1. BIT 286: Web Applications Test Plans

2. Test Plan  Three major parts:  Coverage – what are you going to test?  Methods – how are you going to test it?  Including what you're going to use to test it  Responsibilities – who will be doing the test?  And once they've done the test, what do they do with that information?  These can be quite detailed, or more free-formquite detailed  We'll be doing a more free-form version  From WikipediaWikipedia 2

3. Designing Good Test Cases Good paper on this by Cem KanerCem Kaner  A test case is "“the specific inputs that you’ll try and the procedures that you’ll follow when you test the software.”  Ron Patton (from "Software Testing", SAMS, p. 65)  The starting state of the program, what you'll do, and what you expect to see happen  For a database-driven web app this may mean pre- loading a specific & limited set of data into the database 3

4. Attributes of good test cases (for tests whose goal is to reveal bugs)  More (statistically) powerful  (more likely to expose a bug if it the bug is there)  More likely to yield significant (motivating, persuasive) results  More credible  E.g., people believe that a normal use might normally do this  Easier to evaluate (did the test pass or fail?)  More useful for troubleshooting  Appropriately complex  Simpler early on (when 'nothing works').  If the tests are too complex they'll be vulnerable to blocking bugs  This is all also from Cem KanerCem Kaner 4

5. Black-Box Testing Paradigms  Function testing  Domain testing  Specification-based testing  Risk-based testing  Stress testing - JMeter  Also: Performance Testing (this isn't listed in the paper)  Regression testing  Author defines this as "tests that will be reused". I've always heard this talked about as "tests that check that a previously-fixed bug stays fixed"  User testing  Scenario testing  State-model based testing  High volume automated testing  Exploratory testing  This is all also from Cem Kaner.Cem Kaner 5

6. Types of Test 6

7. Ad-Hoc / Manual Tests  As you're doing this build up a checklist of what you're testing  If someone needs to re-test they've got a list  It'll serve as documentation of what you tried out  Setting this up as an outline in MS Word is good  Put in a numbered list, and then use tab (at the start of the line) to indent rightward, and shift-tab to move the indentation left 1 tab stop  This might be best for some types of tests  For example, the Specification-Based Testing  This is not from Cem Kaner. Here ends the 'Cem Kaner' section of the presentationCem Kaner 7

8. Automated tests  We'll look at using Selenium next class  Interesting advice on testing: http://docs.seleniumhq.org/docs/06_test_design_considerations.jsp http://docs.seleniumhq.org/docs/06_test_design_considerations.jsp 8

9. Today  Let's work together to come up with a test plan for a single page / feature  Example of a feature: the 'upload student homework' on the student- course mega-page  Let's try for at least two paradigms  For Milestone 2:  Come up with a test plan for another 2 or so features/pages  Make a reasonable decision about how may paradigms to use for each feature – this might be just 1  For Milestone 3:  Come up with a test plan for each feature that you've implemented  You need to have some automated tests  You can have some manual test reports for the rest 9