Adjust audi_line_count.py to take into account doc strings

scripts/audit_line_count.py

@@ -17,8 +17,8 @@ import os
 import sys
 from collections import defaultdict
 from dataclasses import dataclass, field
-from pathlib import Path
+from pathlib     import Path
-from typing import Iterator
+from typing      import Iterator
 
 TARGET_DIRS = ["src", "simulation", "tests", "scripts"]
 IGNORED_PATHS = {"__pycache__", ".git", "node_modules", "venv", ".venv", "env", ".env"}
@@ -100,23 +100,34 @@ def analyze_file(path: Path) -> FileStats | None:
  lines = content.splitlines()
  stats.total_lines = len(lines)
 
- for line in lines:
+ docstring_lines = set()
+ tree            = None
+ try:
+  tree = ast.parse(content, filename=str(path))
+  
+  # Extract line numbers for all standalone string expressions (docstrings & block comments)
+  for node in ast.walk(tree):
+   if isinstance(node, ast.Expr) and isinstance(node.value, ast.Constant) and isinstance(node.value.value, str):
+    if hasattr(node, "lineno") and hasattr(node, "end_lineno") and node.end_lineno is not None:
+     for ln in range(node.lineno, node.end_lineno + 1):
+      docstring_lines.add(ln)
+ except Exception:
+  pass
+
+ for i, line in enumerate(lines, 1):
   stripped = line.strip()
   if not stripped:               stats.blank_lines   += 1
+  elif i in docstring_lines:     stats.comment_lines += 1
   elif stripped.startswith("#"): stats.comment_lines += 1
   else:                          stats.code_lines    += 1
 
- try:
+ if tree is not None:
-  tree = ast.parse(content, filename=str(path))
+  visitor = CodeAnalyzer()
- except Exception:
+  visitor.visit(tree)
-  return stats
+  stats.classes         = visitor.classes
-
+  stats.functions       = visitor.functions
- visitor = CodeAnalyzer()
+  stats.methods         = visitor.methods
- visitor.visit(tree)
+  stats.top_level_decls = visitor.top_level_decls
- stats.classes         = visitor.classes
- stats.functions       = visitor.functions
- stats.methods         = visitor.methods
- stats.top_level_decls = visitor.top_level_decls
 
  return stats
 
@@ -146,12 +157,9 @@ def gather_dir_stats(root: Path) -> DirStats:
 
 
 def format_bytes(num_bytes: int) -> str:
- if num_bytes < 1024:
+ if   num_bytes < 1024:        return f"{num_bytes}B"
-  return f"{num_bytes}B"
+ elif num_bytes < 1024 * 1024: return f"{num_bytes / 1024:.1f}KB"
- elif num_bytes < 1024 * 1024:
+ else:                         return f"{num_bytes / (1024 * 1024):.1f}MB"
-  return f"{num_bytes / 1024:.1f}KB"
- else:
-  return f"{num_bytes / (1024 * 1024):.1f}MB"
 
 
 def print_stats(d: DirStats) -> None:
@@ -212,7 +220,7 @@ def main() -> None:
   print(f"  Files:            {combined.file_count:,}")
   print(f"  Lines:            {combined.total_lines:,} (code: {combined.code_lines:,} | comment: {combined.comment_lines:,} | blank: {combined.blank_lines:,})")
   print(f"  Classes:          {combined.class_count:,}")
-  print(f"  Functions:       {combined.function_count:,}")
+  print(f"  Functions:        {combined.function_count:,}")
   print(f"  Methods:          {combined.method_count:,}")
   total_decls = combined.class_count + combined.function_count + combined.method_count
   print(f"  Total decls:      {total_decls:,}")