spark: add file_hashes.py utility for SHA256/MD5 checksums

wizard-work/spark/file_hashes.py

@@ -0,0 +1,43 @@
+#!/usr/bin/env python3
+"""Calculate SHA256 and MD5 hashes for files. Utility snippet."""
+
+import hashlib
+import sys
+from pathlib import Path
+
+
+def sha256_hash(filepath: str) -> str:
+    """Return SHA256 hash of file contents."""
+    h = hashlib.sha256()
+    with open(filepath, 'rb') as f:
+        for chunk in iter(lambda: f.read(8192), b''):
+            h.update(chunk)
+    return h.hexdigest()
+
+
+def md5_hash(filepath: str) -> str:
+    """Return MD5 hash of file contents."""
+    h = hashlib.md5()
+    with open(filepath, 'rb') as f:
+        for chunk in iter(lambda: f.read(8192), b''):
+            h.update(chunk)
+    return h.hexdigest()
+
+
+def main():
+    if len(sys.argv) < 2:
+        print("Usage: file_hashes.py <file1> [file2...]")
+        sys.exit(1)
+
+    for path in sys.argv[1:]:
+        p = Path(path)
+        if not p.exists():
+            print(f"File not found: {p}")
+            continue
+        print(f"{p}")
+        print(f"  SHA256: {sha256_hash(path)}")
+        print(f"  MD5:    {md5_hash(path)}")
+
+
+if __name__ == "__main__":
+    main()