add vulture hook

Author: MarcoGorelli

.pre-commit-config.yaml

@@ -8,6 +8,13 @@ repos:
     hooks:
     -   id: absolufy-imports
         files: ^pandas/
+-   repo: https://github.com/jendrikseipp/vulture
+    rev: 'v2.3'
+    hooks:
+      - id: vulture
+        entry: python scripts/run_vulture.py
+        pass_filenames: true
+        require_serial: false
 -   repo: https://github.com/python/black
     rev: 21.12b0
     hooks:

pandas/tests/frame/test_constructors.py

@@ -1285,8 +1285,7 @@ def test_constructor_empty_list(self):
 
         # Empty generator: list(empty_gen()) == []
         def empty_gen():
-            return
-            yield
+            yield from ()
 
         df = DataFrame(empty_gen(), columns=["A", "B"])
         tm.assert_frame_equal(df, expected)

pandas/tests/io/test_common.py

@@ -556,8 +556,6 @@ def test_encoding_errors(encoding_errors, format):
 
     if format == "csv":
         return
-        content = bad_encoding + b"\n" + bad_encoding
-        reader = pd.read_csv
     else:
         content = (
             b'{"'

scripts/run_vulture.py

@@ -0,0 +1,32 @@
+"""
+Run Vulture, but filter out "unused variable" errors - only find unreachable code.
+
+Might be able to get rid of this if in Vulture they let us filter out errors by
+their codes, see
+https://github.com/jendrikseipp/vulture/issues/269#issuecomment-1004048279.
+"""
+
+import argparse
+import subprocess
+import sys
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("files", nargs="*")
+    args = parser.parse_args()
+
+    output = subprocess.run(
+        ["vulture", "--min-confidence", "100", *args.files],
+        text=True,
+        capture_output=True,
+    )
+    stdout = "".join(
+        [
+            line
+            for line in output.stdout.splitlines(keepends=True)
+            if "unused variable" not in line
+        ]
+    )
+    sys.stdout.write(stdout)
+    ret = bool(stdout)
+    sys.exit(ret)