minidusen 0.8.0 → 0.8.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 5bd84ad91feb76221d8d3f3b23995fceebf512b9
-  data.tar.gz: cf0b8b10d9caab23d4b4f4b7bb5c9d22ed185502
+  metadata.gz: 53ae4e4778bb465b76e102597bcde71ea9a0db5a
+  data.tar.gz: f2edf0821b4675e6be42805345e1cff779a42e43
 SHA512:
-  metadata.gz: a2002af4871106833bd6d2e861fa6cf78b879837e493668cfb5490749a8e99a5ed065d375a55b0cb1e7d6640be2863447b584574230306eca1e7497704353102
-  data.tar.gz: 9e22eb209770e120e23de9d8ab3c19c07ecb019bbb6f7aaa8ee4dc8d230c6a4a3b77ce47cec6eeb4735491615a64ef6998e28fb50684c742c75f2986d533af75
+  metadata.gz: dd3747985637690f5ee1f6504518c7e4a79a32c31ed9d5e99c54e0354c38b6e672b517ac518a0042308dbee49b08740c0b2517fecc32142ca61b8f6bf457c734
+  data.tar.gz: 72a9c8759ba4379793959ebd83dcd694076a146e81a4d0d95ff7635e3f0a6cff789381040f68b3b20267ea0dd7457ca709b0f334ebcf5c4c6173de6594dba42a

data/lib/minidusen/parser.rb

@@ -1,34 +1,63 @@
 module Minidusen
   class Parser
 
+    class CannotParse < StandardError; end
+
     TEXT_QUERY = /(?:(\-)?"([^"]+)"|(\-)?([\S]+))/
     FIELD_QUERY = /(\-)?(\w+)\:#{TEXT_QUERY}/
 
-    def self.parse(query_string)
-      query_string = query_string.dup # we are going to delete substrings in-place
-      query = Query.new
-      extract_field_query_tokens(query_string, query)
-      extract_text_query_tokens(query_string, query)
-      query
-    end
+    class << self
 
-    def self.extract_text_query_tokens(query_string, query)
-      while query_string.sub!(TEXT_QUERY, '')
-        value = "#{$2}#{$4}"
-        exclude = "#{$1}#{$3}" == "-"
-        options = { :field => 'text', :value => value, :exclude => exclude }
-        query << Token.new(options)
+      def parse(object)
+        case object
+        when Query
+          object
+        when String
+          parse_string(object)
+        when Array
+          parse_array(object)
+        else
+          raise CannotParse, "Cannot parse #{object.inspect}"
+        end
+      end
+
+      private
+
+      def parse_string(string)
+        string = string.dup # we are going to delete substrings in-place
+        query = Query.new
+        extract_field_query_tokens(string, query)
+        extract_text_query_tokens(string, query)
+        query
       end
-    end
 
-    def self.extract_field_query_tokens(query_string, query)
-      while query_string.sub!(FIELD_QUERY, '')
-        field = $2
-        value = "#{$4}#{$6}"
-        exclude = "#{$1}" == "-"
-        options = { :field => field, :value => value, :exclude => exclude }
-        query << Token.new(options)
+      def parse_array(array)
+        tokens = array.map { |string|
+          string.is_a?(String) or raise CannotParse, "Cannot parse an array of #{string.class}"
+          Token.new(:field => 'text', :value => string)
+        }
+        Query.new(tokens)
       end
+
+      def extract_text_query_tokens(query_string, query)
+        while query_string.sub!(TEXT_QUERY, '')
+          value = "#{$2}#{$4}"
+          exclude = "#{$1}#{$3}" == "-"
+          options = { :field => 'text', :value => value, :exclude => exclude }
+          query << Token.new(options)
+        end
+      end
+
+      def extract_field_query_tokens(query_string, query)
+        while query_string.sub!(FIELD_QUERY, '')
+          field = $2
+          value = "#{$4}#{$6}"
+          exclude = "#{$1}" == "-"
+          options = { :field => field, :value => value, :exclude => exclude }
+          query << Token.new(options)
+        end
+      end
+
     end
 
   end

data/lib/minidusen/query.rb

@@ -17,6 +17,10 @@ module Minidusen
       tokens[index]
     end
 
+    def size
+      tokens.size
+    end
+
     def to_s
       collect(&:to_s).join(" + ")
     end

data/lib/minidusen/syntax.rb

@@ -11,7 +11,7 @@ module Minidusen
     end
 
     def search(instance, root_scope, query)
-      query = parse(query) if query.is_a?(String)
+      query = parse(query)
       query = query.condensed
       matches = apply_query(instance, root_scope, query.include)
       if query.exclude.any?

data/lib/minidusen/token.rb

@@ -5,7 +5,7 @@ module Minidusen
 
     def initialize(options)
       @value = options.fetch(:value)
-      @exclude = options.fetch(:exclude)
+      @exclude = options.fetch(:exclude, false)
       @field = options.fetch(:field).to_s
     end
 

data/lib/minidusen/version.rb

@@ -1,3 +1,3 @@
 module Minidusen
-  VERSION = '0.8.0'
+  VERSION = '0.8.1'
 end

data/spec/minidusen/parser_spec.rb

@@ -2,19 +2,57 @@ describe Minidusen::Parser do
 
   describe '.parse' do
 
-    it 'should parse field tokens first, because they usually give maximum filtering at little cost' do
-      query = Minidusen::Parser.parse('word1 field1:field1-value word2 field2:field2-value')
-      query.collect(&:value).should == ['field1-value', 'field2-value', 'word1', 'word2']
+    describe 'when called with a String' do
+
+      it 'parses the given string into tokens' do
+        query = Minidusen::Parser.parse('fieldname:fieldvalue word "a phrase"')
+        query.size.should == 3
+        query[0].field.should == 'fieldname'
+        query[0].value.should == 'fieldvalue'
+        query[1].field.should == 'text'
+        query[1].value.should == 'word'
+        query[2].field.should == 'text'
+        query[2].value.should == 'a phrase'
+      end
+
+      it 'should parse field tokens first, because they usually give maximum filtering at little cost' do
+        query = Minidusen::Parser.parse('word1 field1:field1-value word2 field2:field2-value')
+        query.collect(&:value).should == ['field1-value', 'field2-value', 'word1', 'word2']
+      end
+
+      it 'should not consider the dash to be a word boundary' do
+        query = Minidusen::Parser.parse('Baden-Baden')
+        query.collect(&:value).should == ['Baden-Baden']
+      end
+
+      it 'should parse umlauts and accents' do
+        query = Minidusen::Parser.parse('field:åöÙÔøüéíÁ "ÄüÊçñÆ ððÿáÒÉ" pulvérisateur pędzić')
+        query.collect(&:value).should == ['åöÙÔøüéíÁ', 'ÄüÊçñÆ ððÿáÒÉ', 'pulvérisateur', 'pędzić']
+      end
+
     end
 
-    it 'should not consider the dash to be a word boundary' do
-      query = Minidusen::Parser.parse('Baden-Baden')
-      query.collect(&:value).should == ['Baden-Baden']
+    describe 'when called with a Query' do
+
+      it 'returns the query' do
+        passed_query = Minidusen::Query.new
+        parsed_query = Minidusen::Parser.parse(passed_query)
+        parsed_query.should == passed_query
+      end
+
     end
 
-    it 'should parse umlauts and accents' do
-      query = Minidusen::Parser.parse('field:åöÙÔøüéíÁ "ÄüÊçñÆ ððÿáÒÉ" pulvérisateur pędzić')
-      query.collect(&:value).should == ['åöÙÔøüéíÁ', 'ÄüÊçñÆ ððÿáÒÉ', 'pulvérisateur', 'pędzić']
+    describe 'when called with an array of strings' do
+
+      it 'returns a query of text tokens' do
+        query = Minidusen::Parser.parse(['word', 'a phrase'])
+        query.size.should == 2
+        query[0].field.should == 'text'
+        query[0].value.should == 'word'
+        query[1].field.should == 'text'
+        query[1].value.should == 'a phrase'
+      end
+
     end
 
   end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: minidusen
 version: !ruby/object:Gem::Version
-  version: 0.8.0
+  version: 0.8.1
 platform: ruby
 authors:
 - Henning Koch
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-08-21 00:00:00.000000000 Z
+date: 2017-10-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport
@@ -117,7 +117,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.5.1
+rubygems_version: 2.6.13
 signing_key: 
 specification_version: 4
 summary: Low-tech search for ActiveRecord with MySQL or PostgreSQL