Mads Kiilerich
b4ff4f356c
Make header files more self-contained - add missing Ardour and lib includes
2024-10-20 03:30:46 +02:00
626a7cb506
manually fix conflicts during merge with ardour
2024-10-19 19:21:25 -06:00
Mads Kiilerich
955e634371
Make header files more self-contained - add missing std includes
2024-10-20 03:11:53 +02:00
Mads Kiilerich
8a5f21d8e3
Drop some unused files - avoid trying to maintain stuff that can't be tested
2024-10-20 03:11:53 +02:00
Mads Kiilerich
140670541f
Use .inc.h for #include files that not are plain headers
...
Regular .h files *should* be self-contained and independent of previous
includes and guarded to only include once. Make it clear which files
that *doesn't* apply for at all.
2024-10-20 03:11:53 +02:00
Mads Kiilerich
38db12c9ea
Drop Frontier Tranzport surface - it has been dead for a decade
2024-10-20 03:08:03 +02:00
0d71876189
signals_generated.h is no longer required
2024-10-19 21:54:34 +02:00
7b74faaf57
Tidy up libs/lua/LuaBridge/detail/Userdata.h
2024-10-19 18:38:59 +02:00
Mads Kiilerich
a7eb3b7341
un-boost: Add missing #include <cmath> in audiographer silence_trimmer.h
2024-10-19 12:56:43 +02:00
Mads Kiilerich
d217b2f8ad
un-boost: Add missing #include <functional> in rt_task.h
2024-10-19 12:42:25 +02:00
f8e05a595d
Remove conflicting include
2024-10-19 06:27:50 +02:00
7b24d38c0f
Add more missing include (fix windows builds)
2024-10-19 04:58:30 +02:00
af4d70c243
Use heap to build peak data
2024-10-19 04:49:13 +02:00
0a36cdedac
Add missing include (fix windows builds)
2024-10-19 04:46:35 +02:00
74c4ca3e52
Reduce reliance on boost - the hard part
...
the rest from `tools/convert_boost.sh`.
* replace boost::function, boost::bind with std::function and std::bind.
This required some manual fixes, notably std::placeholders,
some static_casts<>, and boost::function::clear -> = {}.
2024-10-19 03:47:21 +02:00
ff95d81612
Reduce reliance on boost - the easy part
...
* boost::unordered_map -> std::unordered_map
* BOOST_STATIC_ASSERT/static_assert
* BOOST_FOREACH -> for
* boost::tuple -> std::tuple/g
* boost::math::isnormal -> std::isnormal
* boost::container::set -> std::set
* boost::none -> std::nullopt
* boost::optional -> std::optional
2024-10-19 03:41:16 +02:00
168b917730
Prepare for std::optional
2024-10-19 03:40:04 +02:00
8eb9263af2
Replace boost::format with PBD::string_compose
2024-10-19 03:39:57 +02:00
Alejandro Domínguez
2436b4df06
Replace boost::noncopyable with C++11 deleted copy constructors
2024-10-19 01:04:15 +02:00
Alejandro Domínguez
915200699b
Replace boost::lexical_cast with std equivalent functions
2024-10-19 01:04:15 +02:00
Alejandro Domínguez
7a0428644f
Replace boost::assign::map_list_of with initializer_list constructor
2024-10-19 01:04:15 +02:00
Alejandro Domínguez
abf7a78984
Replace boost::ptr_list with std::list
2024-10-19 01:04:14 +02:00
Alejandro Domínguez
e326426dbc
Replace boost type traits with std version
2024-10-19 01:04:14 +02:00
Alejandro Domínguez
88e38b2699
Replace boost atomics with std version
2024-10-19 01:04:14 +02:00
Alejandro Domínguez
1ee044c5fc
Replace boost::erase_first with std equivalent
2024-10-19 01:04:14 +02:00
Alejandro Domínguez
1ffb70f670
Replace boost::scoped_ptr with std::unique_ptr (1/2)
2024-10-19 01:04:14 +02:00
30dc9ccc86
Replace boost::scoped_array<T>
2024-10-19 01:04:02 +02:00
2c503eab09
Remove old Coreaudio OSX 10.5 support
2024-10-18 22:52:34 +02:00
99906a2215
Drop build-system support for macOS < 10.13 (require C++17)
2024-10-18 22:52:29 +02:00
bd8525e5f1
Remove deprecated PBD::SignalN
2024-10-18 20:46:31 +02:00
2d7cce44f1
Replace PBD::Signals (1/2)
2024-10-18 20:41:08 +02:00
Alejandro Domínguez
0ade0b2212
Deduplicate SignalWithCombiner<Combiner, void, A...> specialization
2024-10-18 20:41:08 +02:00
Alejandro Domínguez
16dd8528c8
Make SignalN template variadic
2024-10-18 20:41:05 +02:00
837ea32898
actally stop doing clip recording visual stuff when disarmed (libs)
2024-10-18 09:50:48 -06:00
92e452609a
Fix I/O Tasklist RR schedule option
2024-10-18 15:11:08 +02:00
7608d4ade2
API changes for clip recording in TriggerBox
...
Also, disarm all triggers when track-level rec-enable is disabled
2024-10-17 07:44:33 -06:00
a8f58105a9
provide a class-level static PBD::Signal for Trigger (Slot) arming
2024-10-17 07:44:33 -06:00
6b1659497d
add #warning
2024-10-17 07:44:33 -06:00
7c944687c9
preparations for clip data display (MIDI) while recording
2024-10-17 07:44:33 -06:00
c03c3dd918
fix unused variable warning from clang
2024-10-17 07:44:32 -06:00
dab9775223
fix bitwise-vs-logic AND (thanks, clang!)
2024-10-17 07:44:32 -06:00
c2deacdd54
fix definition of Trigger::playable()
...
The recent change to only use the data object breaks GUI handling
of a PropertyChange signal after _region is set following a drag
and drop
2024-10-17 07:44:32 -06:00
62a4df52c7
tentative steps in infrastructure for displaying clip data during capture
2024-10-17 07:44:32 -06:00
6193535a31
apply a fade in & out to captured audio clip data, before writing to disk
...
This policy may need revising later.
2024-10-17 07:44:32 -06:00
87ab0a9e0f
add config vars to control max size of captured MIDI and audio clips
2024-10-17 07:44:32 -06:00
8305ea5762
during MIDI clip recording, convert audio->beat time at capture time
...
This distributes whatever cost there is associated with the domain conversion across
multiple process() calls, rather than doing it all in the process() call where
capture finishes.
2024-10-17 07:44:32 -06:00
Paul Davis
cbc0021182
fix typo (| instead of ||)
2024-10-17 07:44:32 -06:00
1e8b2c521c
provide a better method to decide if a trigger(slot) is occupied
...
This is necessary with clip recording because for some short time after recording,
a trigger may be playable despite not yet having a region.
libs edition.
2024-10-17 07:44:32 -06:00
596c54d742
fix ups for record state in TriggerBox
2024-10-17 07:44:32 -06:00
d7d208dc34
move RecordState enum from ARDOUR::Session to ARDOUR so we can use it in TriggerBox (libs)
2024-10-17 07:44:32 -06:00
6ef4a74728
cleanup of bits of clip recording code
2024-10-17 07:44:32 -06:00
4466563a89
avoid double delete of clip recording audio buffers
2024-10-17 07:44:32 -06:00
25400a4936
fix appending new audio data during clip recording
2024-10-17 07:44:32 -06:00
7455b99b45
account for clip recording when computing monitoring state
2024-10-17 07:44:32 -06:00
a357575954
less debugging
2024-10-17 07:44:32 -06:00
423148c2af
tweaks to cleanup and get much closer to full audio clip recording
2024-10-17 07:44:32 -06:00
26fb50d1a9
a bit more type-safety when handling audio data during clip recording
2024-10-17 07:44:32 -06:00
d7c424c440
attempt to write and use audio data to disk for audio clip recording
2024-10-17 07:44:32 -06:00
0959696070
NO-OP: remove pointless embedded whitespace
2024-10-17 07:44:32 -06:00
b4d0b06d95
const-ify first "src" argument in AudioSource::write() and derivatives
2024-10-17 07:44:32 -06:00
74132b60a0
get the data collection and initial setup part of audio clip recording working
...
Not yet implemented: writing the data to disk and creating a new Region
2024-10-17 07:44:32 -06:00
ac4bb55f13
first functioning MIDI clip recording (libs)
2024-10-17 07:44:32 -06:00
a2d44ba97d
remove ClipRecProcessor, whose functionality will move into TriggerBox where it should have been
2024-10-17 07:44:32 -06:00
1fa42dcb95
provide rec-enable state for TriggerBoxen
2024-10-17 07:44:32 -06:00
052ef18c4a
cliprec: add processor to tracks, and get MIDITrigger prepared for capture
2024-10-17 07:44:32 -06:00
86b01a5d2f
provide an interesting method to convert an RTMidiBuffer<samples> to RTMidiBuffer<beats> without any memory reallocation
2024-10-17 07:44:32 -06:00
88c326aee0
more clip recording refactoring
2024-10-17 07:44:32 -06:00
aa9a998f60
small steps towards clip recording
2024-10-17 07:44:32 -06:00
107706af8a
some tiny steps towards using new data structures for clip recording
2024-10-17 07:44:32 -06:00
75f56cd3e1
the great onceification (libs)
...
Replace use of #ifndef header guards with #pragma once
Modern C++, baby!
2024-10-17 07:44:31 -06:00
a9205ffd55
use nullptr, not NULL
2024-10-17 07:44:31 -06:00
8baaa7eb66
TriggerReference requires some sort of lifetime tracking
...
For now we use std::weak_ptr and std::enable_shared_from_this to accomplish tracking. There
may be an argument for using our own (PBD::Destructible) mechanisms instead.
2024-10-17 07:44:31 -06:00
e3ff81efcb
remove dangling line of code that does nothing
2024-10-17 07:44:31 -06:00
7901b4119f
we're gonna need a bigger debug set
2024-10-17 07:44:31 -06:00
c8924f4133
fix lollipop drawing
...
Cairo coordinate/argument limits are much smaller than the canvas. Trying
to clip to a gigantic rectangle throws cairo into a weird state, so
make sure we only clip to a part an exposed area
2024-10-17 07:44:31 -06:00
e67a3387fd
Fix Editor Lua bindings
...
Methods that have been moved to EditingContext need
to reference the abstract definition there, and PublicEditor's
first parent class also need to be EditingContext (same
memory address)
2024-10-17 07:44:31 -06:00
7e7440bc33
Fix Lua bindings, Session is-a HistoryOwner
...
For derived classes to work in Lua w/o explicit cast,
it needs to be the first parent.
2024-10-17 07:44:31 -06:00
2fceb66193
manual fixups for errors during rebasing against master
2024-10-17 07:44:31 -06:00
911ad78c06
add API to access PBD::UndoHistory member of HistoryOwner
2024-10-17 07:44:31 -06:00
a61f49ad4f
show Bindings _name in DEBUG_TRACE output
2024-10-17 07:44:31 -06:00
42959b1313
adjust MidiModel API to require HistoryOwner not Session for commands
2024-10-17 07:44:31 -06:00
8c086693af
show HistoryOwner::_name in DEBUG_TRACE messages
2024-10-17 07:44:31 -06:00
b989464914
NO-OP: add useful comment
2024-10-17 07:44:31 -06:00
e8fdbb8cd9
Session IS-A history owner
2024-10-17 07:44:31 -06:00
d30c8a1286
abstract concept of a history owner from ARDOUR::Session into libpbd
2024-10-17 07:44:31 -06:00
fd6afb30e6
use MidiModel::ContentsChanged to drive MIDI cue edit swaps
...
No need for a special method called post-edit
2024-10-17 07:44:31 -06:00
62d36832c6
more changes to get MIDI clip editing working
...
MIDITrigger now has a direct reference to a MidiModel, and uses
that as the basis for discovering what state has changed and needs
updated after an edit operation pushes a new state to the trigger
2024-10-17 07:44:30 -06:00
7bf464795a
add ::render() to MidiModel since that's what we'll be editing
2024-10-17 07:44:30 -06:00
06d5496f70
further preparation for MIDI cue editing in triggerbox code
2024-10-17 07:44:30 -06:00
9060a32c34
MIDI state trackers: dump() is const
...
Plus minor fixes to ::resolve_diff()
2024-10-17 07:44:30 -06:00
7dee98279a
extend MidiModel and MIDITrigger in preparation for edit-change handling
2024-10-17 07:44:30 -06:00
5da8de05ca
NO-OP: internal whitespace and newline cleanup
2024-10-17 07:44:30 -06:00
4267d5b0d7
MidiStateTracker: extend API to allow "diffs" between two MidiStateTrackers
2024-10-17 07:44:30 -06:00
d550292f8f
additional DEBUG_TRACE (Destruction) output
2024-10-17 07:44:30 -06:00
e7731f2e89
add -D actions to gtkmm2ext
2024-10-17 07:44:30 -06:00
009a0e6ffb
stacktraces to help track down missing actions
2024-10-17 07:44:30 -06:00
59ed3d7138
cleanup canvas piano roll header so that it can exist with no current MidiView
2024-10-17 07:44:30 -06:00
10244c0360
tempo bar and BBT ruler in MIDI cue editor
2024-10-17 07:44:30 -06:00
b35b30c230
factor out code to extract a TempoMap from an SMF
2024-10-17 07:44:29 -06:00
e3205bded0
do something to make MIDI bindings accessible in any EditingContext
2024-10-17 07:44:29 -06:00
10b48d2cc4
lock in some major steps for the midi cue/pianoroll editor
2024-10-17 07:44:29 -06:00
0eeb281bd6
remove debug output
2024-10-17 07:44:29 -06:00
8e4edd0c07
Revert "separate out all bounds/position info from Region into "Slice""
...
This reverts commit f3da2cfd8b9dc077ee35fc3bbaf133adec23d463.
2024-10-17 07:44:29 -06:00
7a5d7cfe69
Revert "add set_* methods to Slice; remove property additions"
...
This reverts commit 81eee23baab2817218c1766c7d3c37fb7435e619.
2024-10-17 07:44:29 -06:00
659382ecd8
convert debug output from printf to type-safe iostreams
2024-10-17 07:44:29 -06:00
a6e02d7a94
add set_* methods to Slice; remove property additions
...
Derived classes (currently only Region) just register the Slice properties
_start and _length.
2024-10-17 07:44:29 -06:00
45328723b0
separate out all bounds/position info from Region into "Slice"
...
The idea here is to be able to describe region size, start and position
independently of an actual Region object.
2024-10-17 07:44:29 -06:00
7799adc8db
make Stateful a virtual base class of StatefulDestructible
...
This permits dual inheritance from Stateful.
2024-10-17 07:44:29 -06:00
204d4237f8
further steps towards MidiRegionView outside the Editor
2024-10-17 07:44:29 -06:00
e3d790207f
modify MIDITrigger to use an RTMidiBuffer<Beats,Beats> for playback, not a MidiModel
...
Creating an iterator on a MidiModel (Sequence) creates a read-lock on the same,
which exists until the iterator is destroyed.
This new designs renders the model/source to an RTMidiBuffer, then atomically
swaps in a new RTMidiBuffer in an RT context.
Not yet implemented in this temporary branch is handling required
state-changing messages when the buffer is swapped.
2024-10-17 07:44:28 -06:00
a4a277c141
templatize RTMidiBuffer to allow use in different time domains
...
Also add ::track_state() method to configue a MidiStateTracker
to correspond to the state implied by an RTMidiBuffer<T,D>
at a given point in time.
2024-10-17 07:44:28 -06:00
b693d07fcb
Add a new ::render() method to MidiSource that writes to an EventSink
2024-10-17 07:44:28 -06:00
9c292ed6f4
Merge branch 'ardour'
2024-10-17 15:38:42 +02:00
0986b8d1e8
Revert "Only enable RegionFX in debug builds for the time being"
...
This reverts commit d525f8d60c
.
2024-10-17 15:36:29 +02:00
John Emmas
b6cb758357
Another minor tweak to fix a build issue with MSVC/c++17
...
Fixes a conflict between 'std::byte' (new in c++17) and earlier declarations / typedef's that defined 'byte'
2024-10-17 12:07:05 +01:00
d525f8d60c
Only enable RegionFX in debug builds for the time being
...
This reverts commit 14ff2f2e68
.
2024-10-16 18:29:51 +02:00
6149771910
fix yet more cases where SMF are imported and named incorrectly
2024-10-15 22:35:04 -06:00
Mads Kiilerich
0d4bce8663
backends: Fix class name in debug and error messages for set_latency_range and get_latency_range traces
2024-10-15 00:41:12 +02:00
Mads Kiilerich
99ba70384a
FaderPort8: Improve DEBUG_TRACE messages
2024-10-15 00:41:12 +02:00
Mads Kiilerich
5c763c99ba
FaderPort8: Use explicit "Port 1" for FP8 too
...
84e38b4c65
made the same change for FP16.
It has apparently not been necessary for FP8 so far, because backends
have happened to report "Port 1" before the control port. But that is
not necessarily the case. It failed for me when playing around.
Change FP8 handling to also not make assumptions about the device order
returned by the backend.
2024-10-15 00:41:12 +02:00
Mads Kiilerich
b7249c8fe7
build: Set _POSIX_C_SOURCE=200809L instead of comment about -std=gnu99 in 8fea1ea42e
...
We are converging towards setting it globally.
2024-10-14 22:57:15 +02:00
eb87b0e2eb
Remove old safety check
...
These days it only produces false positives (locate at start
of session export).
We leave `ENSURE_PROCESS_THREAD` in place, since it may
come in handy at some point (eg. static analysis, documentation)
2024-10-14 22:52:37 +02:00
8115578d4e
Add RT thread priority debugging
2024-10-14 21:49:56 +02:00
63d3d1ff3d
Initialize fader's numeric entry adjustment
...
Previously the internal adjustment used for numeric entry
was left at 0 if the BarController's adjustment was never modified.
2024-10-14 17:58:00 +02:00
John Emmas
fd8808d7b8
Minor declaration issue that upsets MSVC/c++17
2024-10-13 11:39:48 +01:00
0aff098541
Fix IO Thread priority
...
This adds `pbd_pthread_priority` indirection to correctly get
the absolute thread priority.
and for consistency a 4 letter enum is used.
2024-10-11 10:06:49 +02:00
be4dab7336
Merge branch 'ardour'
2024-10-10 03:18:28 +02:00
5efa5b4529
Optimize audio region read when no regionFX are used
2024-10-10 02:45:03 +02:00
43dd75b10e
invalidate audio region cache when the region _start changes
2024-10-09 16:53:55 -06:00
14ff2f2e68
Revert "Only enable RegionFX in debug builds for the time being"
...
This reverts commit cae710cd7c
.
2024-10-09 20:49:36 +02:00
2a96c9ce98
fix issues with MIDI playback (and who knows what else) near start of roll
2024-10-08 08:36:23 -06:00
cae710cd7c
Only enable RegionFX in debug builds for the time being
2024-10-08 16:23:43 +02:00
59e98bd824
Fix reading peak-file after EOF
...
This issue was introduced in b28090c64c
2024-10-08 11:54:14 +02:00
33806a2735
Revert "Only enable RegionFX in debug builds for the time being"
...
This reverts commit 2d076cccb1
.
2024-10-03 17:28:09 +02:00
e3e014bfe6
NO-OP: comment work
2024-10-02 11:03:08 -06:00
52336eb2f0
fix MIDI Clock output times by using the correct call to get tempo at a position
...
Could be worth a double check of the codebase/hiding the API that ignores ramping
2024-10-02 09:13:28 -06:00
adc9d9e0af
VST3: work around plugins that do not heed ContextInfo::kSendCount
...
see also c5618f01d6
2024-10-01 23:03:19 +02:00
John Emmas
07c79ce92c
Small changes to make pthread_utils.cc buildable again with MSVC
2024-10-01 20:28:16 +01:00
44b2377e72
VST3: Fix possible deadlock when using PSL extension for sends
...
see also c5618f01d6
2024-10-01 19:24:36 +02:00
ca7ac7027b
VST3: add more debug messages for PSL extensions
2024-10-01 01:21:35 +02:00
22a2cb0624
Ignore inactive routes for pre-roll sub-cycles
2024-09-30 23:36:36 +02:00
1aad6805b3
Fix count-in/preroll recording offsets
...
Notably `Route::process_output_buffers` uses
```
output_latency = speed * _output_latency;
```
here, speed already needs to be non-zero during count-in
and pre-roll.
2024-09-30 23:36:06 +02:00
dab22a7c70
Explicitly set Windows Process Scheduling Class
2024-09-30 19:57:53 +02:00
ed437afda7
Fix thread priorities for Windows builds
2024-09-30 19:07:53 +02:00
11f71a3297
Fix playback of sessions with low sample-rate
...
previously, _chunksize =
```
minimum_disk_read_bytes / sizeof (Sample)
```
can become larger then actual allocated ringbuffer:
```
audio_playback_buffer_seconds * sample-rate
```
In which case the buffer was never filled.
The disk writer has similar issues
2024-09-29 18:25:33 +02:00
46f61d7662
Fix and simplify Playlist::fade_range for multiple ranges
2024-09-28 21:16:55 +02:00
28605b5351
Fix Audio region fade property mapping
...
This fixes an issue when undoing region fades.
Particularly but not limited to the following:
1. select a region and split it in the middle
2. switch to range tool. make a range selection across the split
3. Edit > Fade > Fade range seection (or press `/`)
4. Undo
Previously the fade-out of the earlier region was not undone.
2024-09-28 21:15:57 +02:00
1d921dec0b
Debug thread sched_priority
2024-09-28 14:47:52 +02:00
2014faaeca
Fix BaseUI thread priority
2024-09-28 14:45:05 +02:00
dd4a1a6d73
Set thread priority relative to backend
...
This also removed direct calls to backend real_time_priority
for good measure.
2024-09-28 04:15:39 +02:00