Chapter 2: Development environment setup¶

Depending on the intended use case, there are multiple ways to install Flectra. This tutorial will stick to the source install (running Flectra from the source code), which is best suited for Flectra developers.

Throughout this document, we assume that you are installing your development environment on a laptop provided by Flectra with Linux Mint installed and up-to-date. If that is not the case, switch to the Windows or Mac OS tab if any section of the installation guide, depending on which OS you are on. The steps remain essentially the same.

Set up Git¶

Install and configure Git¶

The very first step of the installation process is to install the Git version control system because the Flectra source code is managed on GitHub.

$ sudo apt install git

Tip

Check if Git is installed by trying to print Git’s version with the following command:

$ git --version

Once installed, register your name and email:

$ git config --global user.name "Your full name (trigram)"
$ git config --global user.email "xyz@flectra.com"

Configure GitHub¶

You need a GitHub account to fetch the sources and contribute to Flectra’s development. If you don’t have one yet, create it. For the username, we recommend using your trigram “xyz” (or quadrigam) followed by ‘-flectra’: ‘xyz-flectra’.

The easiest way to authenticate with GitHub is to use an SSH connection. Using SSH authentication allows you to connect to GitHub without supplying your username and password every time you type a command.

Note

The following step-by-step procedure is based on the official GitHub documentation.

Generate a new SSH key, add it to the ssh-agent, and copy the SSH key to your clipboard.

$ ssh-keygen -t ed25519 -C "xyz@flectra.com"
$ ssh-add ~/.ssh/id_ed25519
$ sudo apt install xclip
$ xclip -sel clip < ~/.ssh/id_ed25519.pub

Go to GitHub.com, then click on your profile picture in the upper-right corner of the page and then on Settings.
On the user settings sidebar, click on SSH and GPG keys.
Click on New SSH key or on Add SSH key.
In the Title field, add a descriptive label for the new key.
Paste your key into the Key field.
Click on Add SSH key.

Fetch the sources¶

It is time to fetch the source code of Flectra. First, let’s create a home for the Git repositories in $HOME/src/.

$ mkdir -p $HOME/src
$ cd $HOME/src

Tip

Cloning the repositories will take a while, enjoy a cup of coffee while you wait.

Configure the Git repositories¶

To contribute to an Flectra repository, you first need to fork it, then create a branch containing your changes on the fork, and finally submit a Pull Request to the repository.

Tip

If you are lucky enough to work at Flectra, the forks already exist. They are hosted on https://github.com/flectra-dev/flectra and https://github.com/flectra-dev/professional.

After your two forks are created, their remote address can be added in your local repositories. In the commands below, replace flectra-dev/flectra and flectra-dev/professional with the name of your forks if needed.

$ cd  $HOME/src/flectra
$ git remote add flectra-dev git@github.com:flectra-dev/flectra.git  # Add flectra-dev as a new remote.
$ git remote rename origin flectra  # Change the name of origin (the flectra repository) to flectra.
$ git remote set-url --push flectra no_push  # Remove the possibility to push directly to flectra (you can only push to flectra-dev).

$ cd  $HOME/src/professional
$ git remote add professional-dev git@github.com:flectra-dev/professional.git
$ git remote rename origin professional
$ git remote set-url --push professional no_push

Install the dependencies¶

As seen in Chapter 1: Architecture Overview, Flectra’s server runs on Python and uses PostgreSQL as an RDBMS. In the context of a development machine, the easiest approach is to install everything locally. To do so, follow once again the Installing Flectra guide.

Tip

Some useful SQL commands:

$ createdb $DB_NAME  # Create a database.
$ dropdb $DB_NAME  # Drop a database.

$ psql $DB_NAME  # Connect to a database.
    \l  #List all the available databases.
    \dt  #List all the tables of the $DB_NAME database.
    \d $TABLE_NAME  #Show the structure of the table $TABLE_NAME.
    \q  #Quit the psql environment (ctrl + d).

Run the server¶

Launch with `flectra-bin`¶

Once all dependencies are set up, Flectra can be launched by running flectra-bin, the command-line interface of the server.

$ cd $HOME/src/flectra/
$ ./flectra-bin --addons-path="addons/,../professional/" -d rd-demo

There are multiple command-line arguments that you can use to run the server. In this training you will only need some of them.

-d <database>¶: The database that is going to be used.

--addons-path <directories>¶: A comma-separated list of directories in which modules are stored. These directories are scanned for modules.

--limit-time-cpu <limit>¶: Prevent the worker from using more than <limit> CPU seconds for each request.

--limit-time-real <limit>¶: Prevent the worker from taking longer than <limit> seconds to process a request.

Tip

The --limit-time-cpu and --limit-time-real arguments can be used to prevent the worker from being killed when debugging the source code.
You may face an error similar to AttributeError: module '<MODULE_NAME>' has no attribute '<$ATTRIBUTE'>. In this case, you may need to re-install the module with $ pip install --upgrade --force-reinstall <MODULE_NAME>.

If this error occurs with more than one module, you may need to re-install all the requirements with $ pip install --upgrade --force-reinstall -r requirements.txt.

You can also clear the python cache to solve the issue:
$ cd $HOME/.local/lib/python3.8/site-packages/ $ find -name '*.pyc' -type f -delete
Other commonly used arguments are:
- -i: Install some modules before running the server (comma-separated list).
- -u: Update some modules before running the server (comma-separated list).

Log in to Flectra¶

Open http://localhost:8069/ on your browser. We recommend using Chrome, Firefox, or any other browser with development tools.

To log in as the administrator user, use the following credentials:

email: admin
password: admin

Enable the developer mode¶

The developer or debug mode is useful for training as it gives access to additional (advanced) tools. In the next chapters, we will always assume that you have enabled the developer mode.

Enable the developer mode now. Choose the method that you prefer; they are all equivalent.

Note

The main page of the Settings screen is only accessible if at least one application is installed. You will be led into installing your own application in the next chapter.

Extra tools¶

Useful Git commands¶

Here are some useful Git commands for your day-to-day work.

Switch branches:

When you switch branches, both repositories (flectra and professional) must be synchronized, i.e. both need to be in the same branch.
```
$ cd $HOME/src/flectra
$ git switch 3.0

$ cd $HOME/src/professional
$ git switch 3.0
```

Fetch and rebase:

$ cd $HOME/src/flectra
$ git fetch --all --prune
$ git rebase --autostash flectra/3.0

$ cd $HOME/src/professional
$ git fetch --all --prune
$ git rebase --autostash professional/3.0

Code Editor¶

If you are working at Flectra, many of your colleagues are using VSCode, VSCodium (the open source equivalent), PyCharm, or Sublime Text. However, you are free to choose your preferred editor.

It is important to configure your linters correctly. Using a linter helps you by showing syntax and semantic warnings or errors. Flectra source code tries to respect Python’s and JavaScript’s standards, but some of them can be ignored.

For Python, we use PEP8 with these options ignored:

E501: line too long
E301: expected 1 blank line, found 0
E302: expected 2 blank lines, found 1

For JavaScript, we use ESLint and you can find a configuration file example here.

Administrator tools for PostgreSQL¶

You can manage your PostgreSQL databases using the command line as demonstrated earlier or using a GUI application such as pgAdmin or DBeaver.

To connect the GUI application to your database we recommend you connect using the Unix socket.

Host name/address: /var/run/postgresql
Port: 5432
Username: $USER

Python Debugging¶

When facing a bug or trying to understand how the code works, simply printing things out can go a long way, but a proper debugger can save a lot of time.

You can use a classic Python library debugger (pdb, pudb or ipdb), or you can use your editor’s debugger.

In the following example we use ipdb, but the process is similar with other libraries.

Install the library:
```
pip install ipdb
```

Place a trigger (breakpoint):

import ipdb; ipdb.set_trace()

Example

def copy(self, default=None):
    import ipdb; ipdb.set_trace()
    self.ensure_one()
    chosen_name = default.get('name') if default else ''
    new_name = chosen_name or _('%s (copy)') % self.name
    default = dict(default or {}, name=new_name)
    return super(Partner, self).copy(default)

Here is a list of commands:

h(elp) [command]¶: Print the list of available commands if not argument is supplied. With a command as an argument, print the help about that command.

pp expression¶: The value of the expression is pretty-printed using the pprint module.

w(here)¶: Print a stack trace with the most recent frame at the bottom.

d(own)¶: Move the current frame one level down in the stack trace (to a newer frame).

u(p)¶: Move the current frame one level up in the stack trace (to an older frame).

n(ext)¶: Continue the execution until the next line in the current function is reached or it returns.

c(ontinue)¶: Continue the execution and only stop when a breakpoint is encountered.

s(tep)¶: Execute the current line. Stop at the first possible occasion (either in a function that is called or on the next line in the current function).

q(uit)¶: Quit the debugger. The program being executed is aborted.

Now that your server is running, it’s time to start writing your own application!

Edit on GitHub