aes67-ESP32-P4/main/player.c

#include "player.h"

#include <math.h>
#include <string.h>

#include "aes67_cfg.h"
#include "aes67_ota.h"
#include "aes67_tx.h"
#include "aes67_web.h"
#include "audio_out.h"
#include "audio_ring.h"
#include "decoder.h"
#include "esp_log.h"
#include "esp_timer.h"
#include "freertos/FreeRTOS.h"
#include "freertos/task.h"
#include "hls.h"
#include "spotify.h"

#define RATE            48000
#define CHANNELS        2
#define RING_FRAMES     (4 * RATE)      // 4 s in PSRAM
#define PREFILL_FRAMES  (RATE)          // 1 s before (re)starting output
#define FADE_FRAMES     (RATE * 30 / 1000)  // 30 ms ramp at a source switch
#define SPOTIFY_WAIT_MS 100             // longest the Spotify data callback may wait (cspot holds a lock)
#define CONTROL_MS      250

static const char *TAG = "player";

typedef enum { MODE_TONE, MODE_OFF, MODE_HLS, MODE_SPOTIFY, MODE_AUTO } mode_t_;
static const char *const SRC_NAME[] = { "tone", "off", "hls", "spotify" };

static volatile mode_t_ s_mode = MODE_OFF;
static volatile player_src_t s_src = PLAYER_SRC_OFF;     // what is playing now
static volatile player_src_t s_target = PLAYER_SRC_OFF;  // switch to this (TX task, with fades)
static volatile bool s_playing;         // ring output running (after prefill)
static volatile bool s_flush;           // consumer drops buffered audio on the next read
static volatile bool s_paused;          // Spotify paused: silence, buffer kept, no underrun
static volatile bool s_sp_playing;      // Spotify has started playback and is not paused
static audio_conv_t *s_spotify_conv;

// Failover (mode auto)
static volatile int s_failover_delay_s = 5;
static volatile bool s_failover_on_pause;

// TX task only
static int s_fade_out;                  // frames left in the fade-out before a switch
static int s_fade_in;                   // frames left in the fade-in after a switch

// Player volume, applied after the ring so a change is heard at once (not a ring length later).
static volatile float s_target_gain = 1.0f;
static float s_gain = 1.0f;             // TX task only; ramps to s_target_gain
static volatile int s_volume_pct = 100;

// 0..100 % -> amplitude: cubic curve (about -18 dB at 50 %, 0 % = mute).
static void set_volume(int pct)
{
    pct = pct < 0 ? 0 : pct > 100 ? 100 : pct;
    float r = pct / 100.0f;
    s_volume_pct = pct;
    s_target_gain = r * r * r;
}

// Apply the volume, ramping linearly from the previous gain over this buffer (no zipper noise).
static void apply_gain(int32_t *buf, size_t frames)
{
    float target = s_target_gain;
    if (s_gain == target && target == 1.0f) {
        return;
    }
    float step = (target - s_gain) / frames;
    for (size_t i = 0; i < frames; i++) {
        s_gain += step;
        buf[i * 2] = (int32_t)(buf[i * 2] * s_gain);
        buf[i * 2 + 1] = (int32_t)(buf[i * 2 + 1] * s_gain);
    }
    s_gain = target;
}

// Linear ramp over a fade: position counts down from FADE_FRAMES; fade-in rises, fade-out falls.
static void apply_fade(int32_t *buf, size_t frames, int *left, bool in)
{
    for (size_t i = 0; i < frames && *left > 0; i++, (*left)--) {
        float g = (float)(*left) / FADE_FRAMES;         // 1 -> 0 over the fade
        if (in) {
            g = 1.0f - g;
        }
        buf[i * 2] = (int32_t)(buf[i * 2] * g);
        buf[i * 2 + 1] = (int32_t)(buf[i * 2 + 1] * g);
    }
}

static mode_t_ mode_of(const char *m)
{
    return !strcmp(m, "tone") ? MODE_TONE : !strcmp(m, "hls") ? MODE_HLS : !strcmp(m, "spotify") ? MODE_SPOTIFY :
           !strcmp(m, "auto") ? MODE_AUTO : MODE_OFF;
}

/* ----- output (AES67 TX pull callback, TX task, must not block) ----- */

// One buffer of the current source: ring audio, or silence while off / paused / buffering.
// Returns the frames produced (fewer = underrun).
static size_t read_current(int32_t *buf, size_t frames)
{
    if (s_flush) {
        audio_ring_flush();
        s_flush = false;
        s_playing = false;
    }
    if (s_src == PLAYER_SRC_OFF || (s_src == PLAYER_SRC_SPOTIFY && s_paused)) {
        memset(buf, 0, frames * CHANNELS * sizeof(int32_t));
        return frames;
    }
    if (!s_playing) {
        if (audio_ring_level() < PREFILL_FRAMES) {
            memset(buf, 0, frames * CHANNELS * sizeof(int32_t));
            return frames;
        }
        s_playing = true;
        s_fade_in = FADE_FRAMES;    // (re)start without a click
    }
    size_t got = audio_ring_read(buf, frames);
    if (got < frames) {
        s_playing = false;          // ran dry: underrun (counted by TX), prefill again
    }
    return got;
}

static size_t player_read(int32_t *buf, size_t frames)
{
    // A switch is due: fade the current source out, then switch (ring flushed, new source prefills).
    if (s_target != s_src && s_fade_out == 0) {
        s_fade_out = FADE_FRAMES;
    }
    size_t got = read_current(buf, frames);
    if (s_fade_in > 0) {
        apply_fade(buf, got, &s_fade_in, true);
    }
    if (s_fade_out > 0) {
        apply_fade(buf, got, &s_fade_out, false);
        if (s_fade_out == 0 || got < frames) {
            s_fade_out = 0;
            audio_ring_flush();     // before s_src changes: writers only write for the active source
            s_playing = false;
            s_flush = false;
            s_src = s_target;
        }
    }
    apply_gain(buf, got);
    return got;
}

// Set the source to play. Tone comes from TX itself (phase-locked to PTP), so switches from/to it
// are immediate; everything else fades through player_read.
static void select_source(player_src_t src)
{
    if (src == s_target) {
        return;
    }
    ESP_LOGI(TAG, "source: %s -> %s", SRC_NAME[s_src], SRC_NAME[src]);
    if (src == PLAYER_SRC_TONE || s_src == PLAYER_SRC_TONE) {
        s_target = s_src = src;
        s_flush = true;
        aes67_tx_set_source(src == PLAYER_SRC_TONE ? NULL : player_read);
        return;
    }
    s_target = src;     // player_read fades out and switches
}

void player_apply(const cJSON *source)
{
    if (s_spotify_conv) {       // after player_init: Spotify follows mode / name / credentials live
        spotify_apply();
    }
    s_failover_delay_s = (int)cJSON_GetObjectItemCaseSensitive(source, "failover_delay_s")->valuedouble;
    s_failover_on_pause = cJSON_IsTrue(cJSON_GetObjectItemCaseSensitive(source, "failover_on_pause"));
    mode_t_ mode = mode_of(cJSON_GetObjectItemCaseSensitive(source, "mode")->valuestring);
    if (mode == s_mode) {
        return;
    }
    s_mode = mode;
    // HLS fetches only in hls mode, or in auto while it is the active source (control task).
    hls_suspend(HLS_SUSPEND_FAILOVER, mode == MODE_AUTO);
    switch (mode) {
    case MODE_TONE:    select_source(PLAYER_SRC_TONE); break;
    case MODE_HLS:     select_source(PLAYER_SRC_HLS); break;
    case MODE_SPOTIFY: select_source(PLAYER_SRC_SPOTIFY); break;
    case MODE_AUTO:    select_source(PLAYER_SRC_SPOTIFY); break;  // Spotify inactive: HLS after the delay
    default:           select_source(PLAYER_SRC_OFF); break;
    }
}

/* ----- source side ----- */

// Write converted 48 kHz frames, waiting for space at playback speed. Frames of a source that is
// neither playing nor about to play are dropped (returns n).
size_t player_write(player_src_t src, const int32_t *frames, size_t n)
{
    size_t done = 0;
    while (done < n) {
        if (s_src != src) {
            if (s_target != src) {
                return n;           // not this source's turn: drop
            }
            vTaskDelay(pdMS_TO_TICKS(5));   // switch in progress (fade-out, ~30 ms)
            continue;
        }
        size_t w = audio_ring_write(frames + done * CHANNELS, n - done);
        if (!w) {
            vTaskDelay(pdMS_TO_TICKS(20));
        }
        done += w;
    }
    return done;
}

/* ----- Spotify ----- */

// cspot calls this holding a lock that stopping the session also needs: never wait long. Take only
// what fits in the ring; return 0 after SPOTIFY_WAIT_MS and cspot retries (it can stop in between).
// When it isn't Spotify's turn, hold cspot back the same way: it doesn't stop decoding on pause, so
// dropping frames would let it race through the queue at full speed (CPU and network), starving HLS.
static size_t spotify_pcm(const int16_t *pcm, size_t frames)
{
    int64_t until = esp_timer_get_time() + SPOTIFY_WAIT_MS * 1000LL;
    while (true) {
        if (s_src == PLAYER_SRC_SPOTIFY) {
            // 44.1 kHz in -> about 1.09x as many 48 kHz frames out; keep a margin
            size_t fit = audio_ring_space() * 44100 / 48000;
            fit = fit > 64 ? fit - 64 : 0;
            size_t n = frames < fit ? frames : fit;
            if (n) {
                audio_conv_write(s_spotify_conv, pcm, n, 2, 44100);
                return n;
            }
        }
        if (esp_timer_get_time() >= until) {
            return 0;
        }
        vTaskDelay(pdMS_TO_TICKS(10));
    }
}

static void spotify_event(spotify_event_t ev, int value)
{
    switch (ev) {
    case SPOTIFY_EV_PLAY:
        s_paused = false;
        s_sp_playing = true;
        break;
    case SPOTIFY_EV_PAUSE:
        s_paused = true;
        s_sp_playing = false;
        break;
    case SPOTIFY_EV_FLUSH:
        if (s_src == PLAYER_SRC_SPOTIFY) {
            s_flush = true;         // skip/seek: don't play out the old buffer
        }
        break;
    case SPOTIFY_EV_VOLUME: {
        // Spotify sends 0..65535; the app's slider stays the master of the player volume.
        int pct = (value * 100 + 32767) / 65535;
        set_volume(pct);
        float g = s_target_gain;
        ESP_LOGI(TAG, "volume %d%% (%.1f dB)", pct, g > 0 ? 20.0f * log10f(g) : -INFINITY);
        break;
    }
    }
}

/* ----- failover (mode auto) ----- */

// Spotify stays advertised; HLS plays whenever Spotify is inactive for failover_delay_s, and Spotify
// takes over as soon as it plays. HLS is stopped while Spotify plays (no bandwidth/CPU).
static void control_task(void *arg)
{
    int64_t inactive_since = esp_timer_get_time();
    while (true) {
        vTaskDelay(pdMS_TO_TICKS(CONTROL_MS));
        if (s_mode != MODE_AUTO) {
            inactive_since = esp_timer_get_time();
            continue;
        }
        bool session = spotify_session_active();
        if (!session) {
            s_sp_playing = false;   // session ended (app left, error, disabled)
        }
        bool playing = session && s_sp_playing;
        // Inactive: no session; with failover_on_pause also a connected but paused/stopped one.
        bool inactive = !session || (s_failover_on_pause && !playing);
        int64_t now = esp_timer_get_time();
        if (!inactive) {
            inactive_since = now;
        }
        if (playing && s_target != PLAYER_SRC_SPOTIFY) {
            ESP_LOGI(TAG, "auto: Spotify is playing, switching to it");
            hls_suspend(HLS_SUSPEND_FAILOVER, true);
            select_source(PLAYER_SRC_SPOTIFY);
        } else if (inactive && s_target == PLAYER_SRC_SPOTIFY &&
                   now - inactive_since >= (int64_t)s_failover_delay_s * 1000000) {
            ESP_LOGI(TAG, "auto: Spotify inactive for %d s, failing over to HLS", s_failover_delay_s);
            hls_suspend(HLS_SUSPEND_FAILOVER, false);
            select_source(PLAYER_SRC_HLS);
        }
    }
}

// Firmware upload: flash writes stall tasks for > 20 ms at a time; with a Spotify session running,
// uploads crawled and sometimes ended in a reset (pausing was not enough). End the session (its
// tasks and connections are gone before the first write) and stop HLS fetching meanwhile. The
// device reboots after a good upload anyway; after a failed one the sources come back (Spotify
// waits for the app to reconnect).
static void on_ota(bool starting)
{
    if (starting) {
        spotify_suspend(true);
        hls_suspend(HLS_SUSPEND_OTA, true);
        ESP_LOGI(TAG, "firmware update: Spotify session ended, HLS suspended");
    } else {
        hls_suspend(HLS_SUSPEND_OTA, false);
        spotify_suspend(false);
        ESP_LOGI(TAG, "firmware update failed: sources resumed");
    }
}

static void player_status(cJSON *st)
{
    player_src_t src = s_src;
    const char *state = src == PLAYER_SRC_TONE ? "playing" : src == PLAYER_SRC_OFF ? "idle" :
                        src == PLAYER_SRC_SPOTIFY && !spotify_session_active() ? "idle" :
                        src == PLAYER_SRC_SPOTIFY && s_paused ? "paused" : s_playing ? "playing" : "buffering";
    cJSON_AddStringToObject(st, "active_source", SRC_NAME[src]);
    cJSON_AddStringToObject(st, "source_state", state);
    cJSON_AddStringToObject(st, "spotify_state", spotify_state());
    cJSON_AddNumberToObject(st, "buffer_ms", (double)(audio_ring_level() * 1000 / RATE));
}

esp_err_t player_init(void)
{
    esp_err_t err = audio_ring_init(RING_FRAMES, CHANNELS);
    if (err != ESP_OK) {
        ESP_LOGE(TAG, "ring buffer: %s", esp_err_to_name(err));
        return err;
    }
    aes67_tx_set_source(player_read);
    cJSON *src = cfg_get("source");
    s_mode = (mode_t_)-1;
    player_apply(src);
    cJSON_Delete(src);
    err = decoder_init();
    if (err != ESP_OK) {
        return err;
    }
    s_spotify_conv = audio_conv_create(PLAYER_SRC_SPOTIFY, "spotify");
    if (!s_spotify_conv) {
        return ESP_ERR_NO_MEM;
    }
    spotify_init(spotify_pcm, spotify_event, audio_ring_written, audio_ring_read_pos);
    aes67_ota_on_update(on_ota);
    err = hls_start(decoder_feed);
    if (err != ESP_OK) {
        return err;
    }
    if (xTaskCreate(control_task, "player_ctl", 3072, NULL, 4, NULL) != pdPASS) {
        return ESP_ERR_NO_MEM;
    }
    return status_register(player_status);
}