Ruby mp3 Id3 parsing

Question

Currently I\'m working on a music project, dealing with user mp3 uploads. The problem is that I can\'t find an id3 library that will work correctly for all files. I have tried

Answer 1

http://id3lib-ruby.rubyforge.org/

I particularly liked this one, you can also write tags to the file.

Answer 2

id3tag is another option. Example:

require "id3tag"

mp3_file = File.open('/path/to/your/favorite_song.mp3', "rb")
tag = ID3Tag.read(mp3_file)
puts "#{tag.artist} - #{tag.title}"

Answer 3

http://www.hakubi.us/ruby-taglib/

I used this for a project and it worked quite well. Wrapper around taglib, which is very portable.

Answer 4

As of 2019, the best answers are:

• Krists Ozols' ID3Tag
• Moumar's ruby-mp3info
• taglib-ruby

All other libraries are long-since unmaintained.

Krists Ozols' ID3Tag distinguishing characteristics

• read only
• Can read v1.x, v2.2.x, v2.3.x, v2.4.x tags
• Supports UTF-8, UTF-16, UTF-16BE and ISO8859-1 encoding
• last updated July 2018
• Pure Ruby

Moumar's ruby-mp3info distinguishing characteristics

• read and write
• Only 2.3 version is supported for writings id3v2 tags
• id3v2 tags are always written in UTF-16 encoding
• last updated April 2017
• Pure Ruby

taglib-ruby distinguishing characteristics

• read and write
• Many formats, not just Mp3
• Reading/writing ID3v1 and ID3v2 including ID3v2.4 and Unicode
• last updated May 2018
• Binding of a well-maintained C++ library

Answer 5

I've used this:

http://ruby-mp3info.rubyforge.org/

or

gem install ruby-mp3info (add the regulation sudo for Mac or *nix)

There's some rdoc documentation, which is nice. On the downside, I don't much like the use of upper-case field names, which seems too concerned to preserve the names from the spec. Maybe I should hack in some aliases. Anyway, this sample script scans my music library and counts words in titles:

require 'mp3info'

count = 0
words = Hash.new { |h, k| h[k] = 0 }
Dir.glob("E:/MUSIC/**/*.mp3") do |f|
  count += 1
  Mp3Info.open(f) do |mp3info|
    title = mp3info.tag2.TIT2
    next unless title
    title.split(/\s/).each { |w| words[w.downcase] += 1 }
  end
end
puts "Examined #{count} files"
words.to_a.sort{ |a, b| b[1] <=> a[1] }[0,100].each { |w| puts "#{w[0]}: #{w[1]}" }