pdfinfo 1.2.0 → 1.3.1

checksums.yaml

@@ -1,7 +1,15 @@
 ---
-SHA1:
-  metadata.gz: 9c599fc8f36c8f2eaad1d072b30b702d1abc9283
-  data.tar.gz: 886c3593bd2bd6159ecca069c748265f14ad126c
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    NGI3ZTIzYzMxYzEyNTkyYmQ2ZDEwNzBjMDg2NWYxOGFkNTk3MTM2YQ==
+  data.tar.gz: !binary |-
+    ZDQ1NzRmMjczZmMyNTI4YWEzNDcwZWUyOWYxYTNhZDAyNDI4NGRlZA==
 SHA512:
-  metadata.gz: 901d3defb09ac65a5e1086e746f8a3d242efffc8e85b99159623f538a273d8b463b49002a95658a53f38ba0e94bb6df764726030111014cc8d1418c0bae40068
-  data.tar.gz: 6547e5457cdd88dfbcf2506208ce8dc2c501431806e5c6dc342aab0de84aef3f508dc0a71c451cb2a39313440ec8b5f4ef41fd5bc5decc65b69e310f2ef8acc9
+  metadata.gz: !binary |-
+    ZTY5YWU4M2Q3Y2VhNzIxYTg3ZDZlNWY4OGRiZjZmMmMwMzcyZTJlMjU3ODA2
+    ZDBiYmRkNDMwNTgwNzgyY2EwNTYzZDNkNmY5MmZjNjIwZmI4OGNmMDMwYzU2
+    M2I1ZTBhMjgwOGM2MzkwMjA1MDVhYWEyOTM1MDM4N2Q4MjE5YjE=
+  data.tar.gz: !binary |-
+    MzVhZjE1MmQwYTNiYjM4NjY5MjY0NTE2NzA3MjRjOTE1MDc3YTdjMzdiYjEw
+    NWFlODJjMzI1MTY1MzA2ZDg0MTMzZmU1NGJjZGQ2ODEzNTEwODQyMTE3NWVh
+    YWJiZWY5YjhkNjUzODcyMDczNTExNzgwMWFhM2FhMjAyNDgzMzQ=

data/lib/pdfinfo.rb

@@ -1,79 +1,62 @@
 require 'open3'
 require 'shellwords'
+require 'date'
 require 'time'
+%w(object_to_hash errors page).each {|f| require File.expand_path("../pdfinfo/#{f}", __FILE__)}
 
 class Pdfinfo
-  DIMENSIONS_REGEXP = /([\d\.]+) x ([\d\.]+)/
+  include ObjectToHash
 
-  class PdfinfoError < ::StandardError
-  end
+  attr_reader :pages, :title, :subject, :keywords, :author, :creator,
+    :creation_date, :modified_date, :usage_rights, :producer,
+    :form, :page_count, :width, :height, :file_size, :pdf_version
 
-  class CommandNotFound < PdfinfoError
-    def initialize(command)
-      super("Command Not Found - '#{command}'")
+  class << self
+    def pdfinfo_command
+      @pdfinfo_command || 'pdfinfo'
     end
-  end
 
+    def pdfinfo_command=(cmd)
+      @pdfinfo_command = cmd
+    end
 
-  attr_reader :title,
-    :subject,
-    :keywords,
-    :author,
-    :creator,
-    :creation_date,
-    :usage_rights,
-    :producer,
-    :form,
-    :page_count,
-    :width,
-    :height,
-    :file_size,
-    :pdf_version
-
-  def self.exec(file_path, opts = {})
-    raise CommandNotFound, 'pdfinfo' unless pdfinfo_command?
-    flags = []
-    flags.concat(['-enc', opts.fetch(:encoding, 'UTF-8')])
-    flags.concat(['-opw', opts[:owner_password]]) if opts[:owner_password]
-    flags.concat(['-upw', opts[:user_password]]) if opts[:user_password]
+    # @return [Boolean]
+    def pdfinfo_command?
+      system("type #{pdfinfo_command} >/dev/null 2>&1")
+    end
 
-    command = Shellwords.join([pdfinfo_command, *flags, file_path.to_s])
-    stdout, status = Open3.capture2(command)
-    stdout.encode('UTF-8', invalid: :replace, replace: '', undef: :replace)
+    attr_accessor :config_path
   end
 
-  def self.pdfinfo_command
-    @pdfinfo_command || 'pdfinfo'
-  end
+  def initialize(source_path, opts = {})
+    @pages = []
 
-  def self.pdfinfo_command=(cmd)
-    @pdfinfo_command = cmd
-  end
+    info_hash = parse_shell_response(exec(source_path, opts))
 
-  def self.pdfinfo_command?
-    system("type #{pdfinfo_command} >/dev/null 2>&1")
-  end
+    info_hash.delete_if do |key, value|
+      @pages << Page.from_string(value) if key.match(/Page\s+\d+\ssize/)
+    end
 
-  def initialize(source_path, opts = {})
-    info_hash = parse_shell_response(Pdfinfo.exec(source_path, opts))
-
-    @title          = presence(info_hash['Title'])
-    @subject        = presence(info_hash['Subject'])
-    @author         = presence(info_hash['Author'])
-    @creator        = presence(info_hash['Creator'])
-    @producer       = presence(info_hash['Producer'])
-    @tagged         = !!(info_hash['Tagged'] =~ /yes/)
-    @encrypted      = !!(info_hash['Encrypted'] =~ /yes/)
-    @page_count     = info_hash['Pages'].to_i
-    @file_size      = info_hash['File size'].to_i
-    @form           = info_hash['Form']
-    @pdf_version    = info_hash['PDF version']
-
-    @keywords       = (info_hash['Keywords'] || '').split(/\s/)
-    @creation_date  = parse_time(info_hash['CreationDate'])
-
-    raw_usage_rights = Hash[info_hash['Encrypted'].scan(/(\w+):(\w+)/)]
-    booleanize_usage_right = lambda {|val| !(raw_usage_rights[val] == 'no') }
+    encrypted_val = info_hash.delete('Encrypted')
+
+    @title          = presence(info_hash.delete('Title'))
+    @subject        = presence(info_hash.delete('Subject'))
+    @author         = presence(info_hash.delete('Author'))
+    @creator        = presence(info_hash.delete('Creator'))
+    @producer       = presence(info_hash.delete('Producer'))
+    @tagged         = !!(info_hash.delete('Tagged') =~ /yes/)
+    @encrypted      = !!(encrypted_val =~ /yes/)
+    @page_count     = info_hash.delete('Pages').to_i
+    @file_size      = info_hash.delete('File size').to_i
+    @form           = info_hash.delete('Form')
+    @pdf_version    = info_hash.delete('PDF version')
+    @optimized      = !!(info_hash.delete('Optimized') =~ /yes/)
+    @keywords       = (info_hash.delete('Keywords') || '').split(/\s/)
+    @creation_date  = parse_time(info_hash.delete('CreationDate'))
+    @modified_date  = parse_time(info_hash.delete('ModDate'))
+
+    raw_usage_rights = Hash[encrypted_val.scan(/(\w+):(\w+)/)]
+    booleanize_usage_right = lambda {|val| raw_usage_rights[val] != 'no' }
 
     @usage_rights = {}.tap do |ur|
       ur[:print]     = booleanize_usage_right.call('print')
@@ -82,47 +65,104 @@ class Pdfinfo
       ur[:add_notes] = booleanize_usage_right.call('addNotes')
     end
 
-    @width, @height = extract_page_dimensions(info_hash['Page size'])
+    # temporarily continue setting #width and #height on Pdfinfo object
+    # to maintain legacy behavior
+    @width  = @pages[0].width
+    @height = @pages[0].height
   end
 
+  # @return [Boolean]
   def tagged?
     @tagged
   end
 
+  # @return [Boolean]
   def encrypted?
     @encrypted
   end
 
+  # @return [Boolean]
+  def optimized?
+    @optimized
+  end
+
   %w(print copy change).each do |ur|
+    # @return [Boolean]
     define_method("#{ur}able?") { @usage_rights[ur.to_sym] }
   end
   alias modifiable? changeable?
 
+  # @return [Boolean]
   def annotatable?
     @usage_rights[:add_notes]
   end
 
-  def to_hash
-    instance_variables.inject({}) {|h, var| h[var[1..-1].to_sym] = instance_variable_get(var); h }
-  end
   private
+  # executes pdfinfo command with supplied options
+  # @param [String,Pathname] file_path
+  # @param [Hash] opts
+  # @return [String] output
+  def exec(file_path, opts = {})
+    validate_pdfinfo_command!
+
+    flags = build_options(opts)
+
+    command = [self.class.pdfinfo_command, *flags, file_path.to_s].shelljoin
+
+    stdout, status = Open3.capture2(command)
+    force_utf8_encoding(stdout)
+  end
+
+  # prepares array of flags to pass as command line options
+  # ---
+  # @todo: add option builder class
+  # +++
+  # @param [Hash] opts of options
+  # @option opts [String] :encoding ('UTF-8')
+  # @option opts [String] :owner_password
+  # @option opts [String] :user_password
+  # @return [Array<String>] array of flags
+  def build_options(opts = {})
+    flags = []
+    xpdfrc_path = opts.fetch(:config_path, self.class.config_path)
+
+    # these flag will always be part of the cli options. Values have defaults and can be overridden
+    flags.concat(['-f',   opts.fetch(:first_page, 0)])
+    flags.concat(['-l',   opts.fetch(:last_page, -1)])
+    flags.concat(['-enc', opts.fetch(:encoding, Encoding::UTF_8)])
+    # optional flags.  if no value, the flag wont be added
+    flags.concat(['-cfg', xpdfrc_path])           if xpdfrc_path
+    flags.concat(['-opw', opts[:owner_password]]) if opts[:owner_password]
+    flags.concat(['-upw', opts[:user_password]])  if opts[:user_password]
+    flags.map(&:to_s)
+  end
+
+  # @param [String] str
+  # @return [String] UTF-8 encoded string
+  def force_utf8_encoding(str)
+    return str if str.valid_encoding?
+    str = str.encode(Encoding::UTF_16, invalid: :replace, undef: :replace, replace: '')
+    str.encode!(Encoding::UTF_8)
+  end
+
   def presence(val)
-    (val.nil? || val.empty?) ? nil : val
+    (val.nil? || val.empty? ) ? nil : val
   end
 
   def parse_shell_response(response_str)
     Hash[response_str.split(/\n/).map {|kv| kv.split(/:/, 2).map(&:strip) }]
   end
 
-  def extract_page_dimensions(str)
-    return unless str
-    str.match(DIMENSIONS_REGEXP).captures.map(&:to_f)
-  end
-
   def parse_time(str)
-    presence(str) ? Time.parse(str) : nil
+    return unless presence(str)
+    DateTime.strptime(str, '%a %b %e %H:%M:%S %Y')
   rescue ArgumentError => e
-    warn(e.message)
     nil
   end
+
+  def validate_pdfinfo_command!
+    unless self.class.pdfinfo_command?
+      raise CommandNotFound, self.class.pdfinfo_command
+    end
+  end
 end

metadata

@@ -1,83 +1,55 @@
 --- !ruby/object:Gem::Specification
 name: pdfinfo
 version: !ruby/object:Gem::Version
-  version: 1.2.0
+  version: 1.3.1
 platform: ruby
 authors:
 - Ryan Venegas
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-11-22 00:00:00.000000000 Z
+date: 2014-12-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.6'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.6'
-- !ruby/object:Gem::Dependency
-  name: prawn
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 3.1.0
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 3.1.0
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: coveralls
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 description: Simple ruby wrapper around the pdfinfo executable
@@ -97,17 +69,17 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: 1.9.3
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.4
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: Simple ruby wrapper around the pdfinfo executable