FIT2107-S2-2020 Assignment 2
Automated Unit Testing, Whitebox testing and Continuous
Integration
This assignment will be completed in pairs

Marks: 20% of your final marks.
Due Date: Week 9, 15/10/2020, 11:55 pm AEST (9:55 pm Malaysia Time).
Submission: READ THE INSTRUCTIONS CAREFULLY.
1. Incremental commits on GIT on the repository provided. We will
use the last commit for marking your assignment.
2. As per the faculty policy a copy of your final commit on GIT for
the assignment MUST be uploaded on Moodle.
3. Download your final commit for the assignment on the GIT,
convert it to a ZIP or RAR file (if not done automatically), and
upload it on Moodle under the Assignment 2 submission link.
The zip file should be named using the repository name (or team
name) i.e. REPOSITORYNAME.zip.
Lateness: Late penalty of 10% per day after the due date, including the weekends.
Student
Statement:
An electronic student statement MUST be accepted as per the faculty
policy while submitting your assignment. Once you press the submit
button, the electronic student statement will appear, accept it and
press continue. If the assignment is left in a DRAFT mode, it will not be
accepted for grading. All team members MUST accept the student
statement.
Goal
In this assignment, you will create a small program in Python 3 (or another language, if you
really insist), and test it using unit testing and mocking frameworks. You will also set up
continuous integration, so that testing occurs as you update your source code repository. In
addition, you will apply Whitebox testing techniques to measure the coverage of the code.
Typically, developers are expected to write unit tests for their own code, hence the need to
write your own test suite. However, you are not going to be marked on the quality of your
implementation - you just have to have one to test against; However, your implementation
should be complete. Comprehensive unit tests and fixing bugs that you find as you commit
changes should result in a module that is as close as possible to functionally correct.
Git Infotech
We will be using Monash FIT's private Gitlab repository at Monash Infotech, accessible at:
https://git.infotech.monash.edu.
All changes must be regularly committed to this system. There are instructions for using Git
and GitLab on Moodle.
Your username on this Gitlab server will be your Monash authcate login. To commit to the
repository, you must create an "access token" and use that instead of a password (just type
it in where it asks for your password – normally with git clients). Git via ssh does not work on
this server, unfortunately.
Groups & Repositories
This assignment will be done in groups of two (or, if there are uneven numbers and there is
no alternative, groups of three). The groups are same as created for assignment 1 so you do
not need to create groups again. If you want to change a group, contact your workshop
facilitator. If there will be a genuine reason, we will make a change.
We will create a repository on a Git Infotech Server for you. Please do all project work in the
repository we have created for you. DO NOT create your own repository, or any other git
hosting service, for this assignment. DO make sure you commit to the repository frequently.
You will be assessed on whether you commit to the repository appropriately. Optionally, you
can add TAGS to the commit that you want us to mark for assignment 2. We will only mark
the MASTER branch.
The repositories will be set up with a "skeleton" module and tests that you will need to
extend, and with a functioning but rudimentary Git Infotech CI setup. The current setup runs
all the tests in CalendarTest.py. If you want to add other steps to your CI (for instance,
run a coverage tool) you will have to edit gitlab-ci.yml.
The Calendar Module
When you clone your git repository, you will note that it contains two files - Calendar.py
and CalendarTest.py. These files are skeletons that you must complete for the
assignment!
To access the Google Calendar API, you must have to register an API key. The details are
provided here:
https://developers.google.com/calendar
You must use Google Calendar Python API to implement and test the functionalities below.
You will lose marks for not using the correct API.
The module should have tests and functionalities for following features.
1. As a user, I can see events and reminders for at least 5 years in past from the today’s
date.
2. As a user, I can see events and reminders for at least next two years (in future).
3. As a user, I can navigate through different days, months, and years in the calendar so
that I can view the details of events. For example, if the year 2019 is selected, all
events, reminders, and notifications will be shown. On selecting the specific event or
reminder I can see the detailed information.
4. As a user, I can search events and reminders using different key words.
5. As a user, I can delete events and reminders.
You do require an authentication with Google Calendar API first time to access it. There is no
need to write test cases for authentication.
Extra Functionalities & Tests for the Groups of Three Students ONLY
In addition to the features mentioned in the Calendar Module section, the groups of three
students must write unit tests for the following additional features.
1. As a user, I can edit events and reminders.
2. As a user, I can cancel events and reminders.
Using Another Language
If you really want to try using another language for this assignment, we will not stop you,
however:
• You will need to find a Google Calendar API for the language of your choice. It is NOT
feasible to write your own functionality yourself without using the API.
• You will have to get Gitlab CI for your language of choice working ON YOUR OWN. We
strongly recommend, under the circumstances, that you use Python.
Discuss with your facilitator before making a final decision.
Test Strategy
You only need to test the functional correctness of the code that you write the module and
run the tests using the unittest module, performing continuous integration using Git
Infotech.
The strategy you use to select your test cases is up to you. You should explicitly document
your test case selection strategy in a Markdown file called "test-strategy.md" and add
it to the repository.
We will be assessing you both on the reasonableness of your test strategy, and whether you
have actually applied it. We expect you to try to use the strategies we have discussed, possibly
including the black-box strategies.
You may assist your marker by adding a section to your test strategy document explaining
how you got your test cases from your strategy. While this is not compulsory, remember that
if we cannot see how your strategy got you your test cases, we will mark you down for it!
Coverage
To measure statement and branch coverage as part of your test strategy, there is a standard
Python tool called coverage.py that can help you achieve this. The documentation and
usage instructions are accessible at:
https://coverage.readthedocs.io/en/coverage-5.2.1/.
If you choose to use it, if you can, you should integrate it into your CI scripts so that it runs
along with the tests and generates the report.
Mocking
The point of this exercise is to test your code, you should "mock out" calls to the Google
Calendar API in your tests. You should use the Python mock module for this where possible.
Mocking can be quite complicated, so use the simplest possible mocks at first, and if and when
you have time you can make your mocks more featureful to test your code more fully.
Your mocks should return sensible dummy data to allow your tests to test your code
effectively. You may also consider whether you should use mock's facilities to check the way
the API is called.
Unit Tests
Based on your test strategy, devise a test suite for the module. It is recommended you do this
before or as you develop your module (TDD), not afterwards. You should use your unit tests
to fix bugs in your module.
Continuous Integration
You must set up and use GitLab's Continuous integration to run your tests. Every time you
commit to the master branch, the (appropriately mocked) tests should be run on Git Infotech.
You will have to figure out how to ensure that the Google Calendar module, and any other
submodules is installed on the Gitlab CI docker instance when you run your tests. There are
two options - using setuptools or invoking pip as part of your test script!
Demonstrations
You will be asked to demonstrate your working program and unit tests in your week 10
tutorial. This will be a brief process and is purely to verify that your program and unit tests
run on your machine.
Your unit tests are still expected to run on the CI server, and your demonstrator may need to
run your program again on their own machine. This will be assessed by the marker later.

Working as a Team
You are expected to work as a pair and contribute close to equal amounts on the group
section of this project.
The primary way we will assess this is by examining your repository and assessing the number
and nature of commits by each member of the group. Therefore, it is important to commit
your work - do not get your partner to commit on your behalf.
If you genuinely work on something together (at the same time), please include the text (PAIR
WORK) in the comments for the commits where this occurs. ONLY do this if both of you
worked on the material at the same time (preferably in the same place).
Plagiarism
Please do not copy code from other teams. You can use code you find on the Internet,
provided that you are not violating copyright and you acknowledge where you got it from,
AND it's not a solution to a large part of the assignment. We will use a code similarity tool to
find code similarities if needed.
If you wish to use a downloadable Python module other than Google Calendar or any of their
dependencies, please ask teaching staff first.
If you find some code on the internet that IS a solution to a large fraction of the assignment,
please contact teaching staff.
Special Consideration
If a student faces exceptional circumstances (serious illness or injury, family emergency etc.)
that prevent them completing the assignment, they may apply for special consideration
according to the policy and procedure on the Faculty website:
https://www.monash.edu/exams/changes/special-consideration.
Assignment Forum
You can ask questions on the discussion forum, which the Lecturer will monitor and respond
to daily (and at least once on weekends). All students should monitor this forum - I may clarify
aspects of the assignment on the forum and any such clarifications are considered "official".
You may of course also email the lecturer or arrange to see him virtually if you prefer.
Marking Criteria
• Completion of module according to specifications.
• Appropriateness of test case selection strategy.
• Application of strategy in actual selection of test cases.
• Quality of test cases (including quality of test case code).
• Effectiveness of mock usage.
• Appropriate usage of CI system.
• Level of individual contribution (should be close to equal between partners)
• Quality of test analysis.
• Readability of test analysis.
A detailed rubric is available on the Moodle submission link.

欢迎咨询51作业君