sweeper 0.1 → 0.2

data/CHANGELOG

@@ -1,2 +1,4 @@
 
+v0.2. Genre tagging; bugfixes.
+
 v0.1. First release.

data/Manifest

@@ -7,6 +7,7 @@ LICENSE
 Manifest
 README
 test/integration/sweeper_test.rb
+TODO
 vendor/lastfm.fpclient.beta2.linux-32/lastfmfpclient
 vendor/lastfm.fpclient.beta2.linux-32/readme.txt
 vendor/lastfm.fpclient.beta2.OSX-intel/lastfmfpclient

data/README

@@ -3,7 +3,7 @@ Automatically tag your music collection with metadata from Last.fm.
 
 == License
 
-Copyright 2008 Cloudburst, LLC. Licensed under the AFL 3. See the included LICENSE file. Unlinked portions copyright 2008 Last.fm Ltd., used by permission, and licensed under the GPL2.
+Copyright 2008 Cloudburst, LLC. Licensed under the AFL 3. See the included LICENSE file. Unlinked portions copyright 2008 Last.fm Ltd., used by permission, and licensed under the GPL 2.
 
 The public certificate for the gem is here[http://rubyforge.org/frs/download.php/25331/evan_weaver-original-public_cert.pem]. 
 
@@ -11,15 +11,27 @@ If you use this software, please {make a donation}[http://blog.evanweaver.com/do
 
 == Features
 
-* options
-* stuff
+* faster than MusicBrainz
+* better results than MusicBrainz
+* smart genre tagging
 
-== Usage
+<b>WARNING: This is beta software. Backup your music collection first.</b>
+
+== Requirements
+
+* Linux, Intel OS X, or Windows
+* Ruby and Rubygems
+* id3lib (Linux and OS X only)
+* some mp3 files
 
-First, install the gem:
+== Installation
+
+Just install the gem:
   sudo gem install sweeper
 
-Now, change to the directory of mp3s you want to tag and run:
+== Usage
+
+Change to the directory of mp3s you want to tag and run:
   sweeper
   
 That's all.
@@ -31,8 +43,10 @@ Sweeper takes a few command line options:
     -d, --dir                        Directory to search (defaults to current).
     -r, --recursive                  Recurse directories.
         --dry-run                    Do a dry run (no files will be changed).
-    -f, --force                      Overwrite existing tags.
-    
+    -f, --force                      Overwrite all existing tags.
+    -g, --genre                      Add genre and genre comments.
+    -e, --force-genre                Add genre and genre comments, overwriting e
+        
 == Reporting problems
 
 The support forum is here[http://rubyforge.org/forum/forum.php?forum_id=23599].

data/TODO

@@ -0,0 +1,3 @@
+
+* Add genres via http://ws.audioscrobbler.com/1.0/artist/ARTIST/toptags.xml service.
+* Some kind of album detection?

data/bin/sweeper

@@ -19,7 +19,7 @@ Choice.options do
     desc 'Recurse directories.'
   end
   
-  option :dry do
+  option :"dry-run" do
     long '--dry-run'
     desc 'Do a dry run (no files will be changed).'
   end
@@ -27,14 +27,20 @@ Choice.options do
   option :force do
     short '-f'
     long '--force'
-    desc 'Overwrite existing tags.'
+    desc 'Overwrite all existing tags.'
   end
   
-#  option :genre do
-#    short '-g'
-#    long '--genre'
-#    desc 'Add genre tags from Last.fm.'
-#  end
+  option :genre do
+    short '-g'
+    long '--genre'
+    desc "Add genre and genre comments."
+  end
+  
+  option :"force-genre" do
+    short '-e'
+    long '--force-genre'
+    desc "Add genre and genre comments, overwriting existing ones."
+  end    
 end
 
 Sweeper.new(Choice.choices).run

data/lib/sweeper.rb

@@ -3,31 +3,59 @@ require 'rubygems'
 require 'id3lib'
 require 'xsd/mapping'
 require 'activesupport'
+require 'open-uri'
+require 'uri'
+require 'amatch'
 
 class ID3Lib::Tag
-  alias :url :comment
-  alias :url= :comment=
+  def url
+    f = frame(:WORS)
+    f ? f[:url] : nil
+  end
+  
+  def url=(s)
+    remove_frame(:WORS)    
+    self << {:id => :WORS, :url => s} if s.any?
+  end
 end
 
 class Sweeper
 
   class Problem < RuntimeError; end
 
-  KEYS = ['artist', 'title', 'url']
+  BASIC_KEYS = ['artist', 'title', 'url']
+  GENRE_KEYS = ['genre', 'comment']
+  GENRES = ID3Lib::Info::Genres.sort
+  GENRE_COUNT = 7
+  DEFAULT_GENRE = {'genre' => 'Other', 'comment' => 'other'}
 
   attr_reader :options
 
   def initialize(options = {})
+    options['genre'] ||= options['force-genre']
     @dir = File.expand_path(options['dir'] || Dir.pwd)
     @options = options
   end
   
-  def run
-    @processed = 0
-    recurse(@dir)
-    if @processed == 0
-      puts "No files found."
-      exec "#{$0} --help"
+  def run      
+    @read = 0
+    @updated = 0
+    @failed = 0
+
+    Kernel.at_exit do
+      if @read == 0
+        puts "No files found. Maybe you meant --recursive?"
+        exec "#{$0} --help"
+      else
+        puts "Read: #{@read}\nUpdated: #{@updated}\nFailed: #{@failed}"
+      end
+    end      
+  
+    begin
+      recurse(@dir)
+    rescue Object => e
+      puts "Unknown error: #{e.inspect}"
+      ENV['DEBUG'] ? raise : exit
     end
   end
   
@@ -38,19 +66,21 @@ class Sweeper
       if File.directory? filename and options['recursive']
         recurse(filename)
       elsif File.extname(filename) == ".mp3"
-        @processed += 1
+        @read += 1
         tries = 0
         begin
-          current = read(filename)
-  
-          if options['force']
-            write(filename, lookup(filename))
-          elsif current.keys.size < KEYS.size            
-            write(filename, lookup(filename).except(*current.keys))
+          current = read(filename)  
+          updated = lookup(filename, current)
+          
+          if updated != current
+            write(filename, updated)
+            @updated += 1
           end
+          
         rescue Problem => e          
           tries += 1 and retry if tries < 2
-          puts "Skipped #{filename}: #{e.message}"
+          puts "Skipped #{File.basename(filename)}: #{e.message}"
+          @failed += 1
         end
       end
     end
@@ -58,44 +88,103 @@ class Sweeper
   
   def read(filename)
     tags = {}
-    song = ID3Lib::Tag.new(filename)
+
+    song = ID3Lib::Tag.new(filename, ID3Lib::V2)
+    if song.empty?
+      song = ID3Lib::Tag.new(filename, ID3Lib::V1)
+    end
     
-    KEYS.each do |key|      
+    (BASIC_KEYS + GENRE_KEYS).each do |key|      
       tags[key] = song.send(key) if !song.send(key).blank?
     end
+    
     tags
   end
   
-  def lookup(filename)
+  def lookup(filename, tags = {})
+    updated = {}
+    if options['force'] or 
+      (BASIC_KEYS - tags.keys).any?
+      updated.merge!(lookup_basic(filename))
+    end
+    if options['genre'] and 
+      (options['force'] or options['force-genre'] or (GENRE_KEYS - tags.keys).any?)
+      updated.merge!(lookup_genre(updated.merge(tags)))
+    end
+
+    if options['force']
+      tags.merge!(updated)      
+    elsif options['force-genre']
+      tags.merge!(updated.slice('genre', 'comment'))
+    end
+
+    updated.merge(tags)    
+  end
+  
+  def lookup_basic(filename)
     Dir.chdir File.dirname(binary) do
-      response = silence { `./#{File.basename(binary)} #{filename}` }
+      response = silence { `./#{File.basename(binary)} #{filename.inspect}` }
       object = begin
         XSD::Mapping.xml2obj(response)
       rescue REXML::ParseException
-        raise Problem, "Invalid response."
+        raise Problem, "Server sent invalid response."
       end              
-      raise Problem, "Not found." unless object
+      raise Problem, "Fingerprint failed or not found." unless object
       
       tags = {}
       song = Array(object.track).first      
       
-      KEYS.each do |key|
+      BASIC_KEYS.each do |key|
         tags[key] = song.send(key) if song.respond_to? key
       end
       tags
     end
   end
   
+  def lookup_genre(tags)
+    return DEFAULT_GENRE if tags['artist'].blank?
+    
+    response = begin 
+      open("http://ws.audioscrobbler.com/1.0/artist/#{URI.encode(tags['artist'])}/toptags.xml").read
+    rescue OpenURI::HTTPError
+      return DEFAULT_GENRE
+    end
+    
+    object = XSD::Mapping.xml2obj(response)
+    return DEFAULT_GENRE if !object.respond_to? :tag
+
+    genres = Array(object.tag)[0..(GENRE_COUNT - 1)].map(&:name)
+    return DEFAULT_GENRE if !genres.any?
+    
+    primary = nil
+    genres.each do |this|
+      match_results = Amatch::Levenshtein.new(this).similar(GENRES)
+      max = match_results.max
+      match = GENRES[match_results.index(max)]
+
+      if ['Rock', 'Pop', 'Rap'].include? match
+        # Penalize useless genres
+        max = max / 3.0
+      end
+      
+      if !primary or primary.first < max
+        primary = [max, match]
+      end
+    end
+    
+    {'genre' => primary.last, 'comment' => genres.join(" ")}
+  end
+  
   def write(filename, tags)
     return if tags.empty?
-    puts filename
+    puts File.basename(filename)
     
-    file = ID3Lib::Tag.new(filename)
+    file = ID3Lib::Tag.new(filename, ID3Lib::V2)
     tags.each do |key, value|
       file.send("#{key}=", value)
-      puts "  #{value}"
+      puts "  #{key.capitalize}: #{value}"
     end
-    file.update! unless options['dry']
+    file.update! unless options['dry-run']
   end    
   
   def binary

data/sweeper.gemspec

@@ -1,10 +1,10 @@
 
-# Gem::Specification for Sweeper-0.1
+# Gem::Specification for Sweeper-0.2
 # Originally generated by Echoe
 
 Gem::Specification.new do |s|
   s.name = %q{sweeper}
-  s.version = "0.1"
+  s.version = "0.2"
 
   s.specification_version = 2 if s.respond_to? :specification_version=
 
@@ -16,19 +16,20 @@ Gem::Specification.new do |s|
   s.email = %q{}
   s.executables = ["sweeper"]
   s.extensions = ["ext/extconf.rb"]
-  s.extra_rdoc_files = ["CHANGELOG", "lib/sweeper.rb", "LICENSE", "README"]
-  s.files = ["bin/sweeper", "CHANGELOG", "ext/extconf.rb", "ext/Makefile", "lib/sweeper.rb", "LICENSE", "Manifest", "README", "test/integration/sweeper_test.rb", "vendor/lastfm.fpclient.beta2.linux-32/lastfmfpclient", "vendor/lastfm.fpclient.beta2.linux-32/readme.txt", "vendor/lastfm.fpclient.beta2.OSX-intel/lastfmfpclient", "vendor/lastfm.fpclient.beta2.OSX-intel/libcurl.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/libmad.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/libsamplerate.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/libtag.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/readme.txt", "vendor/lastfm.fpclient.beta2.win32/lastfmfpclient.exe", "vendor/lastfm.fpclient.beta2.win32/libcurl.dll", "vendor/lastfm.fpclient.beta2.win32/libfftw3f-3.dll", "vendor/lastfm.fpclient.beta2.win32/Microsoft.VC80.CRT.manifest", "vendor/lastfm.fpclient.beta2.win32/msvcm80.dll", "vendor/lastfm.fpclient.beta2.win32/msvcp80.dll", "vendor/lastfm.fpclient.beta2.win32/msvcr80.dll", "vendor/lastfm.fpclient.beta2.win32/taglib.dll", "vendor/lastfm.fpclient.beta2.win32/zlib1.dll", "sweeper.gemspec"]
+  s.extra_rdoc_files = ["CHANGELOG", "lib/sweeper.rb", "LICENSE", "README", "TODO"]
+  s.files = ["bin/sweeper", "CHANGELOG", "ext/extconf.rb", "ext/Makefile", "lib/sweeper.rb", "LICENSE", "Manifest", "README", "test/integration/sweeper_test.rb", "TODO", "vendor/lastfm.fpclient.beta2.linux-32/lastfmfpclient", "vendor/lastfm.fpclient.beta2.linux-32/readme.txt", "vendor/lastfm.fpclient.beta2.OSX-intel/lastfmfpclient", "vendor/lastfm.fpclient.beta2.OSX-intel/libcurl.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/libmad.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/libsamplerate.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/libtag.dylib", "vendor/lastfm.fpclient.beta2.OSX-intel/readme.txt", "vendor/lastfm.fpclient.beta2.win32/lastfmfpclient.exe", "vendor/lastfm.fpclient.beta2.win32/libcurl.dll", "vendor/lastfm.fpclient.beta2.win32/libfftw3f-3.dll", "vendor/lastfm.fpclient.beta2.win32/Microsoft.VC80.CRT.manifest", "vendor/lastfm.fpclient.beta2.win32/msvcm80.dll", "vendor/lastfm.fpclient.beta2.win32/msvcp80.dll", "vendor/lastfm.fpclient.beta2.win32/msvcr80.dll", "vendor/lastfm.fpclient.beta2.win32/taglib.dll", "vendor/lastfm.fpclient.beta2.win32/zlib1.dll", "sweeper.gemspec"]
   s.has_rdoc = true
   s.homepage = %q{http://blog.evanweaver.com/files/doc/fauna/sweeper/}
   s.rdoc_options = ["--line-numbers", "--inline-source", "--title", "Sweeper", "--main", "README"]
   s.require_paths = ["lib", "ext"]
   s.rubyforge_project = %q{fauna}
-  s.rubygems_version = %q{1.0.1}
+  s.rubygems_version = %q{1.1.0}
   s.summary = %q{Automatically tag your music collection with metadata from Last.fm.}
   s.test_files = ["test/integration/sweeper_test.rb"]
 
   s.add_dependency(%q<id3lib-ruby>, [">= 0"])
   s.add_dependency(%q<choice>, [">= 0"])
+  s.add_dependency(%q<amatch>, [">= 0"])
   s.add_dependency(%q<activesupport>, [">= 0"])
 end
 
@@ -43,7 +44,7 @@ end
 #   p.summary = "Automatically tag your music collection with metadata from Last.fm."
 #   p.url = "http://blog.evanweaver.com/files/doc/fauna/sweeper/"
 #   p.docs_host = "blog.evanweaver.com:~/www/bax/public/files/doc/"
-#   p.dependencies = ['id3lib-ruby', 'choice', 'activesupport']
+#   p.dependencies = ['id3lib-ruby', 'choice', 'amatch', 'activesupport']
 #   p.clean_pattern = ['doc', 'pkg', 'test/integration/songs']
 #   p.rdoc_pattern = ['README', 'LICENSE', 'CHANGELOG', 'TODO', 'lib/*']
 # end

data/test/integration/sweeper_test.rb

@@ -11,25 +11,36 @@ class SweeperTest < Test::Unit::TestCase
     @found_many = "#{@dir}/1_001.mp3"
     @found_one = "#{@dir}/1_010.mp3"
     @not_found = "#{@dir}/1_003.mp3"
-    @s = Sweeper.new('dir' => @dir)
+    @s = Sweeper.new('dir' => @dir, 'genre' => true)
   end
   
-  def test_lookup
+  def test_lookup_basic
     assert_equal(
       {"artist"=>"Photon Band", 
         "title"=>"To Sing For You", 
         "url"=>"http://www.last.fm/music/Photon+Band/_/To+Sing+For+You"},
-      @s.lookup(@found_many))
+      @s.lookup_basic(@found_many))
     assert_equal(
       {"artist"=>"Various Artists - Vagabond Productions",
         "title"=>"Sugar Man - Tom Heyman",
         "url"=> "http://www.last.fm/music/Various+Artists+-+Vagabond+Productions/_/Sugar+Man+-+Tom+Heyman"},
-      @s.lookup(@found_one))
+      @s.lookup_basic(@found_one))
     assert_raises(Sweeper::Problem) do
-      @s.lookup(@not_found)
+      @s.lookup_basic(@not_found)
     end
   end
   
+  def test_lookup_genre
+    assert_equal(
+       {"genre"=>"Psychadelic", "comment"=>"rock psychedelic mod Philly"},
+      @s.lookup_genre(@s.lookup_basic(@found_many))    
+    )
+    assert_equal(
+      Sweeper::DEFAULT_GENRE,
+      @s.lookup_genre({})
+    )
+  end
+  
   def test_read
     assert_equal({}, 
       @s.read(@found_many))

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: sweeper
 version: !ruby/object:Gem::Version 
-  version: "0.1"
+  version: "0.2"
 platform: ruby
 authors: 
 - Evan Weaver
@@ -51,6 +51,15 @@ dependencies:
       - !ruby/object:Gem::Version 
         version: "0"
     version: 
+- !ruby/object:Gem::Dependency 
+  name: amatch
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: "0"
+    version: 
 - !ruby/object:Gem::Dependency 
   name: activesupport
   version_requirement: 
@@ -71,6 +80,7 @@ extra_rdoc_files:
 - lib/sweeper.rb
 - LICENSE
 - README
+- TODO
 files: 
 - bin/sweeper
 - CHANGELOG
@@ -81,6 +91,7 @@ files:
 - Manifest
 - README
 - test/integration/sweeper_test.rb
+- TODO
 - vendor/lastfm.fpclient.beta2.linux-32/lastfmfpclient
 - vendor/lastfm.fpclient.beta2.linux-32/readme.txt
 - vendor/lastfm.fpclient.beta2.OSX-intel/lastfmfpclient
@@ -127,7 +138,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: fauna
-rubygems_version: 1.0.1
+rubygems_version: 1.1.0
 signing_key: 
 specification_version: 2
 summary: Automatically tag your music collection with metadata from Last.fm.