1
0
Fork 0
mirror of https://github.com/juce-framework/JUCE.git synced 2026-01-24 01:54:22 +00:00
JUCE/modules/juce_dsp/processors/juce_ProcessorChain.h
2022-04-04 12:36:32 +01:00

178 lines
6.4 KiB
C++

/*
==============================================================================
This file is part of the JUCE 7 technical preview.
Copyright (c) 2022 - Raw Material Software Limited
You may use this code under the terms of the GPL v3
(see www.gnu.org/licenses).
For the technical preview this file cannot be licensed commercially.
JUCE IS PROVIDED "AS IS" WITHOUT ANY WARRANTY, AND ALL WARRANTIES, WHETHER
EXPRESSED OR IMPLIED, INCLUDING MERCHANTABILITY AND FITNESS FOR PURPOSE, ARE
DISCLAIMED.
==============================================================================
*/
namespace juce
{
namespace dsp
{
//==============================================================================
#ifndef DOXYGEN
/** The contents of this namespace are used to implement ProcessorChain and should
not be used elsewhere. Their interfaces (and existence) are liable to change!
*/
namespace detail
{
template <typename Fn, typename Tuple, size_t... Ix>
constexpr void forEachInTuple (Fn&& fn, Tuple&& tuple, std::index_sequence<Ix...>)
{
(void) std::initializer_list<int> { ((void) fn (std::get<Ix> (tuple), std::integral_constant<size_t, Ix>()), 0)... };
}
template <typename T>
using TupleIndexSequence = std::make_index_sequence<std::tuple_size<std::remove_cv_t<std::remove_reference_t<T>>>::value>;
template <typename Fn, typename Tuple>
constexpr void forEachInTuple (Fn&& fn, Tuple&& tuple)
{
forEachInTuple (std::forward<Fn> (fn), std::forward<Tuple> (tuple), TupleIndexSequence<Tuple>{});
}
// This could be a template variable, but that code causes an internal compiler error in MSVC 19.00.24215
template <typename Context, size_t Ix>
struct UseContextDirectly
{
static constexpr auto value = ! Context::usesSeparateInputAndOutputBlocks() || Ix == 0;
};
}
#endif
/** This variadically-templated class lets you join together any number of processor
classes into a single processor which will call process() on them all in sequence.
@tags{DSP}
*/
template <typename... Processors>
class ProcessorChain
{
public:
/** Get a reference to the processor at index `Index`. */
template <int Index> auto& get() noexcept { return std::get<Index> (processors); }
/** Get a reference to the processor at index `Index`. */
template <int Index> const auto& get() const noexcept { return std::get<Index> (processors); }
/** Set the processor at index `Index` to be bypassed or enabled. */
template <int Index>
void setBypassed (bool b) noexcept { bypassed[(size_t) Index] = b; }
/** Query whether the processor at index `Index` is bypassed. */
template <int Index>
bool isBypassed() const noexcept { return bypassed[(size_t) Index]; }
/** Prepare all inner processors with the provided `ProcessSpec`. */
void prepare (const ProcessSpec& spec)
{
detail::forEachInTuple ([&] (auto& proc, auto) { proc.prepare (spec); }, processors);
}
/** Reset all inner processors. */
void reset()
{
detail::forEachInTuple ([] (auto& proc, auto) { proc.reset(); }, processors);
}
/** Process `context` through all inner processors in sequence. */
template <typename ProcessContext>
void process (const ProcessContext& context) noexcept
{
detail::forEachInTuple ([this, &context] (auto& proc, auto index) noexcept { this->processOne (context, proc, index); },
processors);
}
private:
template <typename Context, typename Proc, size_t Ix, std::enable_if_t<! detail::UseContextDirectly<Context, Ix>::value, int> = 0>
void processOne (const Context& context, Proc& proc, std::integral_constant<size_t, Ix>) noexcept
{
jassert (context.getOutputBlock().getNumChannels() == context.getInputBlock().getNumChannels());
ProcessContextReplacing<typename Context::SampleType> replacingContext (context.getOutputBlock());
replacingContext.isBypassed = (bypassed[Ix] || context.isBypassed);
proc.process (replacingContext);
}
template <typename Context, typename Proc, size_t Ix, std::enable_if_t<detail::UseContextDirectly<Context, Ix>::value, int> = 0>
void processOne (const Context& context, Proc& proc, std::integral_constant<size_t, Ix>) noexcept
{
auto contextCopy = context;
contextCopy.isBypassed = (bypassed[Ix] || context.isBypassed);
proc.process (contextCopy);
}
std::tuple<Processors...> processors;
std::array<bool, sizeof...(Processors)> bypassed { {} };
};
/** Non-member equivalent of ProcessorChain::get which avoids awkward
member template syntax.
*/
template <int Index, typename... Processors>
inline auto& get (ProcessorChain<Processors...>& chain) noexcept
{
return chain.template get<Index>();
}
/** Non-member equivalent of ProcessorChain::get which avoids awkward
member template syntax.
*/
template <int Index, typename... Processors>
inline auto& get (const ProcessorChain<Processors...>& chain) noexcept
{
return chain.template get<Index>();
}
/** Non-member equivalent of ProcessorChain::setBypassed which avoids awkward
member template syntax.
*/
template <int Index, typename... Processors>
inline void setBypassed (ProcessorChain<Processors...>& chain, bool bypassed) noexcept
{
chain.template setBypassed<Index> (bypassed);
}
/** Non-member equivalent of ProcessorChain::isBypassed which avoids awkward
member template syntax.
*/
template <int Index, typename... Processors>
inline bool isBypassed (const ProcessorChain<Processors...>& chain) noexcept
{
return chain.template isBypassed<Index>();
}
} // namespace dsp
} // namespace juce
#ifndef DOXYGEN
namespace std
{
JUCE_BEGIN_IGNORE_WARNINGS_GCC_LIKE ("-Wmismatched-tags")
/** Adds support for C++17 structured bindings. */
template <typename... Processors>
struct tuple_size<::juce::dsp::ProcessorChain<Processors...>> : integral_constant<size_t, sizeof... (Processors)> {};
/** Adds support for C++17 structured bindings. */
template <size_t I, typename... Processors>
struct tuple_element<I, ::juce::dsp::ProcessorChain<Processors...>> : tuple_element<I, tuple<Processors...>> {};
JUCE_END_IGNORE_WARNINGS_GCC_LIKE
} // namespace std
#endif