* We need to have unique feature names for our Behat tests.
* I also removed the custom Behat tag @role_visibility as we only allow
Behat tags with the component name in Frankenstyle format.
Similarly to what the previous patch does, we now change the default
sorting in functions enrol_get_users_courses() and
enrol_get_all_users_courses() too.
The patch also adds missing phpDocs for the functions, improves the
readability of some existing bits and mentions the changes in the
enrol/upgrade.txt file.
LTS (3.5) requires previous LTS (3.1)
This just deletes all the upgrade steps previous to 3.1.0. Some
small adjustments, like adding missing MOODLE_INTERNAL or tweaking
globals can also be applied when needed.
Also includes an upgrade step to prevent upgrading from any
version < 2016052300 (v3.1.0) as anti-cheating measure.
Next commit will get rid of/deprecate all the upgradelib functions
not used anymore in codebase.
Sometimes the "theme" and "lang" fields in the user and course tables
in the database are set to incorrect values (uninstalled or
non-existent themes and language packs).
This makes Web Services functions to fail because the WS server
validate the returned data using the validate_param function that clean
parameters.
Web Services from user and enrol components has been updated to change
the user preference name field type from PARAM_ALPHANUMEXT to
PARAM_RAW.
The old type was not matching the core functionality.
Before this change the nested select did a full table scan of
either the enrol or user_enrolments table. This meant that
on large sites the query could be very slow.
This change ensures that there are no full table scans which
makes the query much more efficient.
In my testing the all courses version of the query changed
from ~15 s to ~ 5 s; the singles course version changed
from ~15 s to > 0.1 s
We want to avoid page jumps wherever possible. This change uses the original select element as a "proxy" to reserve the space
in the DOM for the enhanced auto-complete when the JS runs.
It uses visibility: hidden to make the select not rendered - but still take up space.
The exact sizing for the CSS was determined by testing and reverse engineering the bootstrap calculations.
The size of user pictures in selectors was reduced to avoid flicker when switching between no selection and a selected user. I
could have reserved a larger amount of space, but it looks worse for pickers with no pictures.
