mindee 1.0.0

data/lib/mindee/fields/list_field.rb

@@ -0,0 +1,71 @@
+# frozen_string_literal: true
+
+module Mindee
+  # Field in a list.
+  class ListFieldItem
+    # The confidence score, value will be between 0.0 and 1.0
+    # @return [Float]
+    attr_accessor :confidence
+    # @return [Array<Array<Float>>]
+    attr_reader :bbox
+    # @return [Array<Array<Float>>]
+    attr_reader :polygon
+    attr_reader :content
+
+    # @param prediction [Hash]
+    def initialize(prediction)
+      @content = prediction['content']
+      @confidence = prediction['confidence']
+      @polygon = prediction['polygon']
+      @bbox = Geometry.get_bbox_as_polygon(@polygon) unless @polygon.nil? || @polygon.empty?
+    end
+
+    # @return [String]
+    def to_s
+      @content.to_s
+    end
+  end
+
+  # Field where actual values are kept in a list (custom docs).
+  class ListField
+    # @return [Array<Mindee::ListFieldItem>]
+    attr_reader :values
+    # @return [Integer, nil]
+    attr_reader :page_id
+    # true if the field was reconstructed or computed using other fields.
+    # @return [Boolean]
+    attr_reader :reconstructed
+    # The confidence score, value will be between 0.0 and 1.0
+    # @return [Float]
+    attr_accessor :confidence
+
+    # @param prediction [Hash]
+    # @param page_id [Integer, nil]
+    # @param reconstructed [Boolean]
+    def initialize(prediction, page_id, reconstructed: false)
+      @values = []
+      @confidence = prediction['confidence']
+      @page_id = page_id || prediction['page_id']
+      @reconstructed = reconstructed
+
+      prediction['values'].each do |field|
+        @values.push(ListFieldItem.new(field))
+      end
+    end
+
+    # @return [Array]
+    def contents_list
+      @values.map(&:content)
+    end
+
+    # @return [String]
+    def contents_str(separator: ' ')
+      @values.map(&:to_s).join(separator)
+    end
+
+    # @return [String]
+    def to_s
+      contents_str(separator: ' ')
+    end
+  end
+end

data/lib/mindee/fields/locale.rb

@@ -0,0 +1,45 @@
+# frozen_string_literal: true
+
+module Mindee
+  # Represents locale information
+  class Locale
+    # The confidence score, value will be between 0.0 and 1.0
+    # @return [Float]
+    attr_reader :confidence
+    #  Language code in ISO 639-1 format.
+    # @return [String]
+    attr_reader :language
+    # Country code in ISO 3166-1 alpha-2 format.
+    # @return [String, nil]
+    attr_reader :country
+    # Currency code in ISO 4217 format.
+    # @return [String]
+    attr_reader :currency
+    # Language code, with country code when available.
+    # @return [String]
+    attr_reader :value
+
+    # @param prediction [Hash]
+    def initialize(prediction)
+      value_key = if prediction.include? 'value'
+                    'value'
+                  else
+                    'language'
+                  end
+      @confidence = prediction['confidence']
+      @value = prediction[value_key]
+      @language = prediction['language']
+      @country = prediction['country']
+      @currency = prediction['currency']
+    end
+
+    def to_s
+      out_str = String.new
+      out_str << "#{@value}; " if @value
+      out_str << "#{@language}; " if @language
+      out_str << "#{@country}; " if @country
+      out_str << "#{@currency}; " if @currency
+      out_str.strip
+    end
+  end
+end

data/lib/mindee/fields/orientation.rb

@@ -0,0 +1,26 @@
+# frozen_string_literal: true
+
+module Mindee
+  # Represents page orientation.
+  class Orientation
+    # @return [Integer]
+    attr_reader :page_id
+    # The confidence score, value will be between 0.0 and 1.0
+    # @return [Float]
+    attr_reader :confidence
+    # A prediction among these 3 possible outputs:
+    #   * 0 degrees: the page is already upright
+    #   * 90 degrees: the page must be rotated clockwise to be upright
+    #   * 270 degrees: the page must be rotated counterclockwise to be upright
+    # @return [Integer]
+    attr_reader :degrees
+
+    # @param prediction [Hash]
+    # @param page_id [Integer]
+    def initialize(prediction, page_id)
+      @degrees = prediction['degrees']
+      @confidence = prediction['confidence']
+      @page_id = page_id
+    end
+  end
+end

data/lib/mindee/fields/payment_details.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+require_relative 'base'
+
+module Mindee
+  # Represents payment details for invoices and receipts
+  class PaymentDetails < Field
+    attr_reader :account_number,
+                :iban,
+                :routing_number,
+                :swift
+
+    # @param prediction [Hash]
+    # @param page_id [Integer, nil]
+    # @param reconstructed [Boolean]
+    def initialize(prediction, page_id, reconstructed: false)
+      super
+      @account_number = prediction['account_number']
+      @iban = prediction['iban']
+      @routing_number = prediction['routing_number']
+      @swift = prediction['swift']
+    end
+
+    def to_s
+      out_str = String.new
+      out_str << "#{@account_number}; " if @account_number
+      out_str << "#{@iban}; " if @iban
+      out_str << "#{@routing_number}; " if @routing_number
+      out_str << "#{@swift}; " if @swift
+      out_str.strip
+    end
+  end
+end

data/lib/mindee/fields/tax.rb

@@ -0,0 +1,35 @@
+# frozen_string_literal: true
+
+require_relative 'base'
+
+module Mindee
+  # Represents tax information.
+  class TaxField < Field
+    # Tax value as 3 decimal float
+    # @return [Float]
+    attr_reader :value
+    # Tax rate percentage
+    # @return [Float]
+    attr_reader :rate
+    # Tax code
+    # @return [String]
+    attr_reader :code
+
+    # @param prediction [Hash]
+    # @param page_id [Integer, nil]
+    def initialize(prediction, page_id)
+      super
+      @value = @value.round(3) unless @value.nil?
+      @rate = prediction['rate'].to_f unless prediction['rate'].nil?
+      @code = prediction['code'] unless prediction['code'] == 'None'
+    end
+
+    def to_s
+      out_str = String.new
+      out_str << "#{@value} " if @value
+      out_str << "#{@rate}% " if @rate
+      out_str << "#{@code} " if @code
+      out_str.strip
+    end
+  end
+end

data/lib/mindee/fields.rb

@@ -0,0 +1,11 @@
+# frozen_string_literal: true
+
+require_relative 'fields/amount'
+require_relative 'fields/base'
+require_relative 'fields/company_registration'
+require_relative 'fields/datefield'
+require_relative 'fields/locale'
+require_relative 'fields/orientation'
+require_relative 'fields/payment_details'
+require_relative 'fields/tax'
+require_relative 'fields/list_field'

data/lib/mindee/geometry.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+module Mindee
+  # Various helper functions for geometry.
+  module Geometry
+    # @return [Array<Float>]
+    def self.get_bbox(vertices)
+      x_min = vertices.map { |v| v[0] }.min
+      x_max = vertices.map { |v| v[0] }.max
+      y_min = vertices.map { |v| v[1] }.min
+      y_max = vertices.map { |v| v[1] }.max
+      [x_min, y_min, x_max, y_max]
+    end
+
+    # @return [Array<Array<Float>>]
+    def self.get_bbox_as_polygon(vertices)
+      x_min, y_min, x_max, y_max = get_bbox(vertices)
+      [[x_min, y_min], [x_max, y_min], [x_max, y_max], [x_min, y_max]]
+    end
+  end
+end

data/lib/mindee/inputs.rb

@@ -0,0 +1,153 @@
+# frozen_string_literal: true
+
+require 'stringio'
+require 'origami'
+require 'marcel'
+
+# Monkey-patching for Origami
+module PDFTools
+  def to_io_stream(params = {})
+    options = {
+      delinearize: true,
+      recompile: true,
+      decrypt: false,
+    }
+    options.update(params)
+
+    if frozen? # incompatible flags with frozen doc (signed)
+      options[:recompile] = nil
+      options[:rebuild_xrefs] = nil
+      options[:noindent] = nil
+      options[:obfuscate] = false
+    end
+    load_all_objects unless @loaded
+
+    intents_as_pdfa1 if options[:intent] =~ %r{pdf[/-]?A1?/i}
+    delinearize! if options[:delinearize] && linearized?
+    compile(options) if options[:recompile]
+
+    io_stream = StringIO.new(output(options))
+    io_stream.set_encoding Encoding::BINARY
+    io_stream
+  end
+end
+
+Origami::PDF.class_eval { include PDFTools }
+
+module Mindee
+  ALLOWED_MIME_TYPES = [
+    'application/pdf',
+    'image/heic',
+    'image/png',
+    'image/jpeg',
+    'image/tiff',
+    'image/webp',
+  ].freeze
+
+  MAX_DOC_PAGES = 3
+
+  # Base class for loading documents.
+  class InputDocument
+    # @return [String]
+    attr_reader :filename
+    # @return [String]
+    attr_reader :filepath
+    # @return [String]
+    attr_reader :file_mimetype
+
+    # @param cut_pages [Boolean]
+    # @param max_pages [Integer]
+    def initialize(cut_pages, max_pages)
+      @file_mimetype = Marcel::MimeType.for @io_stream, name: @filename
+
+      unless ALLOWED_MIME_TYPES.include? @file_mimetype
+        raise "File type not allowed, must be one of #{ALLOWED_MIME_TYPES.join(', ')}"
+      end
+
+      merge_pdf_pages(max_pages) if cut_pages && pdf?
+    end
+
+    def pdf?
+      @file_mimetype == 'application/pdf'
+    end
+
+    # @return [Integer]
+    def page_count
+      if pdf?
+        current_pdf = open_pdf
+        return current_pdf.pages.size
+      end
+      1
+    end
+
+    # @param close [Boolean]
+    def read_document(close: true)
+      @io_stream.seek(0)
+      data = @io_stream.read
+      @io_stream.close if close
+      [data].pack('m')
+    end
+
+    private
+
+    # @param max_pages [Integer]
+    def merge_pdf_pages(max_pages)
+      current_pdf = open_pdf
+      return if current_pdf.pages.size <= MAX_DOC_PAGES
+
+      new_pdf = Origami::PDF.new
+
+      to_insert = [current_pdf.pages[0], current_pdf.pages[-2], current_pdf.pages[-1]].take(max_pages)
+      to_insert.each_with_index do |page, idx|
+        new_pdf.insert_page(idx, page)
+      end
+      @io_stream = new_pdf.to_io_stream
+    end
+
+    # @return [Origami::PDF]
+    def open_pdf
+      pdf_parser = Origami::PDF::LinearParser.new({})
+      @io_stream.seek(0)
+      pdf_parser.parse(@io_stream)
+    end
+  end
+
+  # Load a document from a path.
+  class PathDocument < InputDocument
+    def initialize(filepath, cut_pages, max_pages: MAX_DOC_PAGES)
+      @io_stream = File.open(filepath, 'rb')
+      @filepath = filepath
+      @filename = File.basename(filepath)
+      super(cut_pages, max_pages)
+    end
+  end
+
+  # Load a document from a base64 string.
+  class Base64Document < InputDocument
+    def initialize(base64_string, filename, cut_pages, max_pages: 3)
+      @io_stream = StringIO.new(base64_string.unpack1('m*'))
+      @io_stream.set_encoding Encoding::BINARY
+      @filename = filename
+      super(cut_pages, max_pages)
+    end
+  end
+
+  # Load a document from raw bytes.
+  class BytesDocument < InputDocument
+    def initialize(raw_bytes, filename, cut_pages, max_pages: MAX_DOC_PAGES)
+      @io_stream = StringIO.new(raw_bytes)
+      @io_stream.set_encoding Encoding::BINARY
+      @filename = filename
+      super(cut_pages, max_pages)
+    end
+  end
+
+  # Load a document from a file handle.
+  class FileDocument < InputDocument
+    def initialize(file_handle, filename, cut_pages, max_pages: MAX_DOC_PAGES)
+      @io_stream = file_handle
+      @filename = filename
+      super(cut_pages, max_pages)
+    end
+  end
+end

data/lib/mindee/response.rb

@@ -0,0 +1,27 @@
+# frozen_string_literal: true
+
+module Mindee
+  # Stores all response attributes.
+  class DocumentResponse
+    # @return [String]
+    attr_reader :document_type
+    # @return [Hash]
+    attr_reader :http_response
+    # @return [Mindee::Document]
+    attr_reader :document
+    # @return [Array<Mindee::Document>]
+    attr_reader :pages
+
+    # @param http_response [Hash]
+    def initialize(http_response, document_type, document, pages)
+      @http_response = http_response
+      @document_type = document_type
+      @document = document
+      @pages = pages
+    end
+
+    def to_s
+      inspect
+    end
+  end
+end

data/lib/mindee/version.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+# Mindee
+module Mindee
+  VERSION = '1.0.0'
+
+  def self.find_platform
+    host = RbConfig::CONFIG['host_os']
+    platforms = {
+      linux: %r{linux|cygwin},
+      windows: %r{mswin|mingw|bccwin|wince|emx|win32},
+      macos: %r{mac|darwin},
+      bsd: %r{bsd},
+      solaris: %r{solaris|sunos},
+    }
+    platforms.each do |os, regexp|
+      return os unless (regexp =~ host).nil?
+    end
+  end
+  PLATFORM = find_platform.freeze
+end

data/lib/mindee.rb

@@ -0,0 +1,8 @@
+# frozen_string_literal: true
+
+require 'mindee/client'
+
+module Mindee
+  class Error < StandardError
+  end
+end

data/mindee.gemspec

@@ -0,0 +1,34 @@
+# frozen_string_literal: true
+
+require_relative 'lib/mindee/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = 'mindee'
+  spec.version       = Mindee::VERSION
+  spec.authors       = ['Mindee']
+  spec.email         = ['devrel@mindee.co']
+
+  spec.summary       = 'Mindee API Helper Library for Ruby'
+  #   spec.description   = %q{TODO: Write a longer description or delete this line.}
+  spec.homepage      = 'https://github.com/mindee/mindee-api-ruby'
+  spec.license       = 'MIT'
+  spec.required_ruby_version = Gem::Requirement.new('>= 2.6.0')
+
+  spec.metadata['homepage_uri'] = 'https://mindee.com/'
+  spec.metadata['source_code_uri'] = 'https://github.com/mindee/mindee-api-ruby'
+  spec.metadata['changelog_uri'] = 'https://github.com/mindee/mindee-api-ruby/blob/main/CHANGELOG.md'
+  spec.metadata['rubygems_mfa_required'] = 'true'
+
+  # Specify which files should be added to the gem when it is released.
+  # The `git ls-files -z` loads the files in the RubyGem that have been added into git.
+  spec.files = Dir.chdir(File.expand_path(__dir__)) do
+    `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(.github|spec|features)/}) }
+  end
+  spec.bindir        = 'bin'
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.require_paths = ['lib']
+
+  spec.add_runtime_dependency 'marcel', '~> 1.0.2'
+  spec.add_runtime_dependency 'mrz', '~> 0.2.0'
+  spec.add_runtime_dependency 'origami', '~> 2.1.0'
+end

metadata

@@ -0,0 +1,128 @@
+--- !ruby/object:Gem::Specification
+name: mindee
+version: !ruby/object:Gem::Version
+  version: 1.0.0
+platform: ruby
+authors:
+- Mindee
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2022-07-26 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: marcel
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.0.2
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.0.2
+- !ruby/object:Gem::Dependency
+  name: mrz
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 0.2.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 0.2.0
+- !ruby/object:Gem::Dependency
+  name: origami
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.1.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.1.0
+description: 
+email:
+- devrel@mindee.co
+executables:
+- console
+- mindee.rb
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitattributes"
+- ".gitignore"
+- ".rubocop.yml"
+- CHANGELOG.md
+- CODE_OF_CONDUCT.md
+- Gemfile
+- LICENSE
+- README.md
+- Rakefile
+- bin/console
+- bin/mindee.rb
+- lib/mindee.rb
+- lib/mindee/client.rb
+- lib/mindee/document_config.rb
+- lib/mindee/documents.rb
+- lib/mindee/documents/base.rb
+- lib/mindee/documents/custom.rb
+- lib/mindee/documents/financial_doc.rb
+- lib/mindee/documents/invoice.rb
+- lib/mindee/documents/passport.rb
+- lib/mindee/documents/receipt.rb
+- lib/mindee/endpoint.rb
+- lib/mindee/fields.rb
+- lib/mindee/fields/amount.rb
+- lib/mindee/fields/base.rb
+- lib/mindee/fields/company_registration.rb
+- lib/mindee/fields/datefield.rb
+- lib/mindee/fields/list_field.rb
+- lib/mindee/fields/locale.rb
+- lib/mindee/fields/orientation.rb
+- lib/mindee/fields/payment_details.rb
+- lib/mindee/fields/tax.rb
+- lib/mindee/geometry.rb
+- lib/mindee/inputs.rb
+- lib/mindee/response.rb
+- lib/mindee/version.rb
+- mindee.gemspec
+homepage: https://github.com/mindee/mindee-api-ruby
+licenses:
+- MIT
+metadata:
+  homepage_uri: https://mindee.com/
+  source_code_uri: https://github.com/mindee/mindee-api-ruby
+  changelog_uri: https://github.com/mindee/mindee-api-ruby/blob/main/CHANGELOG.md
+  rubygems_mfa_required: 'true'
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 2.6.0
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.1.2
+signing_key: 
+specification_version: 4
+summary: Mindee API Helper Library for Ruby
+test_files: []