spark: add file hash validator for SHA256 verification

wizard-work/spark/file_hash_validator.py

@@ -0,0 +1,84 @@
+#!/usr/bin/env python3
+"""File Hash Validator - verify SHA256 checksums for file integrity.
+
+Usage:
+    python file_hash_validator.py <file> <expected_sha256>
+    python file_hash_validator.py --verify <checksum_file>
+
+A checksum file format: "<sha256_hash>  <filename>" (same as sha256sum output)
+"""
+
+import hashlib
+import sys
+from pathlib import Path
+
+
+def compute_sha256(filepath: str) -> str:
+    """Compute SHA256 hash of a file."""
+    sha256_hash = hashlib.sha256()
+    with open(filepath, "rb") as f:
+        for chunk in iter(lambda: f.read(8192), b""):
+            sha256_hash.update(chunk)
+    return sha256_hash.hexdigest()
+
+
+def verify_file(filepath: str, expected_hash: str) -> bool:
+    """Verify a file's hash matches expected value."""
+    if not Path(filepath).exists():
+        print(f"ERROR: File not found: {filepath}")
+        return False
+    actual = compute_sha256(filepath)
+    if actual.lower() == expected_hash.lower():
+        print(f"OK: {filepath}")
+        print(f"  SHA256: {actual}")
+        return True
+    else:
+        print(f"FAIL: {filepath}")
+        print(f"  Expected: {expected_hash}")
+        print(f"  Actual:   {actual}")
+        return False
+
+
+def verify_checksum_file(checksum_file: str) -> int:
+    """Verify all files in a checksum file. Returns count of failed verifications."""
+    if not Path(checksum_file).exists():
+        print(f"ERROR: Checksum file not found: {checksum_file}")
+        return 1
+
+    failed = 0
+    with open(checksum_file, "r") as f:
+        for line in f:
+            line = line.strip()
+            if not line or line.startswith("#"):
+                continue
+            parts = line.split()
+            if len(parts) >= 2:
+                expected_hash = parts[0]
+                filepath = parts[1]
+                if not verify_file(filepath, expected_hash):
+                    failed += 1
+            else:
+                print(f"SKIP: Malformed line: {line}")
+
+    if failed == 0:
+        print(f"\nAll files verified successfully.")
+    else:
+        print(f"\n{failed} file(s) failed verification.")
+    return failed
+
+
+def main():
+    if len(sys.argv) < 3:
+        print(__doc__)
+        sys.exit(1)
+
+    if sys.argv[1] == "--verify":
+        exit_code = verify_checksum_file(sys.argv[2])
+    else:
+        exit_code = 0 if verify_file(sys.argv[1], sys.argv[2]) else 1
+
+    sys.exit(exit_code)
+
+
+if __name__ == "__main__":
+    main()