Fetch test data from Zenodo via zenodo_get instead of tracking it in git

.github/workflows/apptainer.yml

@@ -13,6 +13,7 @@ on:
       - "pyproject.toml"
       - "uv.lock"
       - "download_daily_files.py"
+      - "download_data_for_tests.sh"
       - ".github/workflows/apptainer.yml"
   pull_request:
     branches: [main]
@@ -22,6 +23,7 @@ on:
       - "pyproject.toml"
       - "uv.lock"
       - "download_daily_files.py"
+      - "download_data_for_tests.sh"
       - ".github/workflows/apptainer.yml"
   workflow_dispatch:
 
@@ -36,6 +38,36 @@ jobs:
       - name: Checkout repo
         uses: actions/checkout@v7
 
+      - name: Install uv
+        uses: astral-sh/setup-uv@v7
+        with:
+          python-version: "3.12"
+
+      - name: Resolve zenodo checksum for cache key
+        id: zenodo-record
+        run: |
+          DOI=$(grep -oP 'DOI="\K[^"]+' download_data_for_tests.sh)
+          uv run zenodo_get --doi "${DOI}" --md5 --output-dir /tmp/zenodo-md5
+          echo "id=$(md5sum /tmp/zenodo-md5/md5sums.txt | cut -d' ' -f1)" >> "$GITHUB_OUTPUT"
+
+      - name: Cache test data
+        id: cache-test-data
+        uses: actions/cache@v6
+        with:
+          path: |
+            tests/io/dst/data
+            tests/io/f10_7/data
+            tests/io/hp/data
+            tests/io/kp/data
+            tests/io/omni/data
+            tests/io/RBMDataSet/data
+            tests/io/solar_wind/data
+          key: zenodo-test-data-${{ steps.zenodo-record.outputs.id }}
+
+      - name: Download test data
+        if: steps.cache-test-data.outputs.cache-hit != 'true'
+        run: ./download_data_for_tests.sh
+
       - name: Get SWVO package version
         id: swvo_version
         run: |

.github/workflows/tests.yml

@@ -66,6 +66,31 @@ jobs:
           uv sync --locked
           uv pip install coveralls coverage pytest-cov
 
+      - name: Resolve zenodo checksum for cache key
+        id: zenodo-record
+        run: |
+          DOI=$(grep -oP 'DOI="\K[^"]+' download_data_for_tests.sh)
+          uv run zenodo_get --doi "${DOI}" --md5 --output-dir /tmp/zenodo-md5
+          echo "id=$(md5sum /tmp/zenodo-md5/md5sums.txt | cut -d' ' -f1)" >> "$GITHUB_OUTPUT"
+
+      - name: Cache test data
+        id: cache-test-data
+        uses: actions/cache@v6
+        with:
+          path: |
+            tests/io/dst/data
+            tests/io/f10_7/data
+            tests/io/hp/data
+            tests/io/kp/data
+            tests/io/omni/data
+            tests/io/RBMDataSet/data
+            tests/io/solar_wind/data
+          key: zenodo-test-data-${{ steps.zenodo-record.outputs.id }}
+
+      - name: Download test data
+        if: steps.cache-test-data.outputs.cache-hit != 'true'
+        run: ./download_data_for_tests.sh
+
       - name: Run tests
         run: uv run python -m coverage run --source swvo -m pytest -rf tests
 

.gitignore

@@ -31,6 +31,8 @@ docs/api/*
 .nfs*
 .env
 
+tests/io/*/data/
+
 
 **/.DS_Store
 *.sif

README.md

@@ -125,3 +125,7 @@ uv pip install swvo
 ```
 
 All the above `uv` commands assume you have `uv` installed, if not then remove `uv` prefix from the commands and run them directly.
+
+## Test Data
+
+The data used by the test suite is hosted on Zenodo and is not tracked in this repository. Install the `dev` dependency group (`uv sync --group dev`), which pulls in `zenodo_get`, then run `./download_data_for_tests.sh` to fetch and unpack it.

download_data_for_tests.sh

@@ -0,0 +1,14 @@
+#!/usr/bin/env bash
+# SPDX-FileCopyrightText: 2026 GFZ Helmholtz Centre for Geosciences
+# SPDX-FileContributor: Sahil Jhawar
+#
+# SPDX-License-Identifier: Apache-2.0
+
+set -euo pipefail
+
+DOI="10.5281/zenodo.22983305"
+OUTPUT_DIR="."
+
+uv run zenodo_get --doi "${DOI}" --output-dir "${OUTPUT_DIR}"
+unzip -o "${OUTPUT_DIR}/swvo-test-data.zip" -d "${OUTPUT_DIR}"
+rm "${OUTPUT_DIR}/swvo-test-data.zip"

pyproject.toml

@@ -1,3 +1,3 @@
 # SPDX-FileCopyrightText: 2025 GFZ Helmholtz Centre for Geosciences
 #
 # SPDX-License-Identifier: Apache-2.0
@@ -135,6 +135,7 @@
     "coverage>=7.14.0",
     "pytest>=9.0.2",
     "pytest-mock>=3.15.1",
+    "zenodo-get>=3.1.0",
 ]
 docs = [
     "sphinx-copybutton",