refactor image module

sebastianMindee · sebastianMindee · commit 7aa4fb5e652b · 2025-01-27T15:13:47.000+01:00
diff --git a/examples/auto_invoice_splitter_extraction.rb b/examples/auto_invoice_splitter_extraction.rb
@@ -22,7 +22,7 @@ def parse_single_page(mindee_client, input_source)
 end
 
 def parse_multi_page(mindee_client, input_source)
-  pdf_extractor = Mindee::Extraction::PdfExtractor::PdfExtractor.new(input_source)
+  pdf_extractor = Mindee::Image::PdfExtractor::PdfExtractor.new(input_source)
   invoice_splitter_response = mindee_client.enqueue_and_parse(
     input_source,
     Mindee::Product::InvoiceSplitter::InvoiceSplitterV1,
diff --git a/examples/auto_multi_receipts_detector_extraction.rb b/examples/auto_multi_receipts_detector_extraction.rb
@@ -13,7 +13,7 @@ def multi_receipts_detection(file_path, mindee_client)
     close_file: false
   )
 
-  images = Mindee::Extraction::MultiReceiptsExtractor.extract_receipts(input_source, result_split.document.inference)
+  images = Mindee::Image::MultiReceiptsExtractor.extract_receipts(input_source, result_split.document.inference)
   images.each do |sub_image|
     # Optional: Save the files locally
     # sub_image.write_to_file("/path/to/my/extracted/file/folder")
diff --git a/lib/mindee.rb b/lib/mindee.rb
@@ -34,7 +34,7 @@ module ImageCompressor
   end
 
   # Custom extraction module
-  module Extraction
+  module Image
   end
 
   # Parsing internals and fields.
diff --git a/lib/mindee/extraction.rb b/lib/mindee/extraction.rb
@@ -2,5 +2,4 @@
 
 require_relative 'extraction/tax_extractor'
 require_relative 'extraction/multi_receipts_extractor'
-require_relative 'extraction/common'
 require_relative 'extraction/pdf_extractor'
diff --git a/lib/mindee/extraction/common.rb b/lib/mindee/extraction/common.rb
diff --git a/lib/mindee/extraction/multi_receipts_extractor/multi_receipts_extractor.rb b/lib/mindee/extraction/multi_receipts_extractor/multi_receipts_extractor.rb
@@ -1,10 +1,10 @@
 # frozen_string_literal: true
 
-require_relative '../common/image_extractor'
+require_relative '../../image/image_extractor'
 
 module Mindee
   # Image Extraction Module.
-  module Extraction
+  module Image
     # Multi-receipts extraction class wrapper.
     class MultiReceiptsExtractor
       def self.extract_receipts(input_source, inference)
@@ -23,8 +23,8 @@ def self.extract_receipts(input_source, inference)
         (0...input_source.count_pdf_pages).each do |page_id|
           receipt_positions = inference.pages[page_id].prediction.receipts.map(&:bounding_box)
           images.concat(
-            Mindee::Extraction::ImageExtractor.extract_multiple_images_from_source(input_source, page_id + 1,
-                                                                                   receipt_positions)
+            Mindee::Image::ImageExtractor.extract_multiple_images_from_source(input_source, page_id + 1,
+                                                                              receipt_positions)
           )
         end
 
diff --git a/lib/mindee/extraction/pdf_extractor/extracted_pdf.rb b/lib/mindee/extraction/pdf_extractor/extracted_pdf.rb
@@ -2,7 +2,7 @@
 
 module Mindee
   # Pdf Extraction Module.
-  module Extraction
+  module Image
     module PdfExtractor
       # An extracted sub-Pdf.
       class ExtractedPdf
diff --git a/lib/mindee/extraction/pdf_extractor/pdf_extractor.rb b/lib/mindee/extraction/pdf_extractor/pdf_extractor.rb
@@ -2,7 +2,7 @@
 
 module Mindee
   # Pdf Extraction Module.
-  module Extraction
+  module Image
     # Pdf Extraction class.
     module PdfExtractor
       # Pdf extraction class.
@@ -13,7 +13,7 @@ def initialize(local_input)
           if local_input.pdf?
             @source_pdf = local_input.io_stream
           else
-            pdf_image = Extraction::ImageExtractor.attach_image_as_new_file(local_input.io_stream)
+            pdf_image = Image::ImageExtractor.attach_image_as_new_file(local_input.io_stream)
             io_buffer = StringIO.new
             pdf_image.save(io_buffer)
 
@@ -40,7 +40,7 @@ def cut_pages(page_indexes)
 
         # Extract the sub-documents from the main pdf, based on the given list of page indexes.
         # @param page_indexes [Array<Array<Integer>>] List of page number to use for merging in the original Pdf.
-        # @return [Array<Mindee::Extraction::PdfExtractor::ExtractedPdf>] The buffer containing the new Pdf.
+        # @return [Array<Mindee::Image::PdfExtractor::ExtractedPdf>] The buffer containing the new Pdf.
         def extract_sub_documents(page_indexes)
           extracted_pdfs = []
           extension = File.extname(@filename)
@@ -59,8 +59,8 @@ def extract_sub_documents(page_indexes)
             formatted_max_index = format('%03d', page_index_list[page_index_list.length - 1] + 1).to_s
             field_filename = "#{basename}_#{format('%03d',
                                                    (page_index_list[0] + 1))}-#{formatted_max_index}#{extension}"
-            extracted_pdf = Mindee::Extraction::PdfExtractor::ExtractedPdf.new(cut_pages(page_index_list),
-                                                                               field_filename)
+            extracted_pdf = Mindee::Image::PdfExtractor::ExtractedPdf.new(cut_pages(page_index_list),
+                                                                          field_filename)
             extracted_pdfs << extracted_pdf
           end
           extracted_pdfs
@@ -72,7 +72,7 @@ def extract_sub_documents(page_indexes)
         # Extracts invoices as complete PDFs from the document.
         # @param page_indexes [Array<Array<Integer>, InvoiceSplitterV1PageGroup>]
         # @param strict [Boolean]
-        # @return [Array<Mindee::Extraction::PdfExtractor::ExtractedPdf>]
+        # @return [Array<Mindee::Image::PdfExtractor::ExtractedPdf>]
         def extract_invoices(page_indexes, strict: false)
           raise Errors::MindeePDFError, 'No indexes provided.' if page_indexes.empty?
           unless page_indexes[0].is_a?(Mindee::Product::InvoiceSplitter::InvoiceSplitterV1PageGroup)
diff --git a/lib/mindee/extraction/tax_extractor/ocr_extractor.rb b/lib/mindee/extraction/tax_extractor/ocr_extractor.rb
@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 module Mindee
-  module Extraction
+  module Image
     # Generic extractor class
     class OcrExtractor
       # Checks for a list of possible matches in a string & returns the index of the first found candidate.
diff --git a/lib/mindee/extraction/tax_extractor/tax_extractor.rb b/lib/mindee/extraction/tax_extractor/tax_extractor.rb
@@ -5,7 +5,7 @@
 # rubocop:disable Metrics/ClassLength
 
 module Mindee
-  module Extraction
+  module Image
     # Tax extractor class
     class TaxExtractor < OcrExtractor
       # Extracts the most relevant candidate.
diff --git a/lib/mindee/image.rb b/lib/mindee/image.rb
@@ -1,4 +1,6 @@
 # frozen_string_literal: true
 
+require_relative 'image/extracted_image'
 require_relative 'image/image_compressor'
+require_relative 'image/image_extractor'
 require_relative 'image/image_utils'
diff --git a/lib/mindee/image/extracted_image.rb b/lib/mindee/image/extracted_image.rb
@@ -1,16 +1,16 @@
 # frozen_string_literal: true
 
-require_relative '../../input/sources'
+require_relative '../input/sources'
 
 module Mindee
   # Image Extraction Module.
-  module Extraction
+  module Image
     # Generic class for image extraction.
     class ExtractedImage
-      # Id of the page the image was extracted from.
+      # ID of the page the image was extracted from.
       attr_reader :page_id
 
-      # Id of the element on a given page.
+      # ID of the element on a given page.
       attr_reader :element_id
 
       # Buffer object of the file's content.
@@ -51,13 +51,15 @@ def save_to_file(output_path, file_format = nil)
 
           file_format = resolved_path.extname.delete('.').upcase
         end
-        @buffer.rewind
-        image = MiniMagick::Image.read(@buffer)
-        image.format file_format.downcase
-        image.write resolved_path.to_s
-      rescue StandardError
-        raise Errors::MindeeImageError, "Could not save file '#{output_path}'. " \
-                                        'Is the provided file path valid?.'
+        begin
+          @buffer.rewind
+          image = MiniMagick::Image.read(@buffer)
+          image.format file_format.downcase
+          image.write resolved_path.to_s
+        rescue StandardError
+          raise Errors::MindeeImageError, "Could not save file '#{output_path}'. " \
+                                          'Is the provided file path valid?.'
+        end
       end
 
       # Return the file as a Mindee-compatible BufferInput source.
diff --git a/lib/mindee/image/image_extractor.rb b/lib/mindee/image/image_extractor.rb
@@ -4,12 +4,12 @@
 require 'origami'
 require 'stringio'
 require 'tempfile'
-require_relative '../../input/sources'
+require_relative '../input/sources'
 require_relative 'extracted_image'
 
 module Mindee
   # Image Extraction Module.
-  module Extraction
+  module Image
     # Image Extraction wrapper class.
     module ImageExtractor
       def self.attach_image_as_new_file(input_buffer, format: 'jpg')
@@ -35,7 +35,7 @@ def self.attach_image_as_new_file(input_buffer, format: 'jpg')
       # @param [Integer] page_id ID of the Page to extract from.
       # @param [Array<Array<Mindee::Geometry::Point>>, Array<Mindee::Geometry::Quadrangle>] polygons List of coordinates
       # to extract.
-      # @return [Array<Mindee::Extraction::ExtractedImage>] Extracted Images.
+      # @return [Array<Mindee::Image::ExtractedImage>] Extracted Images.
       def self.extract_multiple_images_from_source(input_source, page_id, polygons)
         new_stream = load_input_source_pdf_page_as_image(input_source, page_id)
         new_stream.seek(0)
@@ -49,7 +49,7 @@ def self.extract_multiple_images_from_source(input_source, page_id, polygons)
       # @param [StringIO] pdf_stream Buffer of the PDF.
       # @param [Integer] page_id Page ID.
       # @param [Array<Mindee::Geometry::Point, Mindee::Geometry::Polygon, Mindee::Geometry::Quadrangle>] polygons
-      # @return [Array<Mindee::Extraction::ExtractedImage>] Extracted Images.
+      # @return [Array<Mindee::Image::ExtractedImage>] Extracted Images.
       def self.extract_images_from_polygons(input_source, pdf_stream, page_id, polygons)
         extracted_elements = []
 
diff --git a/spec/extraction/extracted_image_spec.rb b/spec/extraction/extracted_image_spec.rb
@@ -7,7 +7,7 @@
 require 'mini_magick'
 require_relative '../data'
 
-describe Mindee::Extraction::ExtractedImage do
+describe Mindee::Image::ExtractedImage do
   let(:dummy_io_content) { 'This is a test file content.' }
   let(:input_source) do
     Mindee::Input::Source::PathInputSource.new("#{DATA_DIR}/products/invoices/default_sample.jpg")
@@ -17,7 +17,7 @@
     it 'should initialize correctly with valid inputs' do
       page_id = 1
       element_id = 2
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, page_id, element_id)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, page_id, element_id)
 
       expect(extracted_image.buffer).to_not be(nil)
       expect(extracted_image.page_id).to eq(page_id)
@@ -27,15 +27,15 @@
 
     it 'should handle nil element_id by setting it to 0' do
       page_id = 1
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, page_id, nil)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, page_id, nil)
 
       expect(extracted_image.element_id).to eq(0)
     end
 
     it 'should save the buffer to a file with valid format' do
       page_id = 1
       element_id = 2
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, page_id, element_id)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, page_id, element_id)
 
       random_sequence = Array.new(8) { rand(0..9) }.join
       extracted_image.save_to_file("#{DATA_DIR}/output/temp-#{random_sequence}.jpg", 'jpg')
@@ -48,7 +48,7 @@
     it 'should infer file format from extension if not provided' do
       page_id = 1
       element_id = 2
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, page_id, element_id)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, page_id, element_id)
 
       Tempfile.create(['output', '.png']) do |tempfile|
         extracted_image.save_to_file(tempfile.path)
@@ -60,7 +60,7 @@
     it 'should raise an error for invalid file format during save' do
       page_id = 1
       element_id = 2
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, page_id, element_id)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, page_id, element_id)
 
       Tempfile.create(['output', '.']) do |tempfile|
         expect do
@@ -72,7 +72,7 @@
     it 'should raise an error for invalid path during save' do
       page_id = 1
       element_id = 2
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, page_id, element_id)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, page_id, element_id)
 
       invalid_path = '/invalid/path/output.jpg'
       expect do
@@ -83,7 +83,7 @@
     it 'should return a valid source object from as_source' do
       page_id = 1
       element_id = 2
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, page_id, element_id)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, page_id, element_id)
 
       source_object = extracted_image.as_source
 
@@ -94,7 +94,7 @@
     it 'should raise an error when MiniMagick fails during save' do
       allow(MiniMagick::Image).to receive(:read).and_raise(StandardError)
 
-      extracted_image = Mindee::Extraction::ExtractedImage.new(input_source, 1, 2)
+      extracted_image = Mindee::Image::ExtractedImage.new(input_source, 1, 2)
 
       Tempfile.create(['output', '.jpg']) do |tempfile|
         expect do
diff --git a/spec/extraction/extracted_pdf_spec.rb b/spec/extraction/extracted_pdf_spec.rb
@@ -7,11 +7,10 @@
   let(:product_data_dir) { File.join(DATA_DIR, 'products') }
 
   it 'should extract a PDF from an image' do
-    jpg_stream = File.open("#{product_data_dir}/invoices/default_sample.jpg", "r")
-    pdf_wrapper = Mindee::Extraction::PdfExtractor::ExtractedPdf.new(jpg_stream, "dummy.pdf")
+    jpg_stream = File.open("#{product_data_dir}/invoices/default_sample.jpg", 'r')
+    pdf_wrapper = Mindee::Image::PdfExtractor::ExtractedPdf.new(jpg_stream, 'dummy.pdf')
     expect do
       pdf_wrapper.page_count
     end.to raise_error Mindee::Errors::MindeePDFError
   end
-
-end
+end
diff --git a/spec/extraction/image_extractor_spec.rb b/spec/extraction/image_extractor_spec.rb
@@ -5,8 +5,8 @@
 require 'mindee/extraction'
 require_relative '../data'
 
-describe Mindee::Extraction do
-  include Mindee::Extraction
+describe Mindee::Image do
+  include Mindee::Image
   let(:barcode_path) do
     File.join(DATA_DIR, 'products', 'barcode_reader', 'default_sample.jpg')
   end
@@ -23,10 +23,10 @@
       barcodes2 = inference.prediction.codes_2d.map(&:polygon)
       input_source = Mindee::Input::Source::PathInputSource.new(barcode_path)
 
-      extracted_barcodes_1d = Mindee::Extraction::ImageExtractor.extract_multiple_images_from_source(input_source, 1,
-                                                                                                     barcodes1)
-      extracted_barcodes_2d = Mindee::Extraction::ImageExtractor.extract_multiple_images_from_source(input_source, 1,
-                                                                                                     barcodes2)
+      extracted_barcodes_1d = Mindee::Image::ImageExtractor.extract_multiple_images_from_source(input_source, 1,
+                                                                                                barcodes1)
+      extracted_barcodes_2d = Mindee::Image::ImageExtractor.extract_multiple_images_from_source(input_source, 1,
+                                                                                                barcodes2)
 
       expect(extracted_barcodes_1d.size).to eq(1)
       expect(extracted_barcodes_2d.size).to eq(2)
diff --git a/spec/extraction/invoice_splitter_extraction_integration.rb b/spec/extraction/invoice_splitter_extraction_integration.rb
@@ -30,7 +30,7 @@ def prepare_invoice_return(rst_file_path, invoice_prediction)
     )
     inference = response.document.inference
 
-    pdf_extractor = Mindee::Extraction::PdfExtractor::PdfExtractor.new(invoice_splitter_input)
+    pdf_extractor = Mindee::Image::PdfExtractor::PdfExtractor.new(invoice_splitter_input)
     expect(pdf_extractor.page_count).to eq(2)
 
     extracted_pdfs_strict = pdf_extractor.extract_invoices(inference.prediction.invoice_page_groups, strict: true)
diff --git a/spec/extraction/multi_receipts_extractor_spec.rb b/spec/extraction/multi_receipts_extractor_spec.rb
@@ -5,8 +5,8 @@
 require 'mindee/extraction'
 require_relative '../data'
 
-describe Mindee::Extraction do
-  include Mindee::Extraction
+describe Mindee::Image do
+  include Mindee::Image
   let(:multi_receipts_single_page_path) do
     File.join(DATA_DIR, 'products', 'multi_receipts_detector', 'default_sample.jpg')
   end
@@ -28,7 +28,7 @@
       input_sample = Mindee::Input::Source::PathInputSource.new(multi_receipts_single_page_path)
       response = load_json(multi_receipts_single_page_json_path, 'complete.json')
       doc = Mindee::Product::MultiReceiptsDetector::MultiReceiptsDetectorV1.new(response['document']['inference'])
-      extracted_receipts = Mindee::Extraction::MultiReceiptsExtractor.extract_receipts(input_sample, doc)
+      extracted_receipts = Mindee::Image::MultiReceiptsExtractor.extract_receipts(input_sample, doc)
 
       expect(extracted_receipts.size).to eq(6)
 
@@ -76,7 +76,7 @@
       input_sample = Mindee::Input::Source::PathInputSource.new(multi_receipts_multi_page_path)
       response = load_json(multi_receipts_multi_page_json_path, 'multipage_sample.json')
       doc = Mindee::Product::MultiReceiptsDetector::MultiReceiptsDetectorV1.new(response['document']['inference'])
-      extracted_receipts = Mindee::Extraction::MultiReceiptsExtractor.extract_receipts(input_sample, doc)
+      extracted_receipts = Mindee::Image::MultiReceiptsExtractor.extract_receipts(input_sample, doc)
 
       expect(extracted_receipts.size).to eq(5)
 
diff --git a/spec/extraction/pdf_extractor_spec.rb b/spec/extraction/pdf_extractor_spec.rb
diff --git a/spec/extraction/tax_extractor_spec.rb b/spec/extraction/tax_extractor_spec.rb
diff --git a/spec/image/extracted_image_spec.rb b/spec/image/extracted_image_spec.rb
diff --git a/spec/image/image_compressor_spec.rb b/spec/image/image_compressor_spec.rb
diff --git a/spec/input/sources/sources_spec.rb b/spec/input/sources/sources_spec.rb
diff --git a/spec/pdf/pdf_compressor_spec.rb b/spec/pdf/pdf_compressor_spec.rb
diff --git a/spec/pdf/pdf_processor_spec.rb b/spec/pdf/pdf_processor_spec.rb

Original file line number	Diff line number	Diff line change
`@@ -13,7 +13,7 @@ def multi_receipts_detection(file_path, mindee_client)`
`13`	`13`	`close_file: false`
`14`	`14`	`)`
`15`	`15`
`16`		`- images = Mindee::Extraction::MultiReceiptsExtractor.extract_receipts(input_source, result_split.document.inference)`
	`16`	`+ images = Mindee::Image::MultiReceiptsExtractor.extract_receipts(input_source, result_split.document.inference)`
`17`	`17`	`images.each do \|sub_image\|`
`18`	`18`	`# Optional: Save the files locally`
`19`	`19`	`# sub_image.write_to_file("/path/to/my/extracted/file/folder")`