[Rosegarden-bugs] [ rosegarden-Bugs-1885520 ] Arnout and I can't figure out refresh problem

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 454-5900

Bugs item #1885520, was opened at 2008-02-02 21:33
Message generated for change (Comment added) made by dmmcintyr
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104932&aid=1885520&group_id=4932

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: gui
Group: None
Status: Open
Resolution: None
>Priority: 8
Private: No
Submitted By: D. Michael McIntyre (dmmcintyr)
Assigned to: Nobody/Anonymous (nobody)
Summary: Arnout and I can't figure out refresh problem

Initial Comment:
Committed revision 8604, which introduces a new bug in the course of fixing one.  If I go into a notation view, use Segment -> Transform notation for... (or whatever godawful thing I called it; I'm not liking that much so far, and can't seem to remember the name either) and load a preset from the database...

If I pick the bottom option, to apply to all segments on the track, that branch updates the TPB widgets with the new parameters.

No refresh occurs, and the only way to get the TPB and SPB to reflect the new state is to switch tracks and switch back.

We just had a similar bug:

http://sourceforge.net/tracker/index.php?func=detail&aid=1885247&group_id=4932&atid=104932

Arnout seemed to fix this by changing the command syntax used in the TPB, but that just doesn't look like the right fix over here in the notation editor, and I wonder why the fix worked where it was too.

There must be some other way, but I can't remember or discover it.  I tried calling

doc->slotDocumentModified()

but it had no effect.

I'm stumped.

----------------------------------------------------------------------

>Comment By: D. Michael McIntyre (dmmcintyr)
Date: 2008-03-31 01:04

Message:
Logged In: YES 
user_id=663564
Originator: YES

Can't solve this in a vacuum, and two months without the requested hint,
so we're going to release with this one.  I can't fix it, and I won't hold
a release over it.

----------------------------------------------------------------------

Comment By: D. Michael McIntyre (dmmcintyr)
Date: 2008-02-19 20:09

Message:
Logged In: YES 
user_id=663564
Originator: YES

Calling TrackParameterBox::slotUpdateControls() in the TrackParameterBox
object known as m_trackParameterBox to RosegardenGUIView would do the
trick, but I can't figure out how to emit and connect what to plumb it up
and make that happen from the notation view.

Totally stumped I'm afraid.  Real programmer please advise.

----------------------------------------------------------------------

Comment By: D. Michael McIntyre (dmmcintyr)
Date: 2008-02-04 18:24

Message:
Logged In: YES 
user_id=663564
Originator: YES

It's the SPB and TPB that need to reflect the new reality of their
contents.  Highlighting a different track and then moving back where you
just came from accomplishes this, for example.

I saw some kind of SPB::updateYourself() while I was digging.  I guess
maybe I just need to call that, and give the TPB a parallel function,
because I don't think it has one.

I can't remember what I'm talking about, but I can probably find it again,
and hopefully it's enough of a hint to get me going.

----------------------------------------------------------------------

Comment By: Chris Cannam (cannam)
Date: 2008-02-04 07:33

Message:
Logged In: YES 
user_id=13489
Originator: NO

So the problem is that a change in the segment parameters is not being
reflected in the segment parameter box, right?  You're not talking about
updating the notation itself, or the segment canvas, or whatever -- or are
you?

I think there is in fact no proper mechanism for automatic updates to the
parameter boxes to happen at the moment.  The SPB and its friends need to
have some sort of notification via an Observer or similar, and I don't
think it does.  At the moment it relies heavily on the fact that such
updates only come from a very limited set of places (e.g. the user
modifying something in the SPB itself, or a new document being loaded, or a
new segment selected) and those places making an explicit call to the SPB
to update.  I don't remember what that call is, offhand.

This is closely related to the crash when deleting a device which I fixed
a short while ago -- the IPB was not being updated at the proper time and
was continuing to refer to the deleted device for a little longer than it
should have.

----------------------------------------------------------------------

Comment By: D. Michael McIntyre (dmmcintyr)
Date: 2008-02-02 21:49

Message:
Logged In: YES 
user_id=663564
Originator: YES

I tried an

emit updateView();

with similarly little effect.

Surely there is something obvious I'm missing.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104932&aid=1885520&group_id=4932