When a YUI dialogue was opened, it was focusing on the boundingBox, with
the browser moving the scroll position to focus on the top of the
boundingBox. This caused a jump. This only happens when the dialogue is
modal and consequentially has a maskNode present as it changes the initial
positioning behaviour of the boundingBox.
To avoid this, when the maskNode is shown, the dialogue is position at 0,0
in the current viewport. For centered dialogues, the dialogue is
automatically re-positioned after the window has shown. For non-centered
dialogues, the original position is stored and the dialogue is restored to
that position after it has been displayed.
This should not interfere with use of the align function as this will be
called later in the proceedings, after the show has run.
The contribution to course total column was changed from points to % but
this new test was submitted using points. It just needs the numbers converted
to %.
The maxeditingtime needs to be a negative value for unit tests to ensure
that the messages are actually sent out.
Whilst fixing this I also discovered that the create_discussion part of the
generator also creates the first post.
Since MDL-38732 we reject changes if another change has been made more
recently. As a result, if process_data is called, and then we roll into the
next millisecond period before process_data is called again, the change is
rejected.
Different errors were seen because the millisecond rollover occurred at
different time within the test.
Different error conditions were seen because not all of the tests check for
an error from the call to process_data(), instead just checking that the
value had been updated correctly.
Variant has two purposes. First to determine which version of the
question the student will see. Hence it is used to set up the state
of the quetsion when the question is started. Then the internal state of
the question is saved in the first step.
Once that has been done, the variant number is purely informative, and
just used to break down the statistics.
In some cases (the one I have in mind is qtype_opaque) then there is a
complex randomisation process, which may lead to several inital variant
numbers acutally giving the same version of the question. In this case
it is nice if the question can update the stored variant number, to make
the statistics more meaningful.
