Skip to content

Create caches for elephant_data #25

Create caches for elephant_data

Create caches for elephant_data #25

name: Create caches for elephant_data
on:
workflow_dispatch: # Workflow can be triggered manually via GH actions webinterface
push: # When something is pushed into master this checks if caches need to re-created
branches:
- master
schedule:
- cron: "11 23 * * *" # Daily at 23:11 UTC
jobs:
create-data-cache-if-missing:
name: Caching data env
runs-on: ubuntu-latest
strategy:
# do not cancel all in-progress jobs if any matrix job fails
fail-fast: false
steps:
- name: Get current hash (SHA) of the elephant_data repo
id: elephant-data
run: |
echo "dataset_hash=$(git ls-remote https://gin.g-node.org/NeuralEnsemble/elephant-data.git HEAD | cut -f1)" >> $GITHUB_OUTPUT
- uses: actions/cache@v4.0.2
# Loading cache of elephant-data
id: cache-datasets
with:
path: ~/elephant-data
key: datasets-${{ steps.elephant-data.outputs.dataset_hash }}
- name: Cache found?
run: echo "Cache-hit == ${{steps.cache-datasets.outputs.cache-hit == 'true'}}"
- name: Configuring git
if: steps.cache-datasets.outputs.cache-hit != 'true'
run: |
git config --global user.email "elephant_ci@fake_mail.com"
git config --global user.name "elephant CI"
git config --global filter.annex.process "git-annex filter-process" # recommended for efficiency
- name: Install Datalad Linux
if: steps.cache-datasets.outputs.cache-hit != 'true'
run: |
python -m pip install -U pip # Official recommended way
pip install datalad-installer
datalad-installer --sudo ok git-annex --method datalad/packages
pip install datalad
- name: Download dataset
id: download-dataset
if: steps.cache-datasets.outputs.cache-hit != 'true'
# Download repository and also fetch data
run: |
cd ~
datalad --version
datalad install --recursive --get-data https://gin.g-node.org/NeuralEnsemble/elephant-data
- name: Show size of the cache to assert data is downloaded
run: |
cd ~
du -hs ~/elephant-data
ls -lh ~/elephant-data