* fix: catch a possible exception in beta course configuration
when the learner is a beta tester, and the beta test has been set up with a large duration, the courseware djangoapp can return an overflow error on a timedelta call. In those circumstances, this defaults to returning an unmodified date.
FIXES: APER-3848
* feat: update preview url to direct to mfe
* fix: use url builder instead of string formatter
* fix: url redirect for never published units
* fix: remove 404 error when not a preview or staff
* feat: update sequence metadata to allow draft branch
We've long rolled out the
`COURSEWARE_COURSE_NOT_STARTED_ENTERPRISE_LEARNER_ERROR` setting toggle so
it should be safe to remove.
I'm completely at a loss as to why course access checks are not
returning the `course_not_started_enterprise_learner` error code when it
should, but it does in stage/local, so this PR is grasping at straws.
ENT-8766
Web scrapers do annoying stuff like visit urls they shouldn't know about and cause xblock handlers to break.
I tested this by:
Making sure video transcripts worked as normal while logged in
making sure that I got no 500s in my logs while attempting to view it logged out.
Normally, the course API would return an access error_code of
`course_not_started` if the course has not started yet. This change
breaks that up into two codes:
* if the course has not started:
* return error_code=`course_not_started_enterprise_learner` if the
learner is enrolled as a subsidized enterprise learner.
* else, return error_code=`course_not_started`.
This supports a change to the frontend which will interpret
`course_not_started_enterprise_learner` differently and trigger a
redirect to the enterprise (B2B) learner dashboard instead of the B2C
dashboard.
ENT-8078
Normally, the course API would return an access error_code of
`course_not_started` if the course has not started yet. This change
breaks that up into two codes:
* if the course has not started:
* return error_code=`course_not_started_enterprise_learner` if the
learner is enrolled as a subsidized enterprise learner.
* else, return error_code=`course_not_started`.
This supports a change to the frontend which will interpret
`course_not_started_enterprise_learner` differently and trigger a
redirect to the enterprise (B2B) learner dashboard instead of the B2C
dashboard.
ENT-8078
This commit leaves behind just enough Old Mongo (DraftModulestore)
functionality to allow read-only access to static assets and the
root CourseBlock. It removes:
* create/update operations
* child/parent traversal
* inheritance related code
It also removes or converts tests for this functionality.
The ability to read from the root CourseBlock was maintained for
backwards compatibility, since top-level course settings are often
stored here, and this is used by various parts of the codebase,
like displaying dashboards and re-building CourseOverview models.
Any attempt to read the contents of a course by getting the
CourseBlock's children will return an empty list (i.e. it will look
empty).
This commit does _not_ delete content on MongoDB or run any sort of
data migration or cleanup.
* feat: ora date config reflected in dates tab
* docs: update docstrings
* style: quality
* feat: Upgrade Python dependency ora2 (#33119)
temp upgrade so tests pass
Commit generated by workflow `openedx/edx-platform/.github/workflows/upgrade-one-python-dependency.yml@refs/heads/master`
Co-authored-by: jansenk <jansenk@users.noreply.github.com>
* fix: unbound variable
* fix: don't show dates for steps with no concept of due
* feat: Upgrade Python dependency ora2 (#33176)
bump to ora version to include ora date config changes
Commit generated by workflow `openedx/edx-platform/.github/workflows/upgrade-one-python-dependency.yml@refs/heads/master`
Co-authored-by: jansenk <jansenk@users.noreply.github.com>
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: jansenk <jansenk@users.noreply.github.com>
The Webpack configuration file for built-in XBlock JS used to be
generated at build time and git-ignored. It lived at
common/static/xmodule/webpack.xmodule.config.js. It was generated
because the JS that it referred to was also generated at build-time, and
the filenames of those JS modules were not static.
Now that its contents have been made entirely static [1], there is no
reason we need to continue generating this Webpack configuration file.
So, we check it into edx-platform under the name
./webpack.builtinblocks.config.js. We choose to put it in the repo's
root directory because the paths contained in the config file are
relative to the repo's root.
This allows us to behead both the xmodule/static_content.py
(`xmodule_assets`) script andthe `process_xmodule_assets` paver task, a
major step in removing the need for Python in the edx-platform asset
build [2]. It also allows us to delete the `HTMLSnippet` class and all
associated attributes, which were exclusively used by
xmodule/static_content.py..
We leave `xmodule_assets` and `process_xmodule_assets` in as stubs for
now in order to avoid breaking external code (like Tutor) which calls
Paver; the entire pavelib/assets.py function will be eventually removed
soon anyway [3]. Further, to avoid extraneous refactoring, we keep one
method of `HTMLSnippet` around on a few of its former subclasses:
`get_html`. This method was originally part of the XModule framework;
now, it is left over on a few classes as a simple internal helper
method.
References:
1. https://github.com/openedx/edx-platform/pull/32480
2. https://github.com/openedx/edx-platform/issues/31800
3. https://github.com/openedx/edx-platform/issues/31895
Part of: https://github.com/openedx/edx-platform/issues/32481
This re-applies commit 36cc415 with handling an invalid context_key in the
`PartitionService`. It can happen when rendering a `LibraryContentBlock` in
Studio because this service is initialized by the modulestore when validating
an XBlock to gather its error messages in the `studio_xblock_wrapper`.
