honeymoose/music-metadata
1
0
Fork 0
mirror of synced 2026-05-22 14:43:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

id3v2: harden frame parsing and fix edge cases

Browse Source 
- Correctly handle missing null terminators in text fields
- Avoid reading text encoding for frames that do not define it
- Decode URL link frames according to ID3v2 specification
- Improve robustness of POPM and GEOB frame parsing
This commit is contained in:
Borewit
2026-01-06 12:57:31 +01:00
committed by Borewit
parent fd8b32b309
commit 64d2148d11
3 changed files with 37 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/common/Util.ts
+1 -1
View File
@@ -26,7 +26,7 @@ export function findZero(uint8Array: Uint8Array, encoding?: StringEncoding): num
const len = uint8Array.length;
if (encoding === 'utf-16le') {
// Look for 0x00 0x00 on even boundary
// Look for 0x00 0x00 on 2-byte boundary
for (let i = 0; i + 1 < len; i += 2) {
if (uint8Array[i] === 0 && uint8Array[i + 1] === 0) return i;
}
lib/id3v2/FrameParser.ts
+35 -42
View File
@@ -51,6 +51,7 @@ export interface IGeneralEncapsulatedObject {
}
const defaultEnc = 'latin1'; // latin1 == iso-8859-1;
const urlEnc: ITextEncoding = {encoding: defaultEnc, bom: false};
export function parseGenre(origVal: string): string[] {
// match everything inside parentheses
@@ -120,11 +121,14 @@ export class FrameParser {
this.warningCollector.addWarning(`id3v2.${this.major} header has empty tag type=${type}`);
return;
}
const {encoding, bom} = TextEncodingToken.get(uint8Array, 0);
const length = uint8Array.length;
let offset = 0;
let output: unknown = []; // ToDo
const nullTerminatorLength = FrameParser.getNullTerminatorLength(encoding);
let fzero: number;
debug(`Parsing tag type=${type}`);
debug(`Parsing tag type=${type}, encoding=${encoding}, bom=${bom}`);
switch (type !== 'TXXX' && type[0] === 'T' ? 'T*' : type) {
case 'T*': // 4.2.1. Text information frames - details
case 'GRP1': // iTunes-specific ID3v2 grouping field
@@ -133,8 +137,6 @@ export class FrameParser {
case 'MVNM':
case 'PCS':
case 'PCST': {
const {encoding, bom} = TextEncodingToken.get(uint8Array, 0);
debug(`Parsing tag type=${type}, encoding=${encoding}, bom=${bom}`);
let text: string;
try {
text = FrameParser.trimNullPadding(util.decodeString(uint8Array.subarray(1), encoding));
@@ -167,7 +169,7 @@ export class FrameParser {
break;
case 'TCO':
case 'TCON':
output = this.splitValue(type, text).flatMap(v => parseGenre(v));
output = this.splitValue(type, text).map(v => parseGenre(v)).reduce((acc, val) => acc.concat(val), []);
break;
case 'PCS':
case 'PCST':
@@ -182,12 +184,10 @@ export class FrameParser {
}
case 'TXXX': {
const {encoding, bom} = TextEncodingToken.get(uint8Array, 0);
debug(`Parsing tag type=${type}, encoding=${encoding}, bom=${bom}`);
const idAndData = FrameParser.readIdentifierAndData(uint8Array.subarray(1), encoding);
const textTag = {
description: idAndData.id,
text: this.splitValue(type, FrameParser.trimNullPadding(util.decodeString(idAndData.data, encoding)))
text: this.splitValue(type, util.decodeString(idAndData.data, encoding).replace(/\x00+$/, ''))
};
output = textTag;
break;
@@ -196,8 +196,6 @@ export class FrameParser {
case 'PIC':
case 'APIC':
if (includeCovers) {
const {encoding, bom} = TextEncodingToken.get(uint8Array, 0);
debug(`Parsing tag type=${type}, encoding=${encoding}, bom=${bom}`);
const pic: IPicture = {};
uint8Array = uint8Array.subarray(1);
@@ -208,11 +206,10 @@ export class FrameParser {
uint8Array = uint8Array.subarray(3);
break;
case 3:
case 4: {
const formatStr = FrameParser.readNullTerminatedString(uint8Array, {encoding: defaultEnc, bom: false});
pic.format = formatStr.text;
uint8Array = uint8Array.subarray(formatStr.len);
}
case 4:
fzero = util.findZero(uint8Array, defaultEnc);
pic.format = util.decodeString(uint8Array.subarray(0, fzero), defaultEnc);
uint8Array = uint8Array.subarray(fzero + 1);
break;
default:
@@ -224,9 +221,9 @@ export class FrameParser {
pic.type = AttachedPictureType[uint8Array[0] as keyof typeof AttachedPictureType];
uint8Array = uint8Array.subarray(1);
const descriptionStr = FrameParser.readNullTerminatedString(uint8Array, {encoding, bom});
pic.description = descriptionStr.text;
uint8Array = uint8Array.subarray(descriptionStr.len);
fzero = util.findZero(uint8Array, encoding);
pic.description = util.decodeString(uint8Array.subarray(0, fzero), encoding);
uint8Array = uint8Array.subarray(fzero + nullTerminatorLength);
pic.data = uint8Array;
output = pic;
}
@@ -276,11 +273,6 @@ export class FrameParser {
case 'COM':
case 'COMM': {
if (uint8Array.length < TextHeader.len) {
this.warningCollector.addWarning(`id3v2.${this.major} type=${type} frame too short for text header`);
break;
}
const textHeader = TextHeader.get(uint8Array, offset);
offset += TextHeader.len;
@@ -300,20 +292,21 @@ export class FrameParser {
}
case 'UFID': {
const ufid = FrameParser.readIdentifierAndData(uint8Array.subarray(offset), defaultEnc);
const ufid = FrameParser.readIdentifierAndData(uint8Array, defaultEnc);
output = {owner_identifier: ufid.id, identifier: ufid.data} as IIdentifierTag;
break;
}
case 'PRIV': { // private frame
const priv = FrameParser.readIdentifierAndData(uint8Array.subarray(offset), defaultEnc);
const priv = FrameParser.readIdentifierAndData(uint8Array, defaultEnc);
output = {owner_identifier: priv.id, data: priv.data} as ICustomDataTag;
break;
}
case 'POPM': { // Popularimeter
uint8Array = uint8Array.subarray(offset);
const emailStr = FrameParser.readNullTerminatedString(uint8Array, {encoding: defaultEnc, bom: false});
const emailStr = FrameParser.readNullTerminatedString(uint8Array, urlEnc);
const email = emailStr.text;
uint8Array = uint8Array.subarray(emailStr.len);
@@ -334,18 +327,19 @@ export class FrameParser {
}
case 'GEOB': { // General encapsulated object
const {encoding, bom} = TextEncodingToken.get(uint8Array, 0);
debug(`Parsing tag type=${type}, encoding=${encoding}, bom=${bom}`);
// [encoding] <MIME> 0x00 <filename> 0x00/0x00 0x00 <description> 0x00/0x00 0x00 <data>
const {encoding: geobEncoding, bom: geobBom} = TextEncodingToken.get(uint8Array, 0);
uint8Array = uint8Array.subarray(1);
const mimeTypeStr = FrameParser.readNullTerminatedString(uint8Array, {encoding: defaultEnc, bom: false});
const mimeTypeStr = FrameParser.readNullTerminatedString(uint8Array, urlEnc);
const mimeType = mimeTypeStr.text;
uint8Array = uint8Array.subarray(mimeTypeStr.len);
const filenameStr = FrameParser.readNullTerminatedString(uint8Array, {encoding, bom});
const filenameStr = FrameParser.readNullTerminatedString(uint8Array, {encoding: geobEncoding, bom: geobBom});
const filename = filenameStr.text;
uint8Array = uint8Array.subarray(filenameStr.len);
const descriptionStr = FrameParser.readNullTerminatedString(uint8Array, {encoding, bom});
const descriptionStr = FrameParser.readNullTerminatedString(uint8Array, {encoding: geobEncoding, bom: geobBom});
const description = descriptionStr.text;
uint8Array = uint8Array.subarray(descriptionStr.len);
@@ -369,27 +363,28 @@ export class FrameParser {
case 'WPAY':
case 'WPUB':
// Decode URL
output = FrameParser.readNullTerminatedString(uint8Array, {encoding: defaultEnc, bom: false}).text;
output = FrameParser.readNullTerminatedString(uint8Array, urlEnc).text;
break;
case 'WXXX': {
const {encoding, bom} = TextEncodingToken.get(uint8Array, 0);
debug(`Parsing tag type=${type}, encoding=${encoding}, bom=${bom}`);
// Decode URL
// [encoding] <description> 0x00/0x00 0x00 <url>
const {encoding: wxxxEncoding, bom: wxxxBom} = TextEncodingToken.get(uint8Array, 0);
uint8Array = uint8Array.subarray(1);
const descriptionStr = FrameParser.readNullTerminatedString(uint8Array, {encoding, bom});
const descriptionStr = FrameParser.readNullTerminatedString(uint8Array, {encoding: wxxxEncoding, bom: wxxxBom});
const description = descriptionStr.text;
uint8Array = uint8Array.subarray(descriptionStr.len);
// URL is always ISO-8859-1
output = {description, url: FrameParser.trimNullPadding(util.decodeString(uint8Array, defaultEnc))};
break;
}
case 'WFD':
case 'WFED': {
const {encoding, bom} = TextEncodingToken.get(uint8Array, 0);
debug(`Parsing tag type=${type}, encoding=${encoding}, bom=${bom}`);
const {encoding: wfdEncoding, bom: wfdBom} = TextEncodingToken.get(uint8Array, 0);
uint8Array = uint8Array.subarray(1);
output = FrameParser.readNullTerminatedString(uint8Array, {encoding, bom}).text;
output = FrameParser.readNullTerminatedString(uint8Array, {encoding: wfdEncoding, bom: wfdBom}).text;
break;
}
@@ -413,9 +408,8 @@ export class FrameParser {
const valueArray = uint8Array.subarray(bomSize);
const zeroIndex = util.findZero(valueArray, encoding.encoding);
// findZero returns valueArray.length when no terminator is found
if (zeroIndex >= valueArray.length) {
// No terminator found, decode full buffer remainder
return {
text: util.decodeString(valueArray, encoding.encoding),
len: originalLen
@@ -423,7 +417,6 @@ export class FrameParser {
}
const txt = valueArray.subarray(0, zeroIndex);
return {
text: util.decodeString(txt, encoding.encoding),
len: bomSize + zeroIndex + FrameParser.getNullTerminatorLength(encoding.encoding)
@@ -503,4 +496,4 @@ export class Id3v2ContentError extends makeUnexpectedFileContentError('id3v2') {
function makeUnexpectedMajorVersionError(majorVer: number) {
throw new Id3v2ContentError(`Unexpected majorVer: ${majorVer}`);
}
}
lib/id3v2/ID3v2Parser.ts
+1 -1
View File
@@ -170,7 +170,7 @@ export class ID3v2Parser {
}
function makeUnexpectedMajorVersionError(majorVer: number) {
function makeUnexpectedMajorVersionError(majorVer: number): never {
throw new Id3v2ContentError(`Unexpected majorVer: ${majorVer}`);
}