add audio buffer class to remove std::span dependency

util/audio_buffer.h

@@ -0,0 +1,77 @@
+#pragma once
+#include <algorithm>
+#include <cassert>
+#include <vector>
+
+template <typename t_sample>
+class audio_buffer {
+public:
+	audio_buffer(const t_sample** input, int channels, int frames)
+		: m_data(channels * frames)
+		, m_channels(channels)
+		, m_frames(frames)
+		, m_channel_ptrs(channels)
+	{
+		for (int ch = 0; ch < channels; ++ch) {
+			std::copy(input[ch], input[ch] + frames, m_data.begin() + ch * frames);
+		}
+		update_channel_ptrs();
+	}
+
+	audio_buffer(int channels, int frames)
+		: m_channels(channels)
+		, m_frames(frames)
+		, m_data(channels * frames)
+		, m_channel_ptrs(channels)
+	{
+		update_channel_ptrs();
+	}
+
+	void set_size(int channels, int frames)
+	{
+		m_channels = channels;
+		m_frames = frames;
+		m_data.resize(channels * frames);
+		m_channel_ptrs.resize(channels);
+		update_channel_ptrs();
+	}
+
+	void set_data(const t_sample** input, int channels, int frames)
+	{
+		set_size(channels, frames);
+		for (int ch = 0; ch < channels; ++ch) {
+			std::copy(input[ch], input[ch] + frames, m_data.begin() + ch * frames);
+		}
+		update_channel_ptrs();
+	}
+
+	int num_samples() const { return m_frames; }
+
+	int num_channels() const { return m_channels; }
+
+	t_sample* data() { return m_data.data(); }
+
+	const t_sample* data() const { return m_data.data(); }
+
+	// t_sample** access, always up-to-date after construction/resize
+	t_sample** write_ptrs() { return m_channel_ptrs.data(); }
+
+	t_sample* write_ptr(int channel)
+	{
+		assert(channel >= 0 && channel < m_channels);
+		return m_channel_ptrs[channel];
+	}
+
+	const t_sample* const* channel_ptrs() const { return m_channel_ptrs.data(); }
+
+private:
+	void update_channel_ptrs()
+	{
+		for (int ch = 0; ch < m_channels; ++ch) { m_channel_ptrs[ch] = m_data.data() + ch * m_frames; }
+	}
+
+	std::vector<t_sample> m_data;
+	int m_channels;
+	int m_frames; // samples per channel
+	std::vector<t_sample*> m_channel_ptrs;
+};