Save and load workflow from the flac files output by SaveAudio.

comfy_extras/nodes_audio.py

@@ -3,6 +3,10 @@ import torch
 import comfy.model_management
 import folder_paths
 import os
+import io
+import json
+import struct
+from comfy.cli_args import args
 
 class EmptyLatentAudio:
     def __init__(self):
@@ -53,6 +57,61 @@ class VAEDecodeAudio:
         audio = vae.decode(samples["samples"]).movedim(-1, 1)
         return ({"waveform": audio, "sample_rate": 44100}, )
 
+
+def create_vorbis_comment_block(comment_dict, last_block):
+    vendor_string = b'ComfyUI'
+    vendor_length = len(vendor_string)
+
+    comments = []
+    for key, value in comment_dict.items():
+        comment = f"{key}={value}".encode('utf-8')
+        comments.append(struct.pack('<I', len(comment)) + comment)
+
+    user_comment_list_length = len(comments)
+    user_comments = b''.join(comments)
+
+    comment_data = struct.pack('<I', vendor_length) + vendor_string + struct.pack('<I', user_comment_list_length) + user_comments
+    if last_block:
+        id = b'\x84'
+    else:
+        id = b'\x04'
+    comment_block = id + struct.pack('>I', len(comment_data))[1:] + comment_data
+
+    return comment_block
+
+def insert_or_replace_vorbis_comment(flac_io, comment_dict):
+    if len(comment_dict) == 0:
+        return flac_io
+
+    flac_io.seek(4)
+
+    blocks = []
+    last_block = False
+
+    while not last_block:
+        header = flac_io.read(4)
+        last_block = (header[0] & 0x80) != 0
+        block_type = header[0] & 0x7F
+        block_length = struct.unpack('>I', b'\x00' + header[1:])[0]
+        block_data = flac_io.read(block_length)
+
+        if block_type == 4 or block_type == 1:
+            pass
+        else:
+            header = bytes([(header[0] & (~0x80))]) + header[1:]
+            blocks.append(header + block_data)
+
+    blocks.append(create_vorbis_comment_block(comment_dict, last_block=True))
+
+    new_flac_io = io.BytesIO()
+    new_flac_io.write(b'fLaC')
+    for block in blocks:
+        new_flac_io.write(block)
+
+    new_flac_io.write(flac_io.read())
+    return new_flac_io
+
+
 class SaveAudio:
     def __init__(self):
         self.output_dir = folder_paths.get_output_directory()
@@ -78,11 +137,27 @@ class SaveAudio:
         filename_prefix += self.prefix_append
         full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir)
         results = list()
+
+        metadata = {}
+        if not args.disable_metadata:
+            if prompt is not None:
+                metadata["prompt"] = json.dumps(prompt)
+            if extra_pnginfo is not None:
+                for x in extra_pnginfo:
+                    metadata[x] = json.dumps(extra_pnginfo[x])
+
         for (batch_number, waveform) in enumerate(audio["waveform"]):
-            #TODO: metadata
             filename_with_batch_num = filename.replace("%batch_num%", str(batch_number))
             file = f"{filename_with_batch_num}_{counter:05}_.flac"
-            torchaudio.save(os.path.join(full_output_folder, file), waveform, audio["sample_rate"], format="FLAC")
+
+            buff = io.BytesIO()
+            torchaudio.save(buff, waveform, audio["sample_rate"], format="FLAC")
+
+            buff = insert_or_replace_vorbis_comment(buff, metadata)
+
+            with open(os.path.join(full_output_folder, file), 'wb') as f:
+                f.write(buff.getbuffer())
+
             results.append({
                 "filename": file,
                 "subfolder": subfolder,

web/scripts/app.js

@@ -3,7 +3,7 @@ import { ComfyWidgets, initWidgets } from "./widgets.js";
 import { ComfyUI, $el } from "./ui.js";
 import { api } from "./api.js";
 import { defaultGraph } from "./defaultGraph.js";
-import { getPngMetadata, getWebpMetadata, importA1111, getLatentMetadata } from "./pnginfo.js";
+import { getPngMetadata, getWebpMetadata, getFlacMetadata, importA1111, getLatentMetadata } from "./pnginfo.js";
 import { addDomClippingSetting } from "./domWidget.js";
 import { createImageHost, calculateImageGrid } from "./ui/imagePreview.js";
 import { ComfyAppMenu } from "./ui/menu/index.js";
@@ -2277,6 +2277,19 @@ export class ComfyApp {
 			const workflow = pngInfo?.workflow || pngInfo?.Workflow;
 			const prompt = pngInfo?.prompt || pngInfo?.Prompt;
 
+			if (workflow) {
+				this.loadGraphData(JSON.parse(workflow), true, true, fileName);
+			} else if (prompt) {
+				this.loadApiJson(JSON.parse(prompt), fileName);
+			} else {
+				this.showErrorOnFileLoad(file);
+			}
+		} else if (file.type === "audio/flac") {
+			const pngInfo = await getFlacMetadata(file);
+			// Support loading workflows from that webp custom node.
+			const workflow = pngInfo?.workflow;
+			const prompt = pngInfo?.prompt;
+
 			if (workflow) {
 				this.loadGraphData(JSON.parse(workflow), true, true, fileName);
 			} else if (prompt) {

web/scripts/pnginfo.js

@@ -163,6 +163,78 @@ export function getLatentMetadata(file) {
 	});
 }
 
+
+function getString(dataView, offset, length) {
+	let string = '';
+	for (let i = 0; i < length; i++) {
+		string += String.fromCharCode(dataView.getUint8(offset + i));
+	}
+	return string;
+}
+
+// Function to parse the Vorbis Comment block
+function parseVorbisComment(dataView) {
+	let offset = 0;
+	const vendorLength = dataView.getUint32(offset, true);
+	offset += 4;
+	const vendorString = getString(dataView, offset, vendorLength);
+	offset += vendorLength;
+
+	const userCommentListLength = dataView.getUint32(offset, true);
+	offset += 4;
+	const comments = {};
+	for (let i = 0; i < userCommentListLength; i++) {
+		const commentLength = dataView.getUint32(offset, true);
+		offset += 4;
+		const comment = getString(dataView, offset, commentLength);
+		offset += commentLength;
+
+		const [key, value] = comment.split('=');
+
+		comments[key] = value;
+	}
+
+	return comments;
+}
+
+// Function to read a FLAC file and parse Vorbis comments
+export function getFlacMetadata(file) {
+	return new Promise((r) => {
+		const reader = new FileReader();
+		reader.onload = function(event) {
+			const arrayBuffer = event.target.result;
+			const dataView = new DataView(arrayBuffer);
+
+			// Verify the FLAC signature
+			const signature = String.fromCharCode(...new Uint8Array(arrayBuffer, 0, 4));
+			if (signature !== 'fLaC') {
+				console.error('Not a valid FLAC file');
+				return;
+			}
+
+			// Parse metadata blocks
+			let offset = 4;
+			let vorbisComment = null;
+			while (offset < dataView.byteLength) {
+				const isLastBlock = dataView.getUint8(offset) & 0x80;
+				const blockType = dataView.getUint8(offset) & 0x7F;
+				const blockSize = dataView.getUint32(offset, false) & 0xFFFFFF;
+				offset += 4;
+
+				if (blockType === 4) { // Vorbis Comment block type
+					vorbisComment = parseVorbisComment(new DataView(arrayBuffer, offset, blockSize));
+				}
+
+				offset += blockSize;
+				if (isLastBlock) break;
+			}
+
+			r(vorbisComment);
+		};
+		reader.readAsArrayBuffer(file);
+	});
+}
+
 export async function importA1111(graph, parameters) {
 	const p = parameters.lastIndexOf("\nSteps:");
 	if (p > -1) {

web/scripts/ui.js

@@ -373,7 +373,7 @@ export class ComfyUI {
 		const fileInput = $el("input", {
 			id: "comfy-file-input",
 			type: "file",
-			accept: ".json,image/png,.latent,.safetensors,image/webp",
+			accept: ".json,image/png,.latent,.safetensors,image/webp,audio/flac",
 			style: {display: "none"},
 			parent: document.body,
 			onchange: () => {