Eliminate delivery of non-MIDI events to Ardour MIDI buffers.
Fix the general mess of LV2 port flags and types.
Unify old MIDI API and new atom API MIDI stuff.
Making this work required changing the way we do LV2 MIDI slightly: instead of
BufferSet::get_lv2_midi doing the translation automatically, it returns an
empty buffer and the caller is responsible for actually filling it. This
allows the LV2 code to iterate over the MIDI and transport changes in one go to
merge them into the buffer at the correct times.
Synth plugins using the new atom API now work and can get accurate transport
information along with the MIDI that drives them.
git-svn-id: svn://localhost/ardour2/branches/3.0@13522 d708f5d6-7413-0410-9779-e7cbd77b26cf
We send the full transport state (frame position, BBT time, transport speed,
meter) to the plugin:
* At the start of a cycle whenever a relocate or transport speed change has
occurred
* On every occurrence of a meter change within a cycle
This means the plugin gets a sample accurate meter/tempo map, even if the meter
changes in the middle of a cycle. However, this is not quite right yet: things
can get wonky if the tempo map is edited while rolling, since this code will
not detect the change and fail to update the plugin at the start of the cycle.
Other changes:
* Factor out TempoMetric::set_metric() and simplify some tempo functions
* Clean up LV2 URID stuff
git-svn-id: svn://localhost/ardour2/branches/3.0@13513 d708f5d6-7413-0410-9779-e7cbd77b26cf
map MIDI event types (fix#4889).
All uri-map contexts are now just ignored, and equivalent to urid (which is
equivalent to uri-map with context NULL). We now just hope that no event types
are mapped after UINT16_MAX URIs have been mapped, and die horribly otherwise.
This is exceedingly unlikely to happen any time in the next several years, if
ever.
git-svn-id: svn://localhost/ardour2/branches/3.0@12462 d708f5d6-7413-0410-9779-e7cbd77b26cf
It's slightly possible that this causes trivial build failures on different
configurations, but otherwise shouldn't cause any problems (i.e. no actual
changes other than include/naming/namespace stuff). I deliberately avoided
removing libardour-config.h since this can mysteriously break things, though a
few of those do seem to be unnecessary.
This commit only targets includes of ardour/*.h. There is also a very large
number of unnecessary includes of stuff in gtk2_ardour; tackling that should
also give a big improvement in build time when things are modified.
git-svn-id: svn://localhost/ardour2/branches/3.0@12420 d708f5d6-7413-0410-9779-e7cbd77b26cf
number of available buffers, even if it is requested. So in that case the old
assertion will fire erroneously.
git-svn-id: svn://localhost/ardour2/branches/3.0@6229 d708f5d6-7413-0410-9779-e7cbd77b26cf
Vimmers, try let c_space_errors = 1 in your .vimrc to highlight this kind of stuff in red. I don't know the emacs equivalent...
git-svn-id: svn://localhost/ardour2/branches/3.0@5773 d708f5d6-7413-0410-9779-e7cbd77b26cf
The vast majority of Route signal processing is now simply in the list of
processors. There are definitely regressions here, but there's also
a lot of things fixed. It's far too much work to let diverge anymore
regardless, so here it is.
The basic model is: A route has a fixed set of input channels (matching
its JACK input ports and diskstream). The first processor takes this
as input. The next processor is configured using the first processor's
output as input, and is allowed to choose whatever output it wants
given that input... and so on, and so on. Finally, the last processor's
requested output is used to set up the panner and create whatever Jack
ports are needed to output the data.
All 'special' internal processors (meter, fader, amp, insert, send) are
currently transparent: they read any input, and return the same set
of channels back (unmodified, except for amp).
User visible changes:
* LV2 Instrument support (tracks with both MIDI and audio channels)
* MIDI in/out plugin support
* Generic plugin replication (for MIDI plugins, MIDI/audio plugins)
* Movable meter point
Known Bugs:
* Things seem to get weird on loaded sessions
* Output delivery is sketchy
* 2.0 session loading was probably already broken...
but it's definitely broken now :)
Please test this and file bugs if you have any time...
git-svn-id: svn://localhost/ardour2/branches/3.0@5055 d708f5d6-7413-0410-9779-e7cbd77b26cf
All #include statements that include a header that is a part of a library
bundled with ardour MUST use quotes, not angle brackets.
Do this:
#include "ardour/types.h"
NOT this:
#include <ardour/types.h>
Rationale:
This is best practice in general, to ensure we include the local version
and not the system version. That quotes mean "local" (in some sense)
and angle brackets mean "system" (in some sense) is a ubiquitous
convention and IIRC right in the C spec somewhere.
More pragmatically, this is required by (my) waf (stuff) for dependencies
to work correctly. That is:
!!! FAILURE TO DO THIS CAN RESULT IN BROKEN BUILDS !!!
Failure to comply is punishable by death by torture. :)
P.S. It's not that dramatic in all cases, but this (in combination with some
GCC flags specific to the include type) is the best way I have found to be
absolutely 100% positive the local ones are being used (and we definitely
want to be absolutely 100% positive on that one).
git-svn-id: svn://localhost/ardour2/branches/3.0@4655 d708f5d6-7413-0410-9779-e7cbd77b26cf
*increased buffer sizes to get rid of overruns (PitchBender still doesnt work though....)
git-svn-id: svn://localhost/ardour2/branches/3.0@3403 d708f5d6-7413-0410-9779-e7cbd77b26cf
