Update copyright year with a precommit hook (#1514)

* Update copyright year with a precommit hook

* Handle testing on all files

* Declare fewer files as binary

* Only update staged files

Author: mdboom

.gitattributes

@@ -6,7 +6,9 @@ cuda/_version.py export-subst
 # we do not own any headers checked in, don't touch them
 *.h binary
 *.hpp binary
-# Exception: headers we own (cuda_core C++ implementation)
+# Exception: headers we own
+cuda_bindings/cuda/bindings/_bindings/*.h -binary text diff
+cuda_bindings/cuda/bindings/_lib/*.h -binary text diff
 cuda_core/cuda/core/_cpp/*.h -binary text diff
 cuda_core/cuda/core/_cpp/*.hpp -binary text diff
 # git should not convert line endings in PNG files

.pre-commit-config.yaml

@@ -1,4 +1,4 @@
-# SPDX-FileCopyrightText: Copyright (c) 2024-2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 #
 # SPDX-License-Identifier: Apache-2.0
 
@@ -30,6 +30,7 @@ repos:
         additional_dependencies:
           - https://files.pythonhosted.org/packages/cc/20/ff623b09d963f88bfde16306a54e12ee5ea43e9b597108672ff3a408aad6/pathspec-0.12.1-py3-none-any.whl
         exclude: '.*pixi\.lock'
+        args: ["--fix"]
 
       - id: no-markdown-in-docs-source
         name: Prevent markdown files in docs/source directories
@@ -89,5 +90,6 @@ repos:
         args: [--no-pycodestyle]
         exclude: ^cuda_bindings/
 
+
 default_language_version:
       python: python3

toolshed/check_spdx.py

@@ -1,7 +1,10 @@
-# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 
+import datetime
 import os
+import re
+import subprocess
 import sys
 
 import pathspec
@@ -26,29 +29,88 @@ def load_spdx_ignore():
     return pathspec.PathSpec.from_lines("gitwildmatch", lines)
 
 
-def has_spdx_or_is_empty(filepath):
+COPYRIGHT_REGEX = (
+    rb"Copyright \(c\) (?P<years>[0-9]{4}(-[0-9]{4})?) "
+    rb"(?P<affiliation>NVIDIA CORPORATION( & AFFILIATES\. All rights reserved\.)?)"
+)
+COPYRIGHT_SUB = r"Copyright (c) {} \g<affiliation>"
+CURRENT_YEAR = str(datetime.date.today().year)
+
+
+def is_staged(filepath):
+    # If the file is staged, we need to update it to the current year
+    process = subprocess.run(  # noqa: S603
+        ["git", "diff", "--staged", "--", filepath],  # noqa: S607
+        capture_output=True,
+        text=True,
+    )
+    return process.stdout.strip() != ""
+
+
+def find_or_fix_spdx(filepath, fix):
     with open(filepath, "rb") as f:
         blob = f.read()
     if len(blob.strip()) == 0:
         return True
+
     good = True
     for expected_bytes in EXPECTED_SPDX_BYTES:
         if expected_bytes not in blob:
             print(f"MISSING {expected_bytes.decode()}{filepath!r}")
             good = False
+            continue
+
+        match = re.search(COPYRIGHT_REGEX, blob)
+        if match is None:
+            print(f"MISSING valid copyright line in {filepath!r}")
+            good = False
+            continue
+
+        years = match.group("years").decode()
+        if "-" in years:
+            start_year, end_year = years.split("-", 1)
+            if int(start_year) > int(end_year):
+                print(f"INVALID copyright years {years!r} in {filepath!r}")
+                good = False
+                continue
+        else:
+            start_year = end_year = years
+
+        staged = is_staged(filepath)
+
+        if staged and int(end_year) < int(CURRENT_YEAR):
+            print(f"OUTDATED copyright {years!r} (expected {CURRENT_YEAR!r}) in {filepath!r}")
+            good = False
+
+            if fix:
+                new_years = f"{start_year}-{CURRENT_YEAR}"
+                blob = re.sub(
+                    COPYRIGHT_REGEX,
+                    COPYRIGHT_SUB.format(new_years).encode("ascii"),
+                    blob,
+                )
+                with open(filepath, "wb") as f:
+                    f.write(blob)
+
     return good
 
 
 def main(args):
     assert args, "filepaths expected to be passed from pre-commit"
 
+    if "--fix" in args:
+        fix = True
+        del args[args.index("--fix")]
+    else:
+        fix = False
+
     ignore_spec = load_spdx_ignore()
 
     returncode = 0
     for filepath in args:
         if ignore_spec.match_file(filepath):
             continue
-        if not has_spdx_or_is_empty(filepath):
+        if not find_or_fix_spdx(filepath, fix):
             returncode = 1
     return returncode