✨ Add support for new filetypes

Author: yzAiden

sdk/nexent/data_process/json_chunk_processor.py

@@ -82,7 +82,12 @@ def dump(v): return orjson.dumps(v).decode("utf-8")
     def _split_plain(self, text: str) -> List[str]:
         """
         Split plain text by max length, preferring punctuation boundaries.
-        Avoid cutting inside escape sequences (e.g., don't cut after a lone backslash).
+
+        Args:
+            text: Input text
+
+        Returns:
+            List of text chunks
         """
         out: List[str] = []
         all_punct = set(string.punctuation)
@@ -118,7 +123,12 @@ def _split_plain(self, text: str) -> List[str]:
     def _split_json_text(self, text: str) -> List[str]:
         """
         Split JSON-derived text while preserving top-level key-value integrity.
-        Falls back to plain splitting if no safe top-level boundary is found.
+
+        Args:
+            text: JSON-derived string
+
+        Returns:
+            List of text chunks
         """
         out: List[str] = []
         cur = text
@@ -140,10 +150,14 @@ def _split_json_text(self, text: str) -> List[str]:
 
     def _find_last_top_kv(self, text: str, max_len: int) -> int | None:
         """
-        Find the rightmost position <= max_len where a top-level key-value pair ends.
+        Find the split position of the last top-level key-value pair.
 
-        A top-level KV ends at a comma when depth == 1 and outside any string.
-        Additionally, ensures the cut does not leave an unescaped backslash at end.
+        Args:
+            text: JSON substring (prefix)
+
+        Returns:
+            Index after the last complete top-level KV pair,
+            or None if no safe split point exists.
         """
         depth = 0
         in_str = False