As the code in this .js will break the url's integrity when there is already a query string inside the video source's url and cause the url to be invalid in some cases (for example, when working with a url from Windows Azure's Media Service, the appended '?' will cause the url being invalid).
I modified the code by first checking whether the url has already had a query string, and if so, then use '&' instead of '?' to prevent the breaking, and it works for the url from Windows Azure's Media Service.
1. In video_module.py, rewrite the get_ext() function to use the built-in urlparse module to parse the input filename first and then get the file's extension name from the parsed path.
2. In test_video.py, add two test cases (one with query string while the other without) in order to test the rewritten get_ext() function.
Sometimes the video url may look like 'http://abc.com/path/video.mp4?xxxx'.
(For example, the Windows Azure's media service will offer such type of url.)
The original code in video_module.py will produce 'mp4?xxxx' instead of 'mp4'
as the extension of a filename for such type of url, and therefore the video
will be non-playable. The fix here uses the built-in urlparse module to retrive
only the path component from a url and therefore the extension 'mp4' will be
correctly fetched.
Check for status in test.
Add test for track url.
Fix test to use cookie.
Use download_track in feature file.
Add new tests for get_transcript.
Add html5 test.
Add content check.
Add tests for available_translations dispatch.
Disable flaky cms test.
Add w/o english transcript tests.
Add to changelog.
This change addresses STUD-1403 where split_test modules could not render in Studio.
It changes Studio to pass a new parameter 'runtime-type' of 'studio' when rendering
an xblock. In split_test, this will then cause all of the children to be rendered,
rather than just the one that matches the conditional.
Bug would occur when comparing Groups or UserPartitions - because these types
lacked comparator methods, different objects would come up as different when
their values were equal.
Without this, the new 'items' field in the course_info xmodule caused
bizarre behavior. If you added the text 'items' to a course_info
section, then the LMS would throw a 500, because the field name isn't
'data', but the stored data is a string, leading to the check of
`key.field_name in self._data`, where key.field_name is the string 'items',
and self._data is the content string.
Currently wiki_slug is set to course number. However, since multiple
courses can have the same number this may lead to clashes. So wiki_slug
will be set to org.course.name. To maintain the active wiki_slugs for
xml courses this cannot be changed in the CourseDescriptor.
LMS-2136
Without this, the new 'items' field in the course_info xmodule caused
bizarre behavior. If you added the text 'items' to a course_info
section, then the LMS would throw a 500, because the field name isn't
'data', but the stored data is a string, leading to the check of
`key.field_name in self._data`, where key.field_name is the string 'items',
and self._data is the content string.
