A lab session, using Git, GitHub and Travis CI to write and run tests on a shared repository.
In the following text, replace USERNAME
with your GitHub user name.
If you do not already have an account, then visit GitHub and create a free account.
If you already have an account, then visit GitHub and sign in.
Visit https://github.com/epcctraining/git-github-travis-lab.
Click Fork.
If asked "Where should we fork this repository?", select your account.
Within a command-line shell, clone your fork:
$ git clone https://github.com/USERNAME/git-github-travis-lab
$ cd git-github-travis-lab
msc/rot13.py
holds an implementation of ROT-13, a substitution cipher. The co de originates from Literate Programs. For more on ROT-13 substitution cipher, see Wikipedia.
Run:
$ python msc/rot13.py hello
uryyb
Travis CI looks for a file called .travis.yml
in a Git repository. This file tells Travis CI how to build and test your software. In addition, this file can be used to specify any dependencies you need installed before building or testing your software.
Look at the one in this repository:
$ cat .travis.yml
language: python
The language
section tells Travis CI which build environment it should use.
python:
- "2.7"
- "3.4"
Depending on the language selected, a language-specific section (e.g. jdk
, php
or python
) can then be used to specify which versions of the language the job is to be run under. Travis CI will run the job for each of these versions.Here, we request that Travis CI runs the job twice, once under Python 2.7 and once under Python 3.4:
before_install:
- pip install coverage
An optional before_install
section lists the steps that need to be executed before we can build, run or test our code. For certain languages, we also need to download, build and install additional dependencies. For Python, nosetests is pre-installed by Travis CI, but here the Python coverage package is also installed, using the Python pip
package installer, so we can generate a code coverage report, a report on the lines of code executed by our tests.
script:
- nosetests -v --with-coverage
Python needs a script
entry to specify what to run e.g. how to run our tests and what tests to run. We use nosetests -v
and --with-coverage
flags to print out information about the tests being run and a code coverage report.
Once you have an account on GitHub, you can use this to sign in to Travis CI.
Visit Travis CI.
Click on Sign in with GitHub.
Now, tell Travis CI to check for changes in your repository.
Click on your name on the top-right of the Travis CI page.
This page shows a list of your GitHub repositories that Travis CI knows about.
If you cannot see USERNAME/git-github-travis-lab
, then click the Sync button.
Once you can see USERNAME/git-github-travis-lab
, then click on the repository switch (the X button) so that Travis CI knows to check that repository for changes.
In the directory msc/tests
, create a file called test_USERNAME.py
.
Add the following imports, to import the functions in msc/rot13.py
and also nose.tools
test functions:
from nose.tools import assert_equal
from msc.rot13 import rot13
from msc.rot13 import rot13_char
Look at the msc/rot13.py
source code and write a test for one of the two functions:
def rot13_char(ch):
def rot13(s):
For example, a test for rot13_char(ch)
is:
def test_rot13_char_a():
assert_equal("n", rot13_char("a"))
To run your tests, do:
$ nosetests
When your test passes, add it to Git:
$ git add msc/tests/test_USERNAME.py
$ git commit -m "USERNAME's first test for ROT13" .
These changes are to our local repository only. Once we push them to GitHub, they will trigger the running of our Travis CI job. So, let's push these changes to our repository on GitHub, denoted by the 'remote', origin
, into its default, master
, branch:
$ git push origin master
Visit https://travis-ci.org/USERNAME/git-github-travis-lab.
You should see a single job with two sub-jobs, one for each version of these languages we are testing under.
Jobs should be coloured green with a tick, indicating that they passed.
Click on one of the jobs.
You should see that your test ran and passed.
Now, submit your code for inclusion in the original repository, https://github.com/epcctraining/git-github-travis-lab. GitHub calls this a "pull request" - a request that the repository's owner pull in your changes.
Visit https://github.com/USERNAME/git-github-travis-lab.
Click Pull Requests
Click New Pull Request
Set base fork: epcctraining/git-github-travis-lab
Set head fork: USERNAME/git-github-travis-lab
Click Create pull request
Enter a title for your pull request, if one hasn't been entered already.
Click Create pull request
The lab demonstrator will merge in your pull request. Once they have done so, you should see your tests being run as part of https://travis-ci.org/epcctraining/git-github-travis-lab.
Add another test to msc/tests/test_USERNAME.py
.
When it passes, commit your changes to this file.
Push your changes to GitHub.
Create a pull request.
If you want to update your repository with the changes that have been made to epcctraining/git-github-travis-lab
since you first forked it, you can pull these into your repository.
Add the original repository that you forked as a 'remote'. This is a short-hand to the location of that repository:
$ git remote add upstream https://github.com/epcctraining/git-github-travis-lab
Pull in the current version of the master
branch of the repository represented by the upstream
remote:
git pull upstream master
We now have the changes in our local repository, and can now push these into our repository on GitHub:
git pull origin master
Copyright (c) 2015 The University of Edinburgh.
Code is licensed under the Apache 2.0 licence. The licence text is also in LICENSE.md.
Documents are licensed under the Creative Commons Attribution-NonCommercial 2.5 UK: Scotland (CC BY-NC 2.5 SCOTLAND).
This directory includes code based on public domain software. Please see msc/rot13.py.