rifle 0.1.7 → 0.1.8

data/README.md

@@ -30,8 +30,24 @@ urn and a payload
 
 # Payloads
 
-Payloads are expected to be hashes, identified by a company-wide unique id (a urn). These are indexed and can
-be searched by metaphone if fuzzy_matching is enabled. That is, the search term need not be exact.
+Payloads are expected to be hashes, identified by a company-wide unique id (a urn). These are indexed on store by exact words, and collapsed words if punctuation
+is in the middle of a word.
+
+E.g, given the following payload
+
+    O'Connor has telephone +(44)798765432?
+
+Any of the following search terms will match
+
+    O'Connor
+    Oconnor
+    +44798765432
+    +(44)798765432
+    0798765432     # <= Special case, UK phone prefixes are ignored
+
+# Fuzzy Matching
+
+Items can be searched by metaphone if fuzzy_matching is enabled. That is, the search term need not be exact.
 
 E.g, given the following payload
 

data/lib/rifle.rb

@@ -20,8 +20,9 @@ module Rifle
     Processor.new.index_resource(urn, hash)
   end
 
-  def self.search(words, urns_only = false)
-    Processor.new.search_for(words, urns_only)
+  def self.search(words, options = {})
+    options = {urns_only: options} if !!options == options # Support old boolean only arg
+    Processor.new.search_for(words, options)
   end
 
   class Processor
@@ -56,29 +57,29 @@ module Rifle
       metaphones
     end
 
-    def search_for(sentence, urns_only)
-      p "Rifle searching for #{sentence}, urns only #{urns_only}"
+    def search_for(sentence, options)
+      p "Rifle searching for #{sentence}, options #{options}"
       words = get_words_array_from_text(sentence)
       metaphones = get_metaphones_from_word_set(Set.new(words))
 
       urns = nil
       metaphones.each do |metaphone|
         new_urns = get_urns_for_metaphone(metaphone)
-        p metaphone
         urns = urns.nil? ? Set.new(new_urns) : urns.intersection(new_urns)
       end
       urns ||= Set.new
 
       p "Rifle found #{urns.size} urns"
-      if urns_only
+      if options[:urns_only]
         urns
       else
-        urns.map { |u|
+        full_results = urns.map { |u|
           {
-              urn: u,
-              payload: get_payload_for_urn(u)
+            urn: u,
+            payload: get_payload_for_urn(u)
           }
         }
+        full_results = full_results.sort! { |a, b| DateTime.parse(b[:payload]['updated_at']) <=> DateTime.parse(a[:payload]['updated_at']) }
       end
     end
 
@@ -104,17 +105,25 @@ module Rifle
 
     def get_words_array_from_text(text)
       return [] if !text.is_a?(String)
-      text.downcase.split(/[^a-zA-Z0-9]/).select { |w| w.length >= Rifle.settings.min_word_length }
-    end
+      text = text.downcase
 
-    def get_metaphones_from_word_set(words)
+      # First get the smallest parts, split by anything that isn't a letter or number
+      results = text.split(/[^a-zA-Z0-9]/)
+      # Now add the text blocks just with punctuation removed. eg O'Connor -> OConnor
+      by_spaces = text.split(' ')
+      results = results + by_spaces.map { |w| w.gsub(/[^a-zA-Z0-9]/, '') }
       # Add extra search terms. EG, other phone number layouts
-      words = Set.new(words.map { |w|
-        # Here we have to strip all the front +44 and replace with 0.
-        w = w.start_with?('+44') ? "0#{w[3..-1]}" : w
-        w = w.start_with?('+') ? w[1..-1] : w # Also strip + signs, as they are ignored by Resque in keys
-      })
+      results = results + by_spaces.select { |w| w.start_with?('+44') }.map { |w|
+        # Here we have to strip all the front +44 and replace with 0. Also, store the one without a prefix.
+        ["0#{w[3..-1]}", w[3..-1]]
+      }
+      # Unique
+      results = results.flatten.uniq
+
+      results.select { |w| w.length >= Rifle.settings.min_word_length }
+    end
 
+    def get_metaphones_from_word_set(words)
       # Removed ignored words
       words.subtract Rifle.settings.ignored_words
       # Get the parts

data/lib/rifle/version.rb

@@ -1,3 +1,3 @@
 module Rifle
-  VERSION = '0.1.7'
+  VERSION = '0.1.8'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: rifle
 version: !ruby/object:Gem::Version
-  version: 0.1.7
+  version: 0.1.8
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-12-20 00:00:00.000000000 Z
+date: 2013-01-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   type: :runtime