linnaeus 1.0.0 → 1.0.2

data/README.rdoc

@@ -2,7 +2,7 @@
 
 https://raw.github.com/djcp/linnaeus/master/images/linnaeus.jpg
 
-Linnaeus is a redis-backed Bayesian classification system. Please see the generated rdocs for more information.
+Linnaeus is a redis-backed Bayesian classification system. Please see the generated rdocs for more information. Ruby 1.9 is required.
 
 == Examples
 
@@ -23,7 +23,6 @@ Linnaeus is a redis-backed Bayesian classification system. Please see the genera
 
 == The Future
 
-* Make sure we're unicode OK
 * Create additional storage backends - sqlite, postgresql, mongodb, etc.
 * Allow for weighting tweaks.
 

data/VERSION

@@ -1 +1 @@
-1.0.0
+1.0.2

data/lib/linnaeus.rb

@@ -10,12 +10,14 @@ class Linnaeus
     options = {
       persistence_class: Persistence,
       stopwords_class: Stopwords,
-      skip_stemming: false
+      skip_stemming: false,
+      encoding: 'UTF-8'
     }.merge(opts)
 
     @db = options[:persistence_class].new(options)
     @stopword_generator = options[:stopwords_class].new
     @skip_stemming = options[:skip_stemming]
+    @encoding = options[:encoding]
   end
 
   # Count occurences of words in a text corpus.
@@ -25,7 +27,7 @@ class Linnaeus
   #   A string representing a document.  Stopwords are removed and words are stemmed using the "Stemmer" gem.
   def count_word_occurrences(text = '')
     count = {}
-    text.downcase.split.each do |word|
+    text.encode(@encoding).downcase.split.each do |word|
       stemmed_word = (@skip_stemming) ? word : word.stem_porter
       unless stopwords.include? stemmed_word
         count[stemmed_word] = count[stemmed_word] ? count[stemmed_word] + 1 : 1

data/lib/linnaeus/classifier.rb

@@ -1,4 +1,26 @@
 # Classify documents against the Bayesian corpus.
+#
+#  lc = Linnaeus::Classifier.new(<options hash>)
+#  lc.classify 'a string of text' #a wild category appears
+#  lc.classification_scores 'a different string of text' #a hash of categories and scores
+#
+# == Constructor Options
+# persistence_class::
+#   A class implementing persistence - the default (Linnaeus::Persistence) uses redis.
+# stopwords_class::
+#   A class that emits a set of stopwords. The default is Linnaeus::Stopwords
+# skip_stemming::
+#   Set to true to skip porter stemming.
+# encoding::
+#   Force text to use this character set. UTF-8 by default.
+# redis_host::
+#   Passed to persistence class constructor. Defaults to "127.0.0.1"
+# redis_port::
+#   Passed to persistence class constructor. Defaults to "6379".
+# redis_db::
+#   Passed to persistence class constructor. Defaults to "0".
+# redis_*::
+#   Please see Linnaeus::Persistence for the rest of the options that're passed through directly to the Redis client connection.
 class Linnaeus::Classifier < Linnaeus
 
   # Returns a hash of scores for each category in the Bayesian corpus.

data/lib/linnaeus/persistence.rb

@@ -11,14 +11,27 @@ class Linnaeus::Persistence < Linnaeus
     options = {
       redis_host: '127.0.0.1',
       redis_port: '6379',
-      redis_db: 0
+      redis_db: 0,
+      redis_scheme: "redis",
+      redis_path: nil,
+      redis_timeout: 5.0,
+      redis_password: nil,
+      redis_id: nil,
+      redis_tcp_keepalive: 0
     }.merge(opts)
 
     @redis = Redis.new(
       host: options[:redis_host],
       port: options[:redis_port],
-      db: options[:redis_db]
+      db: options[:redis_db],
+      scheme: options[:redis_scheme],
+      path: options[:redis_path],
+      timeout: options[:redis_timeout],
+      password: options[:redis_password],
+      id: options[:redis_id],
+      tcp_keepalive: options[:redis_tcp_keepalive]
     )
+
     self
   end
 
@@ -74,10 +87,8 @@ class Linnaeus::Persistence < Linnaeus
   # word_occurrences::
   #   A hash containing a count of the number of word occurences in a document
   def increment_word_counts_for_category(category, word_occurrences)
-    @redis.multi do |multi|
-      word_occurrences.each do|word,count|
-        multi.hincrby BASE_CATEGORY_KEY + category, word, count
-      end
+    word_occurrences.each do|word,count|
+      @redis.hincrby BASE_CATEGORY_KEY + category, word, count
     end
   end
 
@@ -89,10 +100,8 @@ class Linnaeus::Persistence < Linnaeus
   # word_occurrences::
   #   A hash containing a count of the number of word occurences in a document
   def decrement_word_counts_for_category(category, word_occurrences)
-    @redis.multi do |multi|
-      word_occurrences.each do|word,count|
-        multi.hincrby BASE_CATEGORY_KEY + category, word, - count
-      end
+    word_occurrences.each do|word,count|
+      @redis.hincrby BASE_CATEGORY_KEY + category, word, - count
     end
   end
 

data/lib/linnaeus/trainer.rb

@@ -1,4 +1,27 @@
 # Train or untrain documents from the Bayesian corpus.
+#
+#  lt = Linnaeus::Trainer.new(<options hash>)
+#  lt.train 'category', 'a string of text' 
+#  lt.train 'differentcategory', 'another string of text' 
+#  lt.untrain 'category', 'a document we just removed'
+#
+# == Constructor Options
+# persistence_class::
+#   A class implementing persistence - the default (Linnaeus::Persistence) uses redis.
+# stopwords_class::
+#   A class that emits a set of stopwords. The default is Linnaeus::Stopwords
+# skip_stemming::
+#   Set to true to skip porter stemming.
+# encoding::
+#   Force text to use this character set. UTF-8 by default.
+# redis_host::
+#   Passed to persistence class constructor. Defaults to "127.0.0.1"
+# redis_port::
+#   Passed to persistence class constructor. Defaults to "6379".
+# redis_db::
+#   Passed to persistence class constructor. Defaults to "0".
+# redis_*::
+#   Please see Linnaeus::Persistence for the rest of the options that're passed through directly to the Redis client connection.
 class Linnaeus::Trainer < Linnaeus
 
   # Add a document to the training corpus.

data/linnaeus.gemspec

@@ -0,0 +1,82 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE DIRECTLY
+# Instead, edit Jeweler::Tasks in Rakefile, and run 'rake gemspec'
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = "linnaeus"
+  s.version = "1.0.2"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["djcp"]
+  s.date = "2012-10-31"
+  s.description = "Linnaeus provides a redis-backed Bayesian classifier. Words are stemmed, stopwords are stopped, and redis is used to allow for persistent and concurrent training and classification."
+  s.email = "dan@collispuro.net"
+  s.extra_rdoc_files = [
+    "LICENSE.txt",
+    "README.rdoc"
+  ]
+  s.files = [
+    ".document",
+    ".rspec",
+    ".travis.yml",
+    "Gemfile",
+    "Gemfile.lock",
+    "LICENSE.txt",
+    "README.rdoc",
+    "Rakefile",
+    "VERSION",
+    "images/linnaeus.jpg",
+    "lib/linnaeus.rb",
+    "lib/linnaeus/classifier.rb",
+    "lib/linnaeus/persistence.rb",
+    "lib/linnaeus/stopwords.rb",
+    "lib/linnaeus/trainer.rb",
+    "linnaeus.gemspec",
+    "spec/linnaeus_classifier_spec.rb",
+    "spec/linnaeus_persistence_spec.rb",
+    "spec/linnaeus_spec.rb",
+    "spec/linnaeus_stopwords_spec.rb",
+    "spec/linnaeus_trainer_spec.rb",
+    "spec/spec_helper.rb"
+  ]
+  s.homepage = "http://github.com/djcp/linnaeus"
+  s.licenses = ["MIT"]
+  s.require_paths = ["lib"]
+  s.rubygems_version = "1.8.24"
+  s.summary = "Another redis-backed Bayesian classifier"
+
+  if s.respond_to? :specification_version then
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
+      s.add_runtime_dependency(%q<redis>, ["~> 3.0.0"])
+      s.add_runtime_dependency(%q<stemmer>, ["~> 1.0.0"])
+      s.add_development_dependency(%q<rspec>, ["~> 2.11.0"])
+      s.add_development_dependency(%q<yard>, ["~> 0.7"])
+      s.add_development_dependency(%q<rdoc>, ["~> 3.12"])
+      s.add_development_dependency(%q<bundler>, [">= 0"])
+      s.add_development_dependency(%q<jeweler>, [">= 0"])
+      s.add_development_dependency(%q<simplecov>, [">= 0"])
+    else
+      s.add_dependency(%q<redis>, ["~> 3.0.0"])
+      s.add_dependency(%q<stemmer>, ["~> 1.0.0"])
+      s.add_dependency(%q<rspec>, ["~> 2.11.0"])
+      s.add_dependency(%q<yard>, ["~> 0.7"])
+      s.add_dependency(%q<rdoc>, ["~> 3.12"])
+      s.add_dependency(%q<bundler>, [">= 0"])
+      s.add_dependency(%q<jeweler>, [">= 0"])
+      s.add_dependency(%q<simplecov>, [">= 0"])
+    end
+  else
+    s.add_dependency(%q<redis>, ["~> 3.0.0"])
+    s.add_dependency(%q<stemmer>, ["~> 1.0.0"])
+    s.add_dependency(%q<rspec>, ["~> 2.11.0"])
+    s.add_dependency(%q<yard>, ["~> 0.7"])
+    s.add_dependency(%q<rdoc>, ["~> 3.12"])
+    s.add_dependency(%q<bundler>, [">= 0"])
+    s.add_dependency(%q<jeweler>, [">= 0"])
+    s.add_dependency(%q<simplecov>, [">= 0"])
+  end
+end
+

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: linnaeus
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.0.2
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-10-30 00:00:00.000000000 Z
+date: 2012-10-31 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: redis
@@ -164,6 +164,7 @@ files:
 - lib/linnaeus/persistence.rb
 - lib/linnaeus/stopwords.rb
 - lib/linnaeus/trainer.rb
+- linnaeus.gemspec
 - spec/linnaeus_classifier_spec.rb
 - spec/linnaeus_persistence_spec.rb
 - spec/linnaeus_spec.rb
@@ -185,7 +186,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 494428062127756217
+      hash: 2735128872899197737
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements: