Skip to content

Commit

Permalink
add id3 frame data parsing
Browse files Browse the repository at this point in the history
  • Loading branch information
@adrums86
adrums86 committed Feb 3, 2023
1 parent da67712 commit 4479c56
Show file tree
Hide file tree
Showing 4 changed files with 296 additions and 192 deletions.
179 changes: 9 additions & 170 deletions lib/m2ts/metadata-stream.js
View file
Original file line number Diff line number Diff line change
Expand Up @@ -12,168 +12,7 @@
var
Stream = require('../utils/stream'),
StreamTypes = require('./stream-types'),
typedArrayIndexOf = require('../utils/typed-array').typedArrayIndexOf,
// Frames that allow different types of text encoding contain a text
// encoding description byte [ID3v2.4.0 section 4.]
textEncodingDescriptionByte = {
Iso88591: 0x00, // ISO-8859-1, terminated with \0.
Utf16: 0x01, // UTF-16 encoded Unicode BOM, terminated with \0\0
Utf16be: 0x02, // UTF-16BE encoded Unicode, without BOM, terminated with \0\0
Utf8: 0x03 // UTF-8 encoded Unicode, terminated with \0
},
// return a percent-encoded representation of the specified byte range
// @see http://en.wikipedia.org/wiki/Percent-encoding
percentEncode = function(bytes, start, end) {
var i, result = '';
for (i = start; i < end; i++) {
result += '%' + ('00' + bytes[i].toString(16)).slice(-2);
}
return result;
},
// return the string representation of the specified byte range,
// interpreted as UTf-8.
parseUtf8 = function(bytes, start, end) {
return decodeURIComponent(percentEncode(bytes, start, end));
},
// return the string representation of the specified byte range,
// interpreted as ISO-8859-1.
parseIso88591 = function(bytes, start, end) {
return unescape(percentEncode(bytes, start, end)); // jshint ignore:line
},
parseSyncSafeInteger = function(data) {
return (data[0] << 21) |
(data[1] << 14) |
(data[2] << 7) |
(data[3]);
},
frameParsers = {
'APIC': function(frame) {
var
i = 1,
mimeTypeEndIndex,
descriptionEndIndex,
LINK_MIME_TYPE = '-->';

if (frame.data[0] !== textEncodingDescriptionByte.Utf8) {
// ignore frames with unrecognized character encodings
return;
}

// parsing fields [ID3v2.4.0 section 4.14.]
mimeTypeEndIndex = typedArrayIndexOf(frame.data, 0, i);
if (mimeTypeEndIndex < 0) {
// malformed frame
return;
}

// parsing Mime type field (terminated with \0)
frame.mimeType = parseIso88591(frame.data, i, mimeTypeEndIndex);
i = mimeTypeEndIndex + 1;

// parsing 1-byte Picture Type field
frame.pictureType = frame.data[i];
i++

descriptionEndIndex = typedArrayIndexOf(frame.data, 0, i);
if (descriptionEndIndex < 0) {
// malformed frame
return;
}

// parsing Description field (terminated with \0)
frame.description = parseUtf8(frame.data, i, descriptionEndIndex);
i = descriptionEndIndex + 1;

if (frame.mimeType === LINK_MIME_TYPE) {
// parsing Picture Data field as URL (always represented as ISO-8859-1 [ID3v2.4.0 section 4.])
frame.url = parseIso88591(frame.data, i, frame.data.length)
} else {
// parsing Picture Data field as binary data
frame.pictureData = frame.data.subarray(i, frame.data.length);
}
},
'T*': function(frame) {
if (frame.data[0] !== textEncodingDescriptionByte.Utf8) {
// ignore frames with unrecognized character encodings
return;
}

// parse text field, do not include null terminator in the frame value
// frames that allow different types of encoding contain terminated text [ID3v2.4.0 section 4.]
frame.value = parseUtf8(frame.data, 1, frame.data.length).replace(/\0*$/, '');
// text information frames supports multiple strings, stored as a terminator separated list [ID3v2.4.0 section 4.2.]
frame.values = frame.value.split('\0');
},
'TXXX': function(frame) {
var descriptionEndIndex;

if (frame.data[0] !== textEncodingDescriptionByte.Utf8) {
// ignore frames with unrecognized character encodings
return;
}

descriptionEndIndex = typedArrayIndexOf(frame.data, 0, 1);

if (descriptionEndIndex === -1) {
return;
}

// parse the text fields
frame.description = parseUtf8(frame.data, 1, descriptionEndIndex);
// do not include the null terminator in the tag value
// frames that allow different types of encoding contain terminated text
// [ID3v2.4.0 section 4.]
frame.value = parseUtf8(
frame.data,
descriptionEndIndex + 1,
frame.data.length
).replace(/\0*$/, '');
frame.data = frame.value;
},
'W*': function(frame) {
// parse URL field; URL fields are always represented as ISO-8859-1 [ID3v2.4.0 section 4.]
// if the value is followed by a string termination all the following information should be ignored [ID3v2.4.0 section 4.3]
frame.url = parseIso88591(frame.data, 0, frame.data.length).replace(/\0.*$/, '');
},
'WXXX': function(frame) {
var descriptionEndIndex;

if (frame.data[0] !== textEncodingDescriptionByte.Utf8) {
// ignore frames with unrecognized character encodings
return;
}

descriptionEndIndex = typedArrayIndexOf(frame.data, 0, 1);

if (descriptionEndIndex === -1) {
return;
}

// parse the description and URL fields
frame.description = parseUtf8(frame.data, 1, descriptionEndIndex);
// URL fields are always represented as ISO-8859-1 [ID3v2.4.0 section 4.]
// if the value is followed by a string termination all the following information
// should be ignored [ID3v2.4.0 section 4.3]
frame.url = parseIso88591(
frame.data,
descriptionEndIndex + 1,
frame.data.length
).replace(/\0.*$/, '');
},
'PRIV': function(frame) {
var i;

for (i = 0; i < frame.data.length; i++) {
if (frame.data[i] === 0) {
// parse the description and URL fields
frame.owner = parseIso88591(frame.data, 0, i);
break;
}
}
frame.privateData = frame.data.subarray(i + 1);
frame.data = frame.privateData;
}
},
id3 = require('../tools/parse-id3'),
MetadataStream;

MetadataStream = function(options) {
Expand Down Expand Up @@ -241,7 +80,7 @@ MetadataStream = function(options) {
// last four bytes of the ID3 header.
// The most significant bit of each byte is dropped and the
// results concatenated to recover the actual value.
tagSize = parseSyncSafeInteger(chunk.data.subarray(6, 10));
tagSize = id3.parseSyncSafeInteger(chunk.data.subarray(6, 10));

// ID3 reports the tag size excluding the header but it's more
// convenient for our comparisons to include it
Expand Down Expand Up @@ -272,17 +111,17 @@ MetadataStream = function(options) {
if (tag.data[5] & 0x40) {
// advance the frame start past the extended header
frameStart += 4; // header size field
frameStart += parseSyncSafeInteger(tag.data.subarray(10, 14));
frameStart += id3.parseSyncSafeInteger(tag.data.subarray(10, 14));

// clip any padding off the end
tagSize -= parseSyncSafeInteger(tag.data.subarray(16, 20));
tagSize -= id3.parseSyncSafeInteger(tag.data.subarray(16, 20));
}

// parse one or more ID3 frames
// http://id3.org/id3v2.3.0#ID3v2_frame_overview
do {
// determine the number of bytes in this frame
frameSize = parseSyncSafeInteger(tag.data.subarray(frameStart + 4, frameStart + 8));
frameSize = id3.parseSyncSafeInteger(tag.data.subarray(frameStart + 4, frameStart + 8));
if (frameSize < 1) {
this.trigger('log', {
level: 'warn',
Expand All @@ -305,15 +144,15 @@ MetadataStream = function(options) {
frame.key = frame.id;

// parse frame values
if (frameParsers[frame.id]) {
if (id3.frameParsers[frame.id]) {
// use frame specific parser
frameParsers[frame.id](frame);
id3.frameParsers[frame.id](frame);
} else if (frame.id[0] === 'T') {
// use text frame generic parser
frameParsers['T*'](frame);
id3.frameParsers['T*'](frame);
} else if (frame.id[0] === 'W') {
// use URL link frame generic parser
frameParsers['W*'](frame);
id3.frameParsers['W*'](frame);
}

// handle the special PRIV frame used to indicate the start
Expand Down
9 changes: 5 additions & 4 deletions lib/mp4/probe.js
View file
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ var getUint64 = require('../utils/numbers.js').getUint64;
var timescale, startTime, compositionStartTime, getVideoTrackIds, getTracks,
getTimescaleFromMediaHeader, getEmsgID3;
var window = require('global/window');
var parseId3Frames = require('../tools/parse-id3.js').parseId3Frames;


/**
Expand Down Expand Up @@ -378,12 +379,12 @@ getTracks = function(init) {
getEmsgID3 = function(segmentData, ptOffset = 0) {
var emsgBoxes = findBox(segmentData, ['emsg']);
return emsgBoxes.map((data) => {
var boxData = new Uint8Array(data);
var parsedBox = emsg.parseEmsgBox(boxData);
var parsedBox = emsg.parseEmsgBox(new Uint8Array(data));
var parsedId3Frames = parseId3Frames(parsedBox.message_data);
return {
pts: emsg.scaleTime(parsedBox.presentation_time, parsedBox.timescale, parsedBox.presentation_time_delta, ptOffset),
cueTime: emsg.scaleTime(parsedBox.presentation_time, parsedBox.timescale, parsedBox.presentation_time_delta, ptOffset),
duration: emsg.scaleTime(parsedBox.event_duration, parsedBox.timescale),
data: parsedBox.message_data
frames: parsedId3Frames
};
});
};
Expand Down
Loading

0 comments on commit 4479c56

Please sign in to comment.