This fixes the problem of the invalid /nonexistent_path path making all
certificate validation fail on Fedora. /etc/ssl/certs only recently
started working like on Debian, but it is enough that ca_info works, as
long as /etc/ssl/certs exists and doesn't make cert validation fail.
In the presence of tempo-changes distinguishing between offsets and
absolute positions is signficant. It is only valid to convert absolute
times using the tempo-map
Furthermore since GUI zoom-factor is time-invariant (samples per pixel),
all GUI operations must explictly use samples (or timecnt). It is not
valid (and problematic) to use use a location dependent timepos.
This is the first step to fix various MIDI edit issues for sessions with
tempo-changes.
The old code, using .earlier returned an absolute position when
calculating a relative distance. This is only valid if the session has a
fixed tempo, and the origin is irrelevant when converting the timepos.
This resulted in follow up issues since there is a difference when
summing two positions (each with an origin) vs adding an offset to a
position.
Note: this API changes breaks compilation until the GUI is updated.
This fixes various rounding issues. Notably superclock to sample
conversion must always round down when playing forward.
`::process (start, end, speed = 1)` uses exclusive end.
Processing begins at `start` and end ends just before `end`.
Next cycle will begin with the current end.
One example where this failed:
- New session at 48kHz
- Change tempo to 130 BPM
- Enable snap to 1/8 note
- Snap playhead to 1|3|0
- Enable Metronome
- Play
`assert (superclock_to_samples ((*i).sclock(), sample_rate()) < end);`
end = 177231 samples == superclock 1042118280
A grid point is found at superclock 1042116920 (that is < 1042118280).
However converting it back to samples rounded it to sample 177231 == end,
while actual location is 1360 super-clock ticks before end.
The metronome click has to be started this cycle, since the same
position will not be found at the beginning of the next cycle, with
start = 177232.
Similarly a samplecnt_t t, converted to music-time and back must not be
later than the given sample.
```
timepos_t tsc (t);
assert (timepos_t::from_ticks (tsc.ticks ()).samples () <= t);
```
IOW. When playing forward, all super-clock time between 1|1|0 and 1|1|1
should round down to 1|1|0. "We have not yet reached the first tick".
This allows to export a session from the commandline tool
```Lua
AudioEngine:set_backend("None (Dummy)", "", "")
s = load_session ("/path/to/session", "snapshot")
assert (s);
e = s:simple_export()
assert (e:check_outputs ())
e:run_export ()
e = nil
```
This fixes an issue when destroying VST plugin UIs
rather than hiding them. Even though is hidden before destroying it,
the vistracker is also destroyed in the same function call, before
the event can be processed. The window state remains visible="1"
Music-time has a significantly reduced resolution.
When evaluating parameter automation using sample-time, different
samples can be mapped back to the same beat-tick. This can result in
endless loops.
The same automation event is found again:
```
start: 112640 next_event.when: b13334 -> dist a5892029@a662323200 in dist.samples: 1002
start: 113642 next_event.when: b13334 -> dist a269@a668214960 dist.samples: 0
```
Note this fix may not be correct, since time-domains are mixed.
"end" is using sample-time. Furthermore The loop is only entered
if there is an event found using timepos_t(start).
Further work will be required (ideally use the same resolution for
all time-domains, or force everything that is evaluated in RT-context
to a single time-domain)
Program-name is not required to indicate "Tracks", "Busses", etc
Those are always by provided by PROGRAM_NAME.
"Other" really indicates external JACK (or Pipewire) applications.
Here "External" is a better label.
"Misc" can potentially be interpreted ambiguously, hence keeping
PROGRAM_NAME is useful.
OS specific versions part of a http query has been Ardour
only. Mixbus never posted any of this information. The section
was #ifdef'ed out. Removing this code for good, simplifies merges.
* Update button when the name of a connected route changes
* Indicate partial internal connections (previously only the
client name ("ardour") was displayed.
* Escape markup in I/O plugin and client names
* Refactor code, use a common base-class for route and
route-less I/O Buttons
This fixes a deadlock, because tempo-map write copy
was not released. A later call to TempoMap::write_copy
will deadlock and in this case also hold the process lock.
This can happen if there is more than one tempo on the
timeline or if the tempo-difference is too large.
Removing the shadow makes toolbar and editor-tool buttons align
when the editor-mixer is hidden. It also improves consistency
with all other tabs that have no shadow on the left.
Increasing the table border width to 1, makes aligns the
editor-mixer button with the main toolbar.
