banktools-se 2.3.0 → 2.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 1c5c3cd7492d627e0bd743ad56328b7d97daad3a
-  data.tar.gz: e83b11dbab0e65165fcbcf2ed490770624023ea9
+  metadata.gz: 2e925fd3ed8cc9466d3349621bbd04324592ef67
+  data.tar.gz: 10ae31638c8fb1951f00c9faec6c5d2a14b9ddd3
 SHA512:
-  metadata.gz: 173c075e3289cd074802e1b0530b2cfa4dcf11a087bec1bc6f1582252ab44c67d785ebf239e77e14930ec3db646d820098bded4fed99646e136b30fccfed7061
-  data.tar.gz: 67767a7305bb6425c53200c8e664e9635ce2162c319f70772bd3747a00a104589c874d78429a25769802468a1097f910ca77c533ee388b40cac2ce48cb249173
+  metadata.gz: f765f7fca94f5590d5b4a123bb020d4ebf3b42d46d5f6e20343544faab6ae80df0dfcbcb17c0dc93a2ea0533391ec00a91b756d58d9fd05c3fecea63aca86a3a
+  data.tar.gz: d19a541703158bf3671ab61f6d44f1af364d23d44c907c96f4d2b14fd15c203af7f63df8b59129d63a2b3dd9d1af771f276f44ecf8924bfe50c942a8708a8165

data/README.markdown

@@ -45,10 +45,10 @@ Inspired by [iulianu/iban-tools](https://github.com/iulianu/iban-tools). Please
 
     # This feature is intended to try to find all OCR numbers in a noisy bank statement string.
     # By design it may find too many numbers (e.g. valid substrings of other numbers), so you should check results against actual outstanding invoices.
-    # By default, it excludes OCRs shorter than 4 digits, but this limit can be specified per below.
+    # By default, it excludes OCRs shorter than 4 digits and longer than 19 digits, but this limit can be specified per below.
     BankTools::SE::OCR.find_all_in_string("OCR1230 and ref4564 and 7890")  # => [ "1230", "4564" ]
     BankTools::SE::OCR.find_all_in_string("1230 and 123067", length_digit: true, pad: "0")  # => [ "123067" ]
-    BankTools::SE::OCR.find_all_in_string("00 and 18", min_length: 2)  # => [ "00", "18" ]
+    BankTools::SE::OCR.find_all_in_string("00 and 18 and 1230", min_length: 2, max_length: 3)  # => [ "00", "18" ]
 
     # Plusgiro
 

data/lib/banktools-se/ocr.rb

@@ -64,13 +64,14 @@ module BankTools
         ocr[0...-digits_to_chop]
       end
 
-      def self.find_all_in_string(string, length_digit: false, pad: "", min_length: 4)
+      # max_length is 19 because that's the longest allowed integer by default in a Postgres integer column with Ruby on Rails. So attempting some queries with longer OCRs may cause exceptions.
+      def self.find_all_in_string(string, length_digit: false, pad: "", min_length: 4, max_length: 19)
         expanded_string = string + " " + string.gsub("\n", "") + " " + string.gsub(";", "")
 
         numbers = expanded_string.scan(/\d+/)
 
         expanded_numbers = with_numbers_found_by_removing_prefix_and_postfix(numbers).
-          reject { |n| n.length < min_length }
+          reject { |n| n.length < min_length || n.length > max_length }
 
         expanded_numbers.select { |candidate|
           begin

data/lib/banktools-se/version.rb

@@ -1,5 +1,5 @@
 module BankTools
   module SE
-    VERSION = "2.3.0"
+    VERSION = "2.4.0"
   end
 end

data/spec/ocr_spec.rb

@@ -120,6 +120,18 @@ describe BankTools::SE::OCR do
       expect(BankTools::SE::OCR.find_all_in_string("1234", min_length: 2)).to eq [ "34" ]
     end
 
+    it "lets you configure the accepted OCR max_length" do
+      ocr_length_19 = "1234567890123456785"
+      ocr_length_20 = "12345678901234567894"
+      string = "#{ocr_length_19} #{ocr_length_20}"
+
+      # Default max_length is 19.
+      expect(BankTools::SE::OCR.find_all_in_string(string)).to include ocr_length_19
+      expect(BankTools::SE::OCR.find_all_in_string(string)).not_to include ocr_length_20
+
+      expect(BankTools::SE::OCR.find_all_in_string(string, max_length: 20)).to include ocr_length_19, ocr_length_20
+    end
+
     it "excludes duplicates" do
       expect(BankTools::SE::OCR.find_all_in_string("1230 1230 4564")).to eq [ "1230", "4564" ]
     end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: banktools-se
 version: !ruby/object:Gem::Version
-  version: 2.3.0
+  version: 2.4.0
 platform: ruby
 authors:
 - Henrik Nyh