Fork 0

221 lines
6.6 KiB

Development Guide
This is a guide for developers who would like to contribute to this project.
GitHub Workflow
If you're interested in contributing to pgcli, first of all my heart felt
thanks. `Fork the project <https://github.com/dbcli/pgcli>`_ on github. Then
clone your fork into your computer (``git clone <url-for-your-fork>``). Make
the changes and create the commits in your local machine. Then push those
changes to your fork. Then click on the pull request icon on github and create
a new pull request. Add a description about the change and send it along. I
promise to review the pull request in a reasonable window of time and get back
to you.
In order to keep your fork up to date with any changes from mainline, add a new
git remote to your local copy called 'upstream' and point it to the main pgcli
$ git remote add upstream git@github.com:dbcli/pgcli.git
Once the 'upstream' end point is added you can then periodically do a ``git
pull upstream master`` to update your local copy and then do a ``git push
origin master`` to keep your own fork up to date.
Check Github's `Understanding the GitHub flow guide
<https://guides.github.com/introduction/flow/>`_ for a more detailed
explanation of this process.
Local Setup
The installation instructions in the README file are intended for users of
pgcli. If you're developing pgcli, you'll need to install it in a slightly
different way so you can see the effects of your changes right away without
having to go through the install cycle every time you change the code.
It is highly recommended to use virtualenv for development. If you don't know
what a virtualenv is, `this guide <http://docs.python-guide.org/en/latest/dev/virtualenvs/#virtual-environments>`_
will help you get started.
Create a virtualenv (let's call it pgcli-dev). Activate it:
source ./pgcli-dev/bin/activate
.\pgcli-dev\scripts\activate (for Windows)
Once the virtualenv is activated, `cd` into the local clone of pgcli folder
and install pgcli using pip as follows:
$ pip install --editable .
$ pip install -e .
This will install the necessary dependencies as well as install pgcli from the
working folder into the virtualenv. By installing it using `pip install -e`
we've linked the pgcli installation with the working copy. Any changes made
to the code are immediately available in the installed version of pgcli. This
makes it easy to change something in the code, launch pgcli and check the
effects of your changes.
Adding PostgreSQL Special (Meta) Commands
If you want to work on adding new meta-commands (such as `\dp`, `\ds`, `dy`),
you need to contribute to `pgspecial <https://github.com/dbcli/pgspecial/>`_
Visual Studio Code Debugging
To set up Visual Studio Code to debug pgcli requires a launch.json file.
Within the project, create a file: .vscode\\launch.json like below.
// Use IntelliSense to learn about possible attributes.
// Hover to view descriptions of existing attributes.
// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
"version": "0.2.0",
"configurations": [
"name": "Python: Module",
"type": "python",
"request": "launch",
"module": "pgcli.main",
"justMyCode": false,
"console": "externalTerminal",
"env": {
"PGUSER": "postgres",
"PGPASS": "password",
"PGHOST": "localhost",
"PGPORT": "5432"
Building RPM and DEB packages
You will need Vagrant 1.7.2 or higher. In the project root there is a
Vagrantfile that is setup to do multi-vm provisioning. If you're setting things
up for the first time, then do:
$ version=x.y.z vagrant up debian
$ version=x.y.z vagrant up centos
If you already have those VMs setup and you're merely creating a new version of
DEB or RPM package, then you can do:
$ version=x.y.z vagrant provision
That will create a .deb file and a .rpm file.
The deb package can be installed as follows:
$ sudo dpkg -i pgcli*.deb # if dependencies are available.
$ sudo apt-get install -f pgcli*.deb # if dependencies are not available.
The rpm package can be installed as follows:
$ sudo yum install pgcli*.rpm
Running the integration tests
Integration tests use `behave package <https://behave.readthedocs.io/>`_ and
Configuration settings for this package are provided via a ``behave.ini`` file
in the ``tests`` directory. An example::
stderr_capture = false
pg_test_user = dbuser
pg_test_host = db.example.com
pg_test_port = 30000
First, install the requirements for testing:
$ pip install -U pip setuptools
$ pip install --no-cache-dir ".[sshtunnel]"
$ pip install -r requirements-dev.txt
Ensure that the database user has permissions to create and drop test databases
by checking your ``pg_hba.conf`` file. The default user should be ``postgres``
at ``localhost``. Make sure the authentication method is set to ``trust``. If
you made any changes to your ``pg_hba.conf`` make sure to restart the postgres
service for the changes to take effect.
$ sudo service postgresql restart
After that, tests in the ``/pgcli/tests`` directory can be run with:
(Note that these ``behave`` tests do not currently work when developing on Windows due to pexpect incompatibility.)
# on directory /pgcli/tests
$ behave
And on the ``/pgcli`` directory:
# on directory /pgcli
$ py.test
To see stdout/stderr, use the following command:
$ behave --no-capture
Troubleshooting the integration tests
- Make sure postgres instance on localhost is running
- Check your ``pg_hba.conf`` file to verify local connections are enabled
- Check `this issue <https://github.com/dbcli/pgcli/issues/945>`_ for relevant information.
- `File an issue <https://github.com/dbcli/pgcli/issues/new>`_.
Coding Style
``pgcli`` uses `black <https://github.com/ambv/black>`_ to format the source code. Make sure to install black.
If you're the person responsible for releasing `pgcli`, `this guide <https://github.com/dbcli/pgcli/blob/main/RELEASES.md>`_ is for you.