Reject duplicate workflow keys instead of trusting safe_load

.github/workflows/build.yml

@@ -167,7 +167,6 @@ jobs:
         env:
           GH_TOKEN: ${{ github.token }}
           TAG: ${{ needs.prepare.outputs.release_tag }}
-        env:
           KERNEL_REF: ${{ needs.prepare.outputs.kernel_ref }}
           KERNEL_SHA: ${{ needs.prepare.outputs.kernel_sha }}
         run: |

.github/workflows/ci.yml

@@ -26,14 +26,8 @@ jobs:
           shellcheck --version
           shellcheck scripts/*.sh packaging/*.sh
 
-      - name: workflows parse
-        run: |
-          python3 - <<'EOF'
-          import glob, yaml
-          for f in sorted(glob.glob(".github/workflows/*.yml")):
-              yaml.safe_load(open(f))
-              print("ok", f)
-          EOF
+      - name: workflows
+        run: ./scripts/lint-workflows.py
 
   # Catches the failure mode this repo cares about most: Kconfig silently
   # dropping a requested option. Takes a few minutes, no compilation.

scripts/lint-workflows.py

@@ -0,0 +1,58 @@
+#!/usr/bin/env python3
+"""Check the workflow files the way GitHub does, not the way PyYAML does.
+
+yaml.safe_load() accepts duplicate mapping keys and silently keeps the last
+one, so a second `env:` in a step parses cleanly here and is then rejected by
+GitHub with "'env' is already defined" - after a push, with zero jobs run and
+no log to read. This loader refuses duplicates instead.
+"""
+
+import glob
+import sys
+
+import yaml
+
+
+class StrictLoader(yaml.SafeLoader):
+    pass
+
+
+def no_duplicates(loader, node, deep=False):
+    mapping = {}
+    for key_node, value_node in node.value:
+        key = loader.construct_object(key_node, deep=deep)
+        if key in mapping:
+            raise yaml.YAMLError(
+                "duplicate key %r at line %d, column %d"
+                % (key, key_node.start_mark.line + 1, key_node.start_mark.column + 1)
+            )
+        mapping[key] = loader.construct_object(value_node, deep=deep)
+    return mapping
+
+
+StrictLoader.add_constructor(
+    yaml.resolver.BaseResolver.DEFAULT_MAPPING_TAG, no_duplicates
+)
+
+
+def main(paths):
+    paths = paths or sorted(glob.glob(".github/workflows/*.yml"))
+    failed = False
+    for path in paths:
+        try:
+            doc = yaml.load(open(path), Loader=StrictLoader)
+        except yaml.YAMLError as exc:
+            print("FAIL %s: %s" % (path, exc))
+            failed = True
+            continue
+        jobs = doc.get("jobs") or {}
+        if not jobs:
+            print("FAIL %s: no jobs defined" % path)
+            failed = True
+            continue
+        print("ok   %s (%s)" % (path, ", ".join(jobs)))
+    return 1 if failed else 0
+
+
+if __name__ == "__main__":
+    sys.exit(main(sys.argv[1:]))