KAFKA-17607: Add CI step to verify LICENSE-binary (#18299)

Add the verify_license.py script to our build to detect missing licenses.

Reviewers: Chia-Ping Tsai <chia7712@gmail.com>, Ken Huang <s7133700@gmail.com>, David Arthur <mumrah@gmail.com>

.github/workflows/build.yml

@@ -140,7 +140,7 @@ jobs:
         # --scan:         Publish the build scan. This will only work on PRs from apache/kafka and trunk
         # --no-scan:      For public fork PRs, we won't attempt to publish the scan
         run: |
-          ./gradlew --build-cache --info $SCAN_ARG check siteDocTar -x test
+          ./gradlew --build-cache --info $SCAN_ARG check releaseTarGz -x test
       - name: Archive check reports
         if: always()
         uses: actions/upload-artifact@v4
@@ -170,6 +170,8 @@ jobs:
             find ./site-docs/generated -type f -exec grep -L "." {} \; >&2
             exit 1
           fi
+      - name: Verify license file
+        run: python committer-tools/verify_license.py --skip-build
 
   test:
     needs: [configure, validate, load-catalog]

committer-tools/verify_license.py

@@ -24,6 +24,7 @@ import sys
 import tarfile
 import tempfile
 import subprocess
+import argparse
 
 # Constant: Regex to extract dependency tokens from the LICENSE file.
 # Matches lines that start with a dash and then a dependency token of the form:
@@ -44,7 +45,7 @@ def get_tarball_path(project_dir):
         print("Error: Distributions directory not found:", distributions_dir)
         sys.exit(1)
     
-    pattern = re.compile(r'^kafka_2\.13-.+\.tgz$', re.IGNORECASE)
+    pattern = re.compile(r'^kafka_2\.13-(?!.*docs).+\.tgz$', re.IGNORECASE)
     candidates = [
         os.path.join(distributions_dir, f)
         for f in os.listdir(distributions_dir)
@@ -74,12 +75,20 @@ def get_license_deps(license_text):
     return set(LICENSE_DEP_PATTERN.findall(license_text))
 
 def main():
+    # Argument parser
+    parser = argparse.ArgumentParser(description="Whether to skip executing ReleaseTarGz.")
+    parser.add_argument("--skip-build", action="store_true", help="skip the build")
+    args = parser.parse_args()
+
     # Assume the current working directory is the project root.
     project_dir = os.getcwd()
     print("Using project directory:", project_dir)
 
-    # Build the tarball.
-    run_gradlew(project_dir)
+    if args.skip_build:
+        print("Skip running './gradlew clean releaseTarGz'")
+    else:
+        # Build the tarball.
+        run_gradlew(project_dir)
     tarball = get_tarball_path(project_dir)
     print("Tarball created at:", tarball)