receipt_data_extraction 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 37d4fb35a9ca3dd44afee22b7e56c3a2e1be62de
+  data.tar.gz: cb0ebcba1f12c4f4f8433de6c0d7fe5df7b01b15
+SHA512:
+  metadata.gz: 38af6da3c6612e984649d16b7b51ca27a441c6e37b9b06d8467799ec3d7f5a9dd06ea164da9cf173f7f034baaac5c5fabdb0f78d08419c27088dbecd75d5251e
+  data.tar.gz: d72d34844cf22e4d135776bab8f0ae2a10401c8152a78372b0096c86d1089eb0a1c14fb171ee0c938f6349306114d727e695f0e668697e212e6bc832bead98a1

data/lib/receipt_data_extraction.rb

@@ -0,0 +1,85 @@
+require "reg_exp"
+require "google/cloud/vision"
+
+module ReceiptDataExtraction
+    class Extract
+    
+      def self.call_vision_api(image)
+          image_annotator = Google::Cloud::Vision::ImageAnnotator.new
+          response = image_annotator.text_detection(
+              image: image,
+              max_results: 1 # optional, defaults to 10
+          )
+          text_detected = ""
+          response.responses.each do |res|
+              text_detected << res.text_annotations[0].description
+          end
+        return text_detected
+      end 
+    end
+  
+    class Specific_data  
+        def self.get_shop_name(text)
+            shop_name = text.split(Name_reg).first
+            text = text.sub!(shop_name, "")
+          return shop_name
+        end  
+    
+        def self.get_address(text)
+            address = text.split(Name_reg)[1]
+            text = text.sub!(address,"")
+            address2 = text.split(Name_reg)[2]
+            address = address + " " + address2
+          return address
+        end   
+    
+        def self.get_date(text)
+            date = text.match(Date_reg)
+            date = Date.parse(date[0]).strftime("%d/%m/%Y")
+          return date
+        end 
+        
+        def self.get_total(text)
+            total = 0.00 
+            text.scan(Total_reg) do |amount|
+                if total < amount[0].to_f
+                    total = amount[0].to_f
+                end
+            end 
+          return total
+        end    
+        
+        def self.get_products(text)
+            text.scan(Product_reg) do |qtde, name_product, price, price_total|
+                if (name_product != "") && (price != nil)
+                    if qtde != nil
+                        puts "-qtde: #{qtde}" 
+                    end    
+                        puts "-name_product: #{name_product}"
+                        puts "Price: #{price}"
+                    if price_total != nil
+                        puts "Price_Total: #{price_total}"
+                    end    
+                end
+            end    
+        end
+        
+        def self.get_data_before_total(text)
+            text =  text.split(Clean_text)[0]
+          return text
+        end  
+        
+        #remove all discounts/cancellations
+        def remove_neg_values(text)
+            text = text.gsub(Negative_values, "")
+          return text
+        end  
+    end        
+end
+
+#text_detected = ReceiptDataExtraction::Extract.call_vision_api('../spar.jpg')
+
+
+
+
+

data/lib/reg_exp.rb

@@ -0,0 +1,19 @@
+module ReceiptDataExtraction
+    Name_reg = /\n/
+    
+    #12/11/2010 or 11 jan 2011 or 12 JAN 2011 or 12 Jan 2011
+    month_1 = "(J|j)an|(F|f)eb|(M|m)ar|(A|a)pr|(M|m)ay|(J|j)un|(J|j)ul|(A|a)ug|(S|s)ep|(O|o)ct|(N|n)ov|(D|d)ec"
+    month_2 = "JAN|FEB|MAR|APR|MAY|JUN|JUL|AUG|SEP|OCT|NOV|DEC"
+    Date_reg = /(\s(?<day>(\d{1,2}))(\s|\/|\-)(?<month>((\d+)|(#{month_1})|(#{month_2})))(\s|\/|\-)(?<year>(\d{2,4}))\s+)/
+    
+    #Qty | Name | Price | Total
+    quantity = /(?<qtde>(\d+))/
+    item_name = /(?<name_product>(.*?))/
+    price = /(€|EUR)?(?<price>(\d{1,3}\,)*\d{1,3}\.\d{2})/
+    price_total = /(?<price_total>(\d{1,3}\,)*\d{1,3}\.\d{2})/
+    Product_reg = /((#{quantity})\s+)?#{item_name}\s*#{price}(\s*#{price_total})?/
+    
+    Negative_values = /(€|EUR)?\-(?<price>(\d{1,3}\,)*\d{1,3}\.\d{2})/
+    Total_reg = /(\s*(\btotal\b)|(\bTotal\b)|(\bTOTAL\b))\D*\s*(?<amount>(\d{1,3}\,)*\d{1,3}\.\d{2})/
+    Clean_text = /.*?(?=#{Total_reg})/
+end

metadata

@@ -0,0 +1,45 @@
+--- !ruby/object:Gem::Specification
+name: receipt_data_extraction
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Marcio Camargo
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2019-04-15 00:00:00.000000000 Z
+dependencies: []
+description: returns
+email: marjoscam@comp.ufla.br
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/receipt_data_extraction.rb
+- lib/reg_exp.rb
+homepage: https://github.com/mjcamargo/receipt_data_extraction
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.6.8
+signing_key: 
+specification_version: 4
+summary: Extract data from receipts using google vision
+test_files: []