Create a PR when there are unused stubtest whitelist entries (#4334)

.github/workflows/stubtest-unused-whitelist.yml

@@ -1,4 +1,4 @@
-name: Find unused stubtest whitelist entries
+name: Remove unused stubtest whitelist entries
 
 on:
   workflow_dispatch:
@@ -39,11 +39,20 @@ jobs:
     runs-on: ubuntu-latest
     needs: stubtest
     steps:
+      - uses: actions/checkout@v2
+      - uses: actions/setup-python@v2
+        with:
+          python-version: 3.8
       - name: Get stubtest outputs
         uses: actions/download-artifact@v2
         with:
           name: stubtest-output
-      - name: Remove duplicates
-        run: |
-          cat stubtest-output-* | sort -u | tee stubtest-output
-          test "$(wc -l stubtest-output | cut -d ' ' -f 1)" = 0
+      - name: Collate duplicates
+        run: cat stubtest-output-* | sort -u | tee stubtest-output
+      - name: Remove entries from whitelists
+        run: python/scripts/update-stubtest-whitelist.py stubtest-output
+      - name: Create pull request
+        uses: peter-evans/create-pull-request@v2
+        with:
+          commit-message: Remove unused stubtest whitelist entries
+          title: "[gh-action] Remove unused stubtest whitelist entries"

.gitignore

@@ -39,6 +39,7 @@ htmlcov/
 .cache
 nosetests.xml
 coverage.xml
+stubtest-output*
 
 # Translations
 *.mo

scripts/update-stubtest-whitelist.py

@@ -0,0 +1,58 @@
+#!/usr/bin/env python3
+
+# This script removes lines from stubtest whitelists, according to
+# an input file. The input file has one entry to remove per line.
+# Each line consists of a whitelist filename and an entry, separated
+# by a colon.
+
+# This script is used by the workflow to remove unused whitelist entries.
+
+import sys
+from collections import defaultdict
+from typing import Dict, List, Set, Tuple
+
+
+def main() -> None:
+    if len(sys.argv) != 2:
+        print(f"Usage: {sys.argv[0]} FILENAME", file=sys.stderr)
+        sys.exit(1)
+
+    to_remove = parse_input_file(sys.argv[1])
+    for filename, entries in to_remove.items():
+        remove_entries_from_whitelist(filename, entries)
+
+
+def parse_input_file(input_file: str) -> Dict[str, Set[str]]:
+    to_remove = defaultdict(set)
+    with open(input_file) as f:
+        for filename, wl_entry in [parse_input_line(li) for li in f if li.strip()]:
+            to_remove[filename].add(wl_entry)
+    return to_remove
+
+
+# Returns a (filename, entry) tuple.
+def parse_input_line(line: str) -> Tuple[str, str]:
+    line = line.strip()
+    return line.split(":", maxsplit=1)
+
+
+def remove_entries_from_whitelist(filename: str, entries: Set[str]) -> None:
+    new_lines: List[str] = []
+    with open(filename) as f:
+        for line in f:
+            entry = line.strip().split(" #")[0]
+            if entry in entries:
+                entries.remove(entry)
+            else:
+                new_lines.append(line)
+    if entries:
+        print(f"WARNING: The following entries were not found in '{filename}':", filename=sys.stderr)
+        for entry in entries:
+            print(f"  * {entry}")
+    with open(filename, "w") as f:
+        for line in new_lines:
+            f.write(line)
+
+
+if __name__ == "__main__":
+    main()