spark: py_loc.py - Python line counter utility

wizard-work/spark/py_loc.py

@@ -0,0 +1,88 @@
+#!/usr/bin/env python3
+"""Count lines of Python code (excluding comments and blank lines)."""
+
+import sys
+import ast
+from pathlib import Path
+
+
+def count_python_loc(filepath: str) -> dict:
+    """
+    Count logical lines of Python code.
+    
+    Returns:
+        dict with total_lines, code_lines, comment_lines, blank_lines
+    """
+    path = Path(filepath)
+    
+    if not path.exists():
+        return {"error": f"File not found: {filepath}", "valid": False}
+    
+    try:
+        with open(filepath, 'r') as f:
+            lines = f.readlines()
+    except Exception as e:
+        return {"error": str(e), "valid": False}
+    
+    total = len(lines)
+    code_lines = 0
+    comment_lines = 0
+    blank_lines = 0
+    
+    in_docstring = False
+    docstring_delim = None
+    
+    for line in lines:
+        stripped = line.strip()
+        
+        # Check for docstring start/end
+        if not in_docstring and (stripped.startswith('"""') or stripped.startswith("'''")):
+            if stripped.count('"""') >= 2 or stripped.count("'''") >= 2:
+                # Single line docstring
+                comment_lines += 1
+            else:
+                in_docstring = True
+                docstring_delim = '"""' if '"""' in stripped else "'''"
+            continue
+        elif in_docstring:
+            if docstring_delim in stripped:
+                in_docstring = False
+            comment_lines += 1
+            continue
+        
+        if not stripped:
+            blank_lines += 1
+        elif stripped.startswith('#'):
+            comment_lines += 1
+        else:
+            code_lines += 1
+    
+    return {
+        "valid": True,
+        "file": filepath,
+        "total_lines": total,
+        "code_lines": code_lines,
+        "comment_lines": comment_lines,
+        "blank_lines": blank_lines,
+        "sum": code_lines + comment_lines + blank_lines
+    }
+
+
+def main():
+    if len(sys.argv) < 2:
+        print("Usage: py_loc.py <file1.py> [file2.py...]", file=sys.stderr)
+        sys.exit(1)
+    
+    files = sys.argv[1:]
+    
+    for filepath in files:
+        result = count_python_loc(filepath)
+        if "valid" in result and result["valid"]:
+            print(f"{filepath}: {result['code_lines']} code, {result['comment_lines']} comments, {result['blank_lines']} blank, {result['total_lines']} total")
+        else:
+            print(f"ERROR: {filepath} - {result.get('error', 'unknown')}", file=sys.stderr)
+            sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()