msfl_visitors 1.2.1.dev1 → 1.2.1.dev2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: f37f5301f25a83e07f38aab1b3b82ee096d3ca1d
-  data.tar.gz: 609b068eff9ace33c8ef8f7c3b917f5392981143
+  metadata.gz: 91f283c200ab6615d376fd0bba263230f4734a4b
+  data.tar.gz: e5a9d85eafd0585f93314e53f5c1198628fbd833
 SHA512:
-  metadata.gz: d0d858070bcd4a52e6f9ab10513b60d19e9df46a92d880b42db0231a08cad0aef29c4015b8a752f5bc863785eb346a2c9d50654874ef87a086675bbf3053caf3
-  data.tar.gz: 5a86c87c0942efce43ca92c1822d02ff3a34bcd1776b49b04bc8f86d4b37a5deef56625ef29244ea4c279794b3ccb0d222429154c3ebf6099db134bd0583e22a
+  metadata.gz: df7283c6be4eb3c4e0e5a98d3321449828c734cfa861189da908a698ab54353c78afd7452b99383ef6cbdd13c64a5532376b6607776102adfbe19abc5ea38a59
+  data.tar.gz: 4dce7a8e50ecde1219743ef727da96ded10eda3bc959cf06450c319390e2eadc9fbc6db2f906ca2db6032117337ba5257e8246bfd2b99ab64b2c95cc04a16dc1

data/lib/msfl_visitors/visitor.rb

@@ -1,5 +1,25 @@
 require 'forwardable'
 module MSFLVisitors
+
+  module VisitorHelpers
+    # 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
+
+    def escaped_regex_helper(escaped_str)
+      exp = 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 composable_expr_for(regex_as_literal_string)
+      regex_as_literal_string[3..-4]
+    end
+  end
+
   class Visitor
 
     attr_accessor :clauses, :current_clause
@@ -49,18 +69,13 @@ 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
 
     class TermFilterVisitor
+      include VisitorHelpers
+
       def initialize(visitor)
         @visitor = visitor
       end
@@ -75,11 +90,7 @@ 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
@@ -100,7 +111,8 @@ module MSFLVisitors
 
           when  Nodes::Match
             if node.right.is_a? Nodes::Set
-              escaped_str = node.right.contents.map { |right_child| MSFLVisitors::Nodes::Regex.new(right_child.value.to_s).accept(visitor).inspect[3..-4] }.join('|')
+              escaped_str_frags = node.right.contents.map { |right_child| composable_expr_for(MSFLVisitors::Nodes::Regex.new(right_child.value.to_s).accept(visitor).inspect) }
+              escaped_str = escaped_str_frags.join('|')
               "#{node.left.accept(visitor)} #{BINARY_OPERATORS[node.class]} " + %r[.*#{escaped_str}.*].inspect
             else
               "#{node.left.accept(visitor)} #{BINARY_OPERATORS[node.class]} " + MSFLVisitors::Nodes::Regex.new(node.right.value.to_s).accept(visitor).inspect
@@ -145,6 +157,7 @@ module MSFLVisitors
     end
 
     class AggregationsVisitor
+      include VisitorHelpers
       def initialize(visitor)
         @visitor = visitor
       end
@@ -192,7 +205,8 @@ module MSFLVisitors
                 Nodes::Dataset
             node.value
           when  Nodes::Regex
-            Regexp.escape(node.value.to_s)
+            esc = Regexp.escape("#{node.value.to_s}")
+            composable_expr_for(escaped_regex_helper(esc).inspect)
 
           when  Nodes::GreaterThan,
                 Nodes::GreaterThanEqual,

data/msfl_visitors.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name        = 'msfl_visitors'
-  s.version     = '1.2.1.dev1'
+  s.version     = '1.2.1.dev2'
   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

@@ -14,6 +14,7 @@ describe MSFLVisitors::Visitor do
 
   describe "#escape_es_special_regex_chars" do
 
+    include MSFLVisitors::VisitorHelpers
     {
         'ab@cd' => 'ab\@cd',
         'ab&cd' => 'ab\&cd',
@@ -23,7 +24,7 @@ describe MSFLVisitors::Visitor do
     }.each do |str, expected|
 
       it "escapes '#{str}' as '#{expected}'" do
-        expect(MSFLVisitors::Visitor.new.escape_es_special_regex_chars str).to eq expected
+        expect(escape_es_special_regex_chars str).to eq expected
       end
     end
   end

data/spec/msfl_visitors/visitors/chewy_term_filter_spec.rb

@@ -235,6 +235,15 @@ describe MSFLVisitors::Visitor do
           expected =  /.*a\ \#sentence\@\ contain\&ing\ \<lucene\>\ cha\~rs.*/
           expect(result).to eq expected
         end
+
+        context "when using the Aggregations visitor" do
+
+          before { visitor.mode = :aggregations }
+
+          it "escapes lucene specific special characters" do
+            expect(result).to eq "a\\ \\#sentence\\@\\ contain\\&ing\\ \\<lucene\\>\\ cha\\~rs"
+          end
+        end
       end
 
       context "when the regex contains characters that require escaping" do
@@ -251,8 +260,8 @@ describe MSFLVisitors::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"
+          it "returns: 'this\\ \\/\\ needs\\ to\\ %\\ \\{be,escaped\\}\\ \\*\\.\\ \\^\\[or\\]\\ \\|\\ \\\\else'" do
+            expect(result).to eq "this\\ \\/\\ needs\\ to\\ %\\ \\{be,escaped\\}\\ \\*\\.\\ \\^\\[or\\]\\ \\|\\ \\\\else"
           end
         end
       end
@@ -289,6 +298,15 @@ describe MSFLVisitors::Visitor do
           end
         end
 
+        context "when using the Aggregations visitor" do
+
+          before { visitor.mode = :aggregations }
+
+          it %(results in: { agg_field_name: :lhs, operator: :match, test_value: "this\\ \\(ne\\&eds\\)\\ to\\ be\\*\\ escaped" }) do
+            expected = { agg_field_name: :lhs, operator: :match, test_value: "this\\ \\(ne\\&eds\\)\\ to\\ be\\*\\ escaped" }
+            expect(result).to eq expected
+          end
+        end
       end
 
       context "when the right hand side is a Set node containing Value nodes" do
@@ -324,6 +342,15 @@ describe MSFLVisitors::Visitor do
           it "results in: { agg_field_name: :lhs, operator: :match, test_value: \"foo|bar|baz\" }" do
             expect(result).to eq({agg_field_name: :lhs, operator: :match, test_value: "foo|bar|baz"})
           end
+
+          context "when one of the members of the Set requires escaping" do
+
+            let(:foo_node) { MSFLVisitors::Nodes::Word.new "please&*escape me" }
+
+            it "results in { agg_field_name: :lhs, operator: :match, test_value: \"please\\&\\*escape\\ me }" do
+              expected = { agg_field_name: :lhs, operator: :match, test_value: "please\\&\\*escape\\ me" }
+            end
+          end
         end
       end
     end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: msfl_visitors
 version: !ruby/object:Gem::Version
-  version: 1.2.1.dev1
+  version: 1.2.1.dev2
 platform: ruby
 authors:
 - Courtland Caldwell