lib/mcollective/matcher/scanner.rb

   1 module MCollective
   2   module Matcher
   3     class Scanner
   4       attr_accessor :arguments, :token_index
   5
   6       def initialize(arguments)
   7         @token_index = 0
   8         @arguments = arguments.split("")
   9         @seperation_counter = 0
  10         @white_spaces = 0
  11       end
  12
  13       # Scans the input string and identifies single language tokens
  14       def get_token
  15         if @token_index >= @arguments.size
  16           return nil
  17         end
  18
  19         case @arguments[@token_index]
  20         when "("
  21           return "(", "("
  22
  23         when ")"
  24           return ")", ")"
  25
  26         when "n"
  27           if (@arguments[@token_index + 1] == "o") && (@arguments[@token_index + 2] == "t") && ((@arguments[@token_index + 3] == " ") || (@arguments[@token_index + 3] == "("))
  28             @token_index += 2
  29             return "not", "not"
  30           else
  31             gen_statement
  32           end
  33
  34         when "!"
  35           return "not", "not"
  36
  37         when "a"
  38           if (@arguments[@token_index + 1] == "n") && (@arguments[@token_index + 2] == "d") && ((@arguments[@token_index + 3] == " ") || (@arguments[@token_index + 3] == "("))
  39             @token_index += 2
  40             return "and", "and"
  41           else
  42             gen_statement
  43           end
  44
  45         when "o"
  46           if (@arguments[@token_index + 1] == "r") && ((@arguments[@token_index + 2] == " ") || (@arguments[@token_index + 2] == "("))
  47             @token_index += 1
  48             return "or", "or"
  49           else
  50             gen_statement
  51           end
  52
  53         when " "
  54           return " ", " "
  55
  56         else
  57           gen_statement
  58         end
  59       end
  60
  61       private
  62       # Helper generates a statement token
  63       def gen_statement
  64         func = false
  65         current_token_value = ""
  66         j = @token_index
  67
  68         begin
  69           if (@arguments[j] == "/")
  70             begin
  71               current_token_value << @arguments[j]
  72               j += 1
  73             end until (j >= @arguments.size) || (@arguments[j] =~ /\s/)
  74           elsif (@arguments[j] =~ /=|<|>/)
  75             while !(@arguments[j] =~ /=|<|>/)
  76               current_token_value << @arguments[j]
  77               j += 1
  78             end
  79
  80             current_token_value << @arguments[j]
  81             j += 1
  82
  83             if @arguments[j] == "/"
  84               begin
  85                 current_token_value << @arguments[j]
  86                 j += 1
  87                 if @arguments[j] == "/"
  88                   current_token_value << "/"
  89                   break
  90                 end
  91               end until (j >= @arguments.size) || (@arguments[j] =~ /\//)
  92             else
  93               while (j < @arguments.size) && ((@arguments[j] != " ") && (@arguments[j] != ")"))
  94                 current_token_value << @arguments[j]
  95                 j += 1
  96               end
  97             end
  98           else
  99             begin
 100               # Identify and tokenize regular expressions by ignoring everything between /'s
 101               if @arguments[j] == '/'
 102                 current_token_value << '/'
 103                 j+=1
 104                 while(j < @arguments.size && @arguments[j] != '/')
 105                   current_token_value << @arguments[j]
 106                   j += 1
 107                 end
 108                 current_token_value << @arguments[j] if @arguments[j]
 109                 break
 110               end
 111               if @arguments[j+1] == "("
 112                 func = true
 113                 be_greedy = true
 114               end
 115               current_token_value << @arguments[j]
 116               if be_greedy
 117                 while !(j+1 >= @arguments.size) && @arguments[j] != ')'
 118                   j += 1
 119                   current_token_value << @arguments[j]
 120                 end
 121                 j += 1
 122                 be_greedy = false
 123               else
 124                 j += 1
 125               end
 126               if(@arguments[j] == ' ')
 127                 break if(is_klass?(j) && !(@arguments[j-1] =~ /=|<|>/))
 128               end
 129               if( (@arguments[j] == ' ') && (@seperation_counter < 2) && !(current_token_value.match(/^.+(=|<|>).+$/)) )
 130                 if((index = lookahead(j)))
 131                   j = index
 132                 end
 133               end
 134             end until (j >= @arguments.size) || (@arguments[j] =~ /\s|\)/)
 135             @seperation_counter = 0
 136           end
 137         rescue Exception => e
 138           raise "An exception was raised while trying to tokenize '#{current_token_value} - #{e}'"
 139         end
 140
 141         @token_index += current_token_value.size + @white_spaces - 1
 142         @white_spaces = 0
 143
 144         # bar(
 145         if current_token_value.match(/.+?\($/)
 146           return "bad_token", [@token_index - current_token_value.size + 1, @token_index]
 147         # /foo/=bar
 148         elsif current_token_value.match(/^\/.+?\/(<|>|=).+/)
 149           return "bad_token", [@token_index - current_token_value.size + 1, @token_index]
 150         elsif current_token_value.match(/^.+?\/(<|>|=).+/)
 151           return "bad_token", [@token_index - current_token_value.size + 1, @token_index]
 152         else
 153           if func
 154             if current_token_value.match(/^.+?\((\s*(')[^']*(')\s*(,\s*(')[^']*('))*)?\)(\.[a-zA-Z0-9_]+)?((!=|<=|>=|=|>|<).+)?$/) ||
 155               current_token_value.match(/^.+?\((\s*(")[^"]*(")\s*(,\s*(")[^"]*("))*)?\)(\.[a-zA-Z0-9_]+)?((!=|<=|>=|=|>|<).+)?$/)
 156               return "fstatement", current_token_value
 157             else
 158               return "bad_token", [@token_index - current_token_value.size + 1, @token_index]
 159             end
 160           else
 161             slash_err = false
 162             current_token_value.split('').each do |c|
 163               if c == '/'
 164                 slash_err = !slash_err
 165               end
 166             end
 167             return "bad_token", [@token_index - current_token_value.size + 1, @token_index] if slash_err
 168             return "statement", current_token_value
 169           end
 170         end
 171       end
 172
 173       # Deal with special puppet class statement
 174       def is_klass?(j)
 175         while(j < @arguments.size && @arguments[j] == ' ')
 176           j += 1
 177         end
 178
 179         if @arguments[j] =~ /=|<|>/
 180           return false
 181         else
 182           return true
 183         end
 184       end
 185
 186       # Eat spaces while looking for the next comparison symbol
 187       def lookahead(index)
 188         index += 1
 189         while(index <= @arguments.size)
 190           @white_spaces += 1
 191           unless(@arguments[index] =~ /\s/)
 192             @seperation_counter +=1
 193             return index
 194           end
 195           index += 1
 196         end
 197         return nil
 198       end
 199     end
 200   end
 201 end