class PDF::Reader::Parser

them into useable ruby objects (hash’s, arrays, true, false, etc)
An internal PDF::Reader class that reads objects from the PDF file and converts
###############################################################################

def array

reads a PDF array from the buffer and converts it to a Ruby Array.
###############################################################################

def array
  a = []
  loop do
    item = parse_token
    break if item.kind_of?(Token) and item == "]"
    raise MalformedPDFError, "unterminated array" if @buffer.empty?
    a << item
  end
  a
end

def dictionary

reads a PDF dict from the buffer and converts it to a Ruby Hash.
###############################################################################

def dictionary
  dict = {}
  loop do
    key = parse_token
    break if key.kind_of?(Token) and key == ">>"
    raise MalformedPDFError, "unterminated dict" if @buffer.empty?
    PDF::Reader::Error.validate_type_as_malformed(key, "Dictionary key", Symbol)
    value = parse_token
    value.kind_of?(Token) and Error.str_assert_not(value, ">>")
    dict[key] = value
  end
  dict
end

def hex_string

Reads a PDF hex string from the buffer and converts it to a Ruby String
###############################################################################

def hex_string
  str = "".dup
  loop do
    token = @buffer.token
    break if token == ">"
    raise MalformedPDFError, "unterminated hex string" if @buffer.empty?
    str << token
  end
  # add a missing digit if required, as required by the spec
  str << "0" unless str.size % 2 == 0
  [str].pack('H*')
end

def initialize(buffer, objects=nil)

objects - a PDF::Reader::ObjectHash object that can return objects from the PDF file
buffer - a PDF::Reader::Buffer object that contains PDF data

Create a new parser around a PDF::Reader::Buffer object
###############################################################################

def initialize(buffer, objects=nil)
  @buffer = buffer
  @objects  = objects
end

def object(id, gen)

gen - the object revision number to return
id - the object ID to return

that describes it
If the object is a content stream, returns both the stream and the dictionary
Reads an entire PDF object from the buffer and returns it as a Ruby String.
###############################################################################

def object(id, gen)
  idCheck = parse_token
  # Sometimes the xref table is corrupt and points to an offset slightly too early in the file.
  # check the next token, maybe we can find the start of the object we're looking for
  if idCheck != id
    Error.assert_equal(parse_token, id)
  end
  Error.assert_equal(parse_token, gen)
  Error.str_assert(parse_token, "obj")
  obj = parse_token
  post_obj = parse_token
  if obj.is_a?(Hash) && post_obj == "stream"
    stream(obj)
  else
    obj
  end
end

def parse_token(operators={})

operators - a hash of supported operators to read from the underlying buffer.

object
Reads the next token from the underlying buffer and convets it to an appropriate
###############################################################################

def parse_token(operators={})
  token = @buffer.token
  if STRATEGIES.has_key? token
    STRATEGIES[token].call(self, token)
  elsif token.is_a? PDF::Reader::Reference
    token
  elsif operators.has_key? token
    Token.new(token)
  elsif token.frozen?
    token
  elsif token =~ /\d*\.\d/
    token.to_f
  else
    token.to_i
  end
end

def pdf_name

reads a PDF name from the buffer and converts it to a Ruby Symbol
###############################################################################

def pdf_name
  tok = @buffer.token
  tok = tok.dup.gsub(/#([A-Fa-f0-9]{2})/) do |match|
    match[1, 2].hex.chr
  end
  tok.to_sym
end

def stream(dict)

Decodes the contents of a PDF Stream and returns it as a Ruby String.
###############################################################################

def stream(dict)
  raise MalformedPDFError, "PDF malformed, missing stream length" unless dict.has_key?(:Length)
  if @objects
    length = @objects.deref_integer(dict[:Length])
    if dict[:Filter]
      dict[:Filter] = @objects.deref_name_or_array(dict[:Filter])
    end
  else
    length = dict[:Length] || 0
  end
  PDF::Reader::Error.validate_type_as_malformed(length, "length", Numeric)
  data = @buffer.read(length, :skip_eol => true)
  Error.str_assert(parse_token, "endstream")
  # We used to assert that the stream had the correct closing token, but it doesn't *really*
  # matter if it's missing, and other readers seems to handle its absence just fine
  # Error.str_assert(parse_token, "endobj")
  PDF::Reader::Stream.new(dict, data)
end

def string

Reads a PDF String from the buffer and converts it to a Ruby String
###############################################################################

def string
  str = @buffer.token
  return "".dup.force_encoding("binary") if str == ")"
  Error.assert_equal(parse_token, ")")
  str.gsub!(/\\(\r\n|[nrtbf()\\\n\r]|([0-7]{1,3}))?|\r\n?/m) do |match|
    if $2.nil? # not octal digits
      MAPPING[match] || "".dup
    else # must be octal digits
      ($2.oct & 0xff).chr # ignore high level overflow
    end
  end
  str.force_encoding("binary")
end

Modules

Classes