spark: enhanced json-validator with directory mode and accurate exit codes

wizard-work/spark/json-validator.py

@@ -1,8 +1,14 @@
 #!/usr/bin/env python3
 """JSON Validator - validates JSON files with proper exit codes.
-Exit 0: Valid JSON
+
-Exit 1: Invalid JSON or file not found
+Exit 0: All valid
+Exit 1: One or more invalid
 Exit 2: Usage error
+
+Features:
+- Validate individual files: json-validator.py <file.json> ...
+- Validate directory: json-validator.py --dir <directory>
+- Returns count of invalid files as exit code for --dir mode
 """
 
 import sys
@@ -10,26 +16,79 @@ import json
 from pathlib import Path
 
 
-def validate_json_file(filepath: str) -> bool:
+def validate_json_file(filepath: str) -> tuple[bool, str]:
-    """Validate a JSON file. Returns True if valid, False otherwise."""
+    """Validate a JSON file.
+    
+    Returns:
+        (True, "") if valid
+        (False, error_message) if invalid
+    """
     try:
         with open(filepath, 'r') as f:
             json.load(f)
         print(f"VALID: {filepath}")
-        return True
+        return True, ""
     except json.JSONDecodeError as e:
-        print(f"INVALID: {filepath} - line {e.lineno}, column {e.colno}: {e.msg}", file=sys.stderr)
+        return False, f"Invalid JSON at line {e.lineno}, column {e.colno}: {e.msg}"
-        return False
     except FileNotFoundError:
-        print(f"NOT FOUND: {filepath}", file=sys.stderr)
+        return False, "File not found"
-        return False
+    except Exception as e:
+        return False, f"Error: {e}"
+
+
+def validate_json_directory(directory: str) -> dict:
+    """Validate all .json files in a directory.
+    
+    Returns:
+        Dict with totals and counts
+    """
+    dir_path = Path(directory)
+    if not dir_path.exists():
+        return {"error": f"Directory not found: {directory}", "total": 0, "valid": 0, "invalid": 0}
+    if not dir_path.is_dir():
+        return {"error": f"Not a directory: {directory}", "total": 0, "valid": 0, "invalid": 0}
+    
+    json_files = list(dir_path.glob("*.json"))
+    valid = 0
+    invalid = 0
+    invalid_files = []
+    
+    for jf in json_files:
+        is_valid, error = validate_json_file(str(jf))
+        if is_valid:
+            valid += 1
+        else:
+            invalid += 1
+            invalid_files.append((jf.name, error))
+            print(f"INVALID: {jf.name} - {error}")
+    
+    return {"total": len(json_files), "valid": valid, "invalid": invalid, "invalid_files": invalid_files[:5]}
 
 
 def main():
     if len(sys.argv) < 2:
         print("Usage: json-validator.py <file-or-directory>", file=sys.stderr)
+        print("       json-validator.py --dir <directory>", file=sys.stderr)
         sys.exit(2)
     
+    # Check for --dir flag
+    if "--dir" in sys.argv:
+        try:
+            dir_idx = sys.argv.index("--dir")
+            directory = sys.argv[dir_idx + 1]
+            result = validate_json_directory(directory)
+            if "error" in result:
+                print(f"Error: {result['error']}", file=sys.stderr)
+                sys.exit(2)
+            print(f"\nDirectory: {directory}")
+            print(f"Total: {result['total']}")
+            print(f"Valid: {result['valid']}")
+            print(f"Invalid: {result['invalid']}")
+            sys.exit(result['invalid'])
+        except IndexError:
+            print("Error: --dir requires a directory argument", file=sys.stderr)
+            sys.exit(2)
+    else:
         paths = sys.argv[1:]
         all_valid = True
         
@@ -37,10 +96,12 @@ def main():
             p = Path(path)
             if p.is_dir():
                 for json_file in p.rglob("*.json"):
-                if not validate_json_file(str(json_file)):
+                    is_valid, error = validate_json_file(str(json_file))
+                    if not is_valid:
                         all_valid = False
             elif p.is_file():
-            if not validate_json_file(str(p)):
+                is_valid, error = validate_json_file(str(p))
+                if not is_valid:
                     all_valid = False
             else:
                 print(f"NOT FOUND: {path}", file=sys.stderr)