Fix CRAN link check condition and clean up trailing commas in package… #92
This file contains hidden or 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
| # Start with a list of packages in packages.csv, and get external data about them (e.g. last commit date) to generate an HTML page with the package information. This workflow is triggered manually from the Actions tab. | |
| # Olly Butters | |
| # 15/5/26 | |
| name: Package list | |
| # Controls when the workflow will run | |
| on: | |
| schedule: | |
| - cron: '30 5 * * 0' | |
| timezone: "Europe/London" | |
| push: | |
| branches: [ "main" ] | |
| workflow_dispatch: | |
| jobs: | |
| build: | |
| runs-on: ubuntu-slim | |
| steps: | |
| - name: Checkout repository | |
| uses: actions/checkout@v4 | |
| # Parse the package list, get the external data, and write to a new CSV file | |
| - name: Build package data | |
| run: | | |
| echo $GITHUB_WORKSPACE | |
| ls $GITHUB_WORKSPACE | |
| # Make a cache directory for some info from the GitHub clones. | |
| mkdir cache | |
| # Create new output file | |
| echo "Name,Description,CRAN link,CRAN version,CRAN license,GitHub link,Last commit,GitHub version,GitHub license,Github owner,Status" > output.csv | |
| # Read the data in from package_list.csv. The name and gh_link fields are required. | |
| while IFS="," read -r name description gh_link cran_link status | |
| do | |
| echo "-----------------------------------" | |
| echo "name: $name" | |
| echo "desc: $description" | |
| echo "github_link: $gh_link" | |
| echo "cran_link: $cran_link" | |
| echo "status: $status" | |
| echo "" | |
| ##### GitHub info | |
| echo -e "\nGitHub repo information" | |
| # Extract the repo name from the GitHub link | |
| gh_repo_name=$(echo $gh_link | awk -F/ '{print $NF}') | |
| echo $gh_repo_name | |
| # clone the GitHub repo and process it | |
| if [ -z "${gh_link}" ]; then | |
| echo "No GitHub link provided, skipping GitHub info for this package." | |
| gh_last_commit="" | |
| gh_last_release="" | |
| gh_license="" | |
| gh_functions="" | |
| gh_owner="" | |
| gh_code_of_conduct="" | |
| else | |
| echo "Cloning GitHub repo: $gh_link" | |
| git clone $gh_link | |
| cd $gh_repo_name | |
| # Last commit date of repo | |
| gh_last_commit=$(git log -1 --format=%cs) | |
| echo "last commit: $gh_last_commit" | |
| # Last release | |
| if [ $(git rev-list --tags --max-count=1) ]; then | |
| gh_last_release=$(git describe --tags $(git rev-list --tags --max-count=1)) | |
| else | |
| gh_last_release="" | |
| fi | |
| echo "last GitHub release: $gh_last_release" | |
| # If there is a NAMES file then parse it to get the functino list and save the output | |
| if [ -f "NAMESPACE" ]; then | |
| echo "NAMESPACE file found" | |
| gh_functions=$(sed -n "s/^export(\(.*\))$/\1/p" NAMESPACE | tr '\n' ',') | |
| echo "GitHub functions: $gh_functions" | |
| if [ -n "${gh_functions}" ]; then | |
| echo "${name}: ${gh_functions}" >> ../functions.txt | |
| else | |
| echo $name >> ../functions.txt | |
| fi | |
| fi | |
| # Use the GitHub API to pull in info about the repoisitory | |
| echo "\nGitHub API information" | |
| gh repo view --json codeOfConduct,description,homepageUrl,licenseInfo,owner,parent,updatedAt > gh_repo_info.json | |
| echo "API response:" | |
| cat gh_repo_info.json | |
| # Get the license as it appears in GitHub | |
| gh_license=$(cat gh_repo_info.json | jq '.licenseInfo.nickname') | |
| if [ -z "${gh_license}" ] || [ "${gh_license}" = '""' ]; then | |
| gh_license=$(cat gh_repo_info.json | jq '.licenseInfo.name') | |
| fi | |
| # If the license is "Other", check for a LICENSE.md file and see if it contains actual license | |
| if [ "${gh_license}" = '"Other"' ]; then | |
| if [ -f "LICENSE.md" ]; then | |
| echo "License file found: LICENSE.md" | |
| if $(grep -q "MIT License" LICENSE.md); then | |
| gh_license="MIT License" | |
| fi | |
| fi | |
| fi | |
| echo "GitHub license: $gh_license" | |
| gh_owner=$(cat gh_repo_info.json | jq '.owner.login') | |
| echo "GitHub owner: $gh_owner" | |
| gh_code_of_conduct=$(cat gh_repo_info.json | jq '.codeOfConduct.name') | |
| echo "GitHub code of conduct: $gh_code_of_conduct" | |
| # Keep some files for processing later | |
| cd .. | |
| mkdir cache/$gh_repo_name | |
| cp $gh_repo_name/DESCRIPTION cache/$gh_repo_name/ 2>/dev/null || true | |
| cp $gh_repo_name/NAMESPACE cache/$gh_repo_name/ 2>/dev/null || true | |
| cp $gh_repo_name/README.md cache/$gh_repo_name/ 2>/dev/null || true | |
| cp $gh_repo_name/LICENSE* cache/$gh_repo_name/ 2>/dev/null || true | |
| # tidy up | |
| rm -rf $gh_repo_name | |
| fi | |
| ##### CRAN info | |
| echo "CRAN information" | |
| if [[ -z "${cran_link}" || "${cran_link}" == " "]]; then | |
| cran_version="" | |
| cran_license="" | |
| else | |
| # Get CRAN package name from cran_link | |
| CRAN_package_name=$(echo $cran_link | awk -F= '{print $NF}') | |
| echo "CRAN package name: $CRAN_package_name" | |
| cran_description_file_link="https://cran.r-project.org/web/packages/${CRAN_package_name}/DESCRIPTION" | |
| wget --content-on-error $cran_description_file_link | |
| # might not be on the CRAN | |
| if [ -f DESCRIPTION ]; then | |
| cran_version=$(grep -oP '^Version:\s*\K([a-zA-Z0-9\. _-]+)\s*$' DESCRIPTION) | |
| cran_license=$(grep -oP '^License:\s*\K([a-zA-Z0-9\. _\-\(\)><=\+]+)\s*$' DESCRIPTION) | |
| # tidy up | |
| rm DESCRIPTION | |
| else | |
| cran_version="" | |
| cran_license="" | |
| fi | |
| fi | |
| echo "CRAN version: $cran_version" | |
| echo "CRAN license: $cran_license" | |
| # write to file | |
| echo "$name,$description,$cran_link,$cran_version,$cran_license,$gh_link,$gh_last_commit,$gh_last_release,$gh_license,$gh_owner,$status" >> output.csv | |
| echo "-----------------------------------" | |
| done < <(tail -n +2 package_list.csv) | |
| # copy files to a new directory for deployment | |
| mkdir gh-pages | |
| cp output.csv cache/ | |
| cp functions.txt cache/ | |
| # print the output files to the console for debugging | |
| cat output.csv | |
| cat functions.txt | |
| env: | |
| GH_TOKEN: ${{ secrets.GITHUB_TOKEN }} | |
| # Parse the output CSV file and generate an HTML page | |
| - name: build html page | |
| run: | | |
| python3 build_web_pages/build_packages_summary_page.py | |
| python3 build_web_pages/build_packages_long_page.py | |
| mv output/* gh-pages/ | |
| mv cache/* gh-pages/ | |
| - name: Upload static files as artifact | |
| id: deployment | |
| uses: actions/upload-pages-artifact@v3 | |
| with: | |
| path: gh-pages | |
| # Deployment job | |
| deploy: | |
| environment: | |
| name: github-pages | |
| url: ${{ steps.deployment.outputs.page_url }} | |
| runs-on: ubuntu-slim | |
| permissions: | |
| pages: write | |
| id-token: write | |
| needs: build | |
| steps: | |
| - name: Deploy to GitHub Pages | |
| id: deployment | |
| uses: actions/deploy-pages@v4 |