slingshot-rb 0.0.6 → 0.0.7

data/.gitignore

@@ -5,3 +5,4 @@ pkg/*
 rdoc/
 coverage/
 scratch/
+examples/*.html

data/Rakefile

@@ -50,3 +50,25 @@ rescue LoadError
     abort "RCov is not available. In order to run rcov, you must: sudo gem install rcov"
   end
 end
+
+namespace :web do
+
+  desc "Update the Github website"
+  task :update => :generate do
+    current_branch = `git branch --no-color`.split("\n").select { |line| line =~ /^\* / }.to_s.gsub(/\* (.*)/, '\1')
+    (puts "Unable to determine current branch"; exit(1) ) unless current_branch
+    system "git stash save && git checkout web"
+    system "cp examples/slingshot-dsl.html index.html"
+    system "git add index.html && git co -m 'Updated Slingshot website'"
+    system "git push origin web:gh-pages -f"
+    system "git checkout #{current_branch} && git stash pop"
+  end
+
+  desc "Generate the Rocco documentation page"
+  task :generate do
+    system "rocco examples/slingshot-dsl.rb"
+    html = File.read('examples/slingshot-dsl.html').gsub!(/slingshot\-dsl\.rb/, 'slingshot.rb')
+    File.open('examples/slingshot-dsl.html', 'w') { |f| f.write html }
+    system "open examples/slingshot-dsl.html"
+  end
+end

data/examples/slingshot-dsl.rb

@@ -0,0 +1,378 @@
+# **Slingshot** is a rich and comfortable Ruby API and DSL for the
+# [_ElasticSearch_](http://www.elasticsearch.org/) search engine/database.
+#
+# _ElasticSearch_ is a scalable, distributed, highly-available,
+# RESTful database communicating by JSON over HTTP, based on [Lucene](http://lucene.apache.org/),
+# written in Java. It manages to be very simple and very powerful at the same time.
+#
+# By following these instructions you should have the search running
+# on a sane operation system in less then 10 minutes.
+
+#### Installation
+
+# Install Slingshot with Rubygems.
+#
+#     gem install slingshot-rb
+#
+require 'rubygems'
+require 'slingshot'
+
+#### Prerequisites
+
+# You'll need a working and running _ElasticSearch_ server. Thankfully, that's easy.
+( puts <<-"INSTALL" ; exit(1) ) unless RestClient.get('http://localhost:9200') rescue false
+ [!] You don’t appear to have ElasticSearch installed. Please install and launch it with the following commands.
+ curl -k -L -o elasticsearch-0.15.0.tar.gz http://github.com/downloads/elasticsearch/elasticsearch/elasticsearch-0.15.0.tar.gz
+ tar -zxvf elasticsearch-0.15.0.tar.gz
+ ./elasticsearch-0.15.0/bin/elasticsearch -f
+INSTALL
+
+### Simple Usage
+
+#### Storing and indexing documents
+
+# Let's initialize an index named “articles”.
+Slingshot.index 'articles' do
+  # To make sure it's fresh, let's delete any existing index with the same name.
+  delete
+  # And then, let's create it.
+  create
+
+  # We want to store and index some articles with title and tags. Simple Hashes are OK.
+  store :title => 'One',   :tags => ['ruby'],           :published_on => '2011-01-01'
+  store :title => 'Two',   :tags => ['ruby', 'python'], :published_on => '2011-01-02'
+  store :title => 'Three', :tags => ['java'],           :published_on => '2011-01-02'
+  store :title => 'Four',  :tags => ['ruby', 'php'],    :published_on => '2011-01-03'
+
+  # We force refreshing the index, so we can query it immediately.
+  refresh
+end
+
+# We may want to define a specific [mapping](http://www.elasticsearch.org/guide/reference/api/admin-indices-create-index.html)
+# for the index.
+
+Slingshot.index 'articles' do
+  # To do so, just pass a Hash containing the specified mapping to the `Index#create` method.
+  create :mappings => {
+    # Specify for which type of documents this mapping should be used (`article` in this case).
+    :article => {
+      :properties => {
+        # Specify the type of the field, whether it should be analyzed, etc.
+        :id       => { :type => 'string', :index => 'not_analyzed', :include_in_all => false },
+        # Set the boost or analyzer settings for the field, et cetera. The _ElasticSearch_ guide
+        # has [more information](http://elasticsearch.org/guide/reference/mapping/index.html).
+        :title    => { :type => 'string', :boost => 2.0,            :analyzer => 'snowball'  },
+        :tags     => { :type => 'string', :analyzer => 'keyword'                             },
+        :content  => { :type => 'string', :analyzer => 'snowball'                            }
+      }
+    }
+  }
+end
+
+
+
+#### Searching
+
+# With the documents indexed and stored in the _ElasticSearch_ database, we want to search for them.
+#
+# Slingshot exposes the search interface via simple domain-specific language.
+
+
+##### Simple Query String Searches
+
+# We can do simple searches, like searching for articles containing “One” in their title.
+s = Slingshot.search('articles') do
+  query do
+    string "title:One"
+  end
+end
+
+# The results:
+#     * One [tags: ruby]
+s.results.each do |document|
+  puts "* #{ document.title } [tags: #{document.tags.join(', ')}]"
+end
+
+# Of course, we may write the blocks in shorter notation.
+
+# Let's search for articles whose titles begin with letter “T”.
+s = Slingshot.search('articles') { query { string "title:T*" } }
+
+# The results:
+#     * Two [tags: ruby, python]
+#     * Three [tags: java]
+s.results.each do |document|
+  puts "* #{ document.title } [tags: #{document.tags.join(', ')}]"
+end
+
+# We can use any valid [Lucene query syntax](http://lucene.apache.org/java/3_0_3/queryparsersyntax.html)
+# for the query string queries.
+
+# For debugging, we can display the JSON which is being sent to _ElasticSearch_.
+#
+#     {"query":{"query_string":{"query":"title:T*"}}}
+#
+puts "", "Query:", "-"*80
+puts s.to_json
+
+# Or better, we may display a complete `curl` command, so we can execute it in terminal
+# to see the raw output, tweak params and debug any problems.
+#
+#     curl -X POST "http://localhost:9200/articles/_search?pretty=true" \
+#          -d '{"query":{"query_string":{"query":"title:T*"}}}'
+#
+puts "", "Try the query in Curl:", "-"*80
+puts s.to_curl
+
+
+##### Other Types of Queries
+
+# Of course, we may want to define our queries more expressively, for instance
+# when we're searching for articles with specific _tags_.
+
+# Let's suppose we want to search for articles tagged “ruby” _or_ “python”.
+# That's a great excuse to use a [_terms_](http://elasticsearch.org/guide/reference/query-dsl/terms-query.html)
+# query.
+s = Slingshot.search('articles') do
+  query do
+    terms :tags, ['ruby', 'python']
+  end
+end
+
+# The search, as expected, returns three articles, all tagged “ruby” — among other tags:
+#
+#     * Two [tags: ruby, python]
+#     * One [tags: ruby]
+#     * Four [tags: ruby, php]
+s.results.each do |document|
+  puts "* #{ document.title } [tags: #{document.tags.join(', ')}]"
+end
+
+# What if we wanted to search for articles tagged both “ruby” _and_ “python”.
+# That's a great excuse to specify `minimum_match` for the query.
+s = Slingshot.search('articles') do
+  query do
+    terms :tags, ['ruby', 'python'], :minimum_match => 2
+  end
+end
+
+# The search, as expected, returns one article, tagged with _both_ “ruby” and “python”:
+#
+#     * Two [tags: ruby, python]
+s.results.each do |document|
+  puts "* #{ document.title } [tags: #{document.tags.join(', ')}]"
+end
+
+# _ElasticSearch_ allows us to do many more types of queries.
+# Eventually, _Slingshot_ will support all of them.
+# So far, only these are supported:
+#
+# * [term](http://elasticsearch.org/guide/reference/query-dsl/term-query.html)
+# * [terms](http://elasticsearch.org/guide/reference/query-dsl/terms-query.html)
+
+##### Faceted Search
+
+# _ElasticSearch_ makes it trivial to retrieve complex aggregated data from our index/database,
+# so called [_facets_](http://www.lucidimagination.com/Community/Hear-from-the-Experts/Articles/Faceted-Search-Solr).
+
+# Let's say we want to display article counts for every tag in the database.
+# For that, we'll use a _terms_ facet.
+
+#
+s = Slingshot.search 'articles' do
+  # We will search for articles whose title begins with letter “T”,
+  query { string 'title:T*' }
+
+  # and retrieve their counts “bucketed” by their `tags`.
+  facet 'tags' do
+    terms :tags
+  end
+end
+
+# As we see, our query has found two articles, and if you recall our articles from above,
+# _Two_ is tagged with “ruby” and “python”, _Three_ is tagged with “java”. So the counts
+# won't surprise us:
+#     Found 2 articles: Three, Two
+#     Counts:
+#     -------
+#     ruby       1
+#     python     1
+#     java       1
+puts "Found #{s.results.count} articles: #{s.results.map(&:title).join(', ')}"
+puts "Counts based on tags:", "-"*25
+s.results.facets['tags']['terms'].each do |f|
+  puts "#{f['term'].ljust(10)} #{f['count']}"
+end
+
+# These counts are based on the scope of our current query (called `main` in _ElasticSearch_).
+# What if we wanted to display aggregated counts by `tags` across the whole database?
+
+#
+s = Slingshot.search 'articles' do
+  query { string 'title:T*' }
+
+  facet 'global-tags' do
+    # That's where the `global` scope for a facet comes in.
+    terms :tags, :global => true
+  end
+
+  # As you can see, we can even combine facets scoped
+  # to the current query with global facets.
+  facet 'current-tags' do
+    terms :tags
+  end
+end
+
+# Aggregated results for the current query are the same as previously:
+#     Current query facets:
+#     -------------------------
+#     ruby       1
+#     python     1
+#     java       1
+puts "Current query facets:", "-"*25
+s.results.facets['current-tags']['terms'].each do |f|
+  puts "#{f['term'].ljust(10)} #{f['count']}"
+end
+
+# As we see, aggregated results for the global scope include also
+# tags for articles not matched by the query, such as “java” or “php”:
+#     Global facets:
+#     -------------------------
+#     ruby       3
+#     python     1
+#     php        1
+#     java       1
+puts "Global facets:", "-"*25
+s.results.facets['global-tags']['terms'].each do |f|
+  puts "#{f['term'].ljust(10)} #{f['count']}"
+end
+
+# The real power of facets lies in their combination with
+# [filters](http://elasticsearch.karmi.cz/guide/reference/api/search/filter.html),
+# though:
+
+# > When doing things like facet navigation,
+# > sometimes only the hits are needed to be filtered by the chosen facet,
+# > and all the facets should continue to be calculated based on the original query.
+
+
+##### Filtered Search
+
+# So, let's make out search a bit complex. Let's search for articles whose titles begin
+# with letter “T”, again, but filter the results, so only the articles tagged “ruby”
+# are returned.
+s = Slingshot.search 'articles' do
+  
+  # We use the same **query** as before.
+  query { string 'title:T*' } 
+
+  # And add a _terms_ **filter** based on tags.
+  filter :terms, :tags => ['ruby']
+
+  # And, of course, our facet definition.
+  facet('tags') { terms :tags }
+
+end
+
+# We see that only the article _Two_ (tagged “ruby” and “python”) was returned,
+# _not_ the article _Three_ (tagged “java”):
+#
+#     * Two [tags: ruby, python]
+s.results.each do |document|
+  puts "* #{ document.title } [tags: #{document.tags.join(', ')}]"
+end
+
+# However, count for article _Three_'s tags, “java”, _is_ in fact included in facets:
+#
+#     Counts based on tags:
+#     -------------------------
+#     ruby       1
+#     python     1
+#     java       1
+puts "Counts based on tags:", "-"*25
+s.results.facets['tags']['terms'].each do |f|
+  puts "#{f['term'].ljust(10)} #{f['count']}"
+end
+
+
+##### Sorting
+
+# By default, the results are sorted according to their relevancy
+# (available as the `_score` property).
+
+# But, what if we want to sort the results based on some other criteria,
+# such as published date, price, etc? We can do that.
+s = Slingshot.search 'articles' do
+  # We search for articles tagged “ruby”
+  query { string 'tags:ruby' } 
+
+   # And sort them by their `title`, in descending order.
+  sort { title 'desc' }
+end
+
+# The results:
+#     * Two
+#     * One
+#     * Four
+s.results.each do |document|
+  puts "* #{ document.title }"
+end
+
+# Of course, it's possible to combine more fields in the sorting definition.
+
+s = Slingshot.search 'articles' do
+  # We will just get all articles for this case.
+  query { string '*' } 
+
+  sort do
+    # We will sort the results by their `published_on` property in ascending (default) order,
+    published_on
+    # and by their `title` property, in descending order.
+    title 'desc'
+  end
+end
+
+# The results:
+#     * One         (Published on: 2011-01-01)
+#     * Two         (Published on: 2011-01-02)
+#     * Three       (Published on: 2011-01-02)
+#     * Four        (Published on: 2011-01-03)
+s.results.each do |document|
+  puts "* #{ document.title.ljust(10) }  (Published on: #{ document.published_on })"
+end
+
+##### Highlighting
+
+# Often, you want to highlight the matched snippets in your text.
+# _ElasticSearch_ provides many features for
+# [highlighting](http://www.elasticsearch.org/guide/reference/api/search/highlighting.html),
+# 
+s = Slingshot.search 'articles' do
+  # Let's search for documents containing “Two” in their titles.
+  query { string 'title:Two' } 
+
+   # And use the `highlight` method.
+  highlight :title
+end
+
+# The results:
+#     Title: Two, highlighted title: <em>Two</em>
+s.results.each do |document|
+  puts "Title: #{ document.title }, highlighted title: #{document.highlight.title}"
+end
+
+# Slingshot allows you to specify options for the highlighting, such as:
+#
+s = Slingshot.search 'articles' do
+  query { string 'title:Two' }
+
+  # • specifying the fields to highlight
+  highlight :title, :body
+
+  # • specifying their options
+  highlight :title, :body => { :number_of_fragments => 0 }
+
+  # • or specifying highlighting options, such as the wrapper tag
+  highlight :title, :body, :options => { :tag => '<strong class="highlight">' }
+end
+

data/lib/slingshot.rb

@@ -2,6 +2,7 @@ require 'rest_client'
 require 'yajl/json_gem'
 
 require 'slingshot/rubyext/hash'
+require 'slingshot/logger'
 require 'slingshot/configuration'
 require 'slingshot/client'
 require 'slingshot/client'
@@ -10,6 +11,7 @@ require 'slingshot/search/query'
 require 'slingshot/search/sort'
 require 'slingshot/search/facet'
 require 'slingshot/search/filter'
+require 'slingshot/search/highlight'
 require 'slingshot/results/collection'
 require 'slingshot/results/item'
 require 'slingshot/index'

data/lib/slingshot/configuration.rb

@@ -14,6 +14,11 @@ module Slingshot
       @wrapper = klass || @wrapper || Results::Item
     end
 
+    def self.logger(device=nil, options={})
+      return @logger = Logger.new(device, options) if device
+      @logger || nil
+    end
+
     def self.reset(*properties)
       reset_variables = properties.empty? ? instance_variables : instance_variables & properties.map { |p| "@#{p}" }
       reset_variables.each { |v| instance_variable_set(v, nil) }

data/lib/slingshot/logger.rb

@@ -0,0 +1,61 @@
+module Slingshot
+  class Logger
+
+    def initialize(device, options={})
+      @device = if device.respond_to?(:write)
+        device
+      else
+        File.open(device, 'a')
+      end
+      @device.sync = true
+      @options = options
+      at_exit { @device.close unless @device.closed? }
+    end
+
+    def level
+      @options[:level] || 'info'
+    end
+
+    def write(message)
+      @device.write message
+    end
+
+    def log_request(endpoint, params=nil, curl='')
+      # [_search] (articles,users) 2001-02-12 18:20:42:32
+      #
+      # curl -X POST ....
+      #
+      content  = "# [#{endpoint}] "
+      content += "(#{params.inspect}) " if params
+      content += time
+      content += "\n#\n"
+      content += curl
+      content += "\n\n"
+      write content
+    end
+
+    def log_response(status, json)
+      # [200 OK] (4 msec) Sat Feb 12 19:20:47 2011
+      #
+      # {
+      #   "took" : 4,
+      #   "hits" : [...]
+      #   ...
+      # }
+      #
+      took = JSON.parse(json)['took'] rescue nil
+      content  = "# [#{status}] "
+      content += "(#{took} msec) " if took
+      content += time
+      content += "\n#\n"
+      json.each_line { |line| content += "# #{line}" }
+      content += "\n\n"
+      write content
+    end
+
+    def time
+      Time.now.strftime('%Y-%m-%d %H:%M:%S:%L')
+    end
+
+  end
+end

data/lib/slingshot/results/collection.rb

@@ -13,6 +13,7 @@ module Slingshot
                        h
                      else
                        document = h['_source'] ? h['_source'] : h['fields']
+                       document['highlight'] = h['highlight'] if h['highlight']
                        h.update document if document
                        Configuration.wrapper.new(h)
                      end

data/lib/slingshot/search.rb

@@ -3,7 +3,7 @@ module Slingshot
   
     class Search
 
-      attr_reader :indices, :url, :results, :response, :query, :facets, :filters
+      attr_reader :indices, :url, :results, :response, :json, :query, :facets, :filters
 
       def initialize(*indices, &block)
         @options = indices.pop if indices.last.is_a?(Hash)
@@ -37,6 +37,15 @@ module Slingshot
         self
       end
 
+      def highlight(*args)
+        unless args.empty?
+          @highlight = Highlight.new(*args)
+          self
+        else
+          @highlight
+        end
+      end
+
       def from(value)
         @from = value
         self
@@ -53,17 +62,33 @@ module Slingshot
       end
 
       def perform
-        @url     = "#{Configuration.url}/#{indices.join(',')}/_search"
-        @response = JSON.parse( Configuration.client.post(@url, self.to_json) )
-        @results = Results::Collection.new(@response)
+        @url      = "#{Configuration.url}/#{indices.join(',')}/_search"
+        @response = Configuration.client.post(@url, self.to_json)
+        @json     = Yajl::Parser.parse(@response)
+        @results  = Results::Collection.new(@json)
         self
-      rescue Exception
-        STDERR.puts "Request failed: \n#{self.to_curl}"
+      rescue Exception => e
+        STDERR.puts "[REQUEST FAILED]\n#{self.to_curl}\n"
         raise
+      ensure
+        if Configuration.logger
+          Configuration.logger.log_request '_search', indices, to_curl
+          if Configuration.logger.level == 'debug'
+            # FIXME: Depends on RestClient implementation
+            if @response
+              code = @response.code
+              body = Yajl::Encoder.encode(@json, :pretty => true)
+            else
+              code = e.message
+              body = e.http_body
+            end
+            Configuration.logger.log_response code, body
+          end
+        end
       end
 
       def to_curl
-        %Q|curl -X POST "http://localhost:9200/#{indices}/_search?pretty=true" -d '#{self.to_json}'|
+        %Q|curl -X POST "#{Configuration.url}/#{indices}/_search?pretty=true" -d '#{self.to_json}'|
       end
 
       def to_json
@@ -72,6 +97,7 @@ module Slingshot
         request.update( { :sort   => @sort } )     if @sort
         request.update( { :facets => @facets } )   if @facets
         @filters.each { |filter| request.update( { :filter => filter } ) } if @filters
+        request.update( { :highlight => @highlight } ) if @highlight
         request.update( { :size => @size } )       if @size
         request.update( { :from => @from } )       if @from
         request.update( { :fields => @fields } )   if @fields

data/lib/slingshot/search/facet.rb

@@ -3,7 +3,7 @@ module Slingshot
 
     #--
     # TODO: Implement all elastic search facets (geo, histogram, range, etc)
-    # https://github.com/elasticsearch/elasticsearch/wiki/Search-API-Facets
+    # http://elasticsearch.org/guide/reference/api/search/facets/
     #++
 
     class Facet
@@ -14,8 +14,13 @@ module Slingshot
         self.instance_eval(&block) if block_given?
       end
 
-      def terms(field, options={})
-        @value = { :terms => { :field => field } }.update(options)
+      def terms(field, size=10, options={})
+        @value = { :terms => { :field => field, :size => size } }.update(options)
+        self
+      end
+
+      def date(field, interval='day', options={})
+        @value = { :date_histogram => { :field => field, :interval => interval } }.update(options)
         self
       end
 

data/lib/slingshot/search/highlight.rb

@@ -0,0 +1,39 @@
+module Slingshot
+  module Search
+
+    # http://www.elasticsearch.org/guide/reference/api/search/highlighting.html
+    #
+    class Highlight
+
+      def initialize(*args)
+        @options  = (args.last.is_a?(Hash) && args.last.delete(:options)) || {}
+        extract_highlight_tags
+        @fields   = args.inject({}) do |result, field|
+          field.is_a?(Hash) ? result.update(field) : result[field.to_sym] = {}; result
+        end
+      end
+
+      def to_json
+        to_hash.to_json
+      end
+
+      def to_hash
+        h = { :fields => @fields }
+        h.update @options
+        return h
+      end
+
+      private
+
+      def extract_highlight_tags
+        if tag = @options.delete(:tag)
+          @options.update \
+            :pre_tags  => [tag],
+            :post_tags => [tag.to_s.gsub(/^<([a-z]+).*/, '</\1>')]
+        end
+      end
+
+    end
+
+  end
+end

data/lib/slingshot/search/query.rb

@@ -23,6 +23,11 @@ module Slingshot
         @value
       end
 
+      def all
+        @value = { :match_all => {} }
+        @value
+      end
+
       def to_json
         @value.to_json
       end

data/lib/slingshot/version.rb

@@ -1,3 +1,3 @@
 module Slingshot
-  VERSION = "0.0.6"
+  VERSION = "0.0.7"
 end

data/test/fixtures/articles/1.json

@@ -1 +1 @@
-{"title" : "One", "tags" : ["ruby"]}
+{"title" : "One", "tags" : ["ruby"], "published_on" : "2011-01-01"}

data/test/fixtures/articles/2.json

@@ -1 +1 @@
-{"title" : "Two", "tags" : ["ruby", "python"]}
+{"title" : "Two", "tags" : ["ruby", "python"], "published_on" : "2011-01-02"}

data/test/fixtures/articles/3.json

@@ -1 +1 @@
-{"title" : "Three", "tags" : ["java"]}
+{"title" : "Three", "tags" : ["java"], "published_on" : "2011-01-02"}

data/test/fixtures/articles/4.json

@@ -1 +1 @@
-{"title" : "Four", "tags" : ["erlang"]}
+{"title" : "Four", "tags" : ["erlang"], "published_on" : "2011-01-03"}

data/test/fixtures/articles/5.json

@@ -1 +1 @@
-{"title" : "Five", "tags" : ["javascript", "java"]}
+{"title" : "Five", "tags" : ["javascript", "java"], "published_on" : "2011-01-04"}

data/test/integration/facets_test.rb

@@ -1,4 +1,5 @@
 require 'test_helper'
+require 'date'
 
 module Slingshot
 
@@ -40,6 +41,23 @@ module Slingshot
         assert_equal 5, global_facets.count
       end
 
+      context "date histogram" do
+
+        should "return aggregated values for all results" do
+          s = Slingshot.search('articles-test') do
+            query { all }
+            facet 'published_on' do
+              date :published_on
+            end
+          end
+
+          facets = s.results.facets['published_on']['entries']
+          assert_equal 4, facets.size, facets.inspect
+          assert_equal 2, facets.entries[1]["count"], facets.inspect
+        end
+
+      end
+
     end
 
   end

data/test/integration/highlight_test.rb

@@ -0,0 +1,26 @@
+require 'test_helper'
+
+module Slingshot
+
+  class HighlightIntegrationTest < Test::Unit::TestCase
+    include Test::Integration
+
+    context "Highlight" do 
+
+      should "add 'highlight' field to the result item" do
+        Slingshot::Configuration.logger STDERR, :level => 'debug'
+        s = Slingshot.search('articles-test') do
+          # query { string '-w' }
+          highlight :title
+        end
+
+        doc = s.results.first
+
+        assert_equal 1, doc.highlight.title.size
+        assert doc.highlight.title.to_s.include?('<em>'), "Highlight does not include default highlight tag"
+      end
+
+    end
+
+  end
+end

data/test/unit/configuration_test.rb

@@ -10,6 +10,10 @@ module Slingshot
         Configuration.instance_variable_set(:@client, nil)
       end
 
+      teardown do
+        Configuration.reset
+      end
+
       should "return default URL" do
         assert_equal 'http://localhost:9200', Configuration.url
       end
@@ -28,6 +32,16 @@ module Slingshot
         assert_equal Client::Base, Configuration.client
       end
 
+      should "return nil as logger by default" do
+        assert_nil Configuration.logger
+      end
+
+      should "return set and return logger" do
+        Configuration.logger STDERR
+        assert_not_nil Configuration.logger
+        assert_instance_of Slingshot::Logger, Configuration.logger
+      end
+
       should "allow to reset the configuration for specific property" do
         Configuration.url 'http://example.com'
         Configuration.client Client::Base

data/test/unit/logger_test.rb

@@ -0,0 +1,114 @@
+require 'test_helper'
+require 'time'
+
+module Slingshot
+
+  class LoggerTest < Test::Unit::TestCase
+    include Slingshot
+
+    context "Logger" do
+
+      context "initialized with an IO object" do
+
+        should "take STDOUT" do
+          assert_nothing_raised do
+            logger = Logger.new STDOUT
+          end
+        end
+
+        should "write to STDERR" do
+          STDERR.expects(:write).with('BOOM!')
+          logger = Logger.new STDERR
+          logger.write('BOOM!')
+        end
+
+      end
+
+      context "initialized with file" do
+        teardown { File.delete('myfile.log') }
+
+        should "create the file" do
+          assert_nothing_raised do
+            logger = Logger.new 'myfile.log'
+            assert File.exists?('myfile.log')
+          end
+        end
+
+        should "write to file" do
+          File.any_instance.expects(:write).with('BOOM!')
+          logger = Logger.new 'myfile.log'
+          logger.write('BOOM!')
+        end
+
+      end
+
+    end
+
+    context "levels" do
+
+      should "have default level" do
+        logger = Logger.new STDERR
+        assert_equal 'info', logger.level
+      end
+
+      should "have set the level" do
+        logger = Logger.new STDERR, :level => 'debug'
+        assert_equal 'debug', logger.level
+      end
+
+    end
+
+    context "tracing requests" do
+      setup do
+        Time.stubs(:now).returns(Time.parse('2011-03-19 11:00'))
+        @logger = Logger.new STDERR
+      end
+
+      should "log request in correct format" do
+        log = (<<-"log;").gsub(/^ +/, '')
+          # [_search] (["articles", "users"]) 2011-03-19 11:00:00:L
+          #
+          curl -X GET http://...
+
+        log;
+        @logger.expects(:write).with(log)
+        @logger.log_request('_search', ["articles", "users"], 'curl -X GET http://...')
+      end
+
+      should "log response in correct format" do
+        json = (<<-"json;").gsub(/^\s*/, '')
+        {
+          "took" : 4,
+          "hits" : {
+            "total" : 20,
+            "max_score" : 1.0,
+            "hits" : [ {
+              "_index" : "articles",
+              "_type" : "article",
+              "_id" : "Hmg0B0VSRKm2VAlsasdnqg",
+              "_score" : 1.0, "_source" : { "title" : "Article 1",  "published_on" : "2011-01-01" }
+            }, {
+              "_index" : "articles",
+              "_type" : "article",
+              "_id" : "booSWC8eRly2I06GTUilNA",
+              "_score" : 1.0, "_source" : { "title" : "Article 2", "published_on" : "2011-01-12" }
+            }
+            ]
+          }
+        }
+        json;
+        log  = (<<-"log;").gsub(/^\s*/, '')
+          # [200 OK] (4 msec) 2011-03-19 11:00:00:L
+          #
+        log;
+        # log += json.split.map { |line| "# #{line}" }.join("\n")
+        json.each_line { |line| log += "# #{line}" }
+        log += "\n\n"
+        @logger.expects(:write).with(log)
+        @logger.log_response('200 OK', json)
+      end
+
+    end
+
+  end
+end

data/test/unit/search_facet_test.rb

@@ -13,26 +13,33 @@ module Slingshot::Search
       context "generally" do
 
         should "encode facets with defaults for current query" do
-          assert_equal( { :foo => { :terms => {:field=>'bar'} } }.to_json, Facet.new('foo').terms(:bar).to_json )
+          assert_equal( { :foo => { :terms => {:field=>'bar',:size=>10} } }.to_json, Facet.new('foo').terms(:bar).to_json )
         end
 
         should "encode facets as global" do
-          assert_equal( { :foo => { :terms => {:field=>'bar'}, :global => true } }.to_json,
+          assert_equal( { :foo => { :terms => {:field=>'bar',:size=>10}, :global => true } }.to_json,
                         Facet.new('foo', :global => true).terms(:bar).to_json )
         end
 
         should "encode facet options" do
-          assert_equal( { :foo => { :terms => {:field=>'bar'}, :size => 5 } }.to_json,
-                        Facet.new('foo', :size => 5).terms(:bar).to_json )
-          assert_equal( { :foo => { :terms => {:field=>'bar'}, :size => 5 } }.to_json,
-                        Facet.new('foo').terms(:bar, :size => 5).to_json )
+          assert_equal( { :foo => { :terms => {:field=>'bar',:size=>5} } }.to_json,
+                        Facet.new('foo').terms(:bar, 5).to_json )
         end
 
         should "encode facets when passed as a block" do
           f = Facet.new('foo') do
             terms :bar
           end
-          assert_equal( { :foo => { :terms => {:field=>'bar'} } }.to_json, f.to_json )
+          assert_equal( { :foo => { :terms => {:field=>'bar',:size=>10} } }.to_json, f.to_json )
+        end
+
+      end
+
+      context "date histogram" do
+
+        should "encode the JSON" do
+          f = Facet.new('date') { date :published_on, 'day' }
+          assert_equal({ :date => { :date_histogram => { :field => 'published_on', :interval => 'day' } } }.to_json, f.to_json)
         end
 
       end

data/test/unit/search_highlight_test.rb

@@ -0,0 +1,46 @@
+require 'test_helper'
+
+module Slingshot::Search
+
+  class HighlightTest < Test::Unit::TestCase
+
+    context "Highlight" do
+
+      should "be serialized to JSON" do
+        assert_respond_to Highlight.new(:body), :to_json
+      end
+
+      should "specify highlight for single field" do
+        assert_equal( {:fields => { :body => {} }}.to_json,
+                      Highlight.new(:body).to_json )
+      end
+
+      should "specify highlight for more fields" do
+        assert_equal( {:fields => { :title => {}, :body => {} }}.to_json,
+                      Highlight.new(:title, :body).to_json )
+      end
+
+      should "specify highlight for more fields with options" do
+        assert_equal( {:fields => { :title => {}, :body => { :a => 1, :b => 2 } }}.to_json,
+                      Highlight.new(:title, :body => { :a => 1, :b => 2 }).to_json )
+      end
+
+      should "specify highlight for more fields with highlight options" do
+        # p Highlight.new(:title, :body => {}, :options => { :tag => '<strong>' }).to_hash
+        assert_equal( {:fields => { :title => {}, :body => {} }, :pre_tags => ['<strong>'], :post_tags => ['</strong>'] }.to_json,
+                      Highlight.new(:title, :body => {}, :options => { :tag => '<strong>' }).to_json )
+      end
+    
+      context "with custom tags" do
+
+        should "properly parse tags with class" do
+          assert_equal( {:fields => { :title => {} }, :pre_tags => ['<strong class="highlight">'], :post_tags => ['</strong>'] }.to_json,
+                        Highlight.new(:title, :options => { :tag => '<strong class="highlight">' }).to_json )
+        end
+
+      end
+
+    end
+
+  end
+end

data/test/unit/search_query_test.rb

@@ -33,6 +33,10 @@ module Slingshot::Search
                       Query.new.string('foo', :default_field => 'title') )
       end
 
+      should "search for all documents" do
+        assert_equal( { :match_all => { } }, Query.new.all )
+      end
+
     end
 
   end

data/test/unit/search_test.rb

@@ -5,6 +5,7 @@ module Slingshot
   class SearchTest < Test::Unit::TestCase
 
     context "Search" do
+      setup { Configuration.reset :logger }
 
       should "be initialized with index/indices" do
         assert_raise(ArgumentError) { Search::Search.new }
@@ -54,6 +55,17 @@ module Slingshot
         assert_raise(RestClient::InternalServerError) { s.perform }
       end
 
+      should "log request, but not response, when logger is set" do
+        Configuration.logger STDERR
+
+        Configuration.client.expects(:post).returns('{"hits":[]}')
+        Results::Collection.expects(:new).returns([])
+        Configuration.logger.expects(:log_request).returns(true)
+        Configuration.logger.expects(:log_response).never
+
+        Search::Search.new('index').perform
+      end
+
       context "sort" do
 
         should "allow sorting by multiple fields" do
@@ -71,7 +83,7 @@ module Slingshot
 
       context "facets" do
 
-        should "allow searching for facets" do
+        should "retrieve terms facets" do
           s = Search::Search.new('index') do
             facet('foo1') { terms :bar, :global => true }
             facet('foo2', :global => true) { terms :bar }
@@ -83,6 +95,14 @@ module Slingshot
           assert_not_nil s.facets['foo3']
         end
 
+        should "retrieve date histogram facets" do
+          s = Search::Search.new('index') do
+            facet('date') { date :published_on }
+          end
+          assert_equal 1, s.facets.keys.size
+          assert_not_nil  s.facets['date']
+        end
+
       end
 
       context "filter" do
@@ -99,6 +119,37 @@ module Slingshot
 
       end
 
+      context "highlight" do
+
+        should "allow to specify highlight for single field" do
+          s = Search::Search.new('index') do
+            highlight :body
+          end
+
+          assert_not_nil s.highlight
+          assert_instance_of Slingshot::Search::Highlight, s.highlight
+        end
+
+        should "allow to specify highlight for more fields" do
+          s = Search::Search.new('index') do
+            highlight :body, :title
+          end
+
+          assert_not_nil s.highlight
+          assert_instance_of Slingshot::Search::Highlight, s.highlight
+        end
+
+        should "allow to specify highlight with for more fields with options" do
+          s = Search::Search.new('index') do
+            highlight :body, :title => { :fragment_size => 150, :number_of_fragments => 3 }
+          end
+
+          assert_not_nil s.highlight
+          assert_instance_of Slingshot::Search::Highlight, s.highlight
+        end
+
+      end
+
       context "with from/size" do
 
         should "set the values in request" do

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: slingshot-rb
 version: !ruby/object:Gem::Version 
-  hash: 19
+  hash: 17
   prerelease: 
   segments: 
   - 0
   - 0
-  - 6
-  version: 0.0.6
+  - 7
+  version: 0.0.7
 platform: ruby
 authors: 
 - Karel Minarik
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-03-13 00:00:00 +01:00
+date: 2011-04-02 00:00:00 +02:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -152,18 +152,21 @@ files:
 - README.markdown
 - Rakefile
 - examples/dsl.rb
+- examples/slingshot-dsl.rb
 - lib/slingshot-rb.rb
 - lib/slingshot.rb
 - lib/slingshot/client.rb
 - lib/slingshot/configuration.rb
 - lib/slingshot/dsl.rb
 - lib/slingshot/index.rb
+- lib/slingshot/logger.rb
 - lib/slingshot/results/collection.rb
 - lib/slingshot/results/item.rb
 - lib/slingshot/rubyext/hash.rb
 - lib/slingshot/search.rb
 - lib/slingshot/search/facet.rb
 - lib/slingshot/search/filter.rb
+- lib/slingshot/search/highlight.rb
 - lib/slingshot/search/query.rb
 - lib/slingshot/search/sort.rb
 - lib/slingshot/version.rb
@@ -176,6 +179,7 @@ files:
 - test/fixtures/articles/5.json
 - test/integration/facets_test.rb
 - test/integration/filters_test.rb
+- test/integration/highlight_test.rb
 - test/integration/index_mapping_test.rb
 - test/integration/index_store_test.rb
 - test/integration/query_string_test.rb
@@ -186,10 +190,12 @@ files:
 - test/unit/client_test.rb
 - test/unit/configuration_test.rb
 - test/unit/index_test.rb
+- test/unit/logger_test.rb
 - test/unit/results_collection_test.rb
 - test/unit/results_item_test.rb
 - test/unit/search_facet_test.rb
 - test/unit/search_filter_test.rb
+- test/unit/search_highlight_test.rb
 - test/unit/search_query_test.rb
 - test/unit/search_sort_test.rb
 - test/unit/search_test.rb
@@ -238,6 +244,7 @@ test_files:
 - test/fixtures/articles/5.json
 - test/integration/facets_test.rb
 - test/integration/filters_test.rb
+- test/integration/highlight_test.rb
 - test/integration/index_mapping_test.rb
 - test/integration/index_store_test.rb
 - test/integration/query_string_test.rb
@@ -248,10 +255,12 @@ test_files:
 - test/unit/client_test.rb
 - test/unit/configuration_test.rb
 - test/unit/index_test.rb
+- test/unit/logger_test.rb
 - test/unit/results_collection_test.rb
 - test/unit/results_item_test.rb
 - test/unit/search_facet_test.rb
 - test/unit/search_filter_test.rb
+- test/unit/search_highlight_test.rb
 - test/unit/search_query_test.rb
 - test/unit/search_sort_test.rb
 - test/unit/search_test.rb