Module: OpenURI::Meta

Defined in:
lib/open-uri.rb

Overview

Mixin for holding meta-information.

Constant Summary collapse

RE_LWS =

:stopdoc:

/[\r\n\t ]+/n
RE_TOKEN =
%r{[^\x00- ()<>@,;:\\"/\[\]?={}\x7f]+}n
RE_QUOTED_STRING =
%r{"(?:[\r\n\t !#-\[\]-~\x80-\xff]|\\[\x00-\x7f])*"}n
RE_PARAMETERS =
%r{(?:;#{RE_LWS}?#{RE_TOKEN}#{RE_LWS}?=#{RE_LWS}?(?:#{RE_TOKEN}|#{RE_QUOTED_STRING})#{RE_LWS}?)*}n

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Instance Attribute Details

#base_uriObject

returns a URI that is the base of relative URIs in the data. It may differ from the URI supplied by a user due to redirection.



436
437
438
# File 'lib/open-uri.rb', line 436

def base_uri
  @base_uri
end

#metaObject (readonly)

returns a Hash that represents header fields. The Hash keys are downcased for canonicalization. The Hash values are a field body. If there are multiple field with same field name, the field values are concatenated with a comma.



443
444
445
# File 'lib/open-uri.rb', line 443

def meta
  @meta
end

#metasObject (readonly)

returns a Hash that represents header fields. The Hash keys are downcased for canonicalization. The Hash value are an array of field values.



448
449
450
# File 'lib/open-uri.rb', line 448

def metas
  @metas
end

#statusObject

returns an Array that consists of status code and message.



432
433
434
# File 'lib/open-uri.rb', line 432

def status
  @status
end

Class Method Details

.init(obj, src = nil) ⇒ Object

:nodoc:



415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
# File 'lib/open-uri.rb', line 415

def Meta.init(obj, src=nil) # :nodoc:
  obj.extend Meta
  obj.instance_eval {
    @base_uri = nil
    @meta = {} # name to string.  legacy.
    @metas = {} # name to array of strings.
  }
  if src
    obj.status = src.status
    obj.base_uri = src.base_uri
    src.metas.each {|name, values|
      obj.meta_add_field2(name, values)
    }
  end
end

Instance Method Details

#charsetObject

returns a charset parameter in Content-Type field. It is downcased for canonicalization.

If charset parameter is not given but a block is given, the block is called and its result is returned. It can be used to guess charset.

If charset parameter and block is not given, nil is returned except text type. In that case, “utf-8” is returned as defined by RFC6838 4.2.1



534
535
536
537
538
539
540
541
542
543
544
545
# File 'lib/open-uri.rb', line 534

def charset
  type, *parameters = content_type_parse
  if pair = parameters.assoc('charset')
    pair.last.downcase
  elsif block_given?
    yield
  elsif type && %r{\Atext/} =~ type
    "utf-8" # RFC6838 4.2.1
  else
    nil
  end
end

#content_encodingObject

Returns a list of encodings in Content-Encoding field as an array of strings.

The encodings are downcased for canonicalization.



551
552
553
554
555
556
557
558
# File 'lib/open-uri.rb', line 551

def content_encoding
  vs = @metas['content-encoding']
  if vs && %r{\A#{RE_LWS}?#{RE_TOKEN}#{RE_LWS}?(?:,#{RE_LWS}?#{RE_TOKEN}#{RE_LWS}?)*}o =~ (v = vs.join(', '))
    v.scan(RE_TOKEN).map {|content_coding| content_coding.downcase}
  else
    []
  end
end

#content_typeObject

returns “type/subtype” which is MIME Content-Type. It is downcased for canonicalization. Content-Type parameters are stripped.



519
520
521
522
# File 'lib/open-uri.rb', line 519

def content_type
  type, *_ = content_type_parse
  type || 'application/octet-stream'
end

#content_type_parseObject

:startdoc:



497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
# File 'lib/open-uri.rb', line 497

def content_type_parse # :nodoc:
  vs = @metas['content-type']
  # The last (?:;#{RE_LWS}?)? matches extra ";" which violates RFC2045.
  if vs && %r{\A#{RE_LWS}?(#{RE_TOKEN})#{RE_LWS}?/(#{RE_TOKEN})#{RE_LWS}?(#{RE_PARAMETERS})(?:;#{RE_LWS}?)?\z}no =~ vs.join(', ')
    type = $1.downcase
    subtype = $2.downcase
    parameters = []
    $3.scan(/;#{RE_LWS}?(#{RE_TOKEN})#{RE_LWS}?=#{RE_LWS}?(?:(#{RE_TOKEN})|(#{RE_QUOTED_STRING}))/no) {|att, val, qval|
      if qval
        val = qval[1...-1].gsub(/[\r\n\t !#-\[\]-~\x80-\xff]+|(\\[\x00-\x7f])/n) { $1 ? $1[1,1] : $& }
      end
      parameters << [att.downcase, val]
    }
    ["#{type}/#{subtype}", *parameters]
  else
    nil
  end
end

#last_modifiedObject

returns a Time that represents the Last-Modified field.



481
482
483
484
485
486
487
488
# File 'lib/open-uri.rb', line 481

def last_modified
  if vs = @metas['last-modified']
    v = vs.join(', ')
    Time.httpdate(v)
  else
    nil
  end
end

#meta_add_field(name, value) ⇒ Object

:nodoc:



476
477
478
# File 'lib/open-uri.rb', line 476

def meta_add_field(name, value) # :nodoc:
  meta_add_field2(name, [value])
end

#meta_add_field2(name, values) ⇒ Object

:nodoc:



469
470
471
472
473
474
# File 'lib/open-uri.rb', line 469

def meta_add_field2(name, values) # :nodoc:
  name = name.downcase
  @metas[name] = values
  @meta[name] = values.join(', ')
  meta_setup_encoding if name == 'content-type'
end

#meta_setup_encodingObject

:nodoc:



450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
# File 'lib/open-uri.rb', line 450

def meta_setup_encoding # :nodoc:
  charset = self.charset
  enc = nil
  if charset
    begin
      enc = Encoding.find(charset)
    rescue ArgumentError
    end
  end
  enc = Encoding::ASCII_8BIT unless enc
  if self.respond_to? :force_encoding
    self.force_encoding(enc)
  elsif self.respond_to? :string
    self.string.force_encoding(enc)
  else # Tempfile
    self.set_encoding enc
  end
end