This change will prevent Library Content from being marked as
complete on view and the corresponding version bump to
edx-completion contains code that will start looking at the
children of the library content for completeness.
Fixing 56 GuessedAtParserWarnings, in commit edx#24098
Background: BeautifulSoup automatically picks the fastest parser available. By default, it picks the "lxml" parser.
Per the [BeautifulSoup](https://www.crummy.com/software/BeautifulSoup/bs4/doc/#installing-a-parser) documentation:
> Beautiful Soup supports the HTML parser included in Python’s standard library, but it also supports a number of third-party Python parsers. One is the lxml parser. Depending on your setup, you might install lxml with one of these commands.
> Another alternative is the pure-Python html5lib parser, which parses HTML the way a web browser does.
Context: We changed two statements, one in lms and another in openedx. Both statements fire up BeautifulSoup. Now we explicitly ask for "lxml," following the recommendation on BeautifulSoup's documentation:
> If you can, I recommend you install and use lxml for speed. If you’re using a very old version of Python – earlier than 2.7.3 or 3.2.2 – it’s essential that you install lxml or html5lib. Python’s built-in HTML parser is just not very good in those old versions.
Before:
`soup = BeautifulSoup(content)`
After:
`soup = BeautifulSoup(markup=content, features="lxml")`
The warnings are gone, tests are passing in local.
1. Created a new celery queue with key `SOFTWARE_SECURE_VERIFICATION_ROUTING_KEY`.
2. Added a celery task with retry logic.
3. sorted imports with isort.
4. Changed deprecated `log.warn` => `log.warning`.
Python3 json can not work with bytes type input so used simplejson
instead of python json which works with both bytes and unicode
type input.
Previously we mistakenly updated StringIO to six.String io. which
caused the tests failure. Used io.BytesIO to fix the issues as
assci character bytes can not be converted to unicode.
Some notes:
* The completion API was pulled out into a new repo, but left behind a
url() entry. That entry used a different namespace than the
pulled-out repo, so I had to fix the one place we used the namespace.
* Two urls couldn't be anchored because they broke tests:
url(r'users/(?P<user_id>\w+)/followed$', views.followed_threads, name='followed_threads'),
url(r'users/(?P<user_id>\w+)$', views.user_profile, name='user_profile'),
