This is my solution to the first three user stories of the original BankOCR Kata
From the project root, run lein test. If you wish to develop the project, run lein test-refresh for a tighter feedback loop
In order to illustrate the approach I took to solving the kata, I made frequent commits. Writing failing tests and specs, then writing the code to conform to the specs and tests.
In order to solve, I started with the simplest things first, then built larger things on top of them.
User story 3 necessitates creating an output file. I decided to generate this file in the project root rather than pollute the filesystem of the user that runs the tests.
Write a program that parses an account number file into actual account numbers.
This requirement is fulfilled in user-story-1-test. An input file in resources/fixtures/accounts.txt is read, then parsed into account numbers.
Check account numbers for validity.
This requirement is fulfilled in user-story-2-test. The approach I took to creating the checksum for the account numbers is to turn the string into a collection and reduce it with a decrementing index in the calculation.
Write a file with results of the OCR. Illegible digits are replaced with ?. If there's an incorrect checksum or one of the digits is illegible, it's indicated in a second column
This requirement is fulfilled in user-story-3-test. The approach to solve was to read in resources/fixtures/user-story-3.txt, read the account numbers, give each number a status, then spit back out to output.txt.
Copyright © 2019 Jason Stewart
Distributed under the Eclipse Public License either version 1.0 or (at your option) any later version.