msfl_visitors 1.2.0 → 1.2.1.dev0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b4fcb0f385bb41b619dad3d25724604bf43796e7
-  data.tar.gz: 3c51e3af9013e32a0b423da17f0b9926108dd9c6
+  metadata.gz: e43425b31d0774b6b7bfef7b7a47dd8914195729
+  data.tar.gz: b3fdf922ea429c522a77258ca40e5955d53373f3
 SHA512:
-  metadata.gz: b6626c6471bfeab9c8d1056fd2d23cbb6ccb2bf14409ef0589b7e0b4687c5fa3a0455f7596f223687a0bd8995449c548a5ee93617351b392491704878c3060be
-  data.tar.gz: 33d2578796d206d74757c9d587a34d1d46a148c777ac005599cebd51d8d28727ff231176000a4eedee15f47e20ab0f559842f92a0f5f2866511bcf62c44bd45d
+  metadata.gz: cc796e4b08f79201048cf37715f50cba54abc9026ebbb96669a5c8f4b3e8ef93505a65e2f5e1790494d3775521a65bfdcb30092ba1c7d1fffec329855684a8e7
+  data.tar.gz: 142acb13b5679b278a9c2487ecace20e8917d3673a715b9ecbfa91ee1ea5ab71f131a49a2237851cdb50855e28b1ee8aa1468a831b548736eeb0700b02ffaa09

data/lib/msfl_visitors/visitor.rb

@@ -49,6 +49,13 @@ module MSFLVisitors
       [{clause: root.accept(self)}].concat(clauses).reject { |c| c[:clause] == "" }
     end
 
+    # Note that the ES documentation also indicates that # is a special character that requires
+    # escaping and that this behavior is not part of the PERL regex; however Ruby automatically escapes
+    # literal hashes when constructing regices
+    def escape_es_special_regex_chars(str)
+      str.gsub(/([@&<>~])/) { |m| "\\#{m}" }
+    end
+
     private
 
     attr_reader :mode
@@ -68,12 +75,19 @@ module MSFLVisitors
           Nodes::Match                  => '=~',
       }
 
+      def escaped_regex_helper(escaped_str)
+        exp = visitor.escape_es_special_regex_chars "#{escaped_str}"
+        # why you must use #inspect, not #to_s. @link http://ruby-doc.org/core-1.9.3/Regexp.html#method-i-3D-7E
+        %r[.*#{exp}.*]
+      end
+
       def visit(node)
         case node
           when  Nodes::Field
             node.value.to_s
           when Nodes::Regex
-            %(Regexp.new( '.*' + Regexp.escape( "#{node.value.to_s}" ) + '.*' ))
+            esc = Regexp.escape("#{node.value.to_s}")
+            escaped_regex_helper esc
           when  Nodes::Word
             "\"#{node.value}\""
           when Nodes::Date, Nodes::Time
@@ -86,10 +100,10 @@ module MSFLVisitors
 
           when  Nodes::Match
             if node.right.is_a? Nodes::Set
-              regex = node.right.contents.map { |right_child| MSFLVisitors::Nodes::Regex.new(right_child.value.to_s).accept(visitor)[15..-6] }.join('|')
-              "#{node.left.accept(visitor)} #{BINARY_OPERATORS[node.class]} Regexp.new( '.*' + #{regex[4..-5]} + '.*' )"
+              escaped_str = node.right.contents.map { |right_child| MSFLVisitors::Nodes::Regex.new(right_child.value.to_s).accept(visitor).inspect[3..-4] }.join('|')
+              "#{node.left.accept(visitor)} #{BINARY_OPERATORS[node.class]} " + escaped_regex_helper(escaped_str).inspect
             else
-              "#{node.left.accept(visitor)} #{BINARY_OPERATORS[node.class]} #{MSFLVisitors::Nodes::Regex.new(node.right.value.to_s).accept(visitor)}"
+              "#{node.left.accept(visitor)} #{BINARY_OPERATORS[node.class]} " + MSFLVisitors::Nodes::Regex.new(node.right.value.to_s).accept(visitor).inspect
             end
           when  Nodes::Containment,
                 Nodes::GreaterThan,

data/msfl_visitors.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name        = 'msfl_visitors'
-  s.version     = '1.2.0'
+  s.version     = '1.2.1.dev0'
   s.date        = '2015-06-19'
   s.summary     = "Convert MSFL to other forms"
   s.description = "Visitor pattern approach to converting MSFL to other forms."

data/spec/msfl_visitors/visitor_spec.rb

@@ -11,4 +11,20 @@ describe MSFLVisitors::Visitor do
       expect(subject.send(:mode)).to eq :term
     end
   end
+
+  describe "#escape_es_special_regex_chars" do
+
+    {
+        'ab@cd' => 'ab\@cd',
+        'ab&cd' => 'ab\&cd',
+        'ab<cd' => 'ab\<cd',
+        'ab>cd' => 'ab\>cd',
+        'ab~cd' => 'ab\~cd',
+    }.each do |str, expected|
+
+      it "escapes '#{str}' as '#{expected}'" do
+        expect(MSFLVisitors::Visitor.new.escape_es_special_regex_chars str).to eq expected
+      end
+    end
+  end
 end

data/spec/msfl_visitors/visitors/chewy_term_filter_spec.rb

@@ -198,12 +198,23 @@ describe MSFLVisitors::Visitor do
 
     describe "a Regex node" do
 
-      let(:node) { MSFLVisitors::Nodes::Regex.new "foobar" }
+      let(:node) { MSFLVisitors::Nodes::Regex.new regex }
+
+      let(:regex) { "foobar" }
 
       context "when using the TermFilter visitor" do
 
-        it "results in: 'Regexp.new( '.*' + Regexp.escape( \"foobar\" ) + '.*' )'" do
-          expect(result).to eq %(Regexp.new( '.*' + Regexp.escape( "foobar" ) + '.*' ))
+        it "results in: /.*foobar.*/" do
+          expect(result).to eq /.*foobar.*/
+        end
+
+        context "when the node requires escaping" do
+
+          let(:node) { MSFLVisitors::Nodes::Regex.new "foo*bar" }
+
+          it "results in: #{/.*foo\*bar.*/.inspect}" do
+            expect(result).to eq /.*foo\*bar.*/
+          end
         end
       end
 
@@ -215,6 +226,36 @@ describe MSFLVisitors::Visitor do
           expect(result).to eq "foobar"
         end
       end
+
+      context "when the expression contains `#`, `@`, `&`, `<`, `>`, or `~`" do
+
+        let(:regex) { "a #sentence@ contain&ing <lucene> cha~rs" }
+
+        it "escapes lucene specific special characters" do
+          expected =  /.*a\ \#sentence\@\ contain\&ing\ \<lucene\>\ cha\~rs.*/
+          expect(result).to eq expected
+        end
+      end
+
+      context "when the regex contains characters that require escaping" do
+
+        let(:regex) { 'this / needs to % {be,escaped} *. ^[or] | \else' }
+
+        let(:node) { MSFLVisitors::Nodes::Regex.new regex }
+
+        it "returns: #{/.*this\ \/\ needs\ to\ %\ \{be,escaped\}\ \*\.\ \^\[or\]\ \|\ \\else.*/.inspect}" do
+          expect(result).to eq /.*this\ \/\ needs\ to\ %\ \{be,escaped\}\ \*\.\ \^\[or\]\ \|\ \\else.*/
+        end
+
+        context "when using the Aggregations visitor" do
+
+          before { visitor.mode = :aggregations }
+
+          it "returns: 'this\\ /\\ needs\\ to\\ %\\ \\{be,escaped\\}\\ \\*\\.\\ \\^\\[or\\]\\ \\|\\ \\\\else'" do
+            expect(result).to eq "this\\ /\\ needs\\ to\\ %\\ \\{be,escaped\\}\\ \\*\\.\\ \\^\\[or\\]\\ \\|\\ \\\\else"
+          end
+        end
+      end
     end
 
     describe "a Match node" do
@@ -223,8 +264,8 @@ describe MSFLVisitors::Visitor do
 
       context "when using the TermFilter visitor" do
 
-        it "results in: 'left =~ Regexp.new( '.*' + Regexp.parse( \"rhs\" ) + '.*' )'" do
-          expect(result).to eq %(lhs =~ Regexp.new( '.*' + Regexp.escape( "rhs" ) + '.*' ))
+        it "results in: 'left =~ /.*rhs.*/'" do
+          expect(result).to eq %(lhs =~ /.*rhs.*/)
         end
       end
 
@@ -243,8 +284,8 @@ describe MSFLVisitors::Visitor do
 
         context "when using the TermFilter visitor" do
 
-          it "results in: 'left =~ Regexp.new( '.*' + Regexp.escape( \"this (needs) to be* escaped\" ) + '.*' )'" do
-            expect(result).to eq %(lhs =~ Regexp.new( '.*' + Regexp.escape( "this (needs) to be* escaped" ) + '.*' ))
+          it "results in: 'left =~ /.*this\ \(needs\)\ to\ be\*\ escaped.*/'" do
+            expect(result).to eq %(lhs =~ ) + /.*this\ \(needs\)\ to\ be\*\ escaped.*/.inspect
           end
         end
 
@@ -262,8 +303,8 @@ describe MSFLVisitors::Visitor do
 
         context "when using the TermFilter visitor" do
 
-          it "results in: 'left =~ Regexp.new( '.*' + Regexp.escape( \"foo\" ) + '|' + Regexp.escape( \"bar\" ) + '|' + Regexp.escape( \"baz\" ) + '.*' )'" do
-            expect(result).to eq %(lhs =~ Regexp.new( '.*' + Regexp.escape( "foo" ) + '|' + Regexp.escape( "bar" ) + '|' + Regexp.escape( "baz" ) + '.*' ))
+          it "results in: 'left =~ /.*foo|bar|baz.*/'" do
+            expect(result).to eq %(lhs =~ ) + /.*foo|bar|baz.*/.inspect
           end
         end
 
@@ -719,46 +760,6 @@ describe MSFLVisitors::Visitor do
           end
         end
       end
-
-      describe "a Regex node" do
-
-        let(:regex) { "content" }
-
-        let(:node) { MSFLVisitors::Nodes::Regex.new regex }
-
-        it "returns: 'Regexp.new( '.*' + Regexp.escape( \"content\" ) + '.*' )'" do
-          expect(result).to eq %(Regexp.new( '.*' + Regexp.escape( "#{regex}" ) + '.*' ))
-        end
-
-        context "when using the Aggregations visitor" do
-
-          before { visitor.mode = :aggregations }
-
-          it "returns: the elasticsearch regular expression formatted string \"content\"" do
-            expect(result).to eq "#{regex}"
-          end
-        end
-
-        context "when the regex contains characters that require escaping" do
-
-          let(:regex) { 'this / needs to % {be,escaped} *. ^[or] | \else' }
-
-          let(:node) { MSFLVisitors::Nodes::Regex.new regex }
-
-          it "returns: 'Regexp.new( '.*' + Regexp.escape( \"this / needs to % {be,escaped} *. ^[or] | \\else\" ) + '.*' )'" do
-            expect(result).to eq %(Regexp.new( '.*' + Regexp.escape( "this / needs to % {be,escaped} *. ^[or] | \\else" ) + '.*' ))
-          end
-
-          context "when using the Aggregations visitor" do
-
-            before { visitor.mode = :aggregations }
-
-            it "returns: 'this\\ /\\ needs\\ to\\ %\\ \\{be,escaped\\}\\ \\*\\.\\ \\^\\[or\\]\\ \\|\\ \\\\else'" do
-              expect(result).to eq "this\\ /\\ needs\\ to\\ %\\ \\{be,escaped\\}\\ \\*\\.\\ \\^\\[or\\]\\ \\|\\ \\\\else"
-            end
-          end
-        end
-      end
     end
 
     describe "range value nodes" do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: msfl_visitors
 version: !ruby/object:Gem::Version
-  version: 1.2.0
+  version: 1.2.1.dev0
 platform: ruby
 authors:
 - Courtland Caldwell
@@ -173,9 +173,9 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - ">"
     - !ruby/object:Gem::Version
-      version: '0'
+      version: 1.3.1
 requirements: []
 rubyforge_project: 
 rubygems_version: 2.2.2