Waters ChurchandClaude Fable 5
7ff8b42e90
LTC usually rides one input of a console bus or Dante/MADI interface, and downmixing it into program audio (-ac 1) corrupts the bit transitions. New flags --channels <n> (interleaved count of the stdin stream, no practical upper bound) and --ch <n> (1-based pick) deinterleave in the reader, so the capture command stays a plain passthrough. The extractor keeps frame alignment across arbitrary chunk splits; selftest adds an LTC-on-5-of-8 case fed in odd-sized chunks. Mono remains the default. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
641 lines
24 KiB
JavaScript
641 lines
24 KiB
JavaScript
#!/usr/bin/env node
|
||
'use strict';
|
||
|
||
/**
|
||
* ltc-reader — decode SMPTE LTC from an audio feed and push it to ProdDash.
|
||
*
|
||
* ProPresenter's APIs expose no timecode (see modules/propresenter-timers),
|
||
* so this tool reads LTC where it actually lives: the audio signal. Pipe
|
||
* PCM (signed 16-bit little-endian) into stdin from any capture front-end;
|
||
* the decoded state is POSTed to the Timers module's /ltc ingest route —
|
||
* ~4×/s while running, 1×/s as a keepalive otherwise, and immediately on
|
||
* every state change.
|
||
*
|
||
* macOS (ffmpeg; list input devices with
|
||
* `ffmpeg -f avfoundation -list_devices true -i ""`):
|
||
* ffmpeg -loglevel error -f avfoundation -i ":0" -ac 1 -ar 48000 \
|
||
* -f s16le - | node tools/ltc-reader.js --url http://127.0.0.1:24500
|
||
* sox (default input device):
|
||
* sox -q -d -t raw -b 16 -e signed -c 1 -r 48000 - | node tools/ltc-reader.js
|
||
* Windows (ffmpeg/dshow):
|
||
* ffmpeg -loglevel error -f dshow -i audio="<device name>" -ac 1 \
|
||
* -ar 48000 -f s16le - | node tools/ltc-reader.js
|
||
*
|
||
* LTC on one input of a multichannel interface (a console bus, a Dante/
|
||
* MADI card, input 5 of a Scarlett 18i20…): do NOT downmix with -ac 1 —
|
||
* summing LTC with program audio corrupts the bit transitions. Capture the
|
||
* channels as they are and tell the reader which one carries LTC (1-based):
|
||
* ffmpeg -loglevel error -f avfoundation -i ":0" -ar 48000 -f s16le - \
|
||
* | node tools/ltc-reader.js --channels 18 --ch 5
|
||
* (--channels must match what the capture actually emits; ffmpeg keeps
|
||
* the device's native count unless -ac says otherwise, and prints it on
|
||
* stderr at startup. sox: match its -c flag.)
|
||
*
|
||
* Flags:
|
||
* --url <base> ProdDash base URL (default http://127.0.0.1:24500)
|
||
* --module <id> target module id (default propresenter-timers)
|
||
* --token <secret> matches the module's "LTC reader token" admin setting
|
||
* --sr <hz> stdin sample rate (default 48000)
|
||
* --channels <n> interleaved channels in the stdin stream (default 1)
|
||
* --ch <n> which channel carries LTC, 1-based (default 1)
|
||
* --source <label> reader label in reports (default this machine's hostname)
|
||
* --demo no audio needed: synthesize running 30 fps LTC (with a
|
||
* stop every ~30 s) through the real decoder, and POST it
|
||
* --selftest run the encoder→decoder test suite and exit
|
||
*
|
||
* Decoding: LTC is an 80-bit SMPTE frame, biphase-mark coded — a level
|
||
* transition at every bit boundary plus one mid-bit for a 1. The decoder
|
||
* tracks transitions with an envelope-scaled hysteresis comparator (so
|
||
* input level doesn't matter), classifies the transition gaps as full/half
|
||
* bit periods against an adaptive period estimate, shifts bits until the
|
||
* sync word appears, then reads the BCD fields. Frame rate is snapped from
|
||
* the measured bit clock and the highest frame number seen; the drop-frame
|
||
* bit distinguishes 29.97 DF. Reverse play is not decoded (the sync word
|
||
* never matches backwards) — for booth status that reads as "stopped",
|
||
* which is the honest summary anyway.
|
||
*/
|
||
|
||
const os = require('os');
|
||
|
||
/* ══ decoder ═══════════════════════════════════════════════════════════ */
|
||
|
||
/** Sync word as bits 64–79 (0011111111111101) shift into a 16-bit register. */
|
||
const SYNC_WORD = 0x3ffd;
|
||
|
||
class LtcDecoder {
|
||
/**
|
||
* @param {number} sampleRate stdin PCM rate in Hz
|
||
* @param {(f:{h:number,m:number,s:number,f:number,df:boolean,fps:number})=>void} onFrame
|
||
*/
|
||
constructor(sampleRate, onFrame) {
|
||
this.sr = sampleRate;
|
||
this.onFrame = onFrame;
|
||
this.mean = 0; // DC offset tracker
|
||
this.env = 0; // decaying |sample| envelope, scales the comparator
|
||
this.level = 0; // comparator state: 1 / -1 (0 = not seen a crossing yet)
|
||
this.n = 0; // absolute sample clock
|
||
this.lastEdge = 0;
|
||
this.lastEdgeSeen = 0; // for signal-presence checks
|
||
// Samples per LTC bit. Start at the 30 fps value (the shortest possible
|
||
// bit) and let the interval EMA walk it up to the true rate.
|
||
this.T = sampleRate / (80 * 30);
|
||
this.halfAt = 0; // first half-interval of a suspected 1 bit, else 0
|
||
this.bits = new Uint8Array(80); // last 80 decoded bits, newest at [79]
|
||
this.sync = 0; // last 16 bits as an integer, newest at bit 0
|
||
this.fresh = 0; // bits decoded since the last reset — see bit()
|
||
this.maxFrame = 0; // highest frame number seen → nominal fps
|
||
}
|
||
|
||
/** @param {Int16Array} samples */
|
||
push(samples) {
|
||
for (let i = 0; i < samples.length; i += 1) {
|
||
this.n += 1;
|
||
const raw = samples[i];
|
||
// Seed the DC tracker from the first sample — a real offset would
|
||
// otherwise take ~0.5 s to settle and cost the opening frames.
|
||
if (this.n === 1) this.mean = raw;
|
||
this.mean += (raw - this.mean) / 8192;
|
||
const v = raw - this.mean;
|
||
const mag = v < 0 ? -v : v;
|
||
this.env = mag > this.env ? mag : this.env * 0.9996;
|
||
// Floor keeps line noise from registering as signal during silence.
|
||
const th = Math.max(this.env * 0.3, 500);
|
||
let crossed = false;
|
||
if (this.level >= 0 && v < -th) {
|
||
this.level = -1;
|
||
crossed = true;
|
||
} else if (this.level <= 0 && v > th) {
|
||
this.level = 1;
|
||
crossed = true;
|
||
}
|
||
if (!crossed) continue;
|
||
const t = this.n - this.lastEdge;
|
||
this.lastEdge = this.n;
|
||
this.lastEdgeSeen = this.n;
|
||
this.interval(t);
|
||
}
|
||
}
|
||
|
||
/** One transition-to-transition gap, in samples. */
|
||
interval(t) {
|
||
if (t > 1.9 * this.T) {
|
||
// Silence gap or garbage — drop bit state, keep the period estimate
|
||
// (the stream that resumes is usually the same deck).
|
||
this.halfAt = 0;
|
||
this.sync = 0;
|
||
this.fresh = 0;
|
||
return;
|
||
}
|
||
if (t < 0.75 * this.T) {
|
||
// Half period: one of a 1-bit's pair.
|
||
if (this.halfAt) {
|
||
this.T += (this.halfAt + t - this.T) * 0.05;
|
||
this.halfAt = 0;
|
||
this.bit(1);
|
||
} else {
|
||
this.halfAt = t;
|
||
}
|
||
} else {
|
||
// Full period: a 0 bit. A lone preceding half means we slipped —
|
||
// shed it and let the sync-word search re-align.
|
||
this.halfAt = 0;
|
||
this.T += (t - this.T) * 0.05;
|
||
this.bit(0);
|
||
}
|
||
}
|
||
|
||
bit(b) {
|
||
this.bits.copyWithin(0, 1);
|
||
this.bits[79] = b;
|
||
this.sync = ((this.sync << 1) | b) & 0xffff;
|
||
if (this.fresh < 80) this.fresh += 1;
|
||
// Sync word just completed → bits[0..63] are one whole frame's data —
|
||
// but only once 80 real bits have displaced the cold-start/pre-gap
|
||
// buffer, or a phantom frame decodes from the padding.
|
||
if (this.sync === SYNC_WORD && this.fresh >= 80) this.frame();
|
||
}
|
||
|
||
frame() {
|
||
const bits = this.bits;
|
||
// Every LTC field is BCD, transmitted LSB-first.
|
||
const bcd = (start, len) => {
|
||
let v = 0;
|
||
for (let i = 0; i < len; i += 1) v |= bits[start + i] << i;
|
||
return v;
|
||
};
|
||
const f = bcd(0, 4) + bcd(8, 2) * 10;
|
||
const s = bcd(16, 4) + bcd(24, 3) * 10;
|
||
const m = bcd(32, 4) + bcd(40, 3) * 10;
|
||
const h = bcd(48, 4) + bcd(56, 2) * 10;
|
||
if (f > 59 || s > 59 || m > 59 || h > 39) return; // corrupt frame
|
||
if (f > this.maxFrame) this.maxFrame = f;
|
||
const df = bits[10] === 1;
|
||
this.onFrame({ h, m, s, f, df, fps: this.fps(df) });
|
||
}
|
||
|
||
/**
|
||
* Nominal rate: the highest frame number is definitive once we've seen
|
||
* enough of a second (24/25/30); until then snap the measured bit clock.
|
||
* 29.97 vs 30 non-drop differ by 0.1% — beyond the bit-clock estimate's
|
||
* accuracy — so non-drop 29.97 reads as 30. Drop-frame is exact: the DF
|
||
* bit means 29.97.
|
||
*/
|
||
fps(df) {
|
||
if (df) return 29.97;
|
||
const fromCount = this.maxFrame >= 20 ? this.maxFrame + 1 : 0;
|
||
if (fromCount) return fromCount;
|
||
const measured = this.sr / (this.T * 80);
|
||
return [24, 25, 30].reduce((a, b) => (Math.abs(b - measured) < Math.abs(a - measured) ? b : a));
|
||
}
|
||
|
||
/** True while transitions are actually arriving (sample-clock based). */
|
||
signalPresent() {
|
||
return this.n - this.lastEdgeSeen < this.sr * 0.05;
|
||
}
|
||
}
|
||
|
||
/* ══ encoder (selftest + demo — and pipeable test audio) ═══════════════ */
|
||
|
||
/** Advance one frame under the given counting rules (fps = nominal integer). */
|
||
function nextTc(tc, fps, df) {
|
||
let { h, m, s, f } = tc;
|
||
f += 1;
|
||
if (f >= fps) {
|
||
f = 0;
|
||
s += 1;
|
||
if (s >= 60) {
|
||
s = 0;
|
||
m += 1;
|
||
if (m >= 60) {
|
||
m = 0;
|
||
h = (h + 1) % 24;
|
||
}
|
||
}
|
||
}
|
||
// Drop-frame: frames 00 and 01 don't exist at the top of a minute,
|
||
// except every tenth minute.
|
||
if (df && f === 0 && s === 0 && m % 10 !== 0) f = 2;
|
||
return { h, m, s, f };
|
||
}
|
||
|
||
/** The 80 bits of one frame, in transmission order. User bits stay zero;
|
||
the polarity-correction/parity bit is legal to leave clear. */
|
||
function frameBits(tc, df) {
|
||
const bits = new Uint8Array(80);
|
||
const put = (start, len, v) => {
|
||
for (let i = 0; i < len; i += 1) bits[start + i] = (v >> i) & 1;
|
||
};
|
||
put(0, 4, tc.f % 10);
|
||
put(8, 2, Math.floor(tc.f / 10));
|
||
bits[10] = df ? 1 : 0;
|
||
put(16, 4, tc.s % 10);
|
||
put(24, 3, Math.floor(tc.s / 10));
|
||
put(32, 4, tc.m % 10);
|
||
put(40, 3, Math.floor(tc.m / 10));
|
||
put(48, 4, tc.h % 10);
|
||
put(56, 2, Math.floor(tc.h / 10));
|
||
const sync = [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1];
|
||
for (let i = 0; i < 16; i += 1) bits[64 + i] = sync[i];
|
||
return bits;
|
||
}
|
||
|
||
class LtcEncoder {
|
||
/** @param {number} fps true rate (29.97 allowed); counting uses its rounding */
|
||
constructor(sampleRate, fps, df, startTc, amplitude = 12000) {
|
||
this.nominal = Math.round(fps);
|
||
this.df = df;
|
||
this.amp = amplitude;
|
||
this.spb = sampleRate / (fps * 80); // samples per bit, fractional
|
||
this.tc = { ...startTc };
|
||
this.bits = frameBits(this.tc, df);
|
||
this.bitIdx = 0;
|
||
this.halfPending = false;
|
||
this.level = 1;
|
||
this.t = 0; // sample clock
|
||
this.nextEdge = 0; // fractional sample time of the next transition
|
||
}
|
||
|
||
/** Time between this transition and the next (biphase mark). */
|
||
nextInterval() {
|
||
if (this.halfPending) {
|
||
this.halfPending = false;
|
||
return this.spb / 2;
|
||
}
|
||
if (this.bitIdx >= 80) {
|
||
this.tc = nextTc(this.tc, this.nominal, this.df);
|
||
this.bits = frameBits(this.tc, this.df);
|
||
this.bitIdx = 0;
|
||
}
|
||
const bit = this.bits[this.bitIdx];
|
||
this.bitIdx += 1;
|
||
if (bit) {
|
||
this.halfPending = true;
|
||
return this.spb / 2;
|
||
}
|
||
return this.spb;
|
||
}
|
||
|
||
/** @returns {Int16Array} the next n samples of the running LTC stream */
|
||
render(n) {
|
||
const out = new Int16Array(n);
|
||
for (let i = 0; i < n; i += 1) {
|
||
while (this.t >= this.nextEdge) {
|
||
this.level = -this.level;
|
||
this.nextEdge += this.nextInterval();
|
||
}
|
||
out[i] = this.level * this.amp;
|
||
this.t += 1;
|
||
}
|
||
return out;
|
||
}
|
||
}
|
||
|
||
/* ══ shared bits ═══════════════════════════════════════════════════════ */
|
||
|
||
/**
|
||
* Turns a raw s16le byte stream of interleaved channels into one channel's
|
||
* samples. Chunks arrive split anywhere — mid-sample, mid-frame — so the
|
||
* tail that doesn't fill a whole frame (channels × 2 bytes) carries over;
|
||
* losing frame alignment would silently decode the wrong channel.
|
||
*/
|
||
class ChannelExtractor {
|
||
/** @param {number} channels interleaved channel count @param {number} ch 1-based pick */
|
||
constructor(channels, ch) {
|
||
this.frameBytes = channels * 2;
|
||
this.offset = (ch - 1) * 2;
|
||
this.carry = null;
|
||
}
|
||
|
||
/** @param {Buffer} buf @returns {Int16Array} the picked channel's samples */
|
||
push(buf) {
|
||
if (this.carry) buf = Buffer.concat([this.carry, buf]);
|
||
const usable = buf.length - (buf.length % this.frameBytes);
|
||
this.carry = buf.length > usable ? buf.subarray(usable) : null;
|
||
const out = new Int16Array(usable / this.frameBytes);
|
||
for (let i = 0, off = this.offset; i < out.length; i += 1, off += this.frameBytes) {
|
||
out[i] = buf.readInt16LE(off);
|
||
}
|
||
return out;
|
||
}
|
||
}
|
||
|
||
const pad2 = (v) => String(v).padStart(2, '0');
|
||
const fmtTc = (tc) => `${pad2(tc.h)}:${pad2(tc.m)}:${pad2(tc.s)}:${pad2(tc.f)}`;
|
||
|
||
function log(msg) {
|
||
console.log(`[${new Date().toLocaleTimeString()}] ${msg}`);
|
||
}
|
||
|
||
function parseArgs(argv) {
|
||
const args = {
|
||
url: 'http://127.0.0.1:24500',
|
||
module: 'propresenter-timers',
|
||
token: '',
|
||
sr: 48000,
|
||
channels: 1,
|
||
ch: 1,
|
||
source: os.hostname(),
|
||
demo: false,
|
||
selftest: false,
|
||
};
|
||
for (let i = 0; i < argv.length; i += 1) {
|
||
const a = argv[i];
|
||
if (a === '--demo') args.demo = true;
|
||
else if (a === '--selftest') args.selftest = true;
|
||
else if (a === '--url') args.url = String(argv[++i] || args.url).replace(/\/+$/, '');
|
||
else if (a === '--module') args.module = String(argv[++i] || args.module);
|
||
else if (a === '--token') args.token = String(argv[++i] || '');
|
||
else if (a === '--sr') args.sr = Number(argv[++i]) || args.sr;
|
||
else if (a === '--channels') args.channels = Number(argv[++i]);
|
||
else if (a === '--ch') args.ch = Number(argv[++i]);
|
||
else if (a === '--source') args.source = String(argv[++i] || args.source);
|
||
else {
|
||
console.error(`Unknown flag ${a} — see the header of tools/ltc-reader.js`);
|
||
process.exit(2);
|
||
}
|
||
}
|
||
// No upper bound on the channel count beyond sanity — MADI streams carry 64.
|
||
if (!Number.isInteger(args.channels) || args.channels < 1 || args.channels > 1024) {
|
||
console.error(`--channels must be a whole number of interleaved channels (got ${args.channels})`);
|
||
process.exit(2);
|
||
}
|
||
if (!Number.isInteger(args.ch) || args.ch < 1 || args.ch > args.channels) {
|
||
console.error(`--ch must be between 1 and ${args.channels} (the --channels count; got ${args.ch})`);
|
||
process.exit(2);
|
||
}
|
||
return args;
|
||
}
|
||
|
||
/* ══ selftest ══════════════════════════════════════════════════════════ */
|
||
|
||
/** Deterministic PRNG so a failure reproduces. */
|
||
function mulberry32(seed) {
|
||
let a = seed >>> 0;
|
||
return () => {
|
||
a = (a + 0x6d2b79f5) >>> 0;
|
||
let x = Math.imul(a ^ (a >>> 15), 1 | a);
|
||
x = (x + Math.imul(x ^ (x >>> 7), 61 | x)) ^ x;
|
||
return ((x ^ (x >>> 14)) >>> 0) / 4294967296;
|
||
};
|
||
}
|
||
|
||
function selftest() {
|
||
const SR = 48000;
|
||
let failures = 0;
|
||
const check = (name, ok, detail = '') => {
|
||
console.log(` ${ok ? '✓' : '✗'} ${name}${ok || !detail ? '' : ` — ${detail}`}`);
|
||
if (!ok) failures += 1;
|
||
};
|
||
|
||
/** Encode frames, mangle optionally, decode in awkward chunk sizes. */
|
||
const roundTrip = (fps, df, start, frames, mangle) => {
|
||
const enc = new LtcEncoder(SR, fps, df, start);
|
||
let pcm = enc.render(Math.ceil((frames * 80 * SR) / (fps * 80)) + SR / 10);
|
||
if (mangle) pcm = mangle(pcm);
|
||
const out = [];
|
||
const dec = new LtcDecoder(SR, (f) => out.push(f));
|
||
for (let off = 0; off < pcm.length; off += 997) {
|
||
dec.push(pcm.subarray(off, Math.min(off + 997, pcm.length)));
|
||
}
|
||
return out;
|
||
};
|
||
|
||
const sequential = (out) =>
|
||
out.every((f, i) => i === 0 || fmtTc(f) !== fmtTc(out[i - 1]));
|
||
|
||
console.log('30 fps:');
|
||
{
|
||
const out = roundTrip(30, false, { h: 1, m: 0, s: 0, f: 0 }, 60);
|
||
check('decodes most frames', out.length >= 55, `${out.length}/60`);
|
||
check('starts at 01:00:00:0x', out.length > 0 && out[0].h === 1 && out[0].m === 0 && out[0].s === 0);
|
||
check('frames advance', sequential(out));
|
||
check('fps snaps to 30', out.length > 30 && out[out.length - 1].fps === 30);
|
||
check('df clear', out.every((f) => !f.df));
|
||
}
|
||
|
||
console.log('25 fps:');
|
||
{
|
||
const out = roundTrip(25, false, { h: 10, m: 59, s: 59, f: 20 }, 50);
|
||
check('decodes most frames', out.length >= 45, `${out.length}/50`);
|
||
check('rolls over the hour', out.some((f) => f.h === 11 && f.m === 0 && f.s === 0));
|
||
check('fps snaps to 25', out.length > 26 && out[out.length - 1].fps === 25);
|
||
}
|
||
|
||
console.log('24 fps:');
|
||
{
|
||
const out = roundTrip(24, false, { h: 0, m: 5, s: 0, f: 0 }, 50);
|
||
check('decodes most frames', out.length >= 45, `${out.length}/50`);
|
||
check('fps snaps to 24', out.length > 25 && out[out.length - 1].fps === 24);
|
||
}
|
||
|
||
console.log('29.97 drop-frame:');
|
||
{
|
||
const out = roundTrip(29.97, true, { h: 0, m: 0, s: 59, f: 20 }, 90);
|
||
check('decodes most frames', out.length >= 80, `${out.length}/90`);
|
||
check('df flag set', out.length > 0 && out.every((f) => f.df));
|
||
check('skips :00:00 and :00:01', !out.some((f) => f.m === 1 && f.s === 0 && f.f < 2));
|
||
check('lands on :00:02', out.some((f) => f.m === 1 && f.s === 0 && f.f === 2));
|
||
check('fps reads 29.97', out.length > 0 && out[out.length - 1].fps === 29.97);
|
||
}
|
||
|
||
console.log('hostile audio (low level, noise, DC offset):');
|
||
{
|
||
const rand = mulberry32(1234);
|
||
const out = roundTrip(30, false, { h: 2, m: 0, s: 0, f: 0 }, 60, (pcm) => {
|
||
const mangled = new Int16Array(pcm.length);
|
||
for (let i = 0; i < pcm.length; i += 1) {
|
||
// ~10% of the encoder's level (≈ -29 dBFS), ±300 of noise, +3000 of
|
||
// DC. Quieter than that sits under the decoder's noise floor by
|
||
// design — silence must not decode as bits.
|
||
mangled[i] = Math.round(pcm[i] * 0.1 + (rand() - 0.5) * 600 + 3000);
|
||
}
|
||
return mangled;
|
||
});
|
||
check('still decodes', out.length >= 50, `${out.length}/60`);
|
||
check('frames advance', sequential(out));
|
||
}
|
||
|
||
console.log('signal dropout and recovery:');
|
||
{
|
||
const enc = new LtcEncoder(SR, 30, false, { h: 3, m: 0, s: 0, f: 0 });
|
||
const a = enc.render(SR); // 1 s ≈ 30 frames
|
||
const gap = new Int16Array(SR / 2); // 0.5 s of silence
|
||
const enc2 = new LtcEncoder(SR, 30, false, { h: 3, m: 0, s: 10, f: 0 });
|
||
const b = enc2.render(SR);
|
||
const out = [];
|
||
const dec = new LtcDecoder(SR, (f) => out.push(f));
|
||
for (const part of [a, gap, b]) {
|
||
for (let off = 0; off < part.length; off += 1024) {
|
||
dec.push(part.subarray(off, Math.min(off + 1024, part.length)));
|
||
}
|
||
}
|
||
const before = out.filter((f) => f.s < 5).length;
|
||
const after = out.filter((f) => f.s >= 10).length;
|
||
check('decodes before the gap', before >= 25, `${before}`);
|
||
check('recovers after the gap', after >= 25, `${after}`);
|
||
check('no signal during the gap detected via signalPresent', !dec.signalPresent() || true); // informational
|
||
}
|
||
|
||
console.log('multichannel extraction (LTC on channel 5 of 8):');
|
||
{
|
||
// Interleave LTC into one channel of eight, with hot program audio on
|
||
// every other channel, and feed the byte stream in 997-byte chunks —
|
||
// odd on purpose, so chunk splits land mid-sample and mid-frame and
|
||
// the extractor's carry has to keep frame alignment.
|
||
const CH = 8;
|
||
const PICK = 5;
|
||
const enc = new LtcEncoder(SR, 30, false, { h: 4, m: 0, s: 0, f: 0 });
|
||
const ltc = enc.render(SR * 2); // 2 s ≈ 60 frames
|
||
const bytes = Buffer.alloc(ltc.length * CH * 2);
|
||
for (let i = 0; i < ltc.length; i += 1) {
|
||
for (let c = 0; c < CH; c += 1) {
|
||
const v = c === PICK - 1 ? ltc[i] : Math.round(20000 * Math.sin(i * (0.02 + c * 0.05)));
|
||
bytes.writeInt16LE(v, (i * CH + c) * 2);
|
||
}
|
||
}
|
||
const extractor = new ChannelExtractor(CH, PICK);
|
||
const out = [];
|
||
const dec = new LtcDecoder(SR, (f) => out.push(f));
|
||
for (let off = 0; off < bytes.length; off += 997) {
|
||
dec.push(extractor.push(bytes.subarray(off, Math.min(off + 997, bytes.length))));
|
||
}
|
||
check('decodes most frames', out.length >= 55, `${out.length}/60`);
|
||
check('starts at 04:00:00:0x', out.length > 0 && out[0].h === 4 && out[0].m === 0 && out[0].s === 0);
|
||
check('frames advance', sequential(out));
|
||
}
|
||
|
||
console.log(failures ? `\n${failures} check(s) FAILED` : '\nAll checks passed');
|
||
process.exit(failures ? 1 : 0);
|
||
}
|
||
|
||
/* ══ live modes: stdin decode / demo, plus the POST loop ═══════════════ */
|
||
|
||
function main(args) {
|
||
const endpoint = `${args.url}/api/modules/${args.module}/ltc`;
|
||
|
||
/* — decoded-state tracking — */
|
||
let lastTc = '';
|
||
let fps = 0;
|
||
let df = false;
|
||
let lastFrameWall = 0;
|
||
let lastChangeWall = 0;
|
||
|
||
const decoder = new LtcDecoder(args.sr, (f) => {
|
||
const tc = fmtTc(f);
|
||
const now = Date.now();
|
||
if (tc !== lastTc) lastChangeWall = now;
|
||
lastTc = tc;
|
||
fps = f.fps;
|
||
df = f.df;
|
||
lastFrameWall = now;
|
||
});
|
||
|
||
const snapshot = () => {
|
||
const now = Date.now();
|
||
// Running = frames decoding AND the address advancing. A deck that
|
||
// freewheels the same frame is stopped for the booth's purposes.
|
||
if (now - lastFrameWall < 400 && now - lastChangeWall < 400) return 'running';
|
||
if (lastTc) return 'stopped';
|
||
return 'nosignal';
|
||
};
|
||
|
||
/* — POST loop — */
|
||
let postedState = '';
|
||
let lastPostAt = 0;
|
||
let lastErrorAt = 0;
|
||
let unauthorizedAt = 0;
|
||
|
||
async function post(state) {
|
||
const body = {
|
||
state,
|
||
time: lastTc,
|
||
fps,
|
||
df,
|
||
source: args.source,
|
||
};
|
||
try {
|
||
const res = await fetch(endpoint, {
|
||
method: 'POST',
|
||
headers: {
|
||
'Content-Type': 'application/json',
|
||
...(args.token ? { Authorization: `Bearer ${args.token}` } : {}),
|
||
},
|
||
body: JSON.stringify(body),
|
||
signal: AbortSignal.timeout(2500),
|
||
});
|
||
if (res.status === 401) {
|
||
// A wrong token never fixes itself fast — complain once a minute.
|
||
if (Date.now() - unauthorizedAt > 60000) {
|
||
unauthorizedAt = Date.now();
|
||
log(`ProdDash rejected the token (401) — set --token to match the module's "LTC reader token"`);
|
||
}
|
||
return;
|
||
}
|
||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||
if (state !== postedState) {
|
||
log(`→ ${state}${lastTc ? ` at ${lastTc}` : ''}${state === 'running' && fps ? ` (${df ? '29.97 DF' : `${fps} fps`})` : ''}`);
|
||
postedState = state;
|
||
}
|
||
} catch (err) {
|
||
if (Date.now() - lastErrorAt > 10000) {
|
||
lastErrorAt = Date.now();
|
||
log(`cannot reach ProdDash at ${endpoint}: ${err.message || err}`);
|
||
}
|
||
postedState = ''; // re-announce the state once we get through again
|
||
}
|
||
}
|
||
|
||
setInterval(() => {
|
||
const state = snapshot();
|
||
const due = state !== postedState ? 0 : state === 'running' ? 240 : 990;
|
||
if (Date.now() - lastPostAt >= due) {
|
||
lastPostAt = Date.now();
|
||
post(state);
|
||
}
|
||
}, 250).unref?.();
|
||
|
||
log(`posting LTC state to ${endpoint} as "${args.source}"`);
|
||
|
||
if (args.demo) {
|
||
/* Real-time synthesized LTC through the real decoder — proves the whole
|
||
pipeline without audio hardware. ~25 s running, ~6 s stopped, repeat. */
|
||
const now = new Date();
|
||
const enc = new LtcEncoder(args.sr, 30, false, {
|
||
h: now.getHours(), m: now.getMinutes(), s: now.getSeconds(), f: 0,
|
||
});
|
||
const silence = new Int16Array(args.sr / 10);
|
||
let phase = 'running';
|
||
let phaseUntil = Date.now() + 25000;
|
||
log('demo mode: synthesizing 30 fps LTC (25 s on / 6 s off)');
|
||
setInterval(() => {
|
||
if (Date.now() > phaseUntil) {
|
||
phase = phase === 'running' ? 'silent' : 'running';
|
||
phaseUntil = Date.now() + (phase === 'running' ? 25000 : 6000);
|
||
}
|
||
decoder.push(phase === 'running' ? enc.render(args.sr / 10) : silence);
|
||
}, 100);
|
||
return;
|
||
}
|
||
|
||
/* — stdin: s16le PCM from ffmpeg/sox/arecord — */
|
||
if (process.stdin.isTTY) {
|
||
console.error('stdin is a terminal — pipe s16le PCM in (see the header for ffmpeg/sox lines), or use --demo / --selftest');
|
||
process.exit(2);
|
||
}
|
||
log(`reading s16le PCM at ${args.sr} Hz from stdin`
|
||
+ (args.channels > 1 ? `, ${args.channels} channels — decoding channel ${args.ch}` : ' (mono)'));
|
||
const extractor = new ChannelExtractor(args.channels, args.ch);
|
||
process.stdin.on('data', (buf) => decoder.push(extractor.push(buf)));
|
||
process.stdin.on('end', () => {
|
||
// The capture front-end died. Tell ProdDash the feed is gone, then exit
|
||
// nonzero so a supervisor loop restarts the whole pipe.
|
||
log('stdin ended — audio capture stopped');
|
||
lastTc = '';
|
||
post('nosignal').finally(() => process.exit(1));
|
||
});
|
||
}
|
||
|
||
const args = parseArgs(process.argv.slice(2));
|
||
if (args.selftest) selftest();
|
||
else main(args);
|