picky 4.20.0 → 4.20.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: c5704e07276164090e6059d313cc0be35daeca92
-  data.tar.gz: 81083e451a334a22e710044eef8b97447670ee5e
+  metadata.gz: f25d9203dd2ec9711b17ff15a7f8efe8e4625e63
+  data.tar.gz: 3422b02dc81bf9af884d0f54db2d94b53eedf284
 SHA512:
-  metadata.gz: 37927091e857d0b01a531154e1b6edcb6c991a8e7099fafd17d9e4062c1dab6d92c1e14a6dddce803a9390ec02185574dcc323cf471a2bfdfd6a748f85a71d03
-  data.tar.gz: f592ef7f8c96e5c80fbfdb29a7f3e1883b74888589bf442320312af6542a64e23657d77b440548251f9d5bdcb13570f84b475a63c2b1e600577dff27d09003c0
+  metadata.gz: 8b6b0d73bbd233558fc64cbb65bd1975c5f37b260071d90c363933ac3863435a30b6067574ce48cdb16e117976da6713f05d5108eb47d0f9620de65e8666ab6b
+  data.tar.gz: 214ca1310b052232bafef670a2dc8013bd0c09c70f9ce7ea2e90338c858d71fd81e7e1a0a080bb91aff64c458b63f7a329abad8f9be543c558aee3747daf3996

data/lib/picky/tokenizer.rb

@@ -59,11 +59,10 @@ Case sensitive?     #{@case_sensitive ? "Yes." : "-"}
 
     # Stopwords.
     #
-    # We only allow regexps (even if string would be okay
-    # too for gsub! - it's too hard to understand)
+    # We even allow Strings even if it's hard to understand.
     #
     def stopwords regexp
-      check_argument_in __method__, Regexp, regexp
+      check_argument_in __method__, [Regexp, String, FalseClass], regexp
       @remove_stopwords_regexp = regexp
     end
     def remove_stopwords text
@@ -83,7 +82,7 @@ Case sensitive?     #{@case_sensitive ? "Yes." : "-"}
     # too for gsub! - it's too hard to understand)
     #
     def removes_characters regexp
-      check_argument_in __method__, Regexp, regexp
+      check_argument_in __method__, [Regexp, FalseClass], regexp
       @removes_characters_regexp = regexp
     end
     def remove_illegals text
@@ -192,8 +191,11 @@ Case sensitive?     #{@case_sensitive ? "Yes." : "-"}
 
     # Checks if the right argument type has been given.
     #
-    def check_argument_in method, type, argument, &condition
-      raise ArgumentError.new "Application##{method} takes a #{type} as argument, not a #{argument.class}." unless type === argument
+    def check_argument_in method, types, argument, &condition
+      types = [*types]
+      unless types.any? { |type| type === argument }
+        raise ArgumentError.new "Application##{method} takes any of #{types.join(', ')} as argument, but not a #{argument.class}."
+      end
     end
 
     attr_reader :substituter, :stemmer

data/spec/lib/tokenizer_spec.rb

@@ -236,11 +236,22 @@ ERROR
           tokenizer.remove_illegals('abcdefghijklmnop').should == 'bcdeghijklmnp'
         end
       end
+      context "with removes_characters called using false" do
+        before(:each) do
+          tokenizer.removes_characters false
+        end
+        it "has remove_illegals" do
+          expect { tokenizer.remove_illegals('abcdefghijklmnop') }.to_not raise_error
+        end
+        it "removes illegal characters" do
+          tokenizer.remove_illegals('abcdefghijklmnop').should == 'abcdefghijklmnop'
+        end
+      end
     end
 
     describe 'stopwords' do
       it 'handles broken arguments' do
-        expect { tokenizer.stopwords("hello") }.to raise_error(ArgumentError)
+        expect { tokenizer.stopwords(1) }.to raise_error(ArgumentError)
       end
       context 'without stopwords given' do
         it 'should define a method remove_stopwords' do
@@ -253,6 +264,26 @@ ERROR
           expect { tokenizer.remove_non_single_stopwords('from this text') }.to_not raise_error
         end
       end
+      context 'with stopwords given' do
+        before(:each) do
+          tokenizer.stopwords('t')
+        end
+        it 'should define a method remove_stopwords' do
+          lambda { tokenizer.remove_stopwords('from this text') }.should_not raise_error
+        end
+        it 'should define a method stopwords that removes stopwords' do
+          tokenizer.remove_stopwords('from this text').should == 'from his ex'
+        end
+        it 'should define a method remove_non_single_stopwords' do
+          expect { tokenizer.remove_non_single_stopwords('from this text') }.to_not raise_error
+        end
+        it 'should define a method remove_non_single_stopwords that removes non-single stopwords' do
+          tokenizer.remove_non_single_stopwords('rerere rerere').should == 'rerere rerere'
+        end
+        it 'should define a method remove_non_single_stopwords that does not single stopwords' do
+          tokenizer.remove_non_single_stopwords('rerere').should == 'rerere'
+        end
+      end
       context 'with stopwords given' do
         before(:each) do
           tokenizer.stopwords(/r|e/)
@@ -273,6 +304,26 @@ ERROR
           tokenizer.remove_non_single_stopwords('rerere').should == 'rerere'
         end
       end
+      context 'with stopwords explicitly not given' do
+        before(:each) do
+          tokenizer.stopwords(false)
+        end
+        it 'should define a method remove_stopwords' do
+          lambda { tokenizer.remove_stopwords('from this text') }.should_not raise_error
+        end
+        it 'should define a method stopwords that removes stopwords' do
+          tokenizer.remove_stopwords('from this text').should == 'from this text'
+        end
+        it 'should define a method remove_non_single_stopwords' do
+          expect { tokenizer.remove_non_single_stopwords('from this text') }.to_not raise_error
+        end
+        it 'should define a method remove_non_single_stopwords that removes non-single stopwords' do
+          tokenizer.remove_non_single_stopwords('rerere rerere').should == 'rerere rerere'
+        end
+        it 'should define a method remove_non_single_stopwords that does not single stopwords' do
+          tokenizer.remove_non_single_stopwords('rerere').should == 'rerere'
+        end
+      end
       context 'error case' do
         before(:each) do
           tokenizer.stopwords(/any/)

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: picky
 version: !ruby/object:Gem::Version
-  version: 4.20.0
+  version: 4.20.1
 platform: ruby
 authors:
 - Florian Hanke
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-02-07 00:00:00.000000000 Z
+date: 2014-02-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport