jasherai-oniguruma 1.1.1

data/lib/oniguruma.rb

@@ -0,0 +1,479 @@
+require 'oregexp'
+
+module Oniguruma
+   OPTION_NONE                 = 0
+   OPTION_IGNORECASE           = 1
+   OPTION_EXTEND               = (OPTION_IGNORECASE         << 1)
+   OPTION_MULTILINE            = (OPTION_EXTEND             << 1)
+   OPTION_SINGLELINE           = (OPTION_MULTILINE          << 1)
+   OPTION_FIND_LONGEST         = (OPTION_SINGLELINE         << 1)
+   OPTION_FIND_NOT_EMPTY       = (OPTION_FIND_LONGEST       << 1)
+   OPTION_NEGATE_SINGLELINE    = (OPTION_FIND_NOT_EMPTY     << 1)
+   OPTION_DONT_CAPTURE_GROUP   = (OPTION_NEGATE_SINGLELINE  << 1)
+   OPTION_CAPTURE_GROUP        = (OPTION_DONT_CAPTURE_GROUP << 1)
+   OPTION_NOTBOL               = (OPTION_CAPTURE_GROUP << 1)
+   OPTION_NOTEOL               = (OPTION_NOTBOL << 1)
+   OPTION_POSIX_REGION         = (OPTION_NOTEOL << 1)
+   OPTION_MAXBIT               = OPTION_POSIX_REGION
+   OPTION_DEFAULT              = OPTION_NONE
+
+   OPTIONS_SHORTCUTS = {
+    'i' => OPTION_IGNORECASE,
+    'x' => OPTION_EXTEND,
+    'm' => OPTION_MULTILINE,
+    's' => OPTION_SINGLELINE,
+    'l' => OPTION_FIND_LONGEST,
+    'E' => OPTION_FIND_NOT_EMPTY,
+    'S' => OPTION_NEGATE_SINGLELINE,
+    'G' => OPTION_DONT_CAPTURE_GROUP,
+    'g' => OPTION_CAPTURE_GROUP,
+    'B' => OPTION_NOTBOL,
+    'E' => OPTION_NOTEOL,
+   }
+
+   SYNTAX_ASIS                 = 0
+   SYNTAX_POSIX_BASIC          = 1
+   SYNTAX_POSIX_EXTENDED       = 2
+   SYNTAX_EMACS                = 3
+   SYNTAX_GREP                 = 4
+   SYNTAX_GNU_REGEX            = 5
+   SYNTAX_JAVA                 = 6
+   SYNTAX_PERL                 = 7
+   SYNTAX_PERL_NG              = 8
+   SYNTAX_RUBY                 = 9
+   SYNTAX_DEFAULT              = 10
+
+   ENCODING_ASCII              = 0
+   ENCODING_ISO_8859_1         = 1
+   ENCODING_ISO_8859_2         = 2
+   ENCODING_ISO_8859_3         = 3
+   ENCODING_ISO_8859_4         = 4
+   ENCODING_ISO_8859_5         = 5
+   ENCODING_ISO_8859_6         = 6
+   ENCODING_ISO_8859_7         = 7
+   ENCODING_ISO_8859_8         = 8
+   ENCODING_ISO_8859_9         = 9
+   ENCODING_ISO_8859_10        = 10
+   ENCODING_ISO_8859_11        = 11
+   ENCODING_ISO_8859_12        = 12
+   ENCODING_ISO_8859_13        = 13
+   ENCODING_ISO_8859_14        = 14
+   ENCODING_ISO_8859_15        = 15
+   ENCODING_ISO_8859_16        = 16
+   ENCODING_UTF8               = 17
+   ENCODING_UTF16_BE           = 18
+   ENCODING_UTF16_LE           = 19
+   ENCODING_UTF32_BE           = 20
+   ENCODING_UTF32_LE           = 21
+   ENCODING_EUC_JP             = 22
+   ENCODING_EUC_TW             = 23
+   ENCODING_EUC_KR             = 24
+   ENCODING_EUC_CN             = 25
+   ENCODING_SJIS               = 26
+   ENCODING_KOI8               = 27
+   ENCODING_KOI8_R             = 28
+   ENCODING_CP1251             = 29
+   ENCODING_BIG5               = 30
+   ENCODING_GB18030            = 31
+   ENCODING_UNDEF              = 32
+
+   
+   class ORegexp
+   
+      class << self
+         # :stopdoc:
+         alias compile new
+         # :startdoc:
+      
+         # call-seq:
+         # ORegexp.escape(str)   => a_str
+         # ORegexp.quote(str)    => a_str
+         # 
+         # Escapes any characters that would have special meaning in a regular
+         # expression. Returns a new escaped string, or self if no characters are
+         # escaped.  For any string,
+         # <code>Regexp.escape(<i>str</i>)=~<i>str</i></code> will be true.
+         # 
+         #    ORegexp.escape('\\*?{}.')   #=> \\\\\*\?\{\}\.
+         # 
+         
+         def escape( *args )
+            Regexp.escape( *args )
+         end
+         # :stopdoc:
+         alias quote escape
+         # :startdoc:
+         
+         # call-seq:
+         #    ORegexp.last_match           => matchdata
+         #    ORegexp.last_match(fixnum)   => str
+         #  
+         # The first form returns the <code>MatchData</code> object generated by the
+         # last successful pattern match. The second form returns the nth field in this
+         # <code>MatchData</code> object.
+         #     
+         #    ORegexp.new( 'c(.)t' ) =~ 'cat'       #=> 0
+         #    ORegexp.last_match                    #=> #<MatchData:0x401b3d30>
+         #    ORegexp.last_match(0)                 #=> "cat"
+         #    ORegexp.last_match(1)                 #=> "a"
+         #    ORegexp.last_match(2)                 #=> nil
+         
+         def last_match( index = nil)
+            if index
+               @@last_match[index]
+            else
+               @@last_match
+            end
+         end
+      end
+      
+      # :stopdoc:
+      alias old_initialize initialize 
+      # :startdoc:
+      
+      # call-seq:
+      #     ORegexp.new( pattern, options_hash )
+      #     ORegexp.new( pattern, option_str, encoding_str=nil, syntax_str=nil) 
+      #
+      # Constructs a new regular expression from <i>pattern</i>, which is a 
+      # <code>String</code>. The second parameter <i></i> may be a <code>Hash</code> 
+      # of the form:
+      #
+      # <code>{ :options => option_value, :encoding => encoding_value, :syntax => syntax_value }</code>
+      #
+      # Where <code>option_value</code> is a bitwise <code>OR</code> of 
+      # <code>Oniguruma::OPTION_XXX</code> constants; <code>encoding_value</code>
+      # is one of <code>Oniguruma::ENCODING_XXX</code> constants; and 
+      # <code>syntax_value</code> is one of <code>Oniguruma::SYNTAX_XXX</code>
+      # constants.
+      #  
+      #     r1 = ORegexp.new('^a-z+:\\s+\w+')                                            #=> /^a-z+:\s+\w+/
+      #     r2 = ORegexp.new('cat', :options => OPTION_IGNORECASE )                      #=> /cat/i
+      #     r3 = ORegexp.new('dog', :options => OPTION_EXTEND )                          #=> /dog/x
+      #
+      #     #Accept java syntax on SJIS encoding:
+      #     r4 = ORegexp.new('ape', :syntax  => SYNTAX_JAVA, :encoding => ENCODING_SJIS) #=> /ape/ 
+      #    
+      # Second form uses string shortcuts to set options and encoding:
+      #     r = ORegexp.new('cat', 'i', 'utf8', 'java')
+      
+      def initialize( pattern, *args )
+         defaults = { :options => OPTION_DEFAULT, :encoding => ENCODING_ASCII, :syntax => SYNTAX_DEFAULT}
+	 if args[0].is_a?(String)
+           options = {}
+	   option_str, encoding_str, syntax_str = *args
+	   opt = 0
+	   option_str.each_byte {|x| opt |= (OPTIONS_SHORTCUTS[x.chr] || 0) }
+	   options[:options] = opt
+	   if encoding_str && Oniguruma::const_defined?("ENCODING_#{encoding_str.upcase}")
+             options[:encoding] = Oniguruma::const_get("ENCODING_#{encoding_str.upcase}")
+	   end
+	   if syntax_str && Oniguruma::const_defined?("SYNTAX_#{syntax_str.upcase}")
+             options[:syntax] = Oniguruma::const_get("SYNTAX_#{syntax_str.upcase}")
+	   end
+	 else
+           options = args[0] || {}
+	 end
+         old_initialize( pattern,  defaults.merge( options ).freeze )
+      end
+      
+      # call-seq:
+      #    rxp == other_rxp      => true or false
+      #    rxp.eql?(other_rxp)   => true or false
+      #  
+      # Equality---Two regexps are equal if their patterns are identical, they have
+      # the same character set code, and their <code>#casefold?</code> values are the
+      # same.
+      
+      def == regexp
+         @pattern == regexp.source && kcode == regexp.kcode && casefold? == regexp.casefold?
+      end
+      alias eql? ==
+      
+      # call-seq:
+      #    rxp.casefold?   => true of false
+      #
+      # Returns the value of the case-insensitive flag.
+      
+      def casefold?
+         (@options[:options] & OPTION_IGNORECASE) > 0
+      end
+      
+      # call-seq:
+      #    rxp.kode        => int
+      #
+      # Returns the character set code for the regexp.
+      def kcode
+         @options[:encoding]
+      end
+      
+      # call-seq:
+      #    rxp.options   => fixnum
+      # 
+      # Returns the set of bits corresponding to the options used when creating this
+      # ORegexp (see <code>ORegexp::new</code> for details. Note that additional bits
+      # may be set in the returned options: these are used internally by the regular
+      # expression code. These extra bits are ignored if the options are passed to
+      # <code>ORegexp::new</code>.
+      #    
+      #    Oniguruma::OPTION_IGNORECASE                                 #=> 1
+      #    Oniguruma::OPTION_EXTEND                                     #=> 2
+      #    Oniguruma::OPTION_MULTILINE                                  #=> 4
+      #    
+      #    Regexp.new(r.source, :options => Oniguruma::OPTION_EXTEND )  #=> 2
+      
+      def options
+         @options[:options]
+      end
+      
+      # call-seq:
+      #    rxp.to_s   => str
+      # 
+      # Returns a string containing the regular expression and its options (using the
+      # <code>(?xxx:yyy)</code> notation. This string can be fed back in to
+      # <code>Regexp::new</code> to a regular expression with the same semantics as
+      # the original. (However, <code>Regexp#==</code> may not return true when
+      # comparing the two, as the source of the regular expression itself may
+      # differ, as the example shows).  <code>Regexp#inspect</code> produces a
+      # generally more readable version of <i>rxp</i>.
+      #    
+      #    r1 = ORegexp.new( 'ab+c', :options OPTION_IGNORECASE | OPTION_EXTEND ) #=> /ab+c/ix
+      #    s1 = r1.to_s                                                           #=> "(?ix-m:ab+c)"
+      #    r2 = ORegexp.new(s1)                                                   #=> /(?ix-m:ab+c)/
+      #    r1 == r2                                                               #=> false
+      #    r1.source                                                              #=> "ab+c"
+      #    r2.source                                                              #=> "(?ix-m:ab+c)"      
+      
+      def to_s
+         opt_str = "(?"
+         opt_str += "i" if (@options[:options] & OPTION_IGNORECASE) > 0
+         opt_str += "m" if (@options[:options] & OPTION_MULTILINE) > 0
+         opt_str += "x" if (@options[:options] & OPTION_EXTEND) > 0
+         unless opt_str == "(?imx"
+            opt_str += "-"
+            opt_str += "i" if (@options[:options] & OPTION_IGNORECASE) == 0
+            opt_str += "m" if (@options[:options] & OPTION_MULTILINE) == 0
+            opt_str += "x" if (@options[:options] & OPTION_EXTEND) == 0
+         end
+         opt_str += ")"
+         opt_str + @pattern
+      end
+      
+      
+      # call-seq:
+      #    rxp.inspect   => string
+      #
+      # Returns a readable version of <i>rxp</i>
+      #
+      #    ORegexp.new( 'cat', :options => OPTION_MULTILINE | OPTION_IGNORECASE ).inspect  => /cat/im
+      #    ORegexp.new( 'cat', :options => OPTION_MULTILINE | OPTION_IGNORECASE ).to_s     => (?im-x)cat
+      
+      def inspect
+         opt_str = ""
+         opt_str += "i" if (@options[:options] & OPTION_IGNORECASE) > 0
+         opt_str += "m" if (@options[:options] & OPTION_MULTILINE) > 0
+         opt_str += "x" if (@options[:options] & OPTION_EXTEND) > 0
+         "/" + @pattern + "/" + opt_str
+      end
+      
+      # call-seq:
+      #    rxp.source   => str
+      # 
+      # Returns the original string of the pattern.
+      # 
+      #    ORegex.new( 'ab+c', 'ix' ).source   #=> "ab+c"
+      def source
+         @pattern.freeze
+      end
+      
+      alias match_all scan
+      
+   end
+   
+end
+
+class ::String
+   # Calls <code>Oniguruma::ORegexp#gsub</code> on this string.
+   def ogsub(*args)
+      Oniguruma::ORegexp.new(args.shift).gsub(self, *args)
+   end
+   
+   # Calls <code>Oniguruma::ORegexp#gsub!</code> on this string.
+   def ogsub!(*args)
+      Oniguruma::ORegexp.new(args.shift).gsub!(self, *args)
+   end
+   
+   # Calls <code>Oniguruma::ORegexp#sub</code> on this string.
+   def osub(re, *args)
+      Oniguruma::ORegexp.new( re ).sub(self, *args)
+   end
+   
+   # Calls <code>Oniguruma::ORegexp#sub!</code> on this string.
+   def osub!(re, *args)
+      Oniguruma::ORegexp.new( re ).sub(self, *args)
+   end
+end
+
+class ::MatchData
+   # call-seq:
+   #    to_index[symbol]      => int or nil
+   # 
+   # Returns the group index for the corresponding named group, or 
+   # <code>nil</code> if the group does not exist.
+   #    
+   #    m = ORegexp.new( '(?<begin>^.*?)(?<middle>\d)(?<end>.*)' ).match("THX1138")
+   #    m.to_index[:begin]    #=> 1
+   #    m.to_index[:unknown]  #=> nil
+   def to_index symbol
+      @named_captures && @named_captures[symbol]
+   end
+
+   alias old_aref :[]
+   
+   # call-seq:
+   #    mtch[i]               => obj
+   #    mtch[start, length]   => array
+   #    mtch[range]           => array
+   #    mtch[symbol]          => obj
+   # 
+   # <code>MatchData</code> acts as an array, and may be
+   # accessed using the normal array indexing techniques.  <i>mtch</i>[0] is
+   # equivalent to the special variable <code>$&</code>, and returns the entire
+   # matched string.  <i>mtch</i>[1], <i>mtch</i>[2], and so on return the values
+   # of the matched backreferences (portions of the pattern between parentheses).
+   #    
+   #    m = ORegexp.new( '(.)(.)(\d+)(\d)' ).match("THX1138.")
+   #    m[0]       #=> "HX1138"
+   #    m[1, 2]    #=> ["H", "X"]
+   #    m[1..3]    #=> ["H", "X", "113"]
+   #    m[-3, 2]   #=> ["X", "113"]
+   #
+   # If a symbol is used as index, the corresponding named group is returned,
+   # or <code>nil</code> if such a group does not exist.
+   #
+   #    m = ORegexp.new( '(?<begin>^.*?)(?<middle>\d)(?<end>.*)' ).match("THX1138")
+   #    m[:begin]  #=> "THX"
+   #    m[:moddle]  #=> "1"
+   #    m[:end]  #=> "138"
+   
+   def [](*idx)
+      if idx[0].is_a?(Symbol) 
+         k = to_index( idx[0] )
+         k && old_aref(k)
+       else
+         old_aref(*idx)
+       end
+   end
+   
+   alias old_begin :begin
+   
+   # call-seq:
+   #    mtch.begin(n)        => integer
+   #    mtch.begin           => integer
+   #    mtch.begin(symbol)   => integer
+   # 
+   # Returns the offset of the start of the <em>n</em>th element of the match
+   # array in the string.
+   #    
+   #    m = ORegexp.new( '(.)(.)(\d+)(\d)' ).match("THX1138.")
+   #    m.begin(0)   #=> 1
+   #    m.begin(2)   #=> 2
+   #
+   # If no arguments are given, the index of the
+   # first matching character is returned.
+   #
+   #    m = ORegexp.new( '(.)(.)(\d+)(\d)' ).match("THX1138.")
+   #    m.begin      #=> 1
+   #
+   # If the argument is a symbol, then the beginning of the 
+   # corresponding named group is returned, or <code>nil</code>
+   # if the group does not exist.
+   #
+   #    m = ORegexp.new( '(?<begin>^.*?)(?<middle>\d)(?<end>.*)' ).match("THX1138")
+   #    m.begin(:middle) #=> 3
+   
+   def begin(*idx)
+      if idx[0].is_a?(Symbol) 
+         k = to_index( idx[0] )
+         k && old_begin(k)
+      elsif idx.empty?
+         old_begin( 0 )
+      else
+         old_begin(*idx)
+      end
+   end
+   
+   alias old_end :end
+   
+   # call-seq:
+   #    mtch.end(n)   => integer
+   # 
+   # Returns the offset of the character immediately following the end of the
+   # <em>n</em>th element of the match array in the string.
+   #    
+   #    m = ORegexp.new( '(.)(.)(\d+)(\d)' ).match("THX1138.")
+   #    m.end(0)   #=> 7
+   #    m.end(2)   #=> 3
+   #
+   # If no arguments are given, the index of the
+   # last matching character is returned.
+   #
+   #    m = ORegexp.new( '(.)(.)(\d+)(\d)' ).match("THX1138.")
+   #    m.last      #=> 7
+   #
+   # If the argument is a symbol, then the beginning of the 
+   # corresponding named group is returned, or <code>nil</code>
+   # if the group does not exist.
+   #
+   #    m = ORegexp.new( '(?<begin>^.*?)(?<middle>\d)(?<end>.*)' ).match("THX1138")
+   #    m.end(:middle) #=> 4
+   
+   def end(*idx)
+      if idx[0].is_a?(Symbol) 
+         k = to_index( idx[0] )
+         k && old_end(k)
+      elsif idx.empty?
+         old_end( 0 )
+      else
+         old_end(*idx)
+      end
+   end
+   
+   alias old_offset :offset
+   
+   # call-seq:
+   #    mtch.offset(n)      => array
+   #    mtch.offset         => array
+   #    mtch.offset(symbol) => array
+   # 
+   # Returns a two-element array containing the beginning and ending offsets of
+   # the <em>n</em>th match.
+   #    
+   #    m = ORegexp.new( '(.)(.)(\d+)(\d)' ).match("THX1138.")
+   #    m.offset(0)   #=> [1, 7]
+   #    m.offset(4)   #=> [6, 7]   
+   #
+   # If no arguments are given, the offsets of the entire
+   # sequence are returned.
+   #
+   #    m = ORegexp.new( '(.)(.)(\d+)(\d)' ).match("THX1138.")
+   #    m.offset      #=> [1, 7]
+   #
+   # If the argument is a symbol, then the offsets of the 
+   # corresponding named group are returned, or <code>nil</code>
+   # if the group does not exist.
+   #
+   #    m = ORegexp.new( '(?<begin>^.*?)(?<middle>\d)(?<end>.*)' ).match("THX1138")
+   #    m.end(:middle) #=> [3, 4]
+   
+   def offset(*idx)
+      if idx[0].is_a?(Symbol) 
+         k = to_index( idx[0] )
+         k && old_offset(k)
+      elsif idx.empty?
+         old_offset( 0 )
+      else
+         old_offset(*idx)
+      end
+   end
+end