til: add preliminary audio backend

This is an early implementation of something resembling an audio
backend for rototiller/libtil.

The assumption for now is that everything will use signed 16-bit
native-endian stereo output @ 44.1khz.

1 files changed, 143 insertions, 0 deletions

diff --git a/src/mem_audio.c b/src/mem_audio.c
new file mode 100644
index 0000000..b8d51df
--- /dev/null
+++ b/src/mem_audio.c
@@ -0,0 +1,143 @@
+#include <assert.h>
+#include <errno.h>
+#include <stdlib.h>
+
+#include "til.h"
+#include "til_audio.h"
+#include "til_audio_context.h"
+#include "til_settings.h"
+#include "til_setup.h"
+
+/* "mem" audio backend */
+
+typedef struct mem_audio_setup_t {
+	til_setup_t	til_setup;
+} mem_audio_setup_t;
+
+typedef struct mem_audio_t {
+	til_audio_context_t	til_audio_context;
+
+	unsigned		n_queued;
+	unsigned		n_queued_start_ticks;
+	unsigned		paused:1;
+} mem_audio_t;
+
+
+static int mem_audio_init(til_setup_t *setup, til_audio_context_t **res_context);
+static int mem_audio_queue(til_audio_context_t *context, int16_t *frames, int n_frames);;
+static unsigned mem_audio_n_queued(til_audio_context_t *context);
+static void mem_audio_drop(til_audio_context_t *context);
+static void mem_audio_pause(til_audio_context_t *context);
+static void mem_audio_unpause(til_audio_context_t *context);
+static int mem_audio_setup(const til_settings_t *settings, til_setting_t **res_setting, const til_setting_desc_t **res_desc, til_setup_t **res_setup);
+
+
+til_audio_ops_t mem_audio_ops = {
+	.init = mem_audio_init,
+	.drop = mem_audio_drop,
+	.pause = mem_audio_pause,
+	.unpause = mem_audio_unpause,
+	.queue = mem_audio_queue,
+	.n_queued = mem_audio_n_queued,
+	.setup = mem_audio_setup,
+};
+
+
+/* this simulates an audio timer grinding through the queued frames when unpaused,
+ * returns the remaining n_queued (if any) for convenience, but it's also maintained
+ * at c->n_queued.
+ */
+static unsigned mem_refresh_n_queued(mem_audio_t *c)
+{
+	if (!c->paused && c->n_queued) {
+		unsigned	now = til_ticks_now();
+		unsigned	n_played = ((float)(now - c->n_queued_start_ticks) * 44.1f);
+
+		c->n_queued -= MIN(c->n_queued, n_played);
+	}
+
+	return c->n_queued;
+}
+
+
+static int mem_audio_init(til_setup_t *setup, til_audio_context_t **res_context)
+{
+	mem_audio_t	*c;
+
+	assert(setup);
+	assert(res_context);
+
+	c = til_audio_context_new(&mem_audio_ops, sizeof(mem_audio_t), setup);
+	if (!c)
+		return -ENOMEM;
+
+	c->paused = 1;
+	*res_context = &c->til_audio_context;
+
+	return 0;
+}
+
+
+static void mem_audio_drop(til_audio_context_t *context)
+{
+	mem_audio_t	*c = (mem_audio_t *)context;
+
+	c->n_queued = 0;
+}
+
+
+static void mem_audio_pause(til_audio_context_t *context)
+{
+	mem_audio_t	*c = (mem_audio_t *)context;
+
+	if (!c->paused) {
+		mem_refresh_n_queued(c);
+		c->paused = 1;
+	}
+}
+
+
+static void mem_audio_unpause(til_audio_context_t *context)
+{
+	mem_audio_t	*c = (mem_audio_t *)context;
+
+	if (c->paused) {
+		c->paused = 0;
+		c->n_queued_start_ticks = til_ticks_now();
+	}
+}
+
+
+static int mem_audio_queue(til_audio_context_t *context, int16_t *frames, int n_frames)
+{
+	mem_audio_t	*c = (mem_audio_t *)context;
+
+	mem_refresh_n_queued(c);
+	c->n_queued += n_frames;
+
+	return 0;
+}
+
+
+static unsigned mem_audio_n_queued(til_audio_context_t *context)
+{
+	mem_audio_t	*c = (mem_audio_t *)context;
+
+	return mem_refresh_n_queued(c);
+}
+
+
+static int mem_audio_setup(const til_settings_t *settings, til_setting_t **res_setting, const til_setting_desc_t **res_desc, til_setup_t **res_setup)
+{
+	if (res_setup) {
+		mem_audio_setup_t	*setup;
+
+		setup = til_setup_new(settings, sizeof(*setup), NULL, &mem_audio_ops);
+		if (!setup)
+			return -ENOMEM;
+
+		*res_setup = &setup->til_setup;
+	}
+
+	return 0;
+}