Module: Mail::Utilities

Extended by:
Utilities
Included in:
Encodings, Utilities
Defined in:
lib/mail/utilities.rb

Defined Under Namespace

Classes: BestEffortCharsetEncoder, StrictCharsetEncoder

Constant Summary collapse

TO_CRLF_REGEX =
Regexp.new("(?<!\r)\n|\r(?!\n)")

Class Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Class Attribute Details

.charset_encoderObject

Returns the value of attribute charset_encoder.


334
335
336
# File 'lib/mail/utilities.rb', line 334

def charset_encoder
  @charset_encoder
end

Class Method Details

.b_value_decode(str) ⇒ Object


414
415
416
417
418
419
420
421
422
423
424
425
# File 'lib/mail/utilities.rb', line 414

def Utilities.b_value_decode(str)
  match = str.match(/\=\?(.+)?\?[Bb]\?(.*)\?\=/m)
  if match
    charset = match[1]
    str = Utilities.decode_base64(match[2])
    str = charset_encoder.encode(str, charset)
  end
  transcode_to_scrubbed_utf8(str)
rescue Encoding::UndefinedConversionError, ArgumentError, Encoding::ConverterNotFoundError, Encoding::InvalidByteSequenceError
  warn "Encoding conversion failed #{$!}"
  str.dup.force_encoding(Encoding::UTF_8)
end

.b_value_encode(str, encoding = nil) ⇒ Object


409
410
411
412
# File 'lib/mail/utilities.rb', line 409

def Utilities.b_value_encode(str, encoding = nil)
  encoding = str.encoding.to_s
  [Utilities.encode_base64(str), encoding]
end

.binary_unsafe_to_crlf(string) ⇒ Object

:nodoc:


243
244
245
# File 'lib/mail/utilities.rb', line 243

def self.binary_unsafe_to_crlf(string) #:nodoc:
  string.gsub(TO_CRLF_REGEX, Constants::CRLF)
end

.binary_unsafe_to_lf(string) ⇒ Object

:nodoc:


233
234
235
# File 'lib/mail/utilities.rb', line 233

def self.binary_unsafe_to_lf(string) #:nodoc:
  string.gsub(/\r\n|\r/, Constants::LF)
end

.bracket(str) ⇒ Object


356
357
358
359
360
# File 'lib/mail/utilities.rb', line 356

def Utilities.bracket( str )
  str = ::Mail::Utilities.unbracket( str )
  str = escape_bracket( str )
  '<' + str + '>'
end

.decode_base64(str) ⇒ Object


362
363
364
365
366
367
# File 'lib/mail/utilities.rb', line 362

def Utilities.decode_base64(str)
  if !str.end_with?("=") && str.length % 4 != 0
    str = str.ljust((str.length + 3) & ~3, "=")
  end
  str.unpack( 'm' ).first
end

.decode_utf7(utf7) ⇒ Object


399
400
401
402
403
404
405
406
407
# File 'lib/mail/utilities.rb', line 399

def Utilities.decode_utf7(utf7)
  utf7.gsub(/&([^-]+)?-/n) do
    if $1
      ($1.tr(",", "/") + "===").unpack("m")[0].encode(Encoding::UTF_8, Encoding::UTF_16BE)
    else
      "&"
    end
  end
end

.encode_base64(str) ⇒ Object


369
370
371
# File 'lib/mail/utilities.rb', line 369

def Utilities.encode_base64(str)
  [str].pack( 'm' )
end

.encode_utf7(string) ⇒ Object

From Ruby stdlib Net::IMAP


388
389
390
391
392
393
394
395
396
397
# File 'lib/mail/utilities.rb', line 388

def Utilities.encode_utf7(string)
  string.gsub(/(&)|[^\x20-\x7e]+/) do
    if $1
      "&-"
    else
      base64 = [$&.encode(Encoding::UTF_16BE)].pack("m0")
      "&" + base64.delete("=").tr("/", ",") + "-"
    end
  end.force_encoding(Encoding::ASCII_8BIT)
end

.escape_bracket(str) ⇒ Object


351
352
353
354
# File 'lib/mail/utilities.rb', line 351

def Utilities.escape_bracket( str )
  re = /(?<!\\)([\<\>])/          # Only match unescaped brackets
  str.gsub(re) { |s| '\\' + s }
end

.escape_paren(str) ⇒ Object

Escapes any parenthesis in a string that are unescaped this uses a Ruby 1.9.1 regexp feature of negative look behind


340
341
342
343
# File 'lib/mail/utilities.rb', line 340

def Utilities.escape_paren( str )
  re = /(?<!\\)([\(\)])/          # Only match unescaped parens
  str.gsub(re) { |s| '\\' + s }
end

.get_constant(klass, string) ⇒ Object


377
378
379
# File 'lib/mail/utilities.rb', line 377

def Utilities.get_constant(klass, string)
  klass.const_get( string )
end

.has_constant?(klass, string) ⇒ Boolean

Returns:

  • (Boolean)

373
374
375
# File 'lib/mail/utilities.rb', line 373

def Utilities.has_constant?(klass, string)
  klass.const_defined?( string, false )
end

.param_decode(str, encoding) ⇒ Object


451
452
453
454
455
456
457
458
# File 'lib/mail/utilities.rb', line 451

def Utilities.param_decode(str, encoding)
  str = uri_parser.unescape(str)
  str = charset_encoder.encode(str, encoding) if encoding
  transcode_to_scrubbed_utf8(str)
rescue Encoding::UndefinedConversionError, ArgumentError, Encoding::ConverterNotFoundError
  warn "Encoding conversion failed #{$!}"
  str.dup.force_encoding(Encoding::UTF_8)
end

.param_encode(str) ⇒ Object


460
461
462
463
464
# File 'lib/mail/utilities.rb', line 460

def Utilities.param_encode(str)
  encoding = str.encoding.to_s.downcase
  language = Configuration.instance.param_encode_language
  "#{encoding}'#{language}'#{uri_parser.escape(str)}"
end

.paren(str) ⇒ Object


345
346
347
348
349
# File 'lib/mail/utilities.rb', line 345

def Utilities.paren( str )
  str = ::Mail::Utilities.unparen( str )
  str = escape_paren( str )
  '(' + str + ')'
end

.pick_encoding(charset) ⇒ Object

Pick a Ruby encoding corresponding to the message charset. Most charsets have a Ruby encoding, but some need manual aliasing here.

TODO: add this as a test somewhere:

Encoding.list.map { |e| [e.to_s.upcase == pick_encoding(e.to_s.downcase.gsub("-", "")), e.to_s] }.select {|a,b| !b}
Encoding.list.map { |e| [e.to_s == pick_encoding(e.to_s), e.to_s] }.select {|a,b| !b}

476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
# File 'lib/mail/utilities.rb', line 476

def Utilities.pick_encoding(charset)
  charset = charset.to_s
  encoding = case charset.downcase

  # ISO-8859-8-I etc. http://en.wikipedia.org/wiki/ISO-8859-8-I
  when /^iso[-_]?8859-(\d+)(-i)?$/
    "ISO-8859-#{$1}"

  # ISO-8859-15, ISO-2022-JP and alike
  when /^iso[-_]?(\d{4})-?(\w{1,2})$/
    "ISO-#{$1}-#{$2}"

  # "ISO-2022-JP-KDDI"  and alike
  when /^iso[-_]?(\d{4})-?(\w{1,2})-?(\w*)$/
    "ISO-#{$1}-#{$2}-#{$3}"

  # UTF-8, UTF-32BE and alike
  when /^utf[\-_]?(\d{1,2})?(\w{1,2})$/
    "UTF-#{$1}#{$2}".gsub(/\A(UTF-(?:16|32))\z/, '\\1BE')

  # Windows-1252 and alike
  when /^windows-?(.*)$/
    "Windows-#{$1}"

  when '8bit'
    Encoding::ASCII_8BIT

  # alternatives/misspellings of us-ascii seen in the wild
  when /^iso[-_]?646(-us)?$/, 'us=ascii'
    Encoding::ASCII

  # Microsoft-specific alias for MACROMAN
  when 'macintosh'
    Encoding::MACROMAN

  # Microsoft-specific alias for CP949 (Korean)
  when 'ks_c_5601-1987'
    Encoding::CP949

  # Wrongly written Shift_JIS (Japanese)
  when 'shift-jis'
    Encoding::Shift_JIS

  # GB2312 (Chinese charset) is a subset of GB18030 (its replacement)
  when 'gb2312'
    Encoding::GB18030

  when 'cp-850'
    Encoding::CP850

  when 'latin2'
    Encoding::ISO_8859_2

  else
    charset
  end

  convert_to_encoding(encoding)
end

.q_value_decode(str) ⇒ Object


432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
# File 'lib/mail/utilities.rb', line 432

def Utilities.q_value_decode(str)
  match = str.match(/\=\?(.+)?\?[Qq]\?(.*)\?\=/m)
  if match
    charset = match[1]
    string = match[2].gsub(/_/, '=20')
    # Remove trailing = if it exists in a Q encoding
    string = string.sub(/\=$/, '')
    str = Encodings::QuotedPrintable.decode(string)
    str = charset_encoder.encode(str, charset)
    # We assume that binary strings hold utf-8 directly to work around
    # jruby/jruby#829 which subtly changes String#encode semantics.
    str.force_encoding(Encoding::UTF_8) if str.encoding == Encoding::ASCII_8BIT
  end
  transcode_to_scrubbed_utf8(str)
rescue Encoding::UndefinedConversionError, ArgumentError, Encoding::ConverterNotFoundError
  warn "Encoding conversion failed #{$!}"
  str.dup.force_encoding(Encoding::UTF_8)
end

.q_value_encode(str, encoding = nil) ⇒ Object


427
428
429
430
# File 'lib/mail/utilities.rb', line 427

def Utilities.q_value_encode(str, encoding = nil)
  encoding = str.encoding.to_s
  [Encodings::QuotedPrintable.encode(str), encoding]
end

.safe_for_line_ending_conversion?(string) ⇒ Boolean

:nodoc:

Returns:

  • (Boolean)

247
248
249
250
251
252
253
# File 'lib/mail/utilities.rb', line 247

def self.safe_for_line_ending_conversion?(string) #:nodoc:
  if string.encoding == Encoding::BINARY
    string.ascii_only?
  else
    string.valid_encoding?
  end
end

.string_byteslice(str, *args) ⇒ Object


536
537
538
# File 'lib/mail/utilities.rb', line 536

def Utilities.string_byteslice(str, *args)
  str.byteslice(*args)
end

.to_crlf(string) ⇒ Object

Convert line endings to rn unless the string is binary. Used for encoding 8bit and base64 Content-Transfer-Encoding and for convenience when parsing emails with n line endings instead of the required rn.


269
270
271
272
273
274
275
276
# File 'lib/mail/utilities.rb', line 269

def self.to_crlf(string)
  string = string.to_s
  if safe_for_line_ending_conversion? string
    binary_unsafe_to_crlf string
  else
    string
  end
end

.to_lf(string) ⇒ Object

Convert line endings to n unless the string is binary. Used for sendmail delivery and for decoding 8bit Content-Transfer-Encoding.


257
258
259
260
261
262
263
264
# File 'lib/mail/utilities.rb', line 257

def self.to_lf(string)
  string = string.to_s
  if safe_for_line_ending_conversion? string
    binary_unsafe_to_lf string
  else
    string
  end
end

.transcode_charset(str, from_encoding, to_encoding = Encoding::UTF_8) ⇒ Object


381
382
383
384
385
# File 'lib/mail/utilities.rb', line 381

def Utilities.transcode_charset(str, from_encoding, to_encoding = Encoding::UTF_8)
  to_encoding = Encoding.find(to_encoding)
  replacement_char = to_encoding == Encoding::UTF_8 ? '' : '?'
  charset_encoder.encode(str.dup, from_encoding).encode(to_encoding, :undef => :replace, :invalid => :replace, :replace => replacement_char)
end

.uri_parserObject


466
467
468
# File 'lib/mail/utilities.rb', line 466

def Utilities.uri_parser
  URI::DEFAULT_PARSER
end

Instance Method Details

#atom_safe?(str) ⇒ Boolean

Returns true if the string supplied is free from characters not allowed as an ATOM

Returns:

  • (Boolean)

11
12
13
# File 'lib/mail/utilities.rb', line 11

def atom_safe?( str )
  not Constants::ATOM_UNSAFE === str
end

#blank?(value) ⇒ Boolean

Returns true if the object is considered blank. A blank includes things like '', ' ', nil, and arrays and hashes that have nothing in them.

This logic is mostly shared with ActiveSupport's blank?

Returns:

  • (Boolean)

283
284
285
286
287
288
289
290
291
# File 'lib/mail/utilities.rb', line 283

def blank?(value)
  if value.kind_of?(NilClass)
    true
  elsif value.kind_of?(String)
    value !~ /\S/
  else
    value.respond_to?(:empty?) ? value.empty? : !value
  end
end

#bracket(str) ⇒ Object

Wraps a string in angle brackets and escapes any that are in the string itself

Example:

bracket( 'This is a string' ) #=> '<This is a string>'

131
132
133
# File 'lib/mail/utilities.rb', line 131

def bracket( str )
  Utilities.bracket( str )
end

#capitalize_field(str) ⇒ Object

Capitalizes a string that is joined by hyphens correctly.

Example:

string = 'resent-from-field'
capitalize_field( string ) #=> 'Resent-From-Field'

188
189
190
# File 'lib/mail/utilities.rb', line 188

def capitalize_field( str )
  str.to_s.split("-").map { |v| v.capitalize }.join("-")
end

#constantize(str) ⇒ Object

Takes an underscored word and turns it into a class name

Example:

constantize("hello") #=> "Hello"
constantize("hello-there") #=> "HelloThere"
constantize("hello-there-mate") #=> "HelloThereMate"

199
200
201
# File 'lib/mail/utilities.rb', line 199

def constantize( str )
  str.to_s.split(/[-_]/).map { |v| v.capitalize }.to_s
end

#dasherize(str) ⇒ Object

Swaps out all underscores (_) for hyphens (-) good for stringing from symbols a field name.

Example:

string = :resent_from_field
dasherize( string ) #=> 'resent-from-field'

210
211
212
# File 'lib/mail/utilities.rb', line 210

def dasherize( str )
  str.to_s.tr(Constants::UNDERSCORE, Constants::HYPHEN)
end

#dquote(str) ⇒ Object

Wraps supplied string in double quotes and applies -escaping as necessary, unless it is already wrapped.

Example:

string = 'This is a string'
dquote(string) #=> '"This is a string"'

string = 'This is "a string"'
dquote(string #=> '"This is \"a string\"'

68
69
70
# File 'lib/mail/utilities.rb', line 68

def dquote( str )
  '"' + unquote(str).gsub(/[\\"]/n) {|s| '\\' + s } + '"'
end

#escape_paren(str) ⇒ Object

Escape parenthesies in a string

Example:

str = 'This is (a) string'
escape_paren( str ) #=> 'This is \(a\) string'

155
156
157
# File 'lib/mail/utilities.rb', line 155

def escape_paren( str )
  Utilities.escape_paren( str )
end

#generate_message_idObject


293
294
295
# File 'lib/mail/utilities.rb', line 293

def generate_message_id
  "<#{Mail.random_tag}@#{::Socket.gethostname}.mail>"
end

#map_lines(str, &block) ⇒ Object


225
226
227
# File 'lib/mail/utilities.rb', line 225

def map_lines( str, &block )
  str.each_line.map(&block)
end

#map_with_index(enum, &block) ⇒ Object


229
230
231
# File 'lib/mail/utilities.rb', line 229

def map_with_index( enum, &block )
  enum.each_with_index.map(&block)
end

#match_to_s(obj1, obj2) ⇒ Object

Matches two objects with their to_s values case insensitively

Example:

obj2 = "This_is_An_object"
obj1 = :this_IS_an_object
match_to_s( obj1, obj2 ) #=> true

178
179
180
# File 'lib/mail/utilities.rb', line 178

def match_to_s( obj1, obj2 )
  obj1.to_s.casecmp(obj2.to_s) == 0
end

#paren(str) ⇒ Object

Wraps a string in parenthesis and escapes any that are in the string itself.

Example:

paren( 'This is a string' ) #=> '(This is a string)'

108
109
110
# File 'lib/mail/utilities.rb', line 108

def paren( str )
  Utilities.paren( str )
end

#quote_atom(str) ⇒ Object

If the string supplied has ATOM unsafe characters in it, will return the string quoted in double quotes, otherwise returns the string unmodified


17
18
19
# File 'lib/mail/utilities.rb', line 17

def quote_atom( str )
  atom_safe?( str ) ? str : dquote(str)
end

#quote_phrase(str) ⇒ Object

If the string supplied has PHRASE unsafe characters in it, will return the string quoted in double quotes, otherwise returns the string unmodified


23
24
25
26
27
28
29
30
31
32
33
34
35
# File 'lib/mail/utilities.rb', line 23

def quote_phrase( str )
  if str.respond_to?(:force_encoding)
    original_encoding = str.encoding
    ascii_str = str.to_s.dup.force_encoding('ASCII-8BIT')
    if Constants::PHRASE_UNSAFE === ascii_str
      dquote(ascii_str).force_encoding(original_encoding)
    else
      str
    end
  else
    Constants::PHRASE_UNSAFE === str ? dquote(str) : str
  end
end

#quote_token(str) ⇒ Object

If the string supplied has TOKEN unsafe characters in it, will return the string quoted in double quotes, otherwise returns the string unmodified


44
45
46
47
48
49
50
51
52
53
54
55
56
# File 'lib/mail/utilities.rb', line 44

def quote_token( str )
  if str.respond_to?(:force_encoding)
    original_encoding = str.encoding
    ascii_str = str.to_s.dup.force_encoding('ASCII-8BIT')
    if token_safe?( ascii_str )
      str
    else
      dquote(ascii_str).force_encoding(original_encoding)
    end
  else
    token_safe?( str ) ? str : dquote(str)
  end
end

#token_safe?(str) ⇒ Boolean

Returns true if the string supplied is free from characters not allowed as a TOKEN

Returns:

  • (Boolean)

38
39
40
# File 'lib/mail/utilities.rb', line 38

def token_safe?( str )
  not Constants::TOKEN_UNSAFE === str
end

#unbracket(str) ⇒ Object

Unwraps a string from being wrapped in parenthesis

Example:

str = '<This is a string>'
unbracket( str ) #=> 'This is a string'

141
142
143
144
145
146
147
# File 'lib/mail/utilities.rb', line 141

def unbracket( str )
  if str.start_with?('<') && str.end_with?('>')
    str.slice(1..-2)
  else
    str
  end
end

#underscoreize(str) ⇒ Object

Swaps out all hyphens (-) for underscores (_) good for stringing to symbols a field name.

Example:

string = :resent_from_field
underscoreize ( string ) #=> 'resent_from_field'

221
222
223
# File 'lib/mail/utilities.rb', line 221

def underscoreize( str )
  str.to_s.downcase.tr(Constants::HYPHEN, Constants::UNDERSCORE)
end

#unescape(str) ⇒ Object

Removes any -escaping.

Example:

string = 'This is \"a string\"'
unescape(string) #=> 'This is "a string"'

string = '"This is \"a string\""'
unescape(string) #=> '"This is "a string""'

99
100
101
# File 'lib/mail/utilities.rb', line 99

def unescape( str )
  str.gsub(/\\(.)/, '\1')
end

#unparen(str) ⇒ Object

Unwraps a string from being wrapped in parenthesis

Example:

str = '(This is a string)'
unparen( str ) #=> 'This is a string'

118
119
120
121
122
123
124
# File 'lib/mail/utilities.rb', line 118

def unparen( str )
  if str.start_with?('(') && str.end_with?(')')
    str.slice(1..-2)
  else
    str
  end
end

#unquote(str) ⇒ Object

Unwraps supplied string from inside double quotes and removes any -escaping.

Example:

string = '"This is a string"'
unquote(string) #=> 'This is a string'

string = '"This is \"a string\""'
unqoute(string) #=> 'This is "a string"'

82
83
84
85
86
87
88
# File 'lib/mail/utilities.rb', line 82

def unquote( str )
  if str =~ /^"(.*?)"$/
    unescape($1)
  else
    str
  end
end

#uri_escape(str) ⇒ Object


159
160
161
# File 'lib/mail/utilities.rb', line 159

def uri_escape( str )
  uri_parser.escape(str)
end

#uri_parserObject


167
168
169
# File 'lib/mail/utilities.rb', line 167

def uri_parser
  @uri_parser ||= URI.const_defined?(:DEFAULT_PARSER) ? URI::DEFAULT_PARSER : URI
end

#uri_unescape(str) ⇒ Object


163
164
165
# File 'lib/mail/utilities.rb', line 163

def uri_unescape( str )
  uri_parser.unescape(str)
end