regexp_parser 0.4.6 → 0.4.7

data/lib/regexp_parser/scanner.rb

@@ -2,15 +2,101 @@
 
 # line 1 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 
-# line 775 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 771 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 
 
 # THIS IS A GENERATED FILE, DO NOT EDIT DIRECTLY
 # This file was generated from lib/regexp_parser/scanner/scanner.rl
 
-module Regexp::Scanner
-  
-# line 13 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
+class Regexp::Scanner
+  # General scanner error (catch all)
+  class ScannerError < StandardError; end
+
+  # Base for all scanner validation errors
+  class ValidationError < StandardError
+    def initialize(reason)
+      super reason
+    end
+  end
+
+  # Unexpected end of pattern
+  class PrematureEndError < ScannerError
+    def initialize(where = '')
+      super "Premature end of pattern at #{where}"
+    end
+  end
+
+  # Invalid sequence format. Used for escape sequences, mainly.
+  class InvalidSequenceError < ValidationError
+    def initialize(what = 'sequence', where = '')
+      super "Invalid #{what} at #{where}"
+    end
+  end
+
+  # Invalid group. Used for named groups.
+  class InvalidGroupError < ValidationError
+    def initialize(what, reason)
+      super "Invalid #{what}, #{reason}."
+    end
+  end
+
+  # Invalid groupOption. Used for inline options.
+  class InvalidGroupOption < ValidationError
+    def initialize(option, text)
+      super "Invalid group option #{option} in #{text}"
+    end
+  end
+
+  # Invalid back reference. Used for name a number refs/calls.
+  class InvalidBackrefError < ValidationError
+    def initialize(what, reason)
+      super "Invalid back reference #{what}, #{reason}"
+    end
+  end
+
+  # The property name was not recognized by the scanner.
+  class UnknownUnicodePropertyError < ValidationError
+    def initialize(name)
+      super "Unknown unicode character property name #{name}"
+    end
+  end
+
+  # Scans the given regular expression text, or Regexp object and collects the
+  # emitted token into an array that gets returned at the end. If a block is
+  # given, it gets called for each emitted token.
+  #
+  # This method may raise errors if a syntax error is encountered.
+  # --------------------------------------------------------------------------
+  def self.scan(input_object, &block)
+    new.scan(input_object, &block)
+  end
+
+  def scan(input_object, &block)
+    @literal, top, stack = nil, 0, []
+
+    if input_object.is_a?(Regexp)
+      input    = input_object.source
+      @free_spacing  = (input_object.options & Regexp::EXTENDED != 0)
+    else
+      input   = input_object
+      @free_spacing = false
+    end
+
+
+    data  = input.unpack("c*") if input.is_a?(String)
+    eof   = data.length
+
+    @tokens = []
+    @block  = block_given? ? block : nil
+
+    @in_group, @group_depth = false, 0
+    @spacing_stack = [{:free_spacing => @free_spacing, :depth => 0}]
+
+    in_set,   set_depth, set_type   = false, 0, :set
+    in_conditional, conditional_depth, conditional_stack = false, 0, []
+
+    
+# line 99 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
 class << self
 	attr_accessor :_re_scanner_trans_keys
 	private :_re_scanner_trans_keys, :_re_scanner_trans_keys=
@@ -1586,92 +1672,9 @@ end
 self.re_scanner_en_main = 142;
 
 
-# line 782 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
-
-  # General scanner error (catch all)
-  class ScannerError < StandardError; end
-
-  # Base for all scanner validation errors
-  class ValidationError < StandardError
-    def initialize(reason)
-      super reason
-    end
-  end
-
-  # Unexpected end of pattern
-  class PrematureEndError < ScannerError
-    def initialize(where = '')
-      super "Premature end of pattern at #{where}"
-    end
-  end
-
-  # Invalid sequence format. Used for escape sequences, mainly.
-  class InvalidSequenceError < ValidationError
-    def initialize(what = 'sequence', where = '')
-      super "Invalid #{what} at #{where}"
-    end
-  end
-
-  # Invalid group. Used for named groups.
-  class InvalidGroupError < ValidationError
-    def initialize(what, reason)
-      super "Invalid #{what}, #{reason}."
-    end
-  end
-
-  # Invalid groupOption. Used for inline options.
-  class InvalidGroupOption < ValidationError
-    def initialize(option, text)
-      super "Invalid group option #{option} in #{text}"
-    end
-  end
-
-  # Invalid back reference. Used for name a number refs/calls.
-  class InvalidBackrefError < ValidationError
-    def initialize(what, reason)
-      super "Invalid back reference #{what}, #{reason}"
-    end
-  end
-
-  # The property name was not recognized by the scanner.
-  class UnknownUnicodePropertyError < ValidationError
-    def initialize(name)
-      super "Unknown unicode character property name #{name}"
-    end
-  end
-
-  # Scans the given regular expression text, or Regexp object and collects the
-  # emitted token into an array that gets returned at the end. If a block is
-  # given, it gets called for each emitted token.
-  #
-  # This method may raise errors if a syntax error is encountered.
-  # --------------------------------------------------------------------------
-  def self.scan(input_object, &block)
-    @literal, top, stack = nil, 0, []
-
-    if input_object.is_a?(Regexp)
-      input    = input_object.source
-      @free_spacing  = (input_object.options & Regexp::EXTENDED != 0)
-    else
-      input   = input_object
-      @free_spacing = false
-    end
-
-
-    data  = input.unpack("c*") if input.is_a?(String)
-    eof   = data.length
-
-    @tokens = []
-    @block  = block_given? ? block : nil
-
-    @in_group, @group_depth = false, 0
-    @in_options, @spacing_stack = false, [[@free_spacing, 0]]
-
-    in_set,   set_depth, set_type   = false, 0, :set
-    in_conditional, conditional_depth, conditional_stack = false, 0, []
-
+# line 864 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
     
-# line 1674 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
+# line 1677 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
 begin
 	p ||= 0
 	pe ||= data.length
@@ -1682,9 +1685,9 @@ begin
 	act = 0
 end
 
-# line 866 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 865 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
     
-# line 1687 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
+# line 1690 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
 begin
 	testEof = false
 	_slen, _trans, _keys, _inds, _acts, _nacts = nil
@@ -1712,7 +1715,7 @@ begin
 		begin
 ts = p
 		end
-# line 1715 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
+# line 1718 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
 	end
 	_keys = cs << 1
 	_inds = _re_scanner_index_offsets[cs]
@@ -3385,7 +3388,7 @@ te = p+1
 		begin
 te = p+1
  begin 
-      if in_conditional and conditional_stack.length > 0 and 
+      if in_conditional and conditional_stack.length > 0 and
          conditional_stack.last[1] == @group_depth
         emit(:conditional, :separator, *text(data, ts, te))
       else
@@ -3547,7 +3550,7 @@ te = p+1
      end
 		end
 	when 13 then
-# line 642 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 638 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p+1
  begin 
@@ -3619,7 +3622,7 @@ te = p+1
      end
 		end
 	when 58 then
-# line 713 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 709 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p+1
  begin 
@@ -3631,7 +3634,7 @@ te = p+1
      end
 		end
 	when 54 then
-# line 721 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 717 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p+1
  begin 
@@ -3643,7 +3646,7 @@ te = p+1
      end
 		end
 	when 56 then
-# line 729 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 725 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p+1
  begin 
@@ -3655,7 +3658,7 @@ te = p+1
      end
 		end
 	when 64 then
-# line 737 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 733 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p+1
  begin 
@@ -3663,7 +3666,7 @@ te = p+1
      end
 		end
 	when 4 then
-# line 747 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 743 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p+1
  begin 
@@ -3684,7 +3687,7 @@ p = p - 1; begin
      end
 		end
 	when 57 then
-# line 713 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 709 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p
 p = p - 1; begin 
@@ -3696,7 +3699,7 @@ p = p - 1; begin
      end
 		end
 	when 53 then
-# line 721 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 717 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p
 p = p - 1; begin 
@@ -3708,7 +3711,7 @@ p = p - 1; begin
      end
 		end
 	when 55 then
-# line 729 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 725 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p
 p = p - 1; begin 
@@ -3720,7 +3723,7 @@ p = p - 1; begin
      end
 		end
 	when 63 then
-# line 737 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 733 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p
 p = p - 1; begin 
@@ -3728,7 +3731,7 @@ p = p - 1; begin
      end
 		end
 	when 59 then
-# line 743 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 739 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p
 p = p - 1; begin 
@@ -3743,7 +3746,7 @@ p = p - 1; begin
      end
 		end
 	when 50 then
-# line 755 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 751 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p
 p = p - 1; begin 
@@ -3755,7 +3758,7 @@ p = p - 1; begin
      end
 		end
 	when 49 then
-# line 770 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 766 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 te = p
 p = p - 1; begin 
@@ -3772,7 +3775,7 @@ p = p - 1; begin
      end
 		end
 	when 12 then
-# line 743 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 739 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
  begin p = ((te))-1; end
  begin 
@@ -3787,7 +3790,7 @@ p = p - 1; begin
      end
 		end
 	when 3 then
-# line 770 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 766 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
  begin p = ((te))-1; end
  begin 
@@ -4085,14 +4088,10 @@ te = p+1
         end
       else
         if @spacing_stack.length > 1 and
-          @spacing_stack.last[1] == (@group_depth + 1)
+          @spacing_stack.last[:depth] == (@group_depth + 1)
           @spacing_stack.pop
 
-          @free_spacing = @spacing_stack.last[0]
-
-          if @spacing_stack.length == 1
-            @in_options = false
-          end
+          @free_spacing = @spacing_stack.last[:free_spacing]
         end
 
         emit(:group, :close, *text(data, ts, te))
@@ -4144,10 +4143,10 @@ act = 22;		end
 		begin
 te = p+1
 		end
-# line 770 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 766 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 		begin
 act = 59;		end
-# line 4150 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
+# line 4149 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
 	end
 	end
 	end
@@ -4165,7 +4164,7 @@ ts = nil;		end
 		begin
 act = 0
 		end
-# line 4168 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
+# line 4167 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
 	end
 
 	if cs == 0
@@ -4199,7 +4198,7 @@ act = 0
     text = ts ? copy(data, ts-1..-1) : data.pack('c*')
     raise PrematureEndError.new( text )
   		end
-# line 4202 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
+# line 4201 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner.rb"
 	  end
 	end
 
@@ -4210,7 +4209,7 @@ act = 0
 end
 	end
 
-# line 867 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
+# line 866 "/Users/jannoschmuller/code/regexp_parser/lib/regexp_parser/scanner/scanner.rl"
 
     if cs == re_scanner_error
       text = ts ? copy(data, ts-1..-1) : data.pack('c*')
@@ -4228,12 +4227,25 @@ end
     @tokens
   end
 
+  # Emits an array with the details of the scanned pattern
+  def emit(type, token, text, ts, te)
+    #puts "EMIT: type: #{type}, token: #{token}, text: #{text}, ts: #{ts}, te: #{te}"
+
+    emit_literal if @literal
+
+    if @block
+      @block.call type, token, text, ts, te
+    end
+
+    @tokens << [type, token, text, ts, te]
+  end
+
   private
 
   # Ragel's regex-based scan of the group options introduced a lot of
   # ambiguity, so we just ask it to find the beginning of what looks
   # like an options run and handle the rest in here.
-  def self.scan_options(p, data, ts, te)
+  def scan_options(p, data, ts, te)
     text = text(data, ts, te).first
 
     options_char, options_length = true, 0
@@ -4285,26 +4297,26 @@ end
   end
 
   # Copy from ts to te from data as text
-  def self.copy(data, range)
+  def copy(data, range)
     data[range].pack('c*')
   end
 
   # Copy from ts to te from data as text, returning an array with the text
   #  and the offsets used to copy it.
-  def self.text(data, ts, te, soff = 0)
+  def text(data, ts, te, soff = 0)
     [copy(data, ts-soff..te-1), ts-soff, te]
   end
 
   # Appends one or more characters to the literal buffer, to be emitted later
   # by a call to emit_literal. Contents can be a mix of ASCII and UTF-8.
-  def self.append_literal(data, ts, te)
+  def append_literal(data, ts, te)
     @literal ||= []
     @literal << text(data, ts, te)
   end
 
   # Emits the literal run collected by calls to the append_literal method,
   # using the total start (ts) and end (te) offsets of the run.
-  def self.emit_literal
+  def emit_literal
     ts, te = @literal.first[1], @literal.last[2]
     text = @literal.map {|t| t[0]}.join
 
@@ -4314,43 +4326,34 @@ end
     emit(:literal, :literal, text, ts, te)
   end
 
-  def self.emit_options(text, ts, te)
-    if text =~ /\(\?([mixdau]+)?-?([mix]+)?:/
-      positive, negative = $1, $2
+  def emit_options(text, ts, te)
+    if text =~ /\(\?([mixdau]*)-?([mix]*)(:)?/
+      positive, negative, group_local = $1, $2, $3
 
-      if positive =~ /x/
+      if positive.include?('x')
         @free_spacing = true
       end
 
       # If the x appears in both, treat it like ruby does, the second cancels
       # the first.
-      if negative =~ /x/
+      if negative.include?('x')
         @free_spacing = false
       end
-    end
 
-    @in_options = true
-    @spacing_stack << [@free_spacing, @group_depth]
-
-    emit(:group, :options, text, ts, te)
-  end
-
-  # Emits an array with the details of the scanned pattern
-  def self.emit(type, token, text, ts, te)
-    #puts "EMIT: type: #{type}, token: #{token}, text: #{text}, ts: #{ts}, te: #{te}"
-
-    emit_literal if @literal
-
-    if @block
-      @block.call type, token, text, ts, te
+      if group_local
+        @spacing_stack << {:free_spacing => @free_spacing, :depth => @group_depth}
+      else
+        # switch for parent group level
+        @spacing_stack.last[:free_spacing] = @free_spacing
+      end
     end
 
-    @tokens << [type, token, text, ts, te]
+    emit(:group, :options, text, ts, te)
   end
 
   # Centralizes and unifies the handling of validation related
   # errors.
-  def self.validation_error(type, what, reason)
+  def validation_error(type, what, reason)
     case type
     when :group
       error = InvalidGroupError.new(what, reason)
@@ -4366,12 +4369,12 @@ end
   end
 
   # Used for references with an empty name or number
-  def self.empty_backref_error(type, what)
+  def empty_backref_error(type, what)
     validation_error(:backref, what, 'ref ID is empty')
   end
 
   # Used for named expressions with an empty name
-  def self.empty_name_error(type, what)
+  def empty_name_error(type, what)
     validation_error(type, what, 'name is empty')
   end