Minify ontologies and keep them fresh, for fast metadata validation (SCP-5719) #22
Workflow file for this run
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Minify ontologies | |
| on: | |
| pull_request: | |
| types: [opened] # Only trigger on PR "opened" event | |
| # push: # Uncomment, update branches to develop / debug | |
| # branches: | |
| # ew-minify-ontologies | |
| jobs: | |
| build: | |
| runs-on: ubuntu-latest | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@v3 | |
| - name: Copy and decompress ontologies in repo | |
| run: cd ingest/validation; mkdir tmp; cp -r *.min.tsv.gz tmp/; gzip -d tmp/*.min.tsv.gz | |
| - name: Minify newest ontologies | |
| run: cd ingest/validation; python3 minify_ontologies.py; gzip -dkf *.min.tsv.gz | |
| - name: Diff and commit changes | |
| run: | | |
| #!/bin/bash | |
| # Revert the default `set -e` in GitHub Actions, to e.g. ensure | |
| # "diff" doesn't throw an error when something is found | |
| set +e | |
| # set -x # Enable debugging | |
| cd ingest/validation | |
| # Define directories | |
| SOURCE_DIR="." | |
| TMP_DIR="tmp" | |
| # Ensure TMP_DIR exists | |
| if [ ! -d "$TMP_DIR" ]; then | |
| echo "Temporary directory $TMP_DIR does not exist." | |
| exit 1 | |
| fi | |
| # Flag to track if there are any changes | |
| CHANGES_DETECTED=false | |
| # Find and diff files | |
| for FILE in $(find "$SOURCE_DIR" -type f -name "*.min.tsv"); do | |
| # Get the base name of the file | |
| BASENAME=$(basename "$FILE") | |
| # Construct the path to the corresponding file in the TMP_DIR | |
| TMP_FILE="$TMP_DIR/$BASENAME" | |
| # Check if the corresponding file exists in TMP_DIR | |
| if [ -f "$TMP_FILE" ]; then | |
| # Run the diff command | |
| echo "Diffing $FILE and $TMP_FILE" | |
| diff "$FILE" "$TMP_FILE" > diff_output.txt | |
| # Check if diff output is not empty | |
| if [ -s diff_output.txt ]; then | |
| echo "Differences found in $BASENAME" | |
| cat diff_output.txt | |
| # Copy the updated file to the source directory (if needed) | |
| cp "$TMP_FILE" "$FILE" | |
| # Mark that changes have been detected | |
| CHANGES_DETECTED=true | |
| # Stage the modified file | |
| git add "$FILE".gz | |
| else | |
| echo "No differences in $BASENAME" | |
| fi | |
| else | |
| echo "No corresponding file found in $TMP_DIR for $BASENAME" | |
| fi | |
| done | |
| if [ "$CHANGES_DETECTED" = true ]; then | |
| # Configure Git | |
| git config --global user.name "github-actions" | |
| git config --global user.email "[email protected]" | |
| # Commit changes | |
| git commit -m "Update minified ontologies via GitHub Actions" | |
| git push origin HEAD | |
| else | |
| echo "No changes to commit." | |
| fi |