EOPF‑Zarr plugin

Overview
This repository contains the EOPF GDAL Plugin, a community-driven effort to extend GDAL with support for the Earth Observation Processing Framework (EOPF). By loading this plugin, users can open, process, and visualize EOPF datasets in GDAL-based applications and workflows—such as QGIS, command-line utilities (e.g., gdalinfo, gdal_translate), and custom scripts.

Note: EOPF data is commonly stored in cloud-native formats (e.g., Zarr) or NetCDF-like containers. This plugin aims to handle those data structures seamlessly, exposing them as standard GDAL raster datasets.

---

Features

• Read-Only Support
  Currently supports reading of EOPF datasets, including chunked data (Zarr-based archives), multi-band imagery, and associated metadata.

• Cloud Compatibility
  Leverages GDAL’s virtual file I/O to access data hosted on local storage or cloud object stores (future enhancement: S3, /vsicurl/).

• Metadata Extraction
  Parses core EOPF metadata (dimensions, data types, partial georeferencing if present). Additional coverage of advanced attributes (e.g. sensor-specific metadata) will be added incrementally.

• Integration
  Once installed, any GDAL-based application automatically recognizes EOPF datasets. For example, QGIS “Add Raster Layer” can ingest EOPF data with no additional configuration.

---

Status

Feature	Status
Zarr-based read	In progress
Multi-band support	Planned
Metadata and georeferencing	Planned
Write support	Planned
Cloud (/vsis3/) access	Planned
QGIS usage validation	Ongoing

All functionality is under continuous development. Refer to the Roadmap or GitHub Issues for the full development timeline.

---

Table of Contents

1. Requirements
2. Building
3. Installing the Plugin
4. Usage
5. Testing
6. Contributing
7. License
8. Contact & Credits

---

Requirements

• GDAL >= 3.x (development headers required, e.g. libgdal-dev on Debian/Ubuntu)
• C++17 or later (recommended for improved JSON parsing, concurrency, etc.)
• CMake >= 3.10 (build system)
• (Optional) Python for extended testing or data generation scripts

On Linux, typical packages might include:

sudo apt-get update && sudo apt-get install -y \
  gdal-bin libgdal-dev \
  build-essential cmake \
  python3 python3-pip

---

Building

1. Clone the Repository

    https://github.com/EOPF-Sample-Service/GDAL-ZARR-EOPF.git
    cd GDAL-ZARR-EOPF

2. Configure with CMake

   mkdir build
   cd build
   cmake .. -DCMAKE_BUILD_TYPE=Release
   

   The CMake script attempts to locate GDAL via find_package(GDAL REQUIRED). If GDAL is not found automatically, set GDAL_DIR or GDAL_INCLUDE_DIR/GDAL_LIBRARY manually.

3. Compile

    cmake --build . -j$(nproc)

   This produces a shared library named something like gdal_EOPFZarr.so (on Linux) or gdal_EOPFZarr.dll (Windows).

---

Installing the Plugin

Option A: Copy to GDAL’s Plugin Directory

Move or copy the resulting library to GDAL’s plugin path, such as:

• Linux: /usr/lib/gdalplugins/ or /usr/local/lib/gdalplugins/
• Windows: C:\Program Files\GDAL\gdalplugins\ (varies depending on your setup)

Option B: Use GDAL_DRIVER_PATH

Set an environment variable so GDAL knows where to find your plugin:

export GDAL_DRIVER_PATH="/path/to/build"

Replace /path/to/build with the actual directory containing gdal_EOPF.so (or .dll on Windows).

---

Usage

1. Check Formats

   gdalinfo --formats | grep EOPFZARR

   You should see something like:

   EOPFZarr (ro): Earth Observation Processing Framework
   

2. Open an EOPF Dataset

   gdalinfo -oo EOPF_PROCESS=YES /path/to/EOPF_Zarr_Dataset/

   or in Python:

   from osgeo import gdal
   ds = gdal.OpenEx(zarr_path, gdal.OF_READONLY, open_options=['EOPF_PROCESS=YES'])
   # This will activate your EOPFZarr plugin.

   If everything is set up correctly, the plugin interprets the EOPF data and displays or processes it like any other GDAL-supported format.

3. In QGIS

   • Ensure the plugin library is discoverable by GDAL (see Installing the Plugin).
   • Launch QGIS → Add Raster Layer → Select the EOPF directory.
   • Confirm it displays the data with correct band(s), georeferencing, and metadata.

---

Testing

We provide two types of tests:

1. C++ Tests (optional)

   • Located in tests/. Built with CMake + CTest.

   cd build
   ctest --verbose

2. Python-based Tests

   • Python scripts or pytest tests that open EOPF data. For example, tests/test_eopf.py might validate chunk reading.

   pip install -r python/requirements.txt
   pytest tests/

3. Sample Data

   • Minimal Zarr/EOPF directories are in tests/sample_data/.
   • Future expansions will include real-world reference data for integration testing.

---

Contributing

1. Fork & Branch
   • Fork this repo, create feature branches, and submit Pull Requests.
2. Coding Style
   • Follow GDAL’s guidelines for driver formatting, error handling, etc.
3. Issues & Roadmap
   • See GitHub Issues for our backlog and planned features (multi-band metadata, HPC performance, S3 integration).
   • We welcome bug reports, feature requests, or suggestions.

Security: If you discover a security vulnerability in the plugin, please contact the maintainers directly before creating a public issue.

---

License

This project is released under the MIT License, compatible with GDAL’s open-source licensing. Please see the LICENSE file for details. By contributing, you agree your contributions will be licensed under the same terms.

---

Contact & Credits

• Maintainers: EURAC, Issues
• Acknowledgments:
  • ESA for EOPF specification, sample data, or project sponsorship
  • GDAL & QGIS communities for providing the geospatial foundation

---