link checking test for /about page

[sashakruglov]

Addresses issue #55

[sashakruglov]

This test doesn't use Webdriver at all and doesn't use PageObjects.

[sashakruglov]

That's probably the weakest place - hardcoded id of main area of the page (to avoid collecting anchors from footer or header).

[m8ttyB]

This looks great to me and runs as expected. I like the use of BeautifulSoup to pull the urls from the page. By not hard coding the urls this is more future proof. Thanks @sashakruglov! This will increase our coverage greatly.

py.test --baseurl=http://mozillians.allizom.org -ktest_that_links_in_the_about_page_return_200_code
====================== test session starts ======================
platform darwin -- Python 2.7.2 -- pytest-2.2.4
collected 18 items 

tests/test_about_page.py .

17 tests deselected by "-ktest_that_links_in_the_about_page_return_200_code -m 'nondestructive'" 
============ 1 passed, 17 deselected in 3.95 seconds ============

[klrmn]

I haven't run it but i like it. i kinda think this test should be put in the templates project and expanded to crawl thru all links with the same base_url.

[m8ttyB]

@klrmn curiously good smelling idea. Perhaps when we tackle that in a refactor when we work on issue #56.

[klrmn]

you might want to explicitly filter out javascript urls here (the ones with href #). this project may not have them on this page, but many projects/pages do.

[sashakruglov]

@klrmn This rule needs additional refinements.

There is couple of links with # on this page, but they lead to different web-site. And this sounds like a bad idea.
We can filter URLs with # but only if they are for the same web page that is under test. So, if we are checking:
https://mozillians-dev.allizom.org/en-US/about, we can omit URL like https://mozillians-dev.allizom.org/en-US/about#LALALA.

I might be wrong, but this is not javascript URL. So maybe I just misunderstood your idea.

[m8ttyB]

I'm going to make the call this pull request good. We can integrate @klrmn's ideas in a later pull reuqest.

@sashakruglov if you would fix a few small nits I'll merge this
Here are a few lines to fix that pep8 identified:

pep8 tests/test_about_page.py 
tests/test_about_page.py:33:80: E501 line too long (86 > 79 characters)
tests/test_about_page.py:38:80: E501 line too long (90 > 79 characters)
tests/test_about_page.py:39:80: E501 line too long (100 > 79 characters)

[sashakruglov]

@m8ttyB fixed PEP8 violations

[bebef1987]

LGTM
Merging it in