fixes for file naming

bdfr/file_name_formatter.py

@@ -36,6 +36,7 @@ class FileNameFormatter:
         directory_format_string: str,
         time_format_string: str,
         restriction_scheme: Optional[str] = None,
+        strip_unicode: bool = True,
     ):
         if not self.validate_string(file_format_string):
             raise BulkDownloaderException(f'"{file_format_string}" is not a valid format string')
@@ -43,6 +44,7 @@ class FileNameFormatter:
         self.directory_format_string: list[str] = directory_format_string.split("/")
         self.time_format_string = time_format_string
         self.restiction_scheme = restriction_scheme.lower().strip() if restriction_scheme else None
+        self.strip_unicode = strip_unicode
         if self.restiction_scheme == "windows":
             self.max_path = self.WINDOWS_MAX_PATH_LENGTH
         else:
@@ -65,12 +67,22 @@ class FileNameFormatter:
 
         result = result.replace("/", "")
 
+        # Strip Unicode characters that cause Windows SMB issues if enabled
+        if self.strip_unicode:
+            result = FileNameFormatter._strip_unicode_chars(result)
+
         if self.restiction_scheme is None:
             if platform.system() == "Windows":
                 result = FileNameFormatter._format_for_windows(result)
+                # Strip emojis on Windows if strip_unicode is enabled (for backward compatibility)
+                if self.strip_unicode:
+                    result = FileNameFormatter._strip_emojis(result)
         elif self.restiction_scheme == "windows":
             logger.debug("Forcing Windows-compatible filenames")
             result = FileNameFormatter._format_for_windows(result)
+            # Strip emojis when forcing Windows compatibility if strip_unicode is enabled
+            if self.strip_unicode:
+                result = FileNameFormatter._strip_emojis(result)
         return result
 
     @staticmethod
@@ -219,9 +231,33 @@ class FileNameFormatter:
         invalid_characters = r'<>:"\/|?*'
         for char in invalid_characters:
             input_string = input_string.replace(char, "")
-        input_string = FileNameFormatter._strip_emojis(input_string)
         return input_string
 
+    @staticmethod
+    def _strip_unicode_chars(input_string: str) -> str:
+        """Strip Unicode characters that cause Windows SMB to create 8.3 short names"""
+        import unicodedata
+
+        # Remove emoji and symbols that cause Windows SMB issues
+        result = []
+        for char in input_string:
+            # Keep ASCII characters
+            if ord(char) < 0x80:
+                result.append(char)
+            # Keep common Unicode letters, numbers, and punctuation
+            elif unicodedata.category(char) in ['Lu', 'Ll', 'Lt', 'Lm', 'Lo', 'Nd', 'Nl', 'No', 'Pc', 'Pd', 'Ps', 'Pe', 'Pi', 'Pf', 'Po']:
+                result.append(char)
+            # Strip emoji, symbols, and other special characters that cause 8.3 names
+            elif unicodedata.category(char).startswith(('S', 'So', 'Sk', 'Sm')):  # Symbols
+                continue
+            elif ord(char) > 0x1F000:  # High Unicode ranges often contain emoji
+                continue
+            else:
+                # Keep other Unicode characters that are generally safe
+                result.append(char)
+
+        return ''.join(result)
+
     @staticmethod
     def _strip_emojis(input_string: str) -> str:
         result = input_string.encode("ascii", errors="ignore").decode("utf-8")