Added regex for sanitizing the file names and excluded reserved words

nirman/parser.py

@@ -1,6 +1,18 @@
 import re
 from typing import List, Tuple
 
+# A set for fast, case-insensitive lookup of Windows reserved names.
+# We also include '.' and '..' which are special on all platforms.
+RESERVED_NAMES = {
+    "con", "prn", "aux", "nul",
+    "com1", "com2", "com3", "com4", "com5", "com6", "com7", "com8", "com9",
+    "lpt1", "lpt2", "lpt3", "lpt4", "lpt5", "lpt6", "lpt7", "lpt8", "lpt9",
+    ".", ".."
+}
+
+# Regex for invalid characters, EXCLUDING slashes which are handled separately.
+INVALID_CHARS_REGEX = r'[<>:"|?*]'
+
 def parse_markdown_tree(lines: List[str]) -> List[Tuple[int, str, bool]]:
     """
     Parse a markdown tree structure into a list of tuples representing the file/folder hierarchy.
@@ -38,10 +50,29 @@ def parse_markdown_tree(lines: List[str]) -> List[Tuple[int, str, bool]]:
             depth = 0
         else:
             depth = (len(prefix) // 4) + 1
+
+        # 1. First, determine if it's a directory from the raw name.
+        is_directory = name.endswith(('/', '\\'))
+
+        # 2. Then, create the clean_name by stripping the slashes.
+        clean_name = name.strip('\\/')
+
+        # 3. Now, sanitize the clean_name for invalid characters.
+        clean_name = re.sub(INVALID_CHARS_REGEX, '_', clean_name)
+
+        # 4. Finally, sanitize for reserved system names.
+        base_name = clean_name.split('.')[0]
+        if base_name.lower() in RESERVED_NAMES:
+            clean_name = "_" + clean_name
 
         is_directory = name.endswith(('/', '\\'))
         clean_name = name.strip('\\/')
 
+        # Get the base name before any extension for the check.
+        base_name = clean_name.split('.')[0]
+        if base_name.lower() in RESERVED_NAMES:
+            clean_name = "_" + clean_name
+
         if clean_name == '.' and depth == 0:
             is_directory = True