rexle-xpath-parser 0.1.0 → 0.1.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 696f378210d7800f7d4c79a48d6b9b12bd641209
-  data.tar.gz: 3118e004676a3976c4685338c469eff34c347748
+  metadata.gz: eab4821ed8ede5bd6583130f1b8010e6be40f006
+  data.tar.gz: 7020ccef47cd7f31489057b3c6a7f27749e4bc3b
 SHA512:
-  metadata.gz: 2b3611aa3dfe406a757bdb5b14c6f5bd47dc9ab14987c4695a6ac03d0ad03c31a35d43f7e3886b26c8e31caf3fd013bb8b521323064752b81a3dc6835106d7cb
-  data.tar.gz: 9ae85c002bee277546550dcf84520ccbf168282b85e9a087417755bcfbbbb1d0302499309736031bafa623b0372793cc968cfb216599f60b004f79b6007e51fc
+  metadata.gz: be752255841e13d4c23bf81b501123df2d07e6887f21d27060d21cc8b92be4fb2d4c85b3ebe48bd04a5ca00a81b67869cd0e99cabd45de31e8321afa15565249
+  data.tar.gz: 16804a1b8ea249903b318a4b39b4e29e6cc0b972c30dc0f073e1a0aedbe3e255728230a2110d52c63d52df51f2ddb4c97fecb647951485ff85a205f5a50b1cfd

data/lib/rexle-xpath-parser.rb

@@ -7,31 +7,139 @@ class RexleXPathParser
 
   attr_reader :to_a
 
-  def initialize(s)
-    @to_a = scan(s.clone)
+  def initialize(string)
+    
+    tokens = tokenise string
+    nested_tokens = tokens.map {|x| scan(x)}
+    @to_a = functionalise nested_tokens
+    
   end
 
-  def scan(s2)
+  private
+  
+  # maps the nested tokens to XPath functions, predicates, operators, 
+  #                                                      and 1 or more elements
+  #  
+  def functionalise(a)
 
-    if s2.lstrip[0] == '|' then    
-      return [s2.slice!(/\s*\|\s*/).strip] 
+    a.map do |x|
+
+      if x =~ /\w+\[/ then
+        epath, predicate = x.match(/^([^\[]+)\[([^\]]+)\]/).captures
+        epath.split('/').map {|e| [:select, e]} + [:predicate, predicate]
+      elsif x =~ /\|/
+        [:union] 
+      elsif x =~ /\w+\(/
+        [x.chop.to_sym]
+      elsif x =~ /\d+/
+        [:index, x[1..-2]]
+      elsif x =~ /[\w\/]+/
+        x.split('/').map {|e| [:select, e]}
+      elsif x.is_a? Array
+        functionalise(x)
+      end
+    
     end
 
-    s1 = s2.slice!(/\w+[\(\[]/)
+  end
+  
+  # matches a left bracket with a right bracket recursively if necessary
+  #
+  def lmatch(a, lchar, rchar)
+
+    token = []
+    c = a.first
+    token << c until (c = a.shift; c == lchar or c == rchar or a.empty?)
+    token << c
 
-    if s1[-1] == '[' then
-      s1 << s2.slice!(/[^\]]+\]|\]/)
-      return [s1]
+    if c == lchar then
+      found, tokenx, remainderx = rmatch(a, lchar, rchar)
+      c = found
+      token << tokenx
+      remainder = remainderx
+    else
+      remainder = a.join
     end
 
-    if s2[/\w+[\(\[]/] then
-      r = [scan(s2)]
-      r << scan(s2) while s2[/\w+[\(\[]/]
-      s3 = s2
+    [c, token.join, remainder]
+  end
+  
+  # tokeniser e.g. "a | d(c)" #=> ["a", " | ", "d(c)"] 
+  #
+  def match(s)
+
+    a = []
+
+    if s =~ /^\w+\(/ then
+
+      found, token, remainder = lmatch(s.chars, '(',')')
+
+      if found == ')' then
+        a << token
+      end
+
+    elsif s =~ /^\w+\[/
+      found, token, remainder = lmatch(s.chars, '[',']') 
+      a << token
     else
-      s1 << s2.slice!(/[^\)\]]+[\)\]]|[\)\]]/)
+      token = s.slice!(/\w+/)
+      a << token
+      remainder = s
+    end
+
+    operator = remainder.slice!(/^\s*\|\s*/)
+
+    if operator then
+      a.concat [operator, *match(remainder)]
+    end
+    
+    a
+  end
+   
+  # matches a right bracket for a left bracket which has already been found. 
+  #
+  def rmatch(a, lchar, rchar)
+
+    token = []
+    c = a.first
+    token << c until (c = a.shift; c == lchar or c == rchar or a.empty?)
+    token << c
+
+    if c == lchar then
+      
+      found, tokenx, remainderx = rmatch(a, lchar, rchar)
+      token << tokenx
+
+      # find the rmatch for the starting token
+      found, tokenx, remainderx = rmatch(a, lchar, rchar)        
+      c = found
+      token << tokenx
+      remainder = remainderx
+      
+    elsif c = rchar
+      remainder = a.join
     end
 
-    [s1,r,s3].compact
+    [c, token.join, remainder]
   end
-end
+
+  # accepts a token and drills into it to identify more tokens beneath it
+  #
+  def scan(s)
+
+    if s =~ /^\w+\(/ then
+      func = s.slice!(/\w+\(/)
+      remainder = s[0..-2]
+      if remainder =~ /^\w+\(/ then
+        scan(remainder)
+      else
+        [func, match(remainder)]
+      end
+    else
+      s
+    end
+  end
+      
+  alias tokenise match  
+
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: rexle-xpath-parser
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.1.1
 platform: ruby
 authors:
 - James Robertson
@@ -31,7 +31,7 @@ cert_chain:
   adFNS7+S3VVpQqBQ7hIMoNW5ldGHINll6Z0dbjsCO8OFGOkrnzt9ovLffOXuQHpT
   Gpk2If0KQMuV2Q==
   -----END CERTIFICATE-----
-date: 2015-05-29 00:00:00.000000000 Z
+date: 2015-06-04 00:00:00.000000000 Z
 dependencies: []
 description: 
 email: james@r0bertson.co.uk