mongoid_fulltext 0.5.4 → 0.5.5

data/README.md

@@ -4,6 +4,12 @@ Mongoid Fulltext Search
 Full-text search using n-gram matching for the Mongoid ODM. Tested on MongoDB 1.6 and above, but
 probably works on earlier versions as well.
 
+MongoDB currently has no native full-text search capabilities, so this gem is a good fit for cases
+where you want something a little less than a full-blown indexing service like Solr. mongoid_fulltext
+lets you do a fuzzy string search across relatively short strings, which makes it good for populating
+autocomplete boxes based on the display names of your Rails models but not appropriate for, say, 
+indexing hundreds of thousands of HTML documents.
+
 Some examples:
 --------------
     

data/VERSION

@@ -1 +1 @@
-0.5.4
+0.5.5

data/lib/mongoid_fulltext.rb

@@ -24,7 +24,7 @@ module Mongoid::FullTextSearch
 
       config = { 
         :alphabet => 'abcdefghijklmnopqrstuvwxyz0123456789 ',
-        :word_separators => ' ',
+        :word_separators => "- \n\t",
         :ngram_width => 3,
         :max_ngrams_to_search => 6,
         :apply_prefix_scoring_to_all_words => true,
@@ -219,11 +219,17 @@ module Mongoid::FullTextSearch
 
       # If an ngram appears multiple times in the query string, keep the max score
       ngram_array = ngram_array.group_by{ |h| h[:ngram] }.map{ |key, values| {:ngram => key, :score => values.map{ |v| v[:score] }.max} }
+      
+      if config[:index_short_prefixes] or config[:index_full_words]
+        split_regex_def = config[:word_separators].keys.map{ |k| Regexp.escape(k) }.join
+        split_regex = Regexp.compile("[#{split_regex_def}]")
+        all_words = filtered_str.split(split_regex)
+      end
 
       # Add 'short prefix' records to the array: prefixes of the string that are length (ngram_width - 1)
       if config[:index_short_prefixes]
         prefixes_seen = {}
-        filtered_str.split(Regexp.compile(config[:word_separators].keys.join)).each do |word|
+        all_words.each do |word|
           next if word.length < config[:ngram_width]-1
           prefix = word[0...config[:ngram_width]-1]
           if prefixes_seen[prefix].nil? and (config[:stop_words][word].nil? or word == filtered_str)
@@ -236,7 +242,7 @@ module Mongoid::FullTextSearch
       # Add records to the array of ngrams for each full word in the string that isn't a stop word
       if config[:index_full_words]
         full_words_seen = {}
-        filtered_str.split(Regexp.compile(config[:word_separators].keys.join)).each do |word|
+        all_words.each do |word|
           if word.length > 1 and full_words_seen[word].nil? and (config[:stop_words][word].nil? or word == filtered_str)
             ngram_array << {:ngram => word, :score => 1 + 1.0/filtered_str.length}
             full_words_seen[word] = true

data/mongoid_fulltext.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{mongoid_fulltext}
-  s.version = "0.5.4"
+  s.version = "0.5.5"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Aaron Windsor"]
-  s.date = %q{2011-11-08}
+  s.date = %q{2011-12-19}
   s.description = %q{Full-text search for the Mongoid ORM, using n-grams extracted from text}
   s.email = %q{aaron.windsor@gmail.com}
   s.extra_rdoc_files = [

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: mongoid_fulltext
 version: !ruby/object:Gem::Version
-  version: 0.5.4
+  version: 0.5.5
   prerelease: 
 platform: ruby
 authors:
@@ -9,12 +9,12 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-11-08 00:00:00.000000000 -05:00
+date: 2011-12-19 00:00:00.000000000 -05:00
 default_executable: 
 dependencies:
 - !ruby/object:Gem::Dependency
   name: unicode_utils
-  requirement: &83097910 !ruby/object:Gem::Requirement
+  requirement: &74785740 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -22,10 +22,10 @@ dependencies:
         version: 1.0.0
   type: :runtime
   prerelease: false
-  version_requirements: *83097910
+  version_requirements: *74785740
 - !ruby/object:Gem::Dependency
   name: mongoid
-  requirement: &83097670 !ruby/object:Gem::Requirement
+  requirement: &74785500 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -33,10 +33,10 @@ dependencies:
         version: 2.0.0
   type: :development
   prerelease: false
-  version_requirements: *83097670
+  version_requirements: *74785500
 - !ruby/object:Gem::Dependency
   name: bson_ext
-  requirement: &83097430 !ruby/object:Gem::Requirement
+  requirement: &74785260 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -44,10 +44,10 @@ dependencies:
         version: 1.3.0
   type: :development
   prerelease: false
-  version_requirements: *83097430
+  version_requirements: *74785260
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &83097190 !ruby/object:Gem::Requirement
+  requirement: &74785020 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -55,10 +55,10 @@ dependencies:
         version: 2.5.0
   type: :development
   prerelease: false
-  version_requirements: *83097190
+  version_requirements: *74785020
 - !ruby/object:Gem::Dependency
   name: jeweler
-  requirement: &83096950 !ruby/object:Gem::Requirement
+  requirement: &74784780 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -66,7 +66,7 @@ dependencies:
         version: 1.5.2
   type: :development
   prerelease: false
-  version_requirements: *83096950
+  version_requirements: *74784780
 description: Full-text search for the Mongoid ORM, using n-grams extracted from text
 email: aaron.windsor@gmail.com
 executables: []
@@ -120,7 +120,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 321098159
+      hash: -329929775
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements: