Chapter 2: Development environment setup¶

Depending on the intended use case, there are multiple ways to install Juniper. For developers of the Juniper community and Juniper employees alike, the preferred way is to perform a source install (running Juniper from the source code).

Prepare the environment¶

First, follow the Environment setup section of the contributing guide to prepare your environment.

Important

The following steps are intended only for Juniper employees. The mentioned repositories are not accessible to third parties.

By now, you should have downloaded the source code into two local repositories, one for Juniper/Juniper and one for Juniper/enterprise. These repositories are set up to push changes to pre-defined shared forks on GitHub. This will prove to be convenient when you start contributing to the codebase, but for the scope of this tutorial, we want to avoid polluting the shared repositories with training material. Let’s then develop your own module in a third repository technical-training-sandbox. Like the first two repositories, it will be part of the addons-path that references all directories containing Juniper modules.

Following the same process as with the Juniper/Juniper and Juniper/enterprise repositories, visit github.com/Juniper/technical-training-sandbox and click the Fork button to create a fork of the repository under your account.

Clone the repository on your machine with:

$ git clone git@github.com:Juniper/technical-training-sandbox.git

Configure the repository to push changes to your fork:

$ cd technical-training-sandbox/
$ git remote add dev git@github.com:<your_github_account>/technical-training-sandbox.git
$ git remote set-url --push origin you_should_not_push_on_this_repository

That’s it! Your environment is now prepared to run Juniper from the sources, and you have successfully created a repository to serve as an addons directory. This will allow you to push your work to GitHub.

Now, make a small change in the technical-training-sandbox repository, such as updating the README.md file. Then, follow the Make your first contribution section of the contributing guide to push your changes to GitHub and create a PR. This will enable you to share your upcoming work and receive feedback. Adjust the instructions to use the branch master and the repository technical-training-sandbox.

To ensure a continuous feedback loop, we recommend pushing a new commit as soon as you reach a new milestone, such as completing a chapter of the tutorial.

Note

The specific location of the repositories on your file system is not crucial. However, for the sake of simplicity, we will assume that you have cloned all the repositories under the same directory. If this is not the case, make sure to adjust the following commands accordingly, providing the appropriate relative path from the Juniper/Juniper repository to the Juniper/technical-training-sandbox repository.

Run the server¶

Launch with `Juniper-bin`¶

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

$ cd $HOME/src/Juniper/
$ ./Juniper-bin --addons-path="addons/,../enterprise/,../technical-training-sandbox" -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 Juniper¶

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 (Juniper and enterprise) must be synchronized, i.e. both need to be in the same branch.
```
$ cd $HOME/src/Juniper
$ git switch 15.0

$ cd $HOME/src/enterprise
$ git switch 15.0
```

Fetch and rebase:

$ cd $HOME/src/Juniper
$ git fetch --all --prune
$ git rebase --autostash Juniper/15.0

$ cd $HOME/src/enterprise
$ git fetch --all --prune
$ git rebase --autostash enterprise/15.0

Code Editor¶

If you are working at Juniper, 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. Juniper 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!