# GSoC 2023 project ideas for Open Science Labs

The Open Science Labs (OSL) community is thrilled to announce our participation in the Google Summer of Code 2023. We warmly welcome and encourage individuals from all around the globe, especially those from the Latin America region, to apply to GSoC with Open Science Labs. With OSL members who are fluent in Spanish and Portuguese, we aim to serve as a bridge between students and professionals who may feel less confident in their English language skills and the Open Source community.

## About Open Science Labs

Open Science Labs is a global community dedicated to creating an open space for teaching, learning, and sharing information about open science and computational tools. While we welcome individuals from all regions, we particularly focus on Latin American countries. Our community develops tools that address real-world problems and collaborates with other projects and workgroups to improve technology and create international opportunities for our community.

Although our focus may seem broad, we initially prioritize supporting Research Software Engineers (RSEs) who often face computational challenges in their work. We recognize that many colleagues in scientific fields may not be familiar with programming languages, computational libraries, version control systems, databases, DevOps, and other computational tools. Therefore, we aim to provide a safe and open space for individuals to learn and share their knowledge. Our community is open to everyone, including students, professors, and industry professionals, as we believe that the challenges and technologies used in these fields are frequently similar, if not the same (in many cases).

### Organization

Open Science Labs is primarily organized by its steering council, collaborators, and interns. You can access the current list of our active members on our website: https://opensciencelabs.org/team/.

To ensure a welcoming and safe environment for all members, OSL has established a Code of Conduct, which you can find here: https://opensciencelabs.org/coc/.

If you're interested in learning more about our organization and governance, please visit our website: https://opensciencelabs.org/governance/.

## Project Idea 1: Improve Scientific Python Cookiecutter

### Abstract

Cookiecutter is a powerful command-line utility that generates projects from project templates. This tool provides many benefits, including:

* Simplifying the creation of a project with an initial structure that includes recommended tools, workflows, and structure.
* Enabling new programming users to create a new package with ease and minimal knowledge.

Open Science Labs has developed a cookiecutter Python template that aims to provide the workflow, libraries, and tools recommended by PyOpenSci. In addition, this template includes extra options that can improve the development workflow.

Note: PyOpenSci is currently conducting an investigation to identify the tools, libraries, best practices, and workflows used by the most notable scientific Python communities.

### Current state

The template currently includes the following features:

* Supports a package slug different than the project slug
* Offers support for the following licenses: MIT, BSD 3 Clause, ISC License, Apache Software License 2.0, and GPL 3
* Documentation engines supported: mkdocs, sphinx, jupyter-book
* Test library supported: pytest
* Auto-format code tool: blue, and black
* Initial integration with git (automatically creates a branch for the new code)
* Supports conda (as the base environment) and poetry as the packaging and dependency management tool (additional build systems are already planned)
* Support for pre-commit hooks
* Integration with CI using Github actions
* Release workflow with semantic release and Github actions.

### Tasks

* Improve documentation
* Add support for more build systems
* Add support for more testing libraries
* Add support for more static analysis tools
* Change template engine from cookiecutter to cookieninja.

### Expected outcomes

* All the tasks should be tested and documented.
* For each new implementation, it is necessary a new smoke test.

### Details

* Prerequisites:
* Experience with Python
* Knowledge about at least one packaging tool/library (setuptools, poetry, pdm, etc)
* Some knowledge about Jinja2 or other Template engine would be very appreciated.
* Expected time: 350 hours
* Potential mentor(s): Ivan Ogasawara, Ever Vino, Alexandre de Siqueira, Gagandeep Singh, Agustina Pesce, Saransh Chopra

### References

* https://opensciencelabs.org/
* https://github.com/osl-incubator/cookiecutter-python
* https://github.com/osl-incubator/cookiecutter-python/issues

## Project Idea 2: Implement C++ backend for data structures and algorithms in PyDataStructs

### Abstract

PyDataStructs project aims to be a Python package for various data structures and algorithms (including their parallel implementations). We are also working on providing C++ backend via Python C-API for high performance use cases. It is a partner project under Open Science Labs (see, https://opensciencelabs.org/partners/).

Why PyDataStructs?

- **Single package for all your data structures and algorithms**

- **Consistent and Clean Interface** - The APIs we have provided are consistent with each other, clean, and easy to use. We make sure of that before adding any new data structure or algorithm.

- **Well Tested** - We thoroughly test our code before making any new addition to PyDataStructs. 99 percent lines of our code have already been tested by us.

### Current state

* The first release i.e., PyDataStructs 0.0.1 is available at https://pydatastructs.readthedocs.io/en/0.0.1/index.html.
* Open issues on Github - https://github.com/codezonediitj/pydatastructs/issues.
* We have participated previously in programs like KWoC (organised by IIT Kharagpur), GSSoC.

### Expected outcomes

Well tested implementations of algorithms and data structures in C++ backend.

### Details

* Prerequisites:
* Experience with Python, C++
* Ability to read and implement algorithms
* Knowledge of Python C-API is not necessary but would be an additional benefit.
* Expected time: 350 hours
* Potential mentor(s): Gagandeep Singh, Ivan Ogasawara, Ever Vino, Alexandre de Siqueira, Agustina Pesce, Saransh Chopra

### References

* https://pydatastructs.readthedocs.io/en/latest/
* https://github.com/codezonediitj/pydatastructs
* https://github.com/codezonediitj/pydatastructs/issues

## Project Idea 3: Improve MakIm

### Abstract

**MakIm**, also known as **makim**, is a powerful tool designed to help scientists and researchers streamline their workflow and improve the way they define targets and dependencies. Unlike the traditional Makefile format, MakIm utilizes yaml format, making it easier to define and manage tasks.

With MakIm, you can define targets and dependencies with ease, while also taking advantage of control options such as conditionals and ifs. This enables you to set parameters and define texts for documentation and other important details for each target.

Whether you're working on a research project, analyzing data, or collaborating with a team, MakIm is the perfect tool to help you save time, increase efficiency, and enhance the quality of your work.

### Current state

Currently, **makim** support the following features:

* Help text as first-class in the .makim.yaml specification. It can be used by targets and arguments.
* Targets have an option for arguments.
* Targets have an option for dependencies.
* Dependencies can call a target with specific arguments.
* Dependencies can have a conditional control flow (if).
* Allow the creation of groups, so the targets can be organized by topics.
* Targets and groups have an option for user defined variables and/or environment variables.
* Access arguments, variables or environment variables via template (using Jinja2).
* Option for using dot environment files using env-file key.

### Tasks

* Add support for targets with no groups
* Add a mechanism for the formal definition of a specification (spec), according to different version
* Add a validation that will check if a `.makim.yaml` format is correct according to the spec defined by the version
* Add functionallity for auto complete when TAB key is pressed

### Expected outcomes

* All the tasks should be tested and documented.

### Details

* Prerequisites:
* Experience with Python
* Some knowledge about Jinja2 or other Template engine would be very appreciated.
* Some knowledge about YAML files would be nice.
* Expected time: 350 hours
* Potential mentor(s): Ivan Ogasawara, Ever Vino, Alexandre de Siqueira, Gagandeep Singh, Agustina Pesce, Saransh Chopra