- Clone the repository.
- Open the website folder.
- Create a
.envfile in the website folder. - Add the following environment variables to the
.envfile:
## Azure Blob Storage
AZURE_STORAGE_CONNECTION_STRING = 'connection_string'
AZURE_STORAGE_CONTAINER_SUBMISSION = 'container_name'
AZURE_STORAGE_CONTAINER_VALIDATOR = 'container_name'
## Backend
SECRET_KEY = "secret_key"
DB_NAME = "benchlab"
DB_HOST = "localhost"
DB_USER = "benchlabuser"
DB_PASSWORD = "password"
DEBUG = "True"
EMAIL_BACKEND = "django.core.mail.backends.smtp.EmailBackend"
EMAIL_HOST = "mail.tptech.nl"
EMAIL_FROM = "[email protected]"
EMAIL_HOST_USER = "[email protected]"
EMAIL_HOST_PASSWORD = "password"
EMAIL_PORT = "587"
EMAIL_USE_TLS = "True"
FRONTEND_URL = "http://localhost:5173"
ALLOWED_HOSTS = "*"
CSRF_TRUSTED_ORIGINS = "http://localhost"
JUDGE_HOST = "localhost"
## Frontend
VITE_API_URL = "http://localhost:8000/api"For the value of EMAIL_HOST_PASSWORD, ask someone who set up the repository already.
- Make sure you have Node.js 20 LTS installed on your machine. If not, please install it. LINUX ONLY: You might need to use Node Version Manager (nvm) to install this specific version of Node.js, as the default install command via
apthas an older version. - Install/Activate yarn (run
corepack enable, this command might require admin priviliges). - Open the frontend folder.
- Run
yarn(this will install all dependencies). DO NOT USE NPM TO INSTALL DEPENDENCIES. - To start vite you can run
yarn vite.
- Install MySQL.
- Create a basic server.
- Open the created database and run
CREATE DATABASE benchlab;(usingsudo mysqlon Linux). - The run
CREATE USER 'benchlabuser'@'localhost' IDENTIFIED WITH mysql_native_password BY 'password';. - Finally run
GRANT ALL PRIVILEGES ON benchlab.* TO 'benchlabuser'@'localhost';. - Adjust the .env file based on the properties defined above, in case you modified any values (e.g. the password).
- Make sure you have Python 3.12 installed. On Linux, this can be done with
sudo add-apt-repository ppa:deadsnakes/ppa, followed bysudo apt install python3.12 python3.12-venv python3.12-dev(since you also need venv and the Python development headers). - On Linux, you also need the following:
sudo apt install pkg-config default-libmysqlclient-dev build-essential(used by mysqlclient). - Create a virtual environment by running
python -m venv .venvin the website(root) directory. - Select the virtual environment as your VS Code interpreter.
- Activate the virtual environment by running (Linux)
source .venv/bin/activateor (Windows).venv\Scripts\activate.bat. - Navigate your command line into the backend folder (e.g. using
cd backend). - Run pip install -r requirements.txt (this will install all dependencies).
- Install the Ruff VSCode extension.
- Run migrations:
python manage.py makemigrationsandpython manage.py migrate. - Run seeder:
python manage.py loaddata XXXX_filename.jsonwhereXXXX_filename.jsonis the file with the highest number inbackend/api/fixtures. - To start the server, use
python manage.py runserver.
After following the steps above, the folder structure should look like this:
website/
├─ .github/
├─ .venv/
├─ backend/
├─ docs/
├─ frontend/
├─ .env
├─ .gitignore
├─ README.md
├─ ruff.toml
├─ SECURITY.md
Use the production branch to deploy the webapp.
To get initial data in the database run python manage.py loaddata XXXX_filename.json.
To remove all data from the database run python manage.py flush this will permanently delete all data from the database so be carefull with this.
Note: the passwords for the currently seeded accounts are admin, staff and user respectively
A convenient way to make new seeder files is by creating new entries in the tables via the django admin panel (localhost:8000/admin) and then running the following command: python manage.py dumpdata api --indent 4 --output=api/fixtures/XXXX_filename.json. Here count up from previous highest number in the folder and maybe give the new file a somewhat descriptive name.
You can also choose to write your own seeder files from scratch with json but that can be a bit of a hassle because you have to manually do all the foreign keys.
NOTE: this is not yet a very streamlined workflow so if you find out stuff doesn't work or if you find a better way, please edit the file and make a pull request for it. When doing migrations, sometimes it's not necessary to change all the data, in those cases make a new fixture file after migrating so that the changes are reflected in the seeded data. If you do need to change data this can be a workflow:
- run
python manage.py makemigrationsto generate the migration files. - run
python manage.py migrateto migrate the changes. - adapt the seeder files to reflect the new changes in the db.
- run
python manage.py flushto empty the database. - run
python manage.py loaddata XXXX_filename.jsonto seed the db with the new data.
If the migrate command gives an error, it might be easier to reset your entire database. This can be done with python manage.py reset_db, rather than the flush command.
Ruff is a tool that checks, based on some rules, the Python code, similar to Java Checkstyle.
To change the settings or add features to ruff check the ruff.toml file. This can be found in the backend folder. This file also contains explanations for each setting/rule.
The following keybinds might be different on your machine:
Pressing Shift + Alt + F will allow you to use Ruff to format and fix the code. Keep in mind that Ruff is not able to fix all issues.
Pressing Shift + Alt + O will use isort to sort the imports.
Navigate to show output as shown below in the Usage section. If show output in the workflow failed and it states Code violations found!!!!! at the bottom then you're code does not adhere to the quality standards. The table in the output above shows which categories have violations (in red). For a better understanding of quality standards you can look at the tables in the Metrics section.
For any questions about this workflow or issues contact Tom.
This worklfow is run whenever a commit is pushed to main or a pr into main is made. It may take between 1 to 3 minutes to complete the analysis. To view the results do the following:
Navigate to the github action and click on the job Code Quality:
The Show output tab shows all the metrics gathered by the tools:
The output starts with some intermediate debug results, which you can ignore. The evaluation of you code is shown in the tables below. Threshold violation percentages is the percentage of files/code that violate the metrics shown in the table below. Ranks shows the combined rank for each maintainability attribute. It shows the grade that our code would get according to the metrics that the TU/e uses.
For each characteristic we compute what percentage of the code is above the threshold. Then a rank is computed for that characteristic. For each Maintainability attribute we compute the average rank (based on its corresponding code characteristics).
| Code characteristic | Metric | Threshold | Maintainability attr. |
|---|---|---|---|
| Module size | SLOC | <=400 | ADT |
| Class complexity | Cyclomatic per method: average; max | <10; <20 | DT |
| Class design (for OO) | WMC | <=20 | MRAD |
| Module design (for non OO) | Functions per module/file | <=20 | MRAD |
| Module internal duplication | % of duplicated SLOC inside modules | N/A | RAD |
| Code commenting | % LOCM | >15% | A |
| Cyclic dependencies | # of between classes | 0 | MRADT |
| Module coupling | CBO | <16 | MR |
| Module external duplication | % of duplicated SLOC between modules | N/A | MR |
| rank | % of modules/classes/packages above threshold |
|---|---|
| +2 | 0-3% |
| +1 | 4-5% |
| 0 | 6-10% |
| -1 | 11-20% |
| -2 | 21–100% |
| Code | Maintainability attribute |
|---|---|
| M | Modularity |
| R | Reusability |
| A | Analyzability |
| D | Modifiability |
| T | Testability |
The code quality assessment document that TU/e uses can be found here.
TU/e uses simian to analyse code duplication. understand is used to derive all the other metrics.
Note: understand incorrectly computes Module coupling, it only accounts for classes in the same file.
In order to run backend unit tests, navigate to the backend folder, then run the following command:
python manage.py test tests
When you want to run frontend unit tests for the first time, navigate to the fontend folder. You might have to run this command first to install the necessary dependencies:
yarn
To run frontend unit tests, navigate to the fontend folder, then run the following command:
yarn run test
If you would like to also see code coverage, run the following command instead:
yarn run test --coverage
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot]")
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent