test0908
/
django
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Added the `wait_until()` and `wait_loaded_tag()` methods to `AdminSeleniumWebDriverTestCase` to prevent some concurrency issues with in-memory SQLite database access in the admin Selenium tests. Thanks to Florian Apolloner, Anssi Kääriäinen and Aymeric Augustin for their help debugging this problem. 

git-svn-id: http://code.djangoproject.com/svn/django/trunk@17283 bcc190cf-cafb-0310-a4f2-bffc1f526a37
This commit is contained in:
Julien Phalip 2011-12-28 22:20:30 +00:00
parent 46c12d1293
commit a343a84ce6
3 changed files with 60 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
django/contrib/admin/tests.py
View File

@ -29,6 +29,26 @@ class AdminSeleniumWebDriverTestCase(LiveServerTestCase):
if hasattr(cls, 'selenium'): if hasattr(cls, 'selenium'):
cls.selenium.quit() cls.selenium.quit()
def wait_until(self, callback, timeout=10):
"""
Helper function that blocks the execution of the tests until the
specified callback returns a value that is not falsy. This function can
be called, for example, after clicking a link or submitting a form.
See the other public methods that call this function for more details.
"""
from selenium.webdriver.support.wait import WebDriverWait
WebDriverWait(self.selenium, timeout).until(callback)
def wait_loaded_tag(self, tag_name, timeout=10):
"""
Helper function that blocks until the element with the given tag name
is found on the page.
"""
self.wait_until(
lambda driver: driver.find_element_by_tag_name(tag_name),
timeout
)
def admin_login(self, username, password, login_url='/admin/'): def admin_login(self, username, password, login_url='/admin/'):
""" """
Helper function to log into the admin. Helper function to log into the admin.
@ -41,6 +61,8 @@ class AdminSeleniumWebDriverTestCase(LiveServerTestCase):
login_text = _('Log in') login_text = _('Log in')
self.selenium.find_element_by_xpath( self.selenium.find_element_by_xpath(
'//input[@value="%s"]' % login_text).click() '//input[@value="%s"]' % login_text).click()
# Wait for the next page to be loaded.
self.wait_loaded_tag('body')
def get_css_value(self, selector, attribute): def get_css_value(self, selector, attribute):
""" """

35
docs/topics/testing.txt
View File

@ -1843,6 +1843,41 @@ out the `full reference`_ for more details.
</howto/static-files>` so you'll need to have your project configured </howto/static-files>` so you'll need to have your project configured
accordingly (in particular by setting :setting:`STATIC_URL`). accordingly (in particular by setting :setting:`STATIC_URL`).
.. note::
When using an in-memory SQLite database to run the tests, the same database
connection will be shared by two threads in parallel: the thread in which
the live server is run, and the thread in which the test case is run. It is
important to prevent simultaneous database queries via this shared
connection by the two threads as that may sometimes cause the tests to
randomly fail. So you need to ensure that the two threads do not access the
database at the same time. In particular, this means that in some cases
(for example just after clicking a link or submitting a form) you might
need to check that a response is received by Selenium and that the next
page is loaded before proceeding further with the execution of the tests.
This can be achieved, for example, by making Selenium wait until the
`<body>` HTML tag is found in the response:
.. code-block:: python
def test_login(self):
from selenium.webdriver.support.wait import WebDriverWait
...
self.selenium.find_element_by_xpath('//input[@value="Log in"]').click()
# Wait until the response is received
WebDriverWait(self.selenium, timeout).until(
lambda driver: driver.find_element_by_tag_name('body'), timeout=10)
The difficult point is that there really is no such thing as a "page load",
especially in modern Web apps that have dynamically-generated page
components that do not exist in the HTML initially received from the
server. So simply checking for the presence of the `<body>` tag in the
response might not necessarily be appropriate for all use cases. Please
refer to the `Selenium FAQ`_ and the `Selenium documentation`_ for more
information on this topic.
.. _Selenium FAQ: http://code.google.com/p/selenium/wiki/FrequentlyAskedQuestions#Q:_WebDriver_fails_to_find_elements_/_Does_not_block_on_page_loa
.. _Selenium documentation: http://seleniumhq.org/docs/04_webdriver_advanced.html#explicit-waits
Using different testing frameworks Using different testing frameworks
================================== ==================================

3
tests/regressiontests/admin_inlines/tests.py
View File

@ -443,6 +443,9 @@ class SeleniumFirefoxTests(AdminSeleniumWebDriverTestCase):
self.selenium.find_element_by_name('profile_set-2-last_name').send_keys('2 last name 2') self.selenium.find_element_by_name('profile_set-2-last_name').send_keys('2 last name 2')
self.selenium.find_element_by_xpath('//input[@value="Save"]').click() self.selenium.find_element_by_xpath('//input[@value="Save"]').click()
# Wait for the next page to be loaded.
self.wait_loaded_tag('body')
# Check that the objects have been created in the database # Check that the objects have been created in the database
self.assertEqual(ProfileCollection.objects.all().count(), 1) self.assertEqual(ProfileCollection.objects.all().count(), 1)
self.assertEqual(Profile.objects.all().count(), 3) self.assertEqual(Profile.objects.all().count(), 3)