Class: Pdfrb::Document

Inherits:
Object
  • Object
show all
Defined in:
lib/pdfrb/document.rb,
lib/pdfrb/document/files.rb,
lib/pdfrb/document/fonts.rb,
lib/pdfrb/document/pages.rb,
lib/pdfrb/document/images.rb,
lib/pdfrb/document/metadata.rb,
lib/pdfrb/document/annotations.rb,
lib/pdfrb/document/destinations.rb

Overview

Top-level PDF document facade. Owns:

* The IO it was read from (or nil for in-memory).
* An oid -> Object table for new/modified objects.
* A wrap() pipeline that upgrades raw Hashes to typed
Dictionary subclasses based on /Type or an explicit type.
* On read: an XrefSection + ObjectReader for lazy resolution.
* A revisions stack (TODO 30 — incremental updates).

Defined Under Namespace

Classes: Annotations, Destinations, Files, Fonts, Images, Metadata, Pages

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(io: nil, config: {}) ⇒ Document

Returns a new instance of Document.



22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
 
# File 'lib/pdfrb/document.rb', line 22

def initialize(io: nil, config: {})
  @config = Configuration.new(config)
  @io = io
  @objects = {}        # oid -> Pdfrb::Model::Object (modified or new)
  @next_oid = 1
  @listeners = {}      # message_name -> [Proc]
  @xref = nil
  @object_reader = nil
  @version = "1.4"
  @empty_trailer = nil

  if io
    read_from_io(io)
  else
    seed_empty_structure
  end
end

Instance Attribute Details

#config ⇒ Object (readonly)

Returns the value of attribute config.



40
41
42
 
# File 'lib/pdfrb/document.rb', line 40

def config
  @config
end

#io ⇒ Object (readonly)

Returns the value of attribute io.



40
41
42
 
# File 'lib/pdfrb/document.rb', line 40

def io
  @io
end

#version ⇒ Object (readonly)

Returns the value of attribute version.



40
41
42
 
# File 'lib/pdfrb/document.rb', line 40

def version
  @version
end

#xref ⇒ Object (readonly)

Returns the value of attribute xref.



40
41
42
 
# File 'lib/pdfrb/document.rb', line 40

def xref
  @xref
end

Class Method Details

.open(path, **opts) ⇒ Object 



42
43
44
45
46
47
48
49
50
 
# File 'lib/pdfrb/document.rb', line 42

def self.open(path, **opts)
  if block_given?
    File.open(path, "rb") { |f| yield new(io: f, **opts) }
  else
    bytes = File.binread(path)
    require "stringio"
    new(io: StringIO.new(bytes), **opts)
  end
end

Instance Method Details

#add(value, type: nil) ⇒ Object 



62
63
64
65
66
67
 
# File 'lib/pdfrb/document.rb', line 62

def add(value, type: nil)
  oid = allocate_oid
  obj = wrap(value, type: type, oid: oid, gen: 0)
  register(obj)
  obj
end

#annotations ⇒ Object 



121
122
123
 
# File 'lib/pdfrb/document.rb', line 121

def annotations
  @annotations ||= Document::Annotations.new(self)
end

#catalog ⇒ Object 



150
151
152
153
154
155
156
157
 
# File 'lib/pdfrb/document.rb', line 150

def catalog
  return @catalog if defined?(@catalog) && @catalog

  ref = trailer ? trailer[:Root] : nil
  return nil unless ref

  @catalog = object(ref)
end

#destinations ⇒ Object 



117
118
119
 
# File 'lib/pdfrb/document.rb', line 117

def destinations
  @destinations ||= Document::Destinations.new(self)
end

#dispatch_message(message, *args) ⇒ Object 



87
88
89
90
91
 
# File 'lib/pdfrb/document.rb', line 87

def dispatch_message(message, *args)
  return unless @listeners.key?(message)

  @listeners[message].each { |blk| blk.call(*args) }
end

#each_indirect_object ⇒ Object

Yield every indirect object in this document: modified/new ones first (they shadow loaded ones), then every loaded entry the xref knows about.



162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
 
# File 'lib/pdfrb/document.rb', line 162

def each_indirect_object
  seen = Set.new
  @objects.each_value do |obj|
    next unless obj.indirect?

    seen << obj.oid
    yield obj
  end
  return if @xref.nil?

  @xref.entries.each_key do |oid|
    next if seen.include?(oid)
    next if oid.zero?

    obj = object(Pdfrb::Model::Reference.new(oid, 0))
    next if obj.nil? || !obj.indirect?

    yield obj
  end
end

#files ⇒ Object 



109
110
111
 
# File 'lib/pdfrb/document.rb', line 109

def files
  @files ||= Document::Files.new(self)
end

#fonts ⇒ Object 



101
102
103
 
# File 'lib/pdfrb/document.rb', line 101

def fonts
  @fonts ||= Document::Fonts.new(self)
end

#images ⇒ Object 



105
106
107
 
# File 'lib/pdfrb/document.rb', line 105

def images
  @images ||= Document::Images.new(self)
end

#metadata ⇒ Object 



113
114
115
 
# File 'lib/pdfrb/document.rb', line 113

def metadata
  @metadata ||= Document::Metadata.new(self)
end

#object(reference) ⇒ Object Also known as: dereference

Resolve a Reference to its Object. New/modified objects take precedence over xref-loaded ones; otherwise consult the ObjectReader (which caches per oid).



72
73
74
75
76
77
78
79
80
 
# File 'lib/pdfrb/document.rb', line 72

def object(reference)
  return reference unless reference.is_a?(Pdfrb::Model::Reference)

  modified = @objects[reference.oid]
  return modified if modified
  return nil if @object_reader.nil?

  @object_reader.load_oid(reference.oid)
end

#pages ⇒ Object

---- Facade accessors (Phase 13) ---- Each returns a memoised helper object that knows how to mutate this document. Defined under Pdfrb::Document::*.



97
98
99
 
# File 'lib/pdfrb/document.rb', line 97

def pages
  @pages ||= Document::Pages.new(self)
end

#register_listener(message, &block) ⇒ Object 



83
84
85
 
# File 'lib/pdfrb/document.rb', line 83

def register_listener(message, &block)
  (@listeners[message] ||= []) << block
end

#register_override(obj) ⇒ Object

Replace an indirect object in the @objects table. Used by the Importer when promoting a Dictionary stub to a Stream (so cycles resolve correctly). Idempotent.



128
129
130
 
# File 'lib/pdfrb/document.rb', line 128

def register_override(obj)
  @objects[obj.oid] = obj
end

#trailer ⇒ Object

Convenience accessor for the document Catalog dict. Loads from trailer's /Root, which itself is loaded lazily from the xref.



144
145
146
147
148
 
# File 'lib/pdfrb/document.rb', line 144

def trailer
  return @trailer if defined?(@trailer) && @trailer

  @trailer = read_trailer
end

#wrap(data, type: nil, oid: 0, gen: 0) ⇒ Object 



52
53
54
55
56
57
58
59
60
 
# File 'lib/pdfrb/document.rb', line 52

def wrap(data, type: nil, oid: 0, gen: 0)
  target_class = resolve_target_class(data, type)
  value = unwrap_value(data)
  if target_class <= Pdfrb::Model::Object
    target_class.new(value, oid: oid, gen: gen, document: self)
  else
    target_class.new(value)
  end
end

#write(path = nil, io: nil) ⇒ Object

Write this document to path (or any IO via +io:).

Raises:

  • (ArgumentError) 


133
134
135
136
137
138
139
140
 
# File 'lib/pdfrb/document.rb', line 133

def write(path = nil, io: nil)
  target = io || (path && File.open(path, "wb"))
  raise ArgumentError, "write needs a path or io:" unless target

  Pdfrb::Writer.write(self, target)
  target.close if path && io.nil? && target.respond_to?(:close)
  self
end