Custom template folders¶

You can create custom template folders and pass them as --template-dir argument to bluprint create. Bluprint will copy the contents of the template folder to the new project and also do the following:

• If files README.md, notebooks/example_jupyter.ipynb or notebooks/example_quarto.qmd exits, any placeholder string {{placeholder_name}} will be replaced with a project name.

• If README.md contains placeholder string {{git_account_name}}, it will

  be replaced by git username.

• If there is pyproject.toml file, it will replace {{placeholder_name}} with project name and {{python_version}} with a Python version from the CLI argument.

Template directory can contain anything. However, for Bluprint configs to work, it needs a folder to store yaml configs (default: conf) and another folder to store data in the project to which relative paths in data.yaml point to (default: data).

By default load_config_yaml() looks for conf/config.yaml and load_data_yaml() looks for conf/data.yaml in the project folder. You can use different names and override these defaults; for example this template folder:

my_custom_template
├── configs
│   ├── my_main_config.yaml
│   └── my_data_config.yaml
├── my_data
│   └── ...
├── {{placeholder_name}}
│   └── my_code.py
├── README.md
└── pyproject.toml

can be used to create a new project:

bluprint create my_project --template-dir /path/to/my_custom_template

which can then be used:

from bluprint.config import load_config_yaml, load_data_yaml

conf = load_config_yaml('configs/my_main_config.yaml')
data = load_data_yaml('configs/my_data_config.yaml', data_dir='my_data')

This loads yaml configs from configs folder and parses locally stored data from my_data in this project structure:

my_project
├── configs
│   ├── my_main_config.yaml
│   └── my_data_config.yaml
├── my_data
│   └── ...
├── my_project
│   └── ...
└── ...