MozillaWiki

QA/Execution/Web Testing/Docs/Automation/StyleGuide: Difference between revisions

Page Discussion

QA/Execution/Web Testing/Docs/Automation/StyleGuide (view source)

Revision as of 16:13, 10 November 2015

4,144 bytes added ,  10 November 2015
Undo revision 1105014 by Jlorenzo (talk) - I edited the wrong page
(Undo revision 1105014 by Jlorenzo (talk) - I edited the wrong page)
 
(45 intermediate revisions by 6 users not shown)
Line 1: Line 1:
The goal of the style guide is to try provide rules to write code that looks the same no matter what project. It is a guide and is always up for discussion by the team. I have created [https://github.com/AutomatedTester/mozwebqa-test-templates templates] () based on the details below.
The goal of the style guide is to try provide rules to write code that looks the same no matter what project. It is a guide and is always up for discussion by the team.


=All Files=
= All Files =
==File headers==


*At the top of each file should have python file header 
== File Headers ==
    #!/usr/bin/env python
* Each file should have a completed copy of the [http://www.mozilla.org/MPL/2.0/ MPL2] license block, immediately followed by an empty line.
* Each file should pass [http://www.python.org/dev/peps/pep-0008/ PEP8] except for line length, see below. 


*Each file should have a copy of the [http://www.mozilla.org/MPL/boilerplate-1.1/mpl-tri-license-sh MPL]
<source lang="python">
# Good
def method(self, parameter)


*Each file should pass [http://www.python.org/dev/peps/pep-0008/ PEP8] except for line length, see below. 
# Bad
def method(self,parameter)
</source>


** I.e. parameters  should have a comma and a space e.g.
* Lines should try not to have more than 100 characters.
    Good: def method(self, parameter)
* Docstrings should conform to [http://www.python.org/dev/peps/pep-0257/ PEP0257] and should be on a single line wherever possible.
    Bad:  def method(self,parameter), etc


   
<source lang="python">
# Good
def click_login():
"""Clicks the login link."""
 
# Bad
def click_login():
"""
Clicks the login link.
"""
</source>
 
Where not possible, the first line should be a summary.
 
<source lang="python">
# Good
def login():
"""Logs in.


    Lines should try not to have more than 100 characters. This allows 2 files to be side by side with no overlap for easier development.(I base this on MacVim on my MBP)
Clicks the login link and then waits for the home page to load.


    Indenting should be a soft tab(4 spaces) as common with in Python. Do not mix tabs and spaces
"""


    My VIM settings https://github.com/AutomatedTester/VimSettings
# Bad
def login():
"""Logs in.
Clicks the login link and then waits for the home page to load."""
</source>


    There should be no whitespace at the end of the file(as per PEP8)
* Indenting should be a soft tab (4 spaces) as common with in Python. Do not mix tabs and spaces!
* There should be no whitespace at the end of the file (as per PEP8).
* Comments should be on the line above. Remember to update comments when changing code so that code matches the comments.
* Class names should be in Pascal style as this is Python idiomatic.


     Comments should be on the line above. Remember to update comments when changing code so that code matches the comments.
<source lang="python">
# Good
class TestThisSite:
      
# Bad
class test_this_site:
</source>


    Class names should be in Pascal style as this is Python idiomatic
= Page Objects =


    Good: class TestThisSite(unittest.TestCase):
== General ==
* All page objects should inherit from <code>Page</code> in page.py.
* Page objects should not do asserts. This should be done within the test.
* Each page should be grouped within one module.
* If using multiple words to describe a module separate them with underscores '_'
* Timeout time should be taken from pytest-mozwebqa via <code>page.py's</code> timeout property.
* Single quotes (') should be used instead of double (") throughout.
* Methods should have a single purpose.


    Bad: class test_this_site(unittest.TestCase):
== Logic ==
* Methods should not contain logic that depends on properties of the page. The logic and expectations should be within the test, and adding this to the page object could guard your tests against genuine failures.


==Page Object Style Guide==
<source lang="python">
# Good
def click_login(self)
    self.selenium.find_element(*self._login_locator).click()


     All Page Objects should inherit from Page in page.py
# Bad
def click_login(self)
     if not self.is_user_logged_in:
        self.selenium.find_element(*self._login_locator).click()
    else:
        pass
</source>


    Page Objects should not do asserts. This should be done within the test
== Locators ==
* Locator variables should be prefixed with <code>_</code> to show that it is [http://docs.python.org/tutorial/classes.html#private-variables private].
* Variables should be descriptive of the area and not clash with any properties.
* Should have a suffix of <code>_locator</code>.
* Accessing locators should be done through a property or method as this keeps the locator as read-only.


     Each Page should be grouped within one module
<source lang="python">
@property
def search_term(self):
     return self.selenium.find_element(*self._search_box_locator).value
</source>


    If using mutliple words to describe a module separate them with underscores '_'
* We should use locators in the following order of preference (there will be exceptions):
** ID
** Name
** Class name
** CSS selector
** XPath
* CSS locators should use whitespace for readability when using direct descendants.


     timeout time should be taken from vars.py
<source lang="python">
# Good
_my_locator = "css=#content > p > a"
      
# Bad
_my_locator = "css=#content>p>a"
</source>


Locators Class Variables
* Use Python tuples to define locators:


    Locator variables should be prefixed with _ to show that it is "private"(http://docs.python.org/tutorial/classes.html#private-variables)
<source lang="python">
# Good
_my_locator = (By.ID, "content")
</source>


    Variables should be descriptive of the area and not clash with any properties
== Actions ==
* Methods that perform actions on the page should indicate the action in the method name.


    Suffix of _locator?
<source lang="python">
# Good
def click_report_with_length(length)


Accessing Locator Variables
# Bad
def report_length(length)
</source>


    Accessing Locators should be done through a property method as this keeps the locator as readonly.
* Actions should wait for the appropriate action to complete. This could be an implicit or explicit wait. For example, clicking a login button might explicitly wait for a username field to be visible.


   
== Advanced: Page Regions ==
    @property
In some circumstances, for example where a header/navigation is common across the website, we will use a page region. The page region is a child class of the base Page object, which is inherited by all page objects. This means that the navigation can be reached from any page object and herein lies the DRY!
    def search_box(self):
 
        return self.search_box_locator
A brief example:
       


    This approach can also be used with get_* calls with Selenium making it more idiomatic for the call.
<source lang="python">
class BasePage(Page):


     @property
     @property
     def page_title(self):
     def header(self):
         return self.selenium.get_title()
         return BasePage.HeaderRegion(self.testsetup)
Action methods
   
    class HeaderRegion(Page):
 
        _login_link = (By.ID, "home")


    Methods that perform actions on the page should indicate the action in the method name
        @def click_login(self):
            self.selenium.find_element(*self._login_link).click()
</source>


    GOOD: def click_report_with_length(length)
Referring to this page region with a property makes it very readable and concise from within the test. Clicking login during a test would be performed like this:
<source lang="python">
my_page.header.click_login()
</source>


    BAD: def report_length(length)
Another example where this might be used is on a search results page, the page region being the search results element.


==Test Style Guide==
= Tests =
* Module names should be called test_ and then behavioral areas.
    test_search.py
* Test method signature should include mozwebqa to use pytest-mozwebqa plugin.


    Module names should be called test_ and then behavioural areas. E.g. test_search_positive.py
<source lang="python">
def test_example(self, mozwebqa):
</source>


    Selenium setup should read in details from vars.py
* Test method names should always show the intent of the test case.


    Test Case names should always show the intent of the test case.
<source lang="python">
# Good
def test_that_advanced_search_does_not_find_item(self, mozwebqa):


    GOOD: def test_that_advanced_search_doesnt_find_item(self):
# Bad
def test_advanced_search(self, mozwebqa):
</source>


    BAD: def test_advanced_search(self):
== Assertions ==


    CAVEAT: setUp and tearDown are exceptions to the rule as they are needed for unittest
* Tests should handle the asserts -- not the page objects.
* Tests should use Python's native [https://docs.python.org/2/reference/simple_stmts.html#the-assert-statement assert] statement.
** Note that this is a change from our previous standard of using the [https://wiki.mozilla.org/Web_Testing/Automation/UnittestZero UnittestZero] package.
* When doing equivalency assertions, put the expected value first, followed by the actual value, for example:
<source lang="python">
assert 'expected' == 'actual'  # good
assert 'actual' == 'expected'  # bad
</source>
* When doing negative equivalency, use != and put the unexpected value first, followed by the actual value, for example:
<source lang="python">
assert 'unexpected' != 'actual'  # good
assert 'actual' != 'unexpected'  # bad
</source>
* To directly cause a test to fail raise an AssertionError with an appropriate message, for example:
<source lang="python">
raise AssertionError('message')
</source>
* See [http://pytest.org/latest/assert.html pytest's documentation on asserts] for more help.


    Tests should handle the asserts not the Page objects
= Size of patches =
To make sure that we can review your patch as quickly and efficiently as possibly we would like patches to have a single test in them and the necessary changes to the page objects. This also limits the chances of merge conflicts later.


    Every Test module will have an main() function caller
== Using new and old standards together ==
As we (and Selenium and automation) develop more knowledge some projects might fall behind the standards in this style guide. It can be tempting to want to fix all of the outdated style but in order to keep patches/pulls small (see above!) we are happy to have new and old standards of code sit side by side. As we regularly review and update tests the project will be brought completely up to our current standards.


        if __name__ == "__main__":
Or if you prefer, log a GitHub issue to have a section of code addressed separately to the job you are doing.
            unittest.main()
Jlorenzo
Confirmed users
213

edits

Retrieved from "https://wiki.mozilla.org/Special:MobileDiff/272866...1105022"