> So all I did was swat at it on the run, but I finally got it built.
> I'll look more deeply another day, I hope.
Thanks for looking at it, Michael.
> 2) I continued flailing around deleting things haphazardly. I never got
> a crash, but if you start deleting a bunch of random notes from the
> 12-tuplet groups, you eventually wind up with random triplet rests
> scattered around, and nothing indicating that the remaining notes are 12
> or any other kind of tupleted.
That is something I can reproduce.
The answer seems to be that when I added code a few days ago to figure out
what parts of the bar need to be changed and what don't, it wasn't smart
about groups. It could sometimes grab just a part of a group, and then
the rest of the code wrongly thought that it had the entire group because
the previous code always grabbed whole bars. Result: Bad groups.
Possible approaches:
* Roll that part back. Much as I hate to just give up, it's causing
Chris's crashes and this weirdness, and it was really just supposed to be
an optimization and minimize what got rewritten.
* Make this smarter and also fix that crash. Making it smart about
groups looks to be about 50% more complex but doable. Fixing that crash
looks tricky. I know how it's dividing by zero - the time signature
m_denominator - but I can't see how that can get to be zero.
* Minimize rewrites in a different way. I can think of one other way,
but that looks problematic.
I'm choosing option 1, at least until there's a reason to do differently.
(So my approach is actually a retreat)
> 3) I thought to mess around and see how you're handling beaming
> differently. It looks slightly different, but mostly the same. I might
> be imagining the differences. I might not even be invoking new code.
Well, the differences in beaming aren't too big. If you put short notes
(say, 16ths) so they cross beat lines, you'll see it's paying more
attention to beats. But mostly the difference is that it's handling
tuplets thru the same path as everything else, rather than invoking
TupletCommand on what it hopes is a tupleted group.
Tom Breton (Tehom)
------------------------------------------------------------------------------
DreamFactory - Open Source REST & JSON Services for HTML5 & Native Apps
OAuth, Users, Roles, SQL, NoSQL, BLOB Storage and External API Access
Free app hosting. Or install the open source package on any LAMP server.
Sign up and see examples for AngularJS, jQuery, Sencha Touch and Native!
http://pubads.g.doubleclick.net/gampad/clk?id=63469471&iu=/4140/ostg.clktrk
_______________________________________________
Rosegarden-devel mailing list
[email protected] - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-devel
