middleman-blog-similar 1.1.1 → 2.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 5ee2d869f0fbef993f19006dbb2a3ffd062aa3e1
-  data.tar.gz: 7051b96407a79365e0959785210a2a3e55b5191a
+  metadata.gz: b4791e37639ec38d3703d84e25feb40184997097
+  data.tar.gz: 11982f6b235436a28ce0addd868d918f90f0f0e8
 SHA512:
-  metadata.gz: 419555674c318f4d018b1c39e9de9012b16f7f6fcb5ee9d95a7aa941fac010dcbc33da4dbbb014161d619cde8c3addfce3155f248472c8c0d7bbc43783eeca61
-  data.tar.gz: 840433955aabe91a21b59701ebe1598d4bec976752ea89e7da40de575d05fb33749e30bec8bc821b70c1901c96c629e8b77d6b8f4f0f04634c414d3f57248fa4
+  metadata.gz: ddef1dfc312319c7907069f11af08ee47dd382df4ac727e24009d3be05b4f60e46b6773a01f3de30707abaad7c401c12596146554662818a4aad4235a7a40bfa
+  data.tar.gz: 61b841719e5338f447912fbf0c804a04f29cead34be5013e66f2a0502e53ccf18f282cdcc931725d77707099e4d59013dcfede5f8a7e1b7e66010f21a3eff115

data/.gitignore

@@ -16,3 +16,4 @@ tmp
 Makefile
 .mm-pid-*
 *.gem
+similar.db

data/.rspec

@@ -0,0 +1,2 @@
+--color
+--format documentation

data/.rubocop.yml

@@ -0,0 +1,33 @@
+AllCops:
+  Exclude:
+    - .gems/**/*
+    - tmp/**/*
+Metrics/LineLength:
+    Enabled: false
+Style/Documentation:
+    Enabled: false
+Rails/HasAndBelongsToMany:
+    Enabled: false
+Metrics/AbcSize:
+    Enabled: false
+MethodLength:
+    Max: 50
+CyclomaticComplexity:
+    Max: 25
+PerceivedComplexity:
+    Max: 20
+Style/RescueModifier:
+    Enabled: false
+Style/PredicateName:
+    Enabled: false
+Metrics/ClassLength:
+    Enabled: false
+GuardClause:
+    MinBodyLength: 3
+CaseIndentation:
+    IndentOneStep: true
+NumericLiterals:
+  MinDigits: 7
+FileName:
+  Exclude:
+    - lib/middleman-blog-similar.rb

data/.travis.yml

@@ -1,25 +1,16 @@
 rvm:
-  - 1.9.3
-  - 2.0.0
-  - 2.1.0
-  - jruby-19mode
-
-# Bug in jRuby w/ Rouge: https://github.com/jruby/jruby/issues/1392
-matrix:
-  allow_failures:
-    - rvm: jruby-19mode
+- 2.2.6
+- 2.3.2
+- 2.4.1
 env:
   global:
-    - TAGGER_DIR=$TRAVIS_BUILD_DIR/tagger
-    - TREETAGGER_COMMAND=$TAGGER_DIR/cmd/tree-tagger-english
-    - TEST=true
+  - TEST=true
 before_install:
-   - sudo apt-get -yV install mecab
-   - sudo apt-get -yV install mecab-ipadic-utf8
-   - mkdir $TAGGER_DIR
-   - curl --progress-bar -o $TAGGER_DIR/tree-tagger-linux-3.2.tar.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/tree-tagger-linux-3.2.tar.gz
-   - curl --progress-bar -o $TAGGER_DIR/tagger-scripts.tar.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/tagger-scripts.tar.gz
-   - curl --progress-bar -o $TAGGER_DIR/install-tagger.sh http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/install-tagger.sh
-   - curl --progress-bar -o $TAGGER_DIR/english-par-linux-3.2.bin.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/english-par-linux-3.2.bin.gz
-   - cd $TAGGER_DIR && sh install-tagger.sh && cd -
-script: "bundle exec rake test"
+- curl -L https://git.io/vSUUX | bash
+script: bundle exec rake test
+deploy:
+  provider: rubygems
+  api_key:
+    secure: PjIBKKWhCcamn3+X/H2bfDn8m0X6nGTEwZw1LpBbWSiXTxJeUlp1uFjwQ82e8yG6Fk3vLlrCZRks3eM3NIzv9V2l1QHQdivLsha/nBkIpnXB4bIRvJG1bhPVmgOSZQCDJjzgUWSlDljzwFjxJ1NOSYQzbPyvOtu+Ks7/p+BJzHI=
+  on:
+    tags: true

data/Gemfile

@@ -1,43 +1,28 @@
 source 'https://rubygems.org'
 
-gem 'middleman-core', '~> 3.2'
-gem 'middleman-blog', '~> 3.5'
-
 # Specify your gem's dependencies in middleman-blog-similar.gemspec
 gemspec
 
-gem 'rake',     '~> 10.1.0', :require => false
-gem 'yard',     '~> 0.8.0', :require => false
-
-# Test tools
-gem 'cucumber', '~> 1.3.1'
-gem 'fivemat'
-gem 'aruba',    '~> 0.5.1'
-gem 'rspec'
-gem 'spork'
-gem 'guard-rake'
+gem 'aruba', '~> 0.7.4', require: false
+gem 'cane', platforms: [:mri_19, :mri_20], require: false
+gem 'capybara', '~> 2.5.0', require: false
+gem 'codeclimate-test-reporter', '~> 1.0', require: false
+gem 'coveralls', require: false
+gem 'cucumber', '~> 2.0', require: false
+gem 'engtagger'
 gem 'guard-cucumber'
+gem 'guard-rake'
 gem 'guard-rspec'
-gem 'guard-spork'
-gem 'slim'
+gem 'guard-rubocop', require: false
+gem 'middleman-cli', '>= 4.0.0'
+gem 'natto', '~> 1.1'
+gem 'pry', '~> 0.10', group: :development, require: false
+gem 'rack', '~> 1.6.5', require: false
+gem 'rake', '~> 10.1.0', require: false
 gem 'redcarpet'
-gem 'sqlite3'
-
-# Code Quality
-gem 'cane', :platforms => [:mri_19, :mri_20], :require => false
-
-# Similarity engines
-gem 'levenshtein-ffi', :require => 'levenshtein'
-gem 'levenshtein'
-gem 'damerau-levenshtein'
-# gem 'gsl'
-# gem 'tf-idf-similarity'
-
-# morphological analyzers
-gem 'engtagger'
-gem 'okura'
-gem 'myaso', '~> 0.4.0.pre2'
-
-if RUBY_VERSION >= '1.9.3'
-  gem "codeclimate-test-reporter", :group => :test, :require => nil
-end
+gem 'rspec', '~> 3.0', require: false
+gem 'rspec-collection_matchers', require: false
+gem 'rspec-its', require: false
+gem 'rubocop', require: false
+gem 'slim'
+gem 'yard', '~> 0.8.0', require: false

data/Guardfile

@@ -1,16 +1,16 @@
-guard 'spork' do
-  watch('Gemfile')
-  watch('Gemfile.lock')
-  watch(/.*\.rb$/)
-end
-
-guard 'rspec' do
+guard 'rspec', cmd: 'bundle exec rspec' do
   watch(%r{^spec/.*\.rb$})
-  watch(%r{^lib/(.+)\.rb$}) {|m| ["spec/#{m[1]}_spec.rb"] + Dir["spec/#{m[1]}/*_spec.rb"] }
+  watch(%r{^lib/(.+)\.rb$}) { |m| ["spec/#{m[1]}_spec.rb"] + Dir["spec/#{m[1]}/*_spec.rb"] }
+  watch(%r{^lib/middleman-blog-similar/tagger/(.+)\.rb$}) { ['spec/middleman-blog-similar/tagger_spec.rb'] }
 end
 
-guard 'cucumber' do
+guard 'cucumber', cmd: 'bundle exec cucumber' do
   watch(%r{^features/.+\.feature$})
   watch(%r{^features/support/.+$}) { 'features' }
   watch(%r{^lib/.*\.rb$})          { 'features' }
 end
+
+guard :rubocop, cmd: 'bundle exec rake rubocop' do
+  watch(/.+\.rb$/)
+  watch(%r{(?:.+/)?\.rubocop\.yml$}) { |m| File.dirname(m[0]) }
+end

data/README.md

@@ -1,9 +1,16 @@
-Middleman-Blog-Similar
+middleman-blog-similar
 ======================
 
+[![Gem Version](https://badge.fury.io/rb/middleman-blog-similar.png)][gem]
+[![Build Status](https://travis-ci.org/ngs/middleman-blog-similar.svg?branch=master)][travis]
+[![Dependency Status](https://gemnasium.com/ngs/middleman-blog-similar.png?travis)][gemnasium]
+[![Code Quality](https://codeclimate.com/github/ngs/middleman-blog-similar.png)][codeclimate]
+[![Coverage Status](https://coveralls.io/repos/github/ngs/middleman-blog-similar/badge.svg)][coveralls]
+
 `middleman-blog-similar` is an extension for [middleman-blog] that adds method to lookup similar article.
 
-### Usage
+Usage
+-----
 
 `Middleman::Blog::BlogArticle#similar_articles` returns an array of `Middleman::Blog::BlogArticle` instances.
 
@@ -30,109 +37,78 @@ Configuration
 
 ```ruby
 gem 'middleman-blog-similar'
-#
-# if you have preferred other algorithm:
-#
-# levenshtein
-#   gem 'levenshtein-ffi', :require => 'levenshtein'
-#
-# damerau levenshtein:
-#   gem 'damerau-levenshtein'
 ```
 
 ### `config.rb`
 
 ```ruby
-# Word frequency sort:
-activate :similar # , :algorithm => :word_frequency by default.
-
-# Use TreeTagger
-activate :similar, :algorithm => :'word_frequency/tree_tagger'
-
-# Use MeCab
-activate :similar, :algorithm => :'word_frequency/mecab'
-
-# Levenshtein distance function:
-activate :similar, :algorithm => :levenshtein
-
-# Damerau–Levenshtein distance function:
-activate :similar, :algorithm => :damerau_levenshtein
+activate :similar
 ```
 
-This library supports [levenshtein-ffi], [levenshtein] and [damerau-levenshtein].
+This extension finds similar articles using those are using tags by default.
 
-## Morphological Analysis
+#### Built-in Tagger
 
-### [MeCab]
+You can set taggers using `tagger:` option. [MeCab] and [EngTagger] adopters are built in this extension.
 
-You need to install `mecab` command in your `PATH`.
-
-#### Mac OS X with Homebrew
+```ruby
+# Find by tags (default)
+activate :similar, tagger: :tags
 
-```bash
-brew install mecab mecab-ipadic
-```
+# Using MeCab / Need to add `gem 'natto'` in Gemfile
+activate :similar, tagger: :mecab
 
-### [TreeTagger]
-
-You need to install [TreeTagger] and export path to the tagger script to `TREETAGGER_COMMAND` environment variable.
-
-```bash
-# Mac OS X
-curl -#o tree-tagger-MacOSX-3.2-intel.tar.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/tree-tagger-MacOSX-3.2-intel.tar.gz
-# or Linux
-curl -#o tree-tagger-linux-3.2.tar.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/tree-tagger-linux-3.2.tar.gz
-# Your language parameter file.
-curl -#o english-par-linux-3.2.bin.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/english-par-linux-3.2.bin.gz
-# Tagger scripts
-curl -#o tagger-scripts.tar.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/tagger-scripts.tar.gz
-# Install script
-curl -#o install-tagger.sh http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/install-tagger.sh
-# Install tagger
-sh install-tagger.sh
-# Append export variable
-echo "export TREETAGGER_COMMAND=\"$(pwd)/cmd/tree-tagger-english\"" >> ~/.bash_profile
+# Using EngTagger / Need to add `gem 'entagger'` in Gemfile
+activate :similar, tagger: :entagger
 ```
 
-## Customizing
+#### Using Lambda
 
-You can use custom algorithm and morphological analyser with implementing modules like this:
+You can use lambda as tagger
 
 ```ruby
-class Middleman::Blog::Similar::Algorithm::MyFastDistance < ::Middleman::Blog::Similar::Algorithm
-  def similar_articles
-    # Do stuff and return scores
-  end
-end
+# Resource is a Middleman::Blog::BlogArticle
+activate :similar, tagger: ->(resource) { [resource.data.category] }
 ```
 
+#### Multiple Taggers
+
+You can configure multiple taggers both built-in and lambda taggers.
+
 ```ruby
-class Middleman::Blog::Similar::Algorithm::WordFrequency::SuperClever < ::Middleman::Blog::Similar::Algorithm
-    def words
-      # Do stuff and return words
-    end
-  end
-end
+activate :similar, tagger: {
+  # key = tagger, value = weight
+  mecab: 1,
+  entagger: 1,
+  tags: 3,
+  # key = (ignored), value[0] = weight, value[1] = lambda
+  custom: [5, ->(resource) { resource.data.category ? [resource.data.category] : [] }]
+}
 ```
 
-then in your `config.rb`
+#### Database Location
+
+This extension uses [SQLite3] to calculate similarity between articles.
+
+Database location is `${PROJECT_ROOT}/.similar.db` by default.
+
+You can specify database location using `db:` option.
 
 ```ruby
-activate :similar, :algorithm => :my_fast_distance
-```
+# Expands to ${HOME}/similar.db
+activate :similar, db: '~/similar.db'
 
-Build & Dependency Status
--------------------------
+# Expands to ${PROJECT_ROOT}/tmp/middleman-blog-similar.db
+activate :similar, db: 'tmp/middleman-blog-similar.db'
 
-[![Gem Version](https://badge.fury.io/rb/middleman-blog-similar.png)][gem]
-[![Build Status](https://travis-ci.org/ngs/middleman-blog-similar.svg?branch=master)][travis]
-[![Dependency Status](https://gemnasium.com/ngs/middleman-blog-similar.png?travis)][gemnasium]
-[![Code Quality](https://codeclimate.com/github/ngs/middleman-blog-similar.png)][codeclimate]
+# Stores in memory database
+activate :similar, db: ':memory:'
+```
 
 License
 -------
 
-Copyright (c) 2014 [Atsushi Nagase]. MIT Licensed, see [LICENSE] for details.
+Copyright (c) 2014-2017 [Atsushi Nagase]. MIT Licensed, see [LICENSE] for details.
 
 [middleman]: http://middlemanapp.com
 [middleman-blog]: https://github.com/middleman/middleman-blog
@@ -140,10 +116,9 @@ Copyright (c) 2014 [Atsushi Nagase]. MIT Licensed, see [LICENSE] for details.
 [travis]: http://travis-ci.org/ngs/middleman-blog-similar
 [gemnasium]: https://gemnasium.com/ngs/middleman-blog-similar
 [codeclimate]: https://codeclimate.com/github/ngs/middleman-blog-similar
-[LICENSE]: https://github.com/ngs/middleman-blog-similar/blob/master/LICENSE.md
-[Atsushi Nagase]: http://ngs.io/
-[levenshtein-ffi]: https://github.com/dbalatero/levenshtein-ffi
-[levenshtein]: https://github.com/schuyler/levenshtein
-[damerau-levenshtein]: https://github.com/GlobalNamesArchitecture/damerau-levenshtein
-[TreeTagger]: http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/
-[MeCab]: http://mecab.googlecode.com/svn/trunk/mecab/doc/index.html
+[LICENSE]: LICENSE.md
+[Atsushi Nagase]: https://ngs.io
+[coveralls]: https://coveralls.io/github/ngs/middleman-blog-similar
+[MeCab]: http://taku910.github.io/mecab/
+[EngTagger]: https://github.com/yohasebe/engtagger
+[SQLite3]: https://www.sqlite.org/

data/Rakefile

@@ -3,33 +3,35 @@ Bundler::GemHelper.install_tasks
 
 require 'cucumber/rake/task'
 require 'rspec/core/rake_task'
+require 'rubocop/rake_task'
+RuboCop::RakeTask.new
 
 RSpec::Core::RakeTask.new(:spec)
 
 Cucumber::Rake::Task.new(:cucumber, 'Run features that should pass') do |t|
-  exempt_tags = ""
-  exempt_tags << "--tags ~@nojava " if RUBY_PLATFORM == "java"
+  exempt_tags = ''
+  exempt_tags << '--tags ~@nojava ' if RUBY_PLATFORM == 'java'
   t.cucumber_opts = "--color --tags ~@wip #{exempt_tags} --strict --format #{ENV['CUCUMBER_FORMAT'] || 'pretty'}"
 end
 
 require 'rake/clean'
 
-task :test => ["spec", "cucumber"]
+task test: %w(rubocop spec cucumber)
 
 begin
   require 'cane/rake_task'
 
-  desc "Run cane to check quality metrics"
+  desc 'Run cane to check quality metrics'
   Cane::RakeTask.new(:quality) do |cane|
     cane.no_style = true
     cane.no_doc = true
-    cane.abc_glob = "lib/middleman-blog-similar/**/*.rb"
+    cane.abc_glob = 'lib/middleman-blog-similar/**/*.rb'
   end
 rescue LoadError
-  # warn "cane not available, quality task not provided."
+  warn 'cane not available, quality task not provided.'
 end
 
-desc "Build HTML documentation"
+desc 'Build HTML documentation'
 task :doc do
   sh 'bundle exec yard'
 end

data/features/default.feature

@@ -0,0 +1,21 @@
+Feature: Default
+
+  Scenario: iterate simlar_articles
+    Given a fixture app "test-app"
+    And a file named "config.rb" with:
+      """
+      activate :blog do|blog|
+        blog.layout = "article"
+      end
+      activate :similar
+      """
+    Given the Server is running at "test-app"
+    When I go to "/2014/05/08/article0.html"
+    Then I should see "<h1>Article 0</h1>"
+    Then I should see '<body data-similar-article-count="4">'
+    Then I should see '<li class="a0"><a href="/2014/05/14/article6.html"><span class="title">Article 6</span><span class="tags">dog, Brown, cat</span></a></li>'
+    Then I should see '<li class="a1"><a href="/2014/05/12/article4.html"><span class="title">Article 4</span><span class="tags">dog, cat, fox</span></a></li>'
+    Then I should see '<li class="a2"><a href="/2014/05/09/article1.html"><span class="title">Article 1</span><span class="tags">dog, cat</span></a></li>'
+    Then I should see '<li class="a3"><a href="/2014/05/13/article5.html"><span class="title">Article 5</span><span class="tags">dog</span></a></li>'
+    When I go to "/page.html"
+    Then I should see '<body data-similar-article-count="0">'

data/features/support/env.rb

@@ -1,14 +1,24 @@
-PROJECT_ROOT_PATH = File.dirname(File.dirname(File.dirname(__FILE__)))
-ENV['TEST'] = 'true'
+require 'codeclimate-test-reporter'
+require 'coveralls'
+require 'simplecov'
 
-require 'rubygems'
-require 'spork'
-$LOAD_PATH.unshift File.join(PROJECT_ROOT_PATH, 'lib')
+SimpleCov.formatter = SimpleCov::Formatter::MultiFormatter[
+  Coveralls::SimpleCov::Formatter,
+  SimpleCov::Formatter::HTMLFormatter,
+  CodeClimate::TestReporter::Formatter
+]
 
-Spork.prefork do
-  require "middleman-blog-similar"
+SimpleCov.start do
+  add_filter '/features/'
 end
 
-require "middleman-core"
-require "middleman-blog"
-require "middleman-core/step_definitions"
+ENV['COVERALLS_REPO_TOKEN'] && Coveralls.wear!
+
+ENV['TEST'] = 'true'
+ENV['AUTOLOAD_SPROCKETS'] = 'false'
+
+PROJECT_ROOT_PATH = File.dirname(File.dirname(File.dirname(__FILE__)))
+require 'middleman-core'
+require 'middleman-core/step_definitions'
+require 'middleman-blog'
+require File.join(PROJECT_ROOT_PATH, 'lib', 'middleman-blog-similar')