class Gini::Api::Document

Contains document related data from uploaded or fetched document

Attributes

duration[RW]

Public Class Methods

new(api, location, from_data = nil) click to toggle source

Instantiate a new Gini::Api::Document object from URL

@param [Gini::Api::Client] api Gini::Api::Client object @param [String] location Document URL @param [Hash] from_data Hash with doc data (from search for example)

# File lib/gini-api/document.rb, line 16
def initialize(api, location, from_data = nil)
  @api      = api
  @location = location

  update(from_data)
end

Public Instance Methods

completed?() click to toggle source

Indicate if the document has been processed

@return [Boolean] true if progress == PENDING

# File lib/gini-api/document.rb, line 71
def completed?
  @progress != 'PENDING'
end
extractions(options = {}) click to toggle source

Initialize extractions from @_links and return Gini::Api::Extractions object

@param [Hash] options Options @option options [Boolean] :refresh Invalidate extractions cache @option options [Boolean] :incubator Return experimental extractions

@return [Gini::Api::Document::Extractions] Return Gini::Api::Document::Extractions object for uploaded document

# File lib/gini-api/document.rb, line 110
def extractions(options = {})
  opts = { refresh: false, incubator: false }.merge(options)
  if opts[:refresh] or @extractions.nil?
    @extractions = Gini::Api::Document::Extractions.new(@api, @_links[:extractions], opts[:incubator])
  else
    @extractions
  end
end
layout() click to toggle source

Initialize layout from @_links and return Gini::Api::Layout object

@return [Gini::Api::Document::Layout] Return Gini::Api::Document::Layout object for uploaded document

# File lib/gini-api/document.rb, line 123
def layout
  @layout ||= Gini::Api::Document::Layout.new(@api, @_links[:layout])
end
pages() click to toggle source

Override @pages instance variable. Removes key :pageNumber, key :images and starts by index 0. Page 1 becomes index 0

# File lib/gini-api/document.rb, line 130
def pages
  @pages.map { |page| page[:images] }
end
poll(interval) { |self| ... } click to toggle source

Poll document progress and return when state equals COMPLETED Known states are PENDING, COMPLETED and ERROR

@param [Float] interval API polling interval

# File lib/gini-api/document.rb, line 58
def poll(interval, &block)
  until @progress =~ /(COMPLETED|ERROR)/ do
    update
    yield self if block_given?
    sleep(interval)
  end
  nil
end
processed() click to toggle source

Get processed document

@return [data] The binary representation of the processed document (pdf, jpg, png, …)

# File lib/gini-api/document.rb, line 87
def processed
  response = @api.request(
    :get,
    @_links[:processed],
    headers: { accept: 'application/octet-stream' }
  )
  unless response.status == 200
    raise Gini::Api::DocumentError.new(
      "Failed to fetch processed document (code=#{response.status})",
      response
    )
  end
  response.body
end
report_error(summary = nil, description = nil) click to toggle source

Submit error report on document

@param [String] summary Short summary on the error found @param [String] description More detailed description of the error found

@return [String] Error ID retured from API

# File lib/gini-api/document.rb, line 165
def report_error(summary = nil, description = nil)
  response = @api.request(
    :post,
    "#{@_links[:document]}/errorreport",
    params: { summary: summary, description: description }
  )
  unless response.status == 200
    raise Gini::Api::DocumentError.new(
      "Failed to submit error report for document #{@id} (code=#{response.status})",
      response
    )
  end
  response.parsed[:errorId]
end
submit_feedback(label, value) click to toggle source

Submit feedback on extraction label

@deprecated Use ‘doc.extractions.LABEL = VALUE’ instead. Will be removed in next version @param [String] label Extraction label to submit feedback on @param [String] value The new value for the given label

# File lib/gini-api/document.rb, line 140
def submit_feedback(label, value)
  unless extractions.send(label.to_sym)
    raise Gini::Api::DocumentError.new("Unknown label #{label}: Not found")
  end
  response = @api.request(
    :put,
    "#{@_links[:extractions]}/#{label}",
    headers: { 'content-type' => @api.version_header[:accept] },
    body: { value: value }.to_json
  )
  unless response.status == 204
    raise Gini::Api::DocumentError.new(
      "Failed to submit feedback for label #{label} (code=#{response.status})",
      response
    )
  end
end
successful?() click to toggle source

Was the document processed successfully?

@return [Boolean] true/false based on @progress

# File lib/gini-api/document.rb, line 79
def successful?
  @progress == 'COMPLETED'
end
update(from_data = nil) click to toggle source

Fetch document resource and populate instance variables

@param [Hash] from_data Ruby hash with doc data

# File lib/gini-api/document.rb, line 27
def update(from_data = nil)
  data = {}

  if from_data.nil?
    response = @api.request(:get, @location)
    unless response.status == 200
      raise Gini::Api::DocumentError.new(
        "Failed to fetch document data (code=#{response.status})",
        response
      )
    end
    data = response.parsed
  else
    data = from_data
  end

  data.each do |k, v|
    instance_variable_set("@#{k}", v)

    # We skip pages as it's rewritted by method pages()
    next if k == :pages

    self.class.send(:attr_reader, k)
  end
end