It was not safe to immediately delete the AutomationPattern attached to
an InlineAutomation object before commit "InlineAutomation: delete
attached AutomationPattern immediately" as the AutomationPattern
destructor sometimes fiddles around with AutomationEditor (GUI!).
However this has changed now and this enables us to delete the
AutomationPattern immediately. This fixes some random crashes with
scheduling AutomationPatterns for later deletion.
Closes#2982696.
It was not possible to call any of the functions of AutomationEditor
from threads other than the main GUI thread as the m_pattern pointer
was not protected by a mutex. It is however desired to be able to call
for example AutomationEditor::setCurrentPattern(...) from different
threads (mainly required when deleting AutomationPatterns in other
threads). Thus made this method and all accesses to the m_pattern
pointer thread-safe.
The global channel link checkbox did not work as expected as of the
M/V-split during 0.3.x -> 0.4.x development. Connected the slot
responsible for updating individual ports after toggling checkbox and
fixed the logic behind.
Closes#2964800.
The per-note-panning support is broken (not sure whether this is a bug
in FluidSynth or our code) and leads to improperly panned sounds all
the time - even with all notes being panned to center. Thus disable it
until we have a proper fix for it.
Due to the FxMixer improvements some plugins such as the FLP import
filter now directly access the FxMixer and its view. Therefore the
symbols of the FxMixer and FxMixerView classes need to be exported for
the win32 builds.
There's currently no need for analyzing the sound data if
SpectrumAnalyzer's view is not visible. This saves some CPU time
if SpectrumAnalyzer's view is not shown.
For some effects it is helpful to know whether their attached EffectView
is visible or not. This mainly concerns effects that actually do not
touch the sound data but analyze it and display the results.
* calf-updates:
Simplify Multiband Compressor code and fix several bugs.
Remove zombie small plugin code.
Remove small plugins together with all the supporting infrastructure.
Add Doxygen documentation for plugin_registry class.
More refactoring.
General refactoring. Start implementing external GUI.
+ LV2: add external UI header (not used for anything yet)
Eliminated some (not all) repetition in the code. Fixed a bug where old value
for Q and separation was taken from the wrong variable. Fixed a bug where
having a band bypassed or muted caused the graph to stop redrawing properly.
(cherry picked from commit 3be583385b0bfb69bb7e52ee7426fe27d515560c)
The small plugins don't really fit into the project and should be moved somewhere
else instead. Unfortunately, I don't have enough time to do it myself, so if
anyone needs them, feel free to make a separate project out of them, or merge
them (if possible, it may be very hard) into some other small plugin library.
(cherry picked from commit e03505a778ac3b8edb7a389a11eae03cbacc79fe)
Make the result of plugin_registry::get_by_uri() const, as no external code
should modify metadata objects.
(cherry picked from commit 2f9ae953b30bc8702d302df58f0c7f1a589a15c3)
Constify metadata functions. Constify other functions that really needed
it. Make variables keeping old state mutable for now. Move filter composition
operations into biquad.h. Replace global function get_all_plugins with
a singleton class that returns the same list (as const) or a single plugin
by URI. Add a skeleton implementation of functions in lv2_external_ui
and skeleton UI related functions for external UI.
(cherry picked from commit ae7f7293961a3ea3b194b73ab35886f94e562cd7)
* calf-updates:
CALF: resync source files due to forgotten commits
+ JACK host: implemented Save/Save as...; Open not implemented yet - who could possibly want THAT?
+ EQ: Fix adding EQ plugin using menu in calfjackhost
Some changes in style, smaller h/v-scale added
+ LADSPA: make plugin IDs unique across Calf plugins, add an install-time check
+ EQ-5: refactoring to eliminate code duplication, side effect: stereo meters
+ EQ-n: shorter names for calfjackhost and others
+ Equalizers: convert lists of per-band parameters using macros (ugly, but effective)
+ EQ-8, EQ-12: reduce copypasta by using a class template
+ Compatibility: workaround for GStreamer crash bug in Ubuntu 9.10 (thanks to Stefan Kost for diagnosing the problem and suggesting the workaround)
+ Compressor simplification, added optional audio ports
Probably I forgot to cherry pick one or two commits from CALF Git repo
resulting in a slight diverge of source codes. Resynced the concerned
header files. Fixes the missing Pulsator plugin.
In Base64::decode() we mess around with pointers so it's more type-safe
to realize everything as a template function. This also silences compiler
warning about dereferencing type-punned pointer in Vibed plugin.
Removed some old projects that did not represent LMMS and its abilities
anymore. We also need to replace most of the "Cool songs" and replace
them by more recent projects from LSP.
Users reported the last buffer played repeatedly while the dialog asking
whether to save project is shown. This behaviour even continued while
loading a project. An easy workaround for this is to stop playing song
(or whichever TrackContainer is active at the moment) in
MainWindow::mayChangeProject().
Closes bug #2904781.
As of Stk 4.4 all classes and types have been moved to the namespace "stk".
This breaks build of the Mallets plugin. This can be easily fixed by
using the namespace "stk". For compatibility reasons we have to declare
the namespace before, because it did not exist in previous versions.
Closes#2870878.
It's not a good idea to allocate the buffer for VST chunk data on the
stack as chunks might become quite big (several megabytes) and thus
could cause a stack overflow. Fix this by using new/delete.
We have to lock Mixer when touching Song's state via Song::startExport()
and Song::stopExport() in ProjectRenderer::run() as the FIFO writer
thread may call Mixer::renderNextBuffer() (which calls Song::doActions())
simultaneously. Fixes a random segfault when exporting project.
This was a new bug as the ProjectRenderer does not operate FIFO-less
anymore.
* auto-save:
don't change the current project when auto-saving
auto-saves every minute and recovers upon crash
don't show WelcomeScreen when importing/loading
Conflicts:
src/gui/MainWindow.cpp
* fx-mixer-rework: (25 commits)
FxMixer: use new CPU::bufMixCoeff()
Cpu: added BufMixCoeff + SSE implementation
MixerWorkerThread: added job queue modes and improved class structure
MixerWorkerThread: exit outer loop in processJobQueue() if finished
Song: do not refresh FxMixerView when loading song in console mode
Mixer/FxMixer: separated MixerWorkerThread and ThreadableJob into files
Rough fix for freeze-up bug
Fixed the audio/visual screwups
NOT WORKING! Fx Mixer uses job threads
Mixer: rewrote/reorganized job queueing for worker threads
Fixed bug - Instruments had wrong channel models
Fixed a mixer bug regarding deleting channels
FxMixerView - Left and right to select channels
Fix FL Import with new mixer
Ability to save mixer sends to disk with the mmp
Channel selector has a max range of num channels
Ability to re-order mixer channels
Ability to delete mixer channels in the FX Mixer
Requested improvements to new FxMixerView$
Scrollbar for the fx mixer channels
...
There's no need to start the ProjectRenderer thread with high priority
anymore as the actual rendering is done on the other side of the FIFO.
The ProjectRenderer just waits for new data in the FIFO and encodes
them as they arrive.
Renamed the ProjectRenderer::OutputSettings structure to
ProjectRenderer::EncoderSettings to better reflect its meaning.
Additionally added some basic Doxygen comments.
Until now, Mixer not only was responsible for rendering audio buffers
but also managed writing them to audio backend (through a FIFO) and
handled various quality related parameters.
All this functionality has been moved into the new AudioOutputContext
class. It glues together AudioBackend (formerly called AudioDevice),
global quality settings and the Mixer.
The AudioOutputContext class creates a FifoWriter which calls
Mixer::renderNextBuffer() and writes the output into the BufferFifo of
the AudioOutputContext it belongs to. The BufferFifo is read by the
according AudioBackend which belongs to the AudioOutputContext as well.
The AudioOutputContext also handles resampling in case the AudioBackend
wants the buffers in a different samplerate.
During this rewrite the Mixer class and the according source files have
been renamed from "mixer" to "Mixer". This results in small changes
all over LMMS' code base.
This reverts commit c517f1fa5a.
The commit was not very helpful and introduced new xrun problems.
Instead I'll be rewriting the part of LMMS where Mixer, Mixer's
quality settings, Buffer FIFO and AudioDevice are sticked together.
