vendor/rails/actionmailer/lib/action_mailer/vendor/tmail-1.2.3/tmail/scanner_r.rb

   1 # scanner_r.rb
   2 #
   3 #--
   4 # Copyright (c) 1998-2003 Minero Aoki <aamine@loveruby.net>
   5 #
   6 # Permission is hereby granted, free of charge, to any person obtaining
   7 # a copy of this software and associated documentation files (the
   8 # "Software"), to deal in the Software without restriction, including
   9 # without limitation the rights to use, copy, modify, merge, publish,
  10 # distribute, sublicense, and/or sell copies of the Software, and to
  11 # permit persons to whom the Software is furnished to do so, subject to
  12 # the following conditions:
  13 #
  14 # The above copyright notice and this permission notice shall be
  15 # included in all copies or substantial portions of the Software.
  16 #
  17 # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  18 # EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
  19 # MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
  20 # NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
  21 # LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
  22 # OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
  23 # WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
  24 #
  25 # Note: Originally licensed under LGPL v2+. Using MIT license for Rails
  26 # with permission of Minero Aoki.
  27 #++
  28 #:stopdoc:
  29 require 'tmail/config'
  30
  31 module TMail
  32
  33   class TMailScanner
  34
  35     Version = '1.2.3'
  36     Version.freeze
  37
  38     MIME_HEADERS = {
  39       :CTYPE        => true,
  40       :CENCODING    => true,
  41       :CDISPOSITION => true
  42     }
  43
  44     alnum      = 'a-zA-Z0-9'
  45     atomsyms   = %q[  _#!$%&`'*+-{|}~^/=?  ].strip
  46     tokensyms  = %q[  _#!$%&`'*+-{|}~^@.    ].strip
  47     atomchars  = alnum + Regexp.quote(atomsyms)
  48     tokenchars = alnum + Regexp.quote(tokensyms)
  49     iso2022str = '\e(?!\(B)..(?:[^\e]+|\e(?!\(B)..)*\e\(B'
  50
  51     eucstr  = "(?:[\xa1-\xfe][\xa1-\xfe])+"
  52     sjisstr = "(?:[\x81-\x9f\xe0-\xef][\x40-\x7e\x80-\xfc])+"
  53     utf8str = "(?:[\xc0-\xdf][\x80-\xbf]|[\xe0-\xef][\x80-\xbf][\x80-\xbf])+"
  54
  55     quoted_with_iso2022  = /\A(?:[^\\\e"]+|#{iso2022str})+/n
  56     domlit_with_iso2022  = /\A(?:[^\\\e\]]+|#{iso2022str})+/n
  57     comment_with_iso2022 = /\A(?:[^\\\e()]+|#{iso2022str})+/n
  58
  59     quoted_without_iso2022  = /\A[^\\"]+/n
  60     domlit_without_iso2022  = /\A[^\\\]]+/n
  61     comment_without_iso2022 = /\A[^\\()]+/n
  62
  63     PATTERN_TABLE = {}
  64     PATTERN_TABLE['EUC'] =
  65       [
  66         /\A(?:[#{atomchars}]+|#{iso2022str}|#{eucstr})+/n,
  67         /\A(?:[#{tokenchars}]+|#{iso2022str}|#{eucstr})+/n,
  68         quoted_with_iso2022,
  69         domlit_with_iso2022,
  70         comment_with_iso2022
  71       ]
  72     PATTERN_TABLE['SJIS'] =
  73       [
  74         /\A(?:[#{atomchars}]+|#{iso2022str}|#{sjisstr})+/n,
  75         /\A(?:[#{tokenchars}]+|#{iso2022str}|#{sjisstr})+/n,
  76         quoted_with_iso2022,
  77         domlit_with_iso2022,
  78         comment_with_iso2022
  79       ]
  80     PATTERN_TABLE['UTF8'] =
  81       [
  82         /\A(?:[#{atomchars}]+|#{utf8str})+/n,
  83         /\A(?:[#{tokenchars}]+|#{utf8str})+/n,
  84         quoted_without_iso2022,
  85         domlit_without_iso2022,
  86         comment_without_iso2022
  87       ]
  88     PATTERN_TABLE['NONE'] =
  89       [
  90         /\A[#{atomchars}]+/n,
  91         /\A[#{tokenchars}]+/n,
  92         quoted_without_iso2022,
  93         domlit_without_iso2022,
  94         comment_without_iso2022
  95       ]
  96
  97
  98     def initialize( str, scantype, comments )
  99       init_scanner str
 100       @comments = comments || []
 101       @debug    = false
 102
 103       # fix scanner mode
 104       @received  = (scantype == :RECEIVED)
 105       @is_mime_header = MIME_HEADERS[scantype]
 106
 107       atom, token, @quoted_re, @domlit_re, @comment_re = PATTERN_TABLE[TMail.KCODE]
 108       @word_re = (MIME_HEADERS[scantype] ? token : atom)
 109     end
 110
 111     attr_accessor :debug
 112
 113     def scan( &block )
 114       if @debug
 115         scan_main do |arr|
 116           s, v = arr
 117           printf "%7d %-10s %s\n",
 118                  rest_size(),
 119                  s.respond_to?(:id2name) ? s.id2name : s.inspect,
 120                  v.inspect
 121           yield arr
 122         end
 123       else
 124         scan_main(&block)
 125       end
 126     end
 127
 128     private
 129
 130     RECV_TOKEN = {
 131       'from' => :FROM,
 132       'by'   => :BY,
 133       'via'  => :VIA,
 134       'with' => :WITH,
 135       'id'   => :ID,
 136       'for'  => :FOR
 137     }
 138
 139     def scan_main
 140       until eof?
 141         if skip(/\A[\n\r\t ]+/n)   # LWSP
 142           break if eof?
 143         end
 144
 145         if s = readstr(@word_re)
 146           if @is_mime_header
 147             yield [:TOKEN, s]
 148           else
 149             # atom
 150             if /\A\d+\z/ === s
 151               yield [:DIGIT, s]
 152             elsif @received
 153               yield [RECV_TOKEN[s.downcase] || :ATOM, s]
 154             else
 155               yield [:ATOM, s]
 156             end
 157           end
 158
 159         elsif skip(/\A"/)
 160           yield [:QUOTED, scan_quoted_word()]
 161
 162         elsif skip(/\A\[/)
 163           yield [:DOMLIT, scan_domain_literal()]
 164
 165         elsif skip(/\A\(/)
 166           @comments.push scan_comment()
 167
 168         else
 169           c = readchar()
 170           yield [c, c]
 171         end
 172       end
 173
 174       yield [false, '$']
 175     end
 176
 177     def scan_quoted_word
 178       scan_qstr(@quoted_re, /\A"/, 'quoted-word')
 179     end
 180
 181     def scan_domain_literal
 182       '[' + scan_qstr(@domlit_re, /\A\]/, 'domain-literal') + ']'
 183     end
 184
 185     def scan_qstr( pattern, terminal, type )
 186       result = ''
 187       until eof?
 188         if    s = readstr(pattern) then result << s
 189         elsif skip(terminal)       then return result
 190         elsif skip(/\A\\/)         then result << readchar()
 191         else
 192           raise "TMail FATAL: not match in #{type}"
 193         end
 194       end
 195       scan_error! "found unterminated #{type}"
 196     end
 197
 198     def scan_comment
 199       result = ''
 200       nest = 1
 201       content = @comment_re
 202
 203       until eof?
 204         if s = readstr(content) then result << s
 205         elsif skip(/\A\)/)      then nest -= 1
 206                                      return result if nest == 0
 207                                      result << ')'
 208         elsif skip(/\A\(/)      then nest += 1
 209                                      result << '('
 210         elsif skip(/\A\\/)      then result << readchar()
 211         else
 212           raise 'TMail FATAL: not match in comment'
 213         end
 214       end
 215       scan_error! 'found unterminated comment'
 216     end
 217
 218     # string scanner
 219
 220     def init_scanner( str )
 221       @src = str
 222     end
 223
 224     def eof?
 225       @src.empty?
 226     end
 227
 228     def rest_size
 229       @src.size
 230     end
 231
 232     def readstr( re )
 233       if m = re.match(@src)
 234         @src = m.post_match
 235         m[0]
 236       else
 237         nil
 238       end
 239     end
 240
 241     def readchar
 242       readstr(/\A./)
 243     end
 244
 245     def skip( re )
 246       if m = re.match(@src)
 247         @src = m.post_match
 248         true
 249       else
 250         false
 251       end
 252     end
 253
 254     def scan_error!( msg )
 255       raise SyntaxError, msg
 256     end
 257
 258   end
 259
 260 end   # module TMail
 261 #:startdoc: