In the latest commit, I've fixed most of those little bugs in
tuplet-rewrite.  I took Michael's advice; for the situation of lengthening
adjacent notes of the same pitch, I just make the whole overlapping mess,
and user can undo it.

Which lets me segue to fix #1.  Previously, user couldn't fully undo some
commands like AddDotCommand.  Some undos were too short, eg AddDotCommand
and SetNoteTypeCommand.

>From my notes:

The bug: We got the undo length from the selection length, but then we
sometimes made the notes of the selection longer.  The fix was quick and
dirty: BasicSelectionCommand got another ctor, to which caller passes the
worst-case length of the last note.  That ctor figures out worst-case
end-time and uses it.

Fixed dumb bug.  The bug: I set metric tupling points as if a tied
sequence could have only one tupling, but really it has both a start and
an end.  This sometimes misinformed the bar-decomposition logic resulting
in funny-timed notes.  The fix: Get tupling for both start and end events
in tied sequences (middles still don't matter)


Fixed a bug where add-dot to a tripleted note can shorten it.  The bug: it
deduced note type from duration, but tupletted notes have shorter
duration.  The fix: Use the duration in tuplet time, not performance time.


Fixed bug.  The bug: I wasn't updating the segment refresh statuses for
the bar before the alterations.  This made it appear that some ties
persisted after notes were erased.  The fix: Update refresh statuses
including the bar before the alterations



        Tom Breton (Tehom)




------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Rosegarden-devel mailing list
Rosegarden-devel@lists.sourceforge.net - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-devel

Reply via email to