fix: sanitize endpoint extraction in installation workflow for improved accuracy

Signed-off-by: Akash <[email protected]>

Author: SkySingh04

Diff for: .github/workflows/installation-chart.yaml

@@ -127,19 +127,15 @@ jobs:
           CMD=$(karmadactl token create --print-register-command --kubeconfig $KUBECONFIG)
           TOKEN=$(echo "$CMD" | grep -o '\--token [^ ]*' | cut -d' ' -f2)
           HASH=$(echo "$CMD" | grep -o '\--discovery-token-ca-cert-hash [^ ]*' | cut -d' ' -f2)
-          # Extract server URL and ensure it's properly formatted
-          ENDPOINT=$(kubectl config view --minify -o jsonpath='{.clusters[0].cluster.server}' | sed 's|//.*@|//|')
-          echo "Extracted endpoint: $ENDPOINT"
-          # Sanitize endpoint (remove any accidental leading/trailing spaces or quotes)
-          ENDPOINT=$(echo "$ENDPOINT" | xargs)
+          ENDPOINT=$(kubectl config view --minify -o jsonpath='{.clusters[0].cluster.server}' | sed 's|^https://||')
           echo "Sanitized endpoint: $ENDPOINT"
           echo "token=$TOKEN" >> $GITHUB_OUTPUT
           echo "hash=$HASH" >> $GITHUB_OUTPUT 
           echo "endpoint=$ENDPOINT" >> $GITHUB_OUTPUT
 
       - name: Register cluster
         run: |
-          karmadactl register https://172.18.0.2:6443 \
+          karmadactl register 172.18.0.2:6443 \
             --token ${{ steps.token.outputs.token }} \
             --discovery-token-ca-cert-hash ${{ steps.token.outputs.hash }} \
             --kubeconfig $HOME/.kube/karmada.config