In theory, the EditorSummary::get_editor()/set_editor() calls
should be no-ops if the values are just passed between them,
but this turns out to be not precisely the case. Rather than
figure out exactly how ensure that this is true, mark the
new rect boundaries for the non-moving axis with -1 so that
we know to leave it alone
Add an option to insert new routes at the top of the list ("First").
Reorder/rename the entries in the dialog.
Session's _order_hint is now the signed int it always wanted to be.
The previous code assumed that dragging up from the drop zone into the last
track is always valid. This is not true when the type of the dragged region(s)
do not match that route, which resulted in a crash and/or bizarre behaviour.
This took entirely too long to figure out, there are some real nightmares in
the region drag code...
This has been tested, but needs testing on more platforms (check for
obscured windows/dialogs.
Also use WIN_POS_CENTER in the "ask about loading session" dialog.
(supplied order keys are correct).
This really shouldn't be needed, but historically there have been races
between the treeview and the editor order keys.
Interesting to note that there have been no reported ordering bugs when
loading sessions.
This is debatable, the "sustained until mouse release" behaviour is handy
sometimes, but this way seems like what most people probably want.
Also, this "fire it and forget it and let it delete itself a bit later" thing
with MidiPlayer makes me nervous. I guess it's unlikely someone manages to
select a note then delete a track within 100ms, but, well...
Fixes bug #6179. Top vs. bottom seems pretty arbitrary to me, and this solves
the obscuring issue (which is quite common since there are often PC events at
the start of MIDI files), so bottom it is.
Fixes bug #0006138. This solution does make the other label move when settings
are changed (presumably what the fixed width stuff was for), but I don't think
this is a big deal. Lesser of two evils, at least.
When building with MSVC, some libraries had PROGRAM_NAME set to "Mixbus" and PROGRAM_VERSION set to "3". Whereas other libraries had PROGRAM_NAME set to "Mixbus3" while PROGRAM_VERSION was an empty string.
I've a horrible feeling there must have been a reason for this but I can't quite think what it was. Hopefully though, this change will standardize everything.
this is not a fix yet, just some comments and
code cleanup done while reading/investigating:
* limit reads to available write-space
* skip inactive tracks
* handle potential unsigned + negative value.
This breaks the build for windows builds that don't use the pthreads_win32
library. Using the opaque pthread_t type like this is probably not a great
idea. Using PBD::pthread_name is another option that I've used elsewhere
that seems more useful.
