@jscontent="name"][text()="Remoting Viewer"]' +

[kkania, 2012/01/20 17:16:28]

don't use jscontent, use class instead

[vivianz, 2012/01/27 23:24:36]

have to use jscontent, that is the only unique label there for detail info.

+        '//ancestor::*[@class="plugin-text"]//a[text()="Disable"]')) == 1))

[kkania, 2012/01/20 17:16:28]

this isn't sufficient to determine if the details are shown or not, since the
Disable link is also present in compact mode

[vivianz, 2012/01/27 23:24:36]

change to locate the detail plugin path for Remote Viewer plugin.

+
+  def testAboutPluginEnableAndDisablePDFPlugin(self):
+    """Verify enable and disable plugins from about:plugins page."""
+    self.NavigateToURL('chrome://plugins/')
+    driver = self.NewWebDriver()
+    # Override the animation for expanding detail info to make sure element
+    # remain at the same location where web driver found it.
+    override_animation_style_js = """
+        style = document.createElement('style');
+        style.innerHTML = "* { -webkit-transition: all 0s ease-in !important}";
+        document.head.appendChild(style);
+        """
+    driver.execute_script(override_animation_style_js)
+
+    # Click Details link to expand details.
+    detail_link = driver.find_element_by_id('details-link')
+    while not detail_link.is_displayed():
+      pass
+    detail_link.click()
+    pdf_disable_path = '//*[@class="plugin-name"][text()="Chrome PDF Viewer"' \
+        ']//ancestor::*[@class="plugin-text"]//a[text()="Disable"]'
+    # Confirm Chrome PDF Viewer plugin is found.
+    self.assertTrue(self.WaitUntil(lambda: len(driver.find_elements_by_xpath(
+        pdf_disable_path)) == 1),
+        msg='Failed to find Chrome PDF Viewer plugin')
+
+    # Disable PDF viewer plugin in about:plugins.
+    pdf_disable_link = driver.find_element_by_xpath(pdf_disable_path)
+
+    # Need to sleep off 100ms for the detail-info expansion webkit transition.

[kkania, 2012/01/20 17:16:28]

?

[vivianz, 2012/01/27 23:24:36]

legacy comment deleted.

+    pdf_disable_link.click()
+    self.assertFalse(self.WaitUntil(lambda:
+        self._IsEnabled('Chrome PDF Viewer')))

[kkania, 2012/01/20 17:16:28]

i don't think this is correct

[vivianz, 2012/01/27 23:24:36]

change to assertTrue wait until chrome PDF viewer is not enabled.

+
+    # Navigate to a PDF file and verify the pdf file is downloaded.
+    self.OpenNewBrowserWindow(True)
+    self._GoPDFSiteAndVerify('fw4.pdf', tab_index=0, windex=1,
+                             is_pdf_enabled=False)
+    id = self.GetDownloadsInfo().Downloads()[0]['id']
+    self.PerformActionOnDownload(id,
+                                 'save_dangerous_download',
+                                 window_index=1)
+    self.WaitForAllDownloadsToComplete(windex=1)
+
+    pdf_download = self.GetDownloadsInfo(1).Downloads()
+    # Verify that download info exists in the correct profile.
+    self.assertEqual(len(pdf_download), 1)
+    download_pkg = os.path.join(self.GetDownloadDirectory().value(),
+                                'fw4.pdf')
+    self._ClearLocalDownloadState(download_pkg)
+
+    # Navigate to plugins settings again and re-enable PDF viewer plugin.
+    self.NavigateToURL('chrome://plugins/')
+    driver = self.NewWebDriver()

[kkania, 2012/01/20 17:16:28]

don't make a new driver

[vivianz, 2012/01/27 23:24:36]

Done. removed.

+    pdf_enable_path = '//*[@class="plugin-name"][text()="Chrome PDF Viewer"]' \
+        '//ancestor::*[@class="plugin-text"]//a[text()="Enable"]'
+    self.assertTrue(self.WaitUntil(lambda: len(driver.find_elements_by_xpath(
+         pdf_enable_path)) == 1))
+    pdf_enable_link = driver.find_element_by_xpath(pdf_enable_path)
+    pdf_enable_link.click()
+    self.CloseBrowserWindow(0)
+    self.OpenNewBrowserWindow(True)
+    self._GoPDFSiteAndVerify('fw4.pdf', tab_index=0, windex=0,
+                             is_pdf_enabled=True)
+
+
+if __name__ == '__main__':
+  pyauto_functional.Main()