ocrsdk 0.1.0

data/coverage/report.css

@@ -0,0 +1,97 @@
+body {
+  font-family: Verdana, Helvetica, Arial, Sans-Serif;
+  font-size: 12px;
+  color: #4C4C4C;
+  background-color: #F4F2ED;
+  padding: 1em;
+}
+
+pre, code {
+  color: #000000;
+  font-family: "Bitstream Vera Sans Mono","Monaco","Courier New",monospace;
+  font-size: 95%;
+  line-height: 1.3em;
+  margin-top: 0;
+  margin-bottom: 0;
+  padding: 0;
+  word-wrap: break-word;
+}
+
+h1, h2, h3, h4, h5, h6 {
+  margin: 0em 0em 1em 0em;
+  color: #666666;
+}
+
+h1 { 
+  display: block;
+  font-size: 2em;
+  letter-spacing: -1px;
+}
+
+h1 a:visited{
+  color: 666666;
+}
+
+h2 {
+  margin-top: -1em;	
+}
+
+.hit {
+  background-color: #BED2BE;
+  color: black;
+}
+
+.never {
+  background-color: #E0DEDB;
+  color: black;
+}
+
+.miss {
+  background-color: #CE8B8C;
+  color: black;
+}
+
+.line_number {
+  color: black;
+  font-weight: normal;
+  width: 1px;
+}
+
+.code_line {
+  padding-left: 5px;
+}
+
+table {
+  width: 100%;
+  border: 2px solid #999;
+}
+
+td {
+  padding: 2px;
+}
+
+tr {
+  font-size: 14px;
+}
+
+th {
+  padding: 2px;
+  background-color: #999;
+  color: white;
+  font-size: 12px;
+  text-align: left;
+}
+
+a:hover {
+  text-decoration: none;
+}
+
+#content .side_by_side {
+  width: 50%;
+  overflow: auto;
+  float: left;
+}
+
+#toggle_test_file {
+  color: #666666;
+}

data/lib/ocrsdk/abstract_entity.rb

@@ -0,0 +1,14 @@
+class OCRSDK::AbstractEntity
+  def initialize(application_id=nil, password=nil)
+    @application_id = application_id || '' # Rails.configuration.ocrsdk.application_id
+    @password       = password       || '' # Rails.configuration.ocrsdk.password
+
+    @url = prepare_url @application_id, @password
+  end
+
+private
+
+  def prepare_url(app_id, pass)
+    URI("http://#{CGI.escape app_id}:#{CGI.escape pass}@#{OCRSDK::SERVICE_URL}")
+  end
+end

data/lib/ocrsdk/document.rb

@@ -0,0 +1,8 @@
+class OCRSDK::Document < OCRSDK::AbstractEntity
+private
+
+  # http://ocrsdk.com/documentation/apireference/processDocument/
+  def api_process_document(document, languages, format='txt')
+
+  end
+end

data/lib/ocrsdk/errors.rb

@@ -0,0 +1,13 @@
+module OCRSDK
+  class OCRSDKError         < RuntimeError; end
+  
+  class NetworkError        < OCRSDKError; end
+  class NotEnoughCredits    < OCRSDKError; end
+  class ProcessingFailed    < OCRSDKError; end
+
+  class UnsupportedFeature  < OCRSDKError; end
+  class UnsupportedLanguage < UnsupportedFeature; end
+  class UnsupportedProfile  < UnsupportedFeature; end
+  class UnsupportedInputFormat  < UnsupportedFeature; end
+  class UnsupportedOutputFormat < UnsupportedFeature; end
+end

data/lib/ocrsdk/image.rb

@@ -0,0 +1,56 @@
+class OCRSDK::Image < OCRSDK::AbstractEntity
+  include OCRSDK::Verifiers::Language
+  include OCRSDK::Verifiers::Format
+  include OCRSDK::Verifiers::Profile
+
+  def initialize(image_path, application_id=nil, password=nil)
+    super(application_id, password)
+    @image_path = image_path
+  end
+
+  def as_text(languages)
+    xml_string = api_process_image @image_path, languages, :txt, :text_extraction
+
+    OCRSDK::Promise.from_response xml_string, @application_id, @password
+  end
+
+  def as_text_sync(languages, wait_interval=OCRSDK::DEFAULT_POLL_TIME)
+    as_text(languages).wait(wait_interval).result.force_encoding('utf-8')
+  end
+
+  def as_pdf(languages)
+    xml_string = api_process_image @image_path, languages, :pdf, :document_conversion
+
+    OCRSDK::Promise.from_response xml_string, @application_id, @password
+  end
+
+  def as_pdf_sync(languages, out_path=nil, wait_interval=OCRSDK::DEFAULT_POLL_TIME)
+    result = as_pdf(languages).wait(wait_interval).result
+
+    if out_path.nil?
+      result
+    else
+      File.open(out_path, 'wb+') {|f| f.write result }
+    end
+  end
+
+private
+
+  # TODO handle 4xx and 5xx responses and errors, file not found error
+  # http://ocrsdk.com/documentation/apireference/processImage/
+  def api_process_image(image_path, languages, format=:txt, profile=:document_conversion)
+    raise OCRSDK::UnsupportedInputFormat   unless supported_input_format? File.extname(image_path)[1..-1]
+    raise OCRSDK::UnsupportedOutputFormat  unless supported_output_format? format
+    raise OCRSDK::UnsupportedProfile       unless supported_profile? (profile)
+
+    params = URI.encode_www_form(
+              language: languages_to_s(languages).join(','),
+              exportFormat: format_to_s(format), 
+              profile: profile_to_s(profile))
+    uri = URI.join @url, '/processImage', "?#{params}"
+
+    RestClient.post uri.to_s, upload: { file: File.new(image_path, 'rb') }
+  rescue RestClient::ExceptionWithResponse
+    raise OCRSDK::NetworkError
+  end
+end

data/lib/ocrsdk/pdf.rb

@@ -0,0 +1,27 @@
+class OCRSDK::PDF < OCRSDK::Image
+  # We're on a shaky ground regarding what kind of pdfs
+  # should be recognized and what shouldn't.
+  # Currently we count that if there are
+  #   images * 20 > length of text
+  # then this document might need recognition.
+  # Assumption is that there might be a title,
+  # page numbers or credits along with images.
+  def recognizeable?
+    reader = PDF::Reader.new @image_path
+
+    images = 0
+    text   = 0
+    chars  = Set.new
+    reader.pages.each do |page|
+      text   += page.text.length
+      chars  += page.text.split('').map(&:ord).uniq
+      images += page.xobjects.map {|k, v| v.hash[:Subtype]}.count(:Image)
+    end
+
+    # count number of distinct characters
+    # in case of "searchable", but incorrectly recognized document
+    images * 20 > text || chars.length < 10
+  rescue PDF::Reader::MalformedPDFError, PDF::Reader::UnsupportedFeatureError
+    false
+  end
+end

data/lib/ocrsdk/promise.rb

@@ -0,0 +1,87 @@
+class OCRSDK::Promise < OCRSDK::AbstractEntity
+  include OCRSDK::Verifiers::Status
+
+  attr_reader :task_id, :status, :result_url, :estimate_processing_time
+
+  def self.from_response(xml_string, application_id=nil, password=nil)
+    OCRSDK::Promise.new(nil, application_id, password).parse_response xml_string
+  end
+
+  def initialize(task_id, application_id=nil, password=nil)
+    super(application_id, password)
+    @task_id = task_id
+  end
+
+  def estimate_completion
+    @registration_time + @estimate_processing_time.seconds
+  end
+
+  def parse_response(xml_string)
+    xml = Nokogiri::XML.parse xml_string
+    begin
+      task = xml.xpath('/response/task').first
+      @task_id = task['id']
+    rescue NoMethodError # if Nokogiri can't find root node
+      raise OCRSDK::OCRSDKError, "Problem parsing provided xml string: #{xml_string}"
+    end
+
+    @status     = status_to_sym task['status']
+    @result_url = task['resultUrl']
+    @registration_time        = DateTime.parse task['registrationTime']    
+    @estimate_processing_time = task['estimatedProcessingTime'].to_i
+
+    # admin should be notified in this case
+    raise OCRSDK::NotEnoughCredits  if @status == :not_enough_credits
+
+    self
+  end
+
+  def update
+    parse_response api_update_status
+  end
+
+  def completed?
+    @status == :completed
+  end
+
+  def failed?
+    [:processing_failed, :deleted, :not_enough_credits].include? @status
+  end
+
+  def processing?
+    [:submitted, :queued, :in_progress].include? @status
+  end
+
+  def result
+    raise OCRSDK::ProcessingFailed  if failed?
+    api_get_result
+  end
+
+  def wait(seconds=OCRSDK::DEFAULT_POLL_TIME)
+    while processing? do
+      sleep seconds
+      update
+    end
+
+    self
+  end
+
+private
+
+  # http://ocrsdk.com/documentation/apireference/getTaskStatus/
+  def api_update_status
+    params = URI.encode_www_form taskId: @task_id
+    uri    = URI.join @url, '/getTaskStatus', "?#{params}"
+
+    RestClient.get uri.to_s
+  rescue RestClient::ExceptionWithResponse
+    raise OCRSDK::NetworkError
+  end
+
+  def api_get_result
+    RestClient.get @result_url.to_s
+  rescue RestClient::ExceptionWithResponse
+    raise OCRSDK::NetworkError
+  end
+
+end

data/lib/ocrsdk/verifiers/format.rb

@@ -0,0 +1,26 @@
+module OCRSDK::Verifiers::Format
+  # http://ocrsdk.com/documentation/specifications/image-formats/
+  INPUT_FORMATS = [:bmp, :dcx, :pcx, :png, :jp2, :jpc, :jpg, :jpeg, :jfif, :pdf, 
+    :tif, :tiff, :gif, :djvu, :djv, :jb2]
+
+  # http://ocrsdk.com/documentation/apireference/processImage/
+  OUTPUT_FORMATS = [:txt, :rtf, :docx, :xlsx, :pptx, :pdf_searchable, 
+    :pdf_text_and_images, :xml, :alto]
+
+  def format_to_s(format)
+    format.to_s.camelize(:lower)
+  end
+
+  def supported_input_format?(format)
+    format = format.downcase.to_sym  if format.kind_of? String
+
+    INPUT_FORMATS.include? format
+  end
+
+  def supported_output_format?(format)
+    format = format.underscore.to_sym  if format.kind_of? String
+
+    OUTPUT_FORMATS.include? format
+  end
+
+end

data/lib/ocrsdk/verifiers/language.rb

@@ -0,0 +1,43 @@
+module OCRSDK::Verifiers::Language
+  # http://ocrsdk.com/documentation/specifications/recognition-languages/
+  LANGUAGES = [:afrikaans, :albanian, :aymara, :azeri_latin, :basque, :bemba, 
+    :blackfoot, :breton, :bugotu, :bulgarian, :buryat, :chamorro, :corsican,
+    :crimean_tatar, :croatian, :crow, :czech, :dutch, :dutch_belgian, :english,
+    :eskimo_latin, :esperanto, :estonian, :evenki, :faeroese, :fijian, :finnish,
+    :french, :frisian, :gaelic_scottish, :gagauz, :galician, :ganda, :german,
+    :german_law, :german_luxembourg, :german_medical, :german_new_spelling_law,
+    :greek, :hani, :hausa, :hebrew, :hungarian, :icelandic, :interlingua, :italian,
+    :japanese, :kabardian, :kasub, :kawa, :kikuyu, :kirgiz, :kongo, :korean_hangul,
+    :koryak, :kpelle, :lak, :lappish, :latvian, :lezgin, :macedonian, :malay, :malinke, 
+    :maltese, :mansi, :maori, :mari, :maya, :miao, :minankabaw, :mohawk, :nenets, :nogay, 
+    :norwegian_bokmal, :norwegian_nynorsk, :nyanja, :occidental, :old_english, :old_french, 
+    :old_german, :papiamento, :pidgin_english, :polish, :portuguese_brazilian, 
+    :portuguese_standard, :provencal, :quechua, :romanian, :romanian_moldavia, :romany, 
+    :rundi, :russian, :samoan, :selkup, :serbian_cyrillic, :shona, :sioux, :slovenian, 
+    :somali, :spanish, :sunda, :tabassaran, :tagalog, :tahitian, :tajik, :tatar, :tinpo, 
+    :tun, :turkish, :uighur_cyrillic, :ukrainian, :uzbek_cyrillic, :visayan]
+
+  def language_to_s(language)
+    language.to_s.camelize
+  end
+
+  def language_to_sym(language)
+    language.underscore.to_sym
+  end
+
+  def supported_language?(language)
+    language = language_to_sym language  if language.kind_of? String
+
+    LANGUAGES.include? language
+  end
+
+  def languages_to_s(languages)
+    languages = languages.map(&method(:language_to_s))
+    
+    unless languages.map(&method(:supported_language?)).all?
+      raise OCRSDK::UnsupportedLanguage
+    else
+      languages
+    end
+  end
+end

data/lib/ocrsdk/verifiers/profile.rb

@@ -0,0 +1,15 @@
+module OCRSDK::Verifiers::Profile
+  # http://ocrsdk.com/documentation/specifications/processing-profiles/
+  PROFILES = [:document_conversion, :document_archiving, :text_extraction,
+    :field_level_recognition, :barcode_recognition]
+
+  def profile_to_s(profile)
+    profile.to_s.camelize(:lower)
+  end
+
+  def supported_profile?(profile)
+    profile = profile.underscore.to_sym  if profile.kind_of? String
+
+    PROFILES.include? profile
+  end
+end

data/lib/ocrsdk/verifiers/status.rb

@@ -0,0 +1,19 @@
+module OCRSDK::Verifiers::Status
+  # http://ocrsdk.com/documentation/specifications/task-statuses/
+  STATUSES = [:submitted, :queued, :in_progress, :completed, 
+    :processing_failed, :deleted, :not_enough_credits]
+
+  def status_to_s(status)
+    status.to_s.camelize
+  end
+
+  def status_to_sym(status)
+    status.underscore.to_sym
+  end
+
+  def supported_status?(status)
+    status = status_to_sym status  if status.kind_of? String
+
+    STATUSES.include? status
+  end
+end

data/lib/ocrsdk/verifiers.rb

@@ -0,0 +1,9 @@
+# Intended to be included into target classes
+# they serve the purpose of validating and converting
+# symbols between internal and outside api representations
+module OCRSDK::Verifiers
+end
+
+Dir[File.dirname(__FILE__) + '/verifiers/*.rb'].each do |file| 
+  require file
+end

data/lib/ocrsdk.rb

@@ -0,0 +1,21 @@
+require 'cgi'
+require 'uri'
+require 'nokogiri'
+require 'pdf-reader'
+require 'rest-client'
+require 'active_support/inflector'
+require 'active_support/time'
+
+# http://ocrsdk.com/documentation/apireference/
+module OCRSDK
+  DEFAULT_POLL_TIME = 3
+  SERVICE_URL = 'cloud.ocrsdk.com'
+end
+
+require 'ocrsdk/errors'
+require 'ocrsdk/verifiers'
+require 'ocrsdk/abstract_entity'
+require 'ocrsdk/image'
+require 'ocrsdk/pdf'
+require 'ocrsdk/document'
+require 'ocrsdk/promise'

data/ocrsdk.gemspec

@@ -0,0 +1,28 @@
+Gem::Specification.new do |s|
+  s.name = "ocrsdk"
+  s.version = File.read("VERSION").delete("\n\r")
+  s.authors = ["Andrew Korzhuev"]
+  s.description = %q{Abbyy's OCR (ocrsdk.com) API wrapper in Ruby.}
+  s.summary = %q{Abbyy's OCR (ocrsdk.com) API wrapper in Ruby.}
+  s.email = "andrew@korzhuev.com"
+  s.extra_rdoc_files = ["LICENSE", "README.md"]
+  s.rdoc_options = ["--charset=UTF-8"]
+  s.homepage = "http://github.com/andrusha/ocrsdk"
+  s.require_paths = ["lib"]
+  s.executables = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
+  s.files = `git ls-files`.split("\n")
+  s.test_files = `git ls-files -- {test,spec,features}/*`.split("\n")
+  s.add_runtime_dependency "rest-client" #git://github.com/babatakao/rest-client.git
+  s.add_runtime_dependency "nokogiri"
+  s.add_runtime_dependency "pdf-reader"
+  s.add_runtime_dependency "activesupport"
+
+  s.add_development_dependency "rake", ">= 0.8"
+  if RUBY_VERSION >= '1.9'
+    s.add_development_dependency "cover_me", ">= 1.2.0"
+  else
+    s.add_development_dependency "rcov", ">= 0.9"
+  end
+  s.add_development_dependency "rspec", ">= 2"
+  s.add_development_dependency "webmock"
+end

data/spec/helpers/ocrsdk_helpers.rb

@@ -0,0 +1,106 @@
+module OCRSDKHelpers
+
+  def process_image_response
+    <<-XML
+      <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+      <response>
+        <task id="22345200-abe8-4f60-90c8-0d43c5f6c0f6"
+          registrationTime="2001-01-01T13:18:22Z"
+          statusChangeTime="2001-01-01T13:18:22Z"
+          status="Submitted"
+          error="{An error message.}"
+          filesCount="10"
+          credits="10"
+          estimatedProcessingTime="3600"
+          resultUrl="http://domain/blob ID"
+          description="My first OCR task"/>
+        <task/>
+      </response>
+    XML
+  end
+
+  def process_image_response_credits
+    <<-XML
+      <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+      <response>
+        <task id="22345200-abe8-4f60-90c8-0d43c5f6c0f6"
+          registrationTime="2001-01-01T13:18:22Z"
+          statusChangeTime="2001-01-01T13:18:22Z"
+          status="NotEnoughCredits"
+          error="{An error message.}"
+          filesCount="10"
+          credits="0"
+          estimatedProcessingTime="3600"
+          resultUrl="http://domain/blob ID"
+          description="My first OCR task"/>
+        <task/>
+      </response>
+    XML
+  end
+
+  def process_image_updated_response
+    <<-XML
+      <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+      <response>
+        <task id="update-task-id"
+          registrationTime="2001-01-01T13:18:22Z"
+          statusChangeTime="2001-02-01T13:18:22Z"
+          status="InProgress"
+          error="{An error message.}"
+          filesCount="10"
+          credits="10"
+          estimatedProcessingTime="3600"
+          resultUrl="http://domain/blob ID"
+          description="My first OCR task"/>
+        <task/>
+      </response>
+    XML
+  end
+
+  def process_image_completed_response
+    <<-XML
+      <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+      <response>
+        <task id="update-task-id"
+          registrationTime="2001-01-01T13:18:22Z"
+          statusChangeTime="2001-03-01T13:18:22Z"
+          status="Completed"
+          error="{An error message.}"
+          filesCount="10"
+          credits="10"
+          estimatedProcessingTime="3600"
+          resultUrl="http://domain/blob ID"
+          description="My first OCR task"/>
+        <task/>
+      </response>
+    XML
+  end
+
+  def process_image_failed_response
+    <<-XML
+      <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+      <response>
+        <task id="update-task-id"
+          registrationTime="2001-01-01T13:18:22Z"
+          statusChangeTime="2001-03-01T13:18:22Z"
+          status="ProcessingFailed"
+          error="{An error message.}"
+          filesCount="10"
+          credits="10"
+          estimatedProcessingTime="3600"
+          resultUrl="http://domain/blob ID"
+          description="My first OCR task"/>
+        <task/>
+      </response>
+    XML
+  end
+
+  def mock_ocrsdk
+    OCRSDK::Image.any_instance.stub(:api_process_image) { |x,y,i,j| process_image_response }
+    OCRSDK::Promise.any_instance.stub(:api_update_status) { process_image_completed_response }
+    OCRSDK::Promise.any_instance.stub(:api_get_result) { "meow" }    
+  end
+
+end
+
+RSpec.configuration.include OCRSDKHelpers