chore: fetch repo metadata daily

Author: alexeagle

.github/workflows/fetch_repo_metadata.sh

@@ -0,0 +1,50 @@
+#!/usr/bin/env bash
+
+set -o errexit -o nounset -o pipefail
+
+# Minimum required version: https://github.com/cli/cli/releases/tag/v2.49.0
+REQUIRED_GH_VERSION="2.49.0"
+
+# Check if gh is available
+if ! command -v gh &> /dev/null; then
+    echo "Error: GitHub CLI (gh) is not installed or not in your PATH."
+    exit 1
+fi
+
+# Ensure gh is logged in
+if ! gh auth status &> /dev/null; then
+    echo "Error: GitHub CLI (gh) is not logged in."
+    exit 1
+fi
+
+OUT=$(mktemp -d)
+# current script directory
+SCRIPT_DIR=$(dirname "$0")
+(
+    cd $SCRIPT_DIR/../../data/bazel-central-registry/modules
+    for module in $(ls); do
+        # GraphQL: Although you appear to have the correct authorization credentials, the `confluentinc` organization has an IP allow list enabled, and your IP address is not permitted to access this resource.
+        if [[ "$module" == "librdkafka" ]]; then
+            echo "  Skipping $module (known access restrictions)"
+            continue
+        fi
+        
+        echo "Fetching metadata for $module"
+        organdrepo=$(jq --raw-output <$module/metadata.json '.repository[] | select(startswith("github:")) | sub("github:"; "")' | head -1)
+        
+        # Skip if no GitHub repository found
+        if [[ -z "$organdrepo" || "$organdrepo" == "null" ]]; then
+            echo "  No GitHub repository found, skipping..."
+            continue
+        fi
+        
+        gh repo view --json description,licenseInfo,repositoryTopics,stargazerCount "$organdrepo" > $OUT/$module.github_metadata.json
+    done
+)
+
+ARCHIVE=github_metadata.tar.gz
+tar --create --auto-compress \
+    --directory "$OUT" \
+    --file "$ARCHIVE" .
+
+echo metadata=$PWD/$ARCHIVE > ${GITHUB_OUTPUT:-/dev/stdout}

.github/workflows/fetch_repo_metadata.yml

@@ -0,0 +1,26 @@
+# triggers on a cron, every day at 12:00 AM UTC
+
+on:
+    schedule:
+      - cron: '0 12 * * *'
+    # also triggered from the github UI
+    workflow_dispatch:
+  
+  jobs:
+    fetch_repo_metadata:  
+      runs-on: ubuntu-latest
+      steps:
+        - uses: actions/checkout@v5
+          with:
+            submodules: true
+        - name: Fetch repo metadata
+          id: fetch_repo_metadata
+          env:
+            GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          run: |
+            ./.github/workflows/fetch_repo_metadata.sh
+        - name: Upload metadata
+          uses: actions/upload-artifact@v4
+          with:
+            name: github_metadata
+            path: ${{ steps.fetch_repo_metadata.outputs.metadata }}