7. Contributing to bitmath

This section describes the guidelines for contributing to bitmath.

7.1. Issue Reporting

If you are encounter an issue with the bitmath library, please use the provided template.

7.2. Code Style/Formatting

Please conform to PEP 0008 for code formatting. This specification outlines the style that is required for patches.

Your code must follow this (or note why it can’t) before patches will be accepted. There is one consistent exception to this rule:

E501
Line too long

The pycodestyle tests for bitmath include a --ignore option to automatically exclude E501 errors from the tests.

7.3. Commit Messages

Please write intelligent commit messages.

For example:

Capitalized, short (50 chars or less) summary

More detailed explanatory text, if necessary.  Wrap it to about 72
characters or so.  In some contexts, the first line is treated as
the subject of an email and the rest of the text as the body.  The
blank line separating the summary from the body is critical (unless
you omit the body entirely); tools like rebase can get confused if
you run the two together.

Write your commit message in the imperative: "Fix bug" and not
"Fixed bug" or "Fixes bug."  This convention matches up with commit
messages generated by commands like git merge and git revert.

Further paragraphs come after blank lines.

- Bullet points are okay, too

- Typically a hyphen or asterisk is used for the bullet, followed
  by a single space, with blank lines in between, but conventions
  vary here

- Use a hanging indent

7.4. Pull Requests

After a pull request is submitted on GitHub two automatic processes are started:

  1. Travis-CI clones the new pull request and runs the automated test suite.
  2. Coveralls clones the new pull request and determines if the request would increase or decrease the overall code test coverage.

Please check back shortly after submitting a pull request to verify that the Travis-CI process passes.

7.4.1. What Happens If The Build Breaks

Pull requests which break the build will be looked at closely and you may be asked to fix the tests.

The bitmath project welcomes all contributors so it’s OK if you’re unable to fix the tests yourself. Just leave a comment in the pull request explaining so if that is the case.

Likewise, if Coveralls indicates the pull request would decrease the overall test-coverage, and you aren’t able to fix it yourself, just leave a comment in the pull request.

7.5. Automated Tests

Write unittests for any new functionality, if you are up to the task. This is not a requirement, but it does get you a lot of karma.

All bitmath code includes unit tests to verify expected functionality. In the rest of this section we’ll learn how the unit tests are put together and how to interact with them.

7.5.1. Components

bitmath unit tests are integrated with/depend on the following items:

  • Travis CI - Free online service providing continuous integration functionality for open source projects. Tests are ran automatically on every git commit. Integrates with GitHub to notify you if a pull request passes or fails all unitests.
  • Coveralls - Free online service providing code test coverage reporting. Integrates with GitHub to notify you if a pull-request would improve/decrease overall code test coverage.
  • unittest - Python unit testing framework. All bitmath tests are written using this framework.
  • nose - Per the nose website: “extends unittest to make testing easier”. nose is used to run our unit tests.
  • coverage - A tool for measuring code coverage of Python programs. For bitmath we require a minimum test coverage of 90%. This is invoked by nose automatically.
  • pycodestyle - A tool to check Python code against some of the style conventions in PEP 0008.
  • pyflakes - A simple program which checks Python source files for errors.
  • virtualenv - A tool to create isolated Python environments. Allows us to install additional package dependencies without requiring access to the system site-packages directory.
  • Makefiles - Utility scripts used for project building and testing. How bitmath uses Makefiles is described later in this section.

7.5.2. Targets

In the scope of this document, we use the term target in the context of makefile targets. For the purpose of this documentation, we can think of these targets as pre-defined commands coded in a makefile. bitmath testing targets include:

  • ci - Run the tests exactly how they are ran in Travis-CI. The ci target automatically calls the pycodestyle, pyflakes, uniquetestnames, and unittests targets.
  • ci3 - Is the same as the ci target, except it runs using the Python 3.x interpreter.
  • unittests - Run the functional test suite.
  • pycodestyle - Run PEP 0008 syntax checks.
  • pyflakes - Run pyflakes error checks.
  • clean - Remove temporary files and build artifacts from the checked-out repository.
  • uniquetestnames - Ensures no unit tests have the same name.
  • tests - A quicker version of ci. Different from ci in that tests uses libraries installed on the local development workstation. tests runs the unittests, pycodestyle, uniquetestnames, and pyflakes tests automatically.

To ensure the highest degree of confidence in test results you should always use the ci and ci3 targets.

When Travis-CI runs an integration test, it calls the ci and ci3 targets.

7.5.3. Running the Tests

The bitmath test suite is invoked via the Makefile. The following is an example of how to run the ci test target manually:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
[~/Projects/bitmath] 17:22:21  (master)
$ make ci
#############################################
# Running Unique TestCase checker
#############################################
./tests/test_unique_testcase_names.sh
#############################################
# Creating a virtualenv
#############################################
virtualenv bitmathenv
New python executable in bitmathenv/bin/python
Installing setuptools, pip...done.
. bitmathenv/bin/activate && pip install -r requirements.txt
Downloading/unpacking python-coveralls (from -r requirements.txt (line 1))
  Downloading python_coveralls-2.4.3-py2.py3-none-any.whl
Downloading/unpacking nose (from -r requirements.txt (line 2))

... snip ...

Convert a bitmath GiB into a Tb ... ok
Convert a bitmath PiB into a TiB ... ok
Convert a bitmath GiB into a Tib ... ok
Convert to kb ... ok
Convert a bitmath Bit into a MiB ... ok
bitmath type converted to the same unit is properly converted ... ok
float(bitmath) returns a float ... ok
int(bitmath) returns an int ... ok
long(bitmath) returns a long ... ok

Name      Stmts   Miss  Cover   Missing
---------------------------------------
bitmath     440      1    99%   1152
----------------------------------------------------------------------
Ran 163 tests in 0.035s

OK
:

On line 2 we see how to call a makefile target. In this case it’s quite straightforward: make ci. Other targets are called in the same way. For example, to run the clean target, you run the command make clean. To run the Python 3.x test suite, you would run the command make ci3.

7.5.4. Troubleshooting

If you find yourself unable to run the unit tests:

  1. Search for relevant error messages
  2. Read the error message closely. The solution could be hidden in the error message output. The problem could be as simple as a missing dependency
  3. If you are unable to figure out all the necessary dependencies to run the tests, file an issue on that specific projects GitHub issue tracker. Include the full error message.