origami 1.2.7 → 2.0.0

data/lib/origami/parsers/fdf.rb

@@ -1,21 +1,20 @@
 =begin
 
-= File
-	parsers/fdf.rb
+    This file is part of Origami, PDF manipulation framework for Ruby
+    Copyright (C) 2016	Guillaume Delugré.
 
-= Info
-	Origami is free software: you can redistribute it and/or modify
-  it under the terms of the GNU Lesser General Public License as published by
-  the Free Software Foundation, either version 3 of the License, or
-  (at your option) any later version.
+    Origami is free software: you can redistribute it and/or modify
+    it under the terms of the GNU Lesser General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
 
-  Origami is distributed in the hope that it will be useful,
-  but WITHOUT ANY WARRANTY; without even the implied warranty of
-  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-  GNU Lesser General Public License for more details.
+    Origami is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU Lesser General Public License for more details.
 
-  You should have received a copy of the GNU Lesser General Public License
-  along with Origami.  If not, see <http://www.gnu.org/licenses/>.
+    You should have received a copy of the GNU Lesser General Public License
+    along with Origami.  If not, see <http://www.gnu.org/licenses/>.
 
 =end
 
@@ -23,26 +22,35 @@ require 'origami/parser'
 
 module Origami
 
-  class FDF
-    class Parser < Origami::Parser
-      def parse(stream) #:nodoc:
-        super
-
-        fdf = Adobe::FDF.new
-        fdf.header = Adobe::FDF::Header.parse(stream)
-        @options[:callback].call(fdf.header)
-        
-        loop do 
-          break if (object = parse_object).nil?
-          fdf << object
-        end
-        
-        fdf.revisions.first.xreftable = parse_xreftable
-        fdf.revisions.first.trailer = parse_trailer
+    class FDF
+        class Parser < Origami::Parser
+            def parse(stream) #:nodoc:
+                super(stream)
+
+                fdf = FDF.new
+                fdf.header = FDF::Header.parse(@data)
+                @options[:callback].call(fdf.header)
+
+                loop do
+                    break if (object = parse_object).nil?
+                    fdf.insert(object)
+                end
+
+                fdf.revisions.first.xreftable = parse_xreftable
+                fdf.revisions.first.trailer = parse_trailer
+
+                if Origami::OPTIONS[:enable_type_propagation]
+                    trailer = fdf.revisions.first.trailer
 
-        fdf
-      end
+                    if trailer[:Root].is_a?(Reference)
+                        fdf.cast_object(trailer[:Root], FDF::Catalog, self)
+                    end
+
+                    propagate_types(fdf)
+                end
+
+                fdf
+            end
+        end
     end
-  end
 end
-

data/lib/origami/parsers/pdf.rb

@@ -1,115 +1,95 @@
 =begin
 
-= File
-	parsers/pdf.rb
-
-= Info
-	This file is part of Origami, PDF manipulation framework for Ruby
-	Copyright (C) 2010	Guillaume Delugré <guillaume AT security-labs DOT org>
-	All right reserved.
-	
-  Origami is free software: you can redistribute it and/or modify
-  it under the terms of the GNU Lesser General Public License as published by
-  the Free Software Foundation, either version 3 of the License, or
-  (at your option) any later version.
-
-  Origami is distributed in the hope that it will be useful,
-  but WITHOUT ANY WARRANTY; without even the implied warranty of
-  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-  GNU Lesser General Public License for more details.
-
-  You should have received a copy of the GNU Lesser General Public License
-  along with Origami.  If not, see <http://www.gnu.org/licenses/>.
+    This file is part of Origami, PDF manipulation framework for Ruby
+    Copyright (C) 2016	Guillaume Delugré.
+
+    Origami is free software: you can redistribute it and/or modify
+    it under the terms of the GNU Lesser General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    Origami is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU Lesser General Public License for more details.
+
+    You should have received a copy of the GNU Lesser General Public License
+    along with Origami.  If not, see <http://www.gnu.org/licenses/>.
 
 =end
 
 require 'origami/parser'
 
 module Origami
-  
-  class PDF
-    class Parser < Origami::Parser
-      def initialize(params = {})
-        options =
-        {
-          :password => '',            # Default password being tried when opening a protected document.
-          :prompt_password => Proc.new { 
-            print "Password: "
-            gets.chomp 
-          },                          # Callback procedure to prompt password when document is encrypted.
-          :force => false             # Force PDF header detection
-        }.update(params)
-
-        super(options)
-      end
-
-      private
-
-      def parse_initialize #:nodoc:
-        if @options[:force] == true
-          @data.skip_until(/%PDF-/).nil?
-          @data.pos = @data.pos - 5
-        end
 
-        pdf = PDF.new(self)
-
-        info "...Reading header..."
-        begin
-          pdf.header = PDF::Header.parse(@data)
-          @options[:callback].call(pdf.header)
-        rescue InvalidHeaderError => e
-          if @options[:ignore_errors] == true
-            warn "PDF header is invalid, ignoring..."
-          else
-            raise e
-          end
-        end
+    class PDF
+        class Parser < Origami::Parser
+            def initialize(params = {})
+                options =
+                {
+                  password: '',                 # Default password being tried when opening a protected document.
+                  prompt_password: lambda do    # Callback procedure to prompt password when document is encrypted.
+                      require 'io/console'
+                      STDERR.print "Password: "
+                      STDIN.noecho(&:gets).chomp
+                  end,
+                  force: false              # Force PDF header detection
+                }.update(params)
+
+                super(options)
+            end
 
-        pdf
-      end
+            private
 
-      def parse_finalize(pdf) #:nodoc:
-        warn "This file has been linearized." if pdf.is_linearized?
+            def parse_initialize #:nodoc:
+                if @options[:force] == true
+                    @data.skip_until(/%PDF-/).nil?
+                    @data.pos = @data.pos - 5
+                end
 
-        if Origami::OPTIONS[:enable_type_propagation]
-          info "...Propagating types..."
-          @deferred_casts.each_pair do |ref, type|
-            type = [ type ] unless type.is_a?(::Array)
-            type.each do |hint|
-              pdf.cast_object(ref, hint)
-            end
-          end
-        end
+                pdf = PDF.new(self)
 
-        #
-        # Decrypt encrypted file contents
-        #
-        if pdf.is_encrypted?
-          warn "This document contains encrypted data!"
-        
-          passwd = @options[:password]
-          begin
-            pdf.decrypt(passwd)
-          rescue EncryptionInvalidPasswordError
-            if passwd.empty?
-              passwd = @options[:prompt_password].call
-              retry unless passwd.empty?
+                info "...Reading header..."
+                begin
+                    pdf.header = PDF::Header.parse(@data)
+                    @options[:callback].call(pdf.header)
+                rescue InvalidHeaderError
+                    raise unless @options[:ignore_errors]
+                    warn "PDF header is invalid, ignoring..."
+                end
+
+                pdf
             end
 
-            raise EncryptionInvalidPasswordError
-          end
-        end
+            def parse_finalize(pdf) #:nodoc:
+                warn "This file has been linearized." if pdf.linearized?
 
-        if pdf.is_signed?
-          warn "This document has been signed!"
-        end
+                propagate_types(pdf) if Origami::OPTIONS[:enable_type_propagation]
 
-        pdf
-      end
-    end
-  end
+                #
+                # Decrypt encrypted file contents
+                #
+                if pdf.encrypted?
+                    warn "This document contains encrypted data!"
 
-end
+                    passwd = @options[:password]
+                    begin
+                        pdf.decrypt(passwd)
+                    rescue EncryptionInvalidPasswordError
+                        if passwd.empty?
+                            passwd = @options[:prompt_password].call
+                            retry unless passwd.empty?
+                        end
 
-require 'origami/parsers/pdf/linear'
+                        raise
+                    end
+                end
 
+                warn "This document has been signed!" if pdf.signed?
+
+                pdf
+            end
+        end
+    end
+
+end

data/lib/origami/parsers/pdf/lazy.rb

@@ -0,0 +1,137 @@
+=begin
+
+    This file is part of Origami, PDF manipulation framework for Ruby
+    Copyright (C) 2016	Guillaume Delugré.
+
+    Origami is free software: you can redistribute it and/or modify
+    it under the terms of the GNU Lesser General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    Origami is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU Lesser General Public License for more details.
+
+    You should have received a copy of the GNU Lesser General Public License
+    along with Origami.  If not, see <http://www.gnu.org/licenses/>.
+
+=end
+
+
+require 'origami/parsers/pdf'
+
+module Origami
+
+    class PDF
+
+        #
+        # Create a new PDF lazy Parser.
+        #
+        class LazyParser < Parser
+            def parse(stream)
+                super
+
+                pdf = parse_initialize
+                revisions = []
+
+                # Set the scanner position at the end.
+                @data.terminate
+
+                # Locate the startxref token.
+                until @data.match?(/#{Trailer::XREF_TOKEN}/)
+                    raise ParsingError, "No xref token found" if @data.pos == 0
+                    @data.pos -= 1
+                end
+
+                # Extract the offset of the last xref section.
+                trailer = Trailer.parse(@data, self)
+                raise ParsingError, "Cannot locate xref section" if trailer.startxref.zero?
+
+                xref_offset = trailer.startxref
+                while xref_offset and xref_offset != 0
+
+                    # Create a new revision based on the xref section offset.
+                    revision = parse_revision(pdf, xref_offset)
+
+                    # Locate the previous xref section.
+                    if revision.xrefstm
+                        xref_offset = revision.xrefstm[:Prev].to_i
+                    else
+                        xref_offset = revision.trailer[:Prev].to_i
+                    end
+
+                    # Prepend the revision.
+                    revisions.unshift(revision)
+                end
+
+                pdf.revisions.clear
+                revisions.each do |rev|
+                    pdf.revisions.push(rev)
+                    pdf.insert(rev.xrefstm) if rev.has_xrefstm?
+                end
+
+                parse_finalize(pdf)
+
+                pdf
+            end
+
+            private
+
+            def parse_revision(pdf, offset)
+                raise ParsingError, "Invalid xref offset" if offset < 0 or offset >= @data.string.size
+
+                @data.pos = offset
+
+                # Create a new revision.
+                revision = PDF::Revision.new(pdf)
+
+                # Regular xref section.
+                if @data.match?(/#{XRef::Section::TOKEN}/)
+                    xreftable = parse_xreftable
+                    raise ParsingError, "Cannot parse xref section" if xreftable.nil?
+
+                    revision.xreftable = xreftable
+                    revision.trailer = parse_trailer
+
+                    # Handle hybrid cross-references.
+                    if revision.trailer[:XRefStm].is_a?(Integer)
+                        begin
+                            offset = revision.trailer[:XRefStm].to_i
+                            xrefstm = parse_object(offset)
+
+                            if xrefstm.is_a?(XRefStream)
+                                revision.xrefstm = xrefstm
+                            else
+                                warn "Invalid xref stream at offset #{offset}"
+                            end
+
+                        rescue
+                            warn "Cannot parse xref stream at offset #{offset}"
+                        end
+                    end
+
+                # The xrefs are stored in a stream.
+                else
+                    xrefstm = parse_object
+                    raise ParsingError, "Invalid xref stream" unless xrefstm.is_a?(XRefStream)
+
+                    revision.xrefstm = xrefstm
+
+                    # Search for the trailer.
+                    if @data.skip_until Regexp.union(Trailer::XREF_TOKEN, *Trailer::TOKENS)
+                        @data.pos -= @data.matched_size
+
+                        revision.trailer = parse_trailer
+                    else
+                        warn "No trailer found."
+                        revision.trailer = Trailer.new
+                    end
+                end
+
+                revision
+            end
+        end
+    end
+
+end

data/lib/origami/parsers/pdf/linear.rb

@@ -1,85 +1,83 @@
 =begin
 
-= File
-	parsers/linear.rb
+    This file is part of Origami, PDF manipulation framework for Ruby
+    Copyright (C) 2016	Guillaume Delugré.
 
-= Info
-	Origami is free software: you can redistribute it and/or modify
-  it under the terms of the GNU Lesser General Public License as published by
-  the Free Software Foundation, either version 3 of the License, or
-  (at your option) any later version.
+    Origami is free software: you can redistribute it and/or modify
+    it under the terms of the GNU Lesser General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
 
-  Origami is distributed in the hope that it will be useful,
-  but WITHOUT ANY WARRANTY; without even the implied warranty of
-  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-  GNU Lesser General Public License for more details.
+    Origami is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU Lesser General Public License for more details.
 
-  You should have received a copy of the GNU Lesser General Public License
-  along with Origami.  If not, see <http://www.gnu.org/licenses/>.
+    You should have received a copy of the GNU Lesser General Public License
+    along with Origami.  If not, see <http://www.gnu.org/licenses/>.
 
 =end
 
 
-require 'origami/parser'
-require 'origami/pdf'
+require 'origami/parsers/pdf'
 
 module Origami
 
-  class PDF
-
-    #
-    # Create a new PDF linear Parser.
-    #
-    class LinearParser < Parser
-      def parse(stream)
-        super
-        
-        pdf = parse_initialize
+    class PDF
 
         #
-        # Parse each revision
+        # Create a new PDF linear Parser.
         #
-        revision = 0
-        until @data.eos? do
-          
-          begin
-            pdf.add_new_revision unless revision.zero?
-            revision = revision + 1
-
-            info "...Parsing revision #{pdf.revisions.size}..."
-            loop do
-              break if (object = parse_object).nil?
-              pdf.insert(object)
+        class LinearParser < Parser
+            def parse(stream)
+                super
+            
+                pdf = parse_initialize
+
+                #
+                # Parse each revision
+                #
+                revision = 0
+                until @data.eos? do
+                    begin
+                        pdf.add_new_revision unless revision.zero?
+                        revision = revision + 1
+
+                        info "...Parsing revision #{pdf.revisions.size}..."
+                        loop do
+                            break if (object = parse_object).nil?
+                            pdf.insert(object)
+                        end
+
+                        pdf.revisions.last.xreftable = parse_xreftable
+
+                        trailer = parse_trailer
+                        pdf.revisions.last.trailer = trailer
+
+                        if trailer.startxref != 0
+                            xrefstm = pdf.get_object_by_offset(trailer.startxref)
+                        elsif trailer[:XRefStm].is_a?(Integer)
+                            xrefstm = pdf.get_object_by_offset(trailer[:XRefStm])
+                        end
+
+                        if xrefstm.is_a?(XRefStream)
+                            warn "Found a XRefStream for this revision at #{xrefstm.reference}"
+                            pdf.revisions.last.xrefstm = xrefstm
+                        end
+
+                    rescue
+                        error "Cannot read : " + (@data.peek(10) + "...").inspect
+                        error "Stopped on exception : " + $!.message
+
+                        break
+                    end
+                end
+
+                pdf.loaded!
+
+                parse_finalize(pdf)
             end
-
-            pdf.revisions.last.xreftable = parse_xreftable
-
-            trailer = parse_trailer
-            pdf.revisions.last.trailer = trailer
-
-            xrefstm = pdf.get_object_by_offset(trailer.startxref) || 
-              (pdf.get_object_by_offset(trailer.XRefStm) if trailer.has_field? :XRefStm)
-
-            if not xrefstm.nil?
-              warn "Found a XRefStream for this revision at #{xrefstm.reference}"
-              pdf.revisions.last.xrefstm = xrefstm
-            end
-
-          rescue SystemExit
-            raise
-          rescue Exception => e
-            error "Cannot read : " + (@data.peek(10) + "...").inspect
-            error "Stopped on exception : " + e.message
-
-            break
-          end
-
         end
-
-        parse_finalize(pdf)
-      end
-
     end
-  end
-end
 
+end