Authors can upload an image (or choose an existing one) from the
settings page, using the in-context uploader from PDF
textbooks. Includes tests for backwards compatibility with XML courses
-- they used a magic filename (images/course_image.jpg) which is
mapped to a location in the Mongo contentstore.
Still needs some UX work, though the backend plumbing is there.
Letting xblocks handle scope rather than separating fields into
different attrs. Although, split still shunts content fields to a
different collection than setting and children fields.
The big difference is that content fields will always be a dict and not
sometimes just a string and there's no special casing of 'data' attr.
The other mind change is no more 'metadata' dict.
Features coming down the pipe will want to be able to:
* Refer to enrollments before they are actually activated (approval step).
* See what courses a user used to be enrolled in for when they re-enroll in
the same course, or a different run of that course.
* Have different "modes" of enrolling in a course, representing things like
honor certificate enrollment, auditing (no certs), etc.
This change adds an is_active flag and mode (with default being "honor").
The commit is only as large as it is because many parts of the codebase were
manipulating enrollments by adding and removing CourseEnrollment objects
directly. It was necessary to create classmethods on CourseEnrollment to
encapsulate this functionality and then port everything over to using them.
The migration to add columns has been tested on a prod replica, and seems to be
fine for running on a live system with single digit millions of rows of
enrollments.
Fixed all common and LMS tests.
The tests were failing because XMLDescriptor adds in some attributes
to _model_data, such as `xml_attributes`, that aren't necessary. The
solution is to handle all XML parsing in VideoDescriptor. There's
still one test failing in CMS, which has to do with metadata being
saved. I'm still working out how to update it in such a way that it
doesn't fail, but still tests something meaningful.
This allows a more general approach to modules such as word_cloud and
video which have no XML data (it's all store in metadata), but use
XmlDescriptor for backwards compatibility. They now generate an empty
tag on export, and clear out empty tags on import.
Also a small change to the video module as a result -- if it's asked
to parse empty XML data, it won't try to parse anything.
