[GSoC Project Proposal]: Flesh out erddapy xarray plugin and xarray-subset-grid with CIs, tests and documentation

[ocefpaf]

Project Description

erdappy

The erddapy xarray plugin (https://github.com/ioos/erddapy) is a thin wrapper to erddapy's to.xarray() that provides a new engine to xarray. With this plugin xarray can load any ERDDAP netcdf-like URL instead of just OPeNDAP ones. It creates some consistency for users coming from ERDDAP by removing the requisite of knowing if the ERDDAP URL is a valid OPeNDAP one or not.

xarray_subset_grid

The xarray_subset_grid package is an xarray extension that provides the ability to generate subsets of large oceanographic model results. Many oceanographic (and meteorologic) models are run on massive grids, with millions of nodes – end users often need only a small subset of the entire model domain (for instance one Bay or Estuary), and having to download and work with huge datasets is untenable. The package is designed to work with rectangular, curvilinear, and unstructured grid systems.

Both of these packages are currently functional, but need improvements to the documentation, test and Continuous Integration (CI) infrastructure, as well as attention to additional issues and bug reports that have been identified.

Detailed info can be found at:

ioos/erddapy#434 [This is a merged PR – doesn’t seem to be the right one? Maybe just point to: https://github.com/ioos/erddapy/issues]

and

https://github.com/ioos/xarray-subset-grid/issues

(GSoC 2026 labels pending)

Expected Outcomes

Test suites with >=90% of coverage running on Continuous Integration.

Major functionality documented.

Major bugs / issues addressed.

Skills Required

Basic Python xarray, pytest
GitHub Actions
Sphinx documentation System

Expected Project Size

350 hours

Project Difficulty

Novice / Intermediate

Mentor Name(s)

Filipe Fernandes, Micah Wengren, Chris Barker

Mentor GitHub Handle(s)

@ocefpaf, @mwengren, @ChrisBarker-NOAA

Mentor Contact Email(s)

ocefpaf@gmail.com

(should we add Micah and Chris ?)

Contributor Evaluation/GSoC GitHub Issue Links

No response

Additional Project Background/Links

https://github.com/ioos/erddapy
https://github.com/ioos/xarray-subset-grid/

[NobleCoder69]

Hi,

I’m interested in this project for GSoC 2026 and have started looking through the xarray-erddap repository.

I understand the main focus is adding tests, CI, and improving documentation. Are there any existing NOAA/IOOS projects you’d recommend as a reference for CI or test structure? Also, is there a preferred ERDDAP server or dataset for automated testing?

I’m planning to set up the project locally and would be happy to start with a small contribution if that helps.

Thanks!

[ocefpaf]

Please wait until we know we will be a selected org for GSoC. At this moment we are just building the idea pages.

[NobleCoder69]

​"Thanks for the reply! I'm going to set up the environment locally and see where I can help with the test suite. I'll reach out if I have specific questions."

[sam-1002]

Hello !!

I recently came across this project and I’m genuinely excited about the direction it’s heading. The integration between xarray and ERDDAP, along with the focus on improving reliability through testing and CI, really caught my attention.
I’m strongly considering applying for GSoC 2026 with this project and have begun reviewing the repository to understand the current structure and open issues. I’m particularly interested in contributing to strengthening the test coverage and CI workflows.
I’d love to get more involved and would be happy to start with a smaller issue or documentation improvement while I continue exploring the codebase.

Looking forward to learning more and contributing meaningfully.

Thank you!!

[staree14]

hello, @ocefpaf, @mwengren, @ChrisBarker-NOAA !
I'm Sreeya, a CS undergrad interested in contributing to IOOS for GSoC 2026. I'm exploring this project and the erddapy/xarray-subset-grid repositories. Are there any specific areas you'd recommend me to focus on first?

[thedgarg31]

Hi @ocefpaf

I’m very interested in this project and would love to contribute as a GSoC 2026 applicant. The goal of strengthening the erddapy xarray plugin and xarray-subset-grid through improved CI workflows, testing coverage, and documentation aligns well with my experience in Python development, GitHub Actions, and open-source contributions.

I’m currently exploring the repositories to understand the existing structure, especially around the plugin engine implementation, test setup, and CI pipelines. I would be happy to start by working on beginner-friendly tasks such as improving test coverage, documentation, or CI configurations while I continue familiarizing myself with the codebase.

Please let me know if there are any recommended starting points or issues that would be most helpful for new contributors to begin with.

Thank you, and I look forward to contributing and learning from the team.

[ocefpaf]

Thanks all for your interested in this project. To help us know you all better can you please, in an issue or email, answer the following questions?

• What is your experience level?
• Please share links for the top 3 projects you created or contribute to.
• In your own words, what do you believe is the best testing strategy for Python projects?
• What are your motivations to work on this project?

Note that is not the selection process per se, we are just trying to get to know you all a little better.

[thedgarg31]

Hi, thank you for reaching out — happy to share more about myself.

1. Experience Level

I would describe myself as an intermediate open-source contributor with hands-on experience in Python, C++, testing practices, and GitHub workflows. I am comfortable working with issues, pull requests, CI checks, rebasing, documentation improvements, and feature implementations.

2. Top 3 Contributions / Projects

• Badge UI DSL Component (Merged)

Implemented a read-only Badge UI component with deterministic style token validation, stable manifest output, and UI tests.
PR: namel3ss-Ai/namel3ss#40

Issue: namel3ss-Ai/namel3ss#14

• Quickstart Mode for AssetOpsBench (IBM)

Developed a lightweight quickstart system including demo, batch, and interactive modes, a minimal agent framework, local evaluation pipeline, and onboarding documentation.
PR: IBM/AssetOpsBench#147

• Documentation Cross-Linking Improvement

Improved discoverability by linking runtime execution and concurrency documentation.
Issue: namel3ss-Ai/namel3ss#12

3. Best Testing Strategy for Python Projects (My View)

From my experience, an effective Python testing strategy includes:

A layered approach — unit, integration, and end-to-end tests

Deterministic validation to ensure predictable behavior

Strong edge-case and failure testing

Use of mocking to isolate components

Automated CI integration with coverage checks

Overall, testing should focus on correctness, reliability, and long-term maintainability.

4. Motivation to Work on This Project

My main motivations are:

Contributing to impactful open-source ecosystems

Improving software reliability through testing

Learning from real-world collaboration and review processes

Working on projects that value clean design, reproducibility, and developer experience

Thank you again — I’m excited about the opportunity to contribute and learn from the community.

Best regards,
Daksh Garg