every write pass deletes existing tracks, which means it also deletes any existing
EOT event. Rather than try to replicate the _length value() that is kept in a
Source object in the SMF object, add a virtual method to SMF that returns
the _length value (or std::numeric_limits<Beats>::max() if not set).
If the _length value is not the max, we will add EOT events to each track
(usually just one) right before writing to disk.
This partially reverts eced764480. In many places Ardour uses
SMF directly, without loading the MIDI model in a libardour SMFSource.
In this case no information (track, channel-count) was available,
after eced764480, and worse various members were not initialized
and showed random numbers.
This fixes various import options.
we screen midi files for some aggregate info:
used channels, used patches, and note-count
you can't do this from open() because there are cases (after importing)
when the source exists but it is not yet written to disk
This significantly speeds up parsing MIDI files with complex
tempo-maps. e.g. "Black MIDI Trilogy_2.mid" has 24134 Tempo
changes. Prior to this commit parsing that file took over 5 minutes.
now it loads in under one seconds (libsmf only; libardour still
add overhead, and now needs about 30-40 seconds, previously
it took about 10 mins).
The problem was that every call to `smf_track_add_event_pulses()`
calls `seconds_from_pulses()` which calls `smf_get_tempo_by_seconds()`
which iterates over the tempo-map:
for every midi-event { for ever tempo until that midi-event {..} }
This does not scale to 3.5M events and 24k tempo-changes.
