From f6a5c4fe409db6811813943f7751ba1c256ea269 Mon Sep 17 00:00:00 2001 From: Emily Bache Date: Wed, 13 Mar 2013 18:25:25 +0100 Subject: [PATCH] improved the description of text-based approval testing --- GildedRose/README.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/GildedRose/README.md b/GildedRose/README.md index 93a1a76d..09d6f98e 100644 --- a/GildedRose/README.md +++ b/GildedRose/README.md @@ -21,15 +21,15 @@ I've also set this kata up on [cyber-dojo](http://cyber-dojo.com) for several la -Whichever testing approach you choose, the idea of the exercise is to do some deliberate practice, and improve your Refactoring skills. The idea is not to re-write the code from scratch, but rather to practice taking small steps, running the tests often, and incrementally improving the design. +Whichever testing approach you choose, the idea of the exercise is to do some deliberate practice, and improve your skills at designing test cases and refactoring. The idea is not to re-write the code from scratch, but rather to practice designing tests, taking small steps, running the tests often, and incrementally improving the design. -## Text-Based Testing +## Text-Based Approval Testing -This is a testing approach which is very useful when refactoring legacy code. The basic idea is to create tests that use the text which the code produces. Before you change the code, you run it, and save the output as a "Golden Copy". Then after you change the code, you run it again, and compare the output against the Golden Copy. Any differences, and the test fails. +This is a testing approach which is very useful when refactoring legacy code. Before you change the code, you run it, and gather the output of the code as a plain text file. You review the text, and if it correctly describes the behaviour as you understand it, you can "approve" it, and save it as a "Golden Master". Then after you change the code, you run it again, and compare the new output against the Golden Master. Any differences, and the test fails. It's basically the same idea as "assertEquals(expected, actual)" in a unit test, except the text you are comparing is typically much longer, and the "expected" value is saved from actual output, rather than being defined in advance. -Typically a piece of legacy code may not produce suitable textual output from the start, so you may need to modify it before you can write your first text-based test. That could involve inserting log statements into the code, or just writing a "main" method that executes the code and prints out what the result is afterwards. It's this latter approach we are using here to test GildedRose. +Typically a piece of legacy code may not produce suitable textual output from the start, so you may need to modify it before you can write your first text-based approval test. That could involve inserting log statements into the code, or just writing a "main" method that executes the code and prints out what the result is afterwards. It's this latter approach we are using here to test GildedRose. The Text-Based tests in this repository are designed to be used with the tool "TextTest" (http://texttest.org). This tool helps you to organize and run text-based tests. There is more information in the README file in the "texttests" subdirectory.