Add chunker.py

chunker.py

@@ -1,46 +1,34 @@
 """
 Chunking utilities for the RAG agent.
 
-Uses RecursiveCharacterTextSplitter from LangChain with chunk_size=500 and overlap=100.
+Uses RecursiveCharacterTextSplitter from LangChain with chunk_size=500 and chunk_overlap=100.
 """
-import os
-from pathlib import Path
-from typing import List
 
 from langchain_text_splitters import RecursiveCharacterTextSplitter
 
 # Default splitter configuration
-CHUNK_SIZE = 500
+DEFAULT_SPLITTER = RecursiveCharacterTextSplitter(
-OVERLAP = 100
+    chunk_size=500,
+    chunk_overlap=100,
+)
 
-splitter = RecursiveCharacterTextSplitter(chunk_size=CHUNK_SIZE, chunk_overlap=OVERLAP)
+def split_text(text: str):
-
-def split_text(text: str) -> List[str]:
     """Split a large string into chunks.
 
     Parameters
     ----------
     text: str
-        The raw document content.
+        Text to split.
 
     Returns
     -------
-    List[str]
+    list[str]
-        A list of chunk strings.
+        List of chunk strings.
     """
-    return splitter.split_text(text)
+    return DEFAULT_SPLITTER.split_text(text)
 
-# Helper to read all files from a directory and split them
+# Example usage (not executed in tests)
-
+if __name__ == "__main__":
-def load_and_split(directory: Path) -> List[tuple]:
+    sample = "\n".join([f"Line {i}" for i in range(1000)])
-    """Load text files from *directory* and split into chunks.
+    chunks = split_text(sample)
-
+    print(f"Generated {len(chunks)} chunks")
-    Returns a list of tuples (chunk, metadata).
-    Metadata contains the source file path.
-    """
-    chunks = []
-    for file_path in directory.rglob("*.txt"):
-        content = file_path.read_text(encoding="utf-8")
-        for chunk in split_text(content):
-            chunks.append((chunk, {"source": str(file_path)}))
-    return chunks